From mat999 at gmail.com  Thu Sep  1 01:57:30 2011
From: mat999 at gmail.com (SplitIce)
Date: Thu, 1 Sep 2011 11:57:30 +1000
Subject: upstream keepalive - call for testing
In-Reply-To: <6a9ebe9133762e47eb7d9b732d7bcd9b.NginxMailingListEnglish@forum.nginx.org>
References: <20110829014709.GT1137@mdounin.ru>
	<6a9ebe9133762e47eb7d9b732d7bcd9b.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <CALFfGYa0wKpusQZMjwsZ=uW1jZgRCibQkXSM37FRfBtAkYjpiw@mail.gmail.com>

I havent had any crashes on 3x10mbit servers (90% utilisation) or 2x100mbit
servers (20% utilisation).

Seems very stable.

On Thu, Sep 1, 2011 at 6:04 AM, magicbear <nginx-forum at nginx.us> wrote:

> OK, It works for 3 days with 40 million request haven't coredump.
> Thanks.
>
> Maxim Dounin Wrote:
> -------------------------------------------------------
> > Hello!
> >
> > On Sun, Aug 28, 2011 at 01:07:25PM -0400,
> > magicbear wrote:
> >
> > > Hello, tested for 2 days, the segmentation
> > failure frequence is
> > > decreased, but sometime still will dead.
> > >
> > > Here is the coredump information. It seems
> > appear too at official
> > > without patch version.
> >
> > [...]
> >
> > > (gdb) bt
> > > #0  0x000000000045bd99 in
> > ngx_http_upstream_handler (ev=0x7f99438901e8)
> > > at src/http/ngx_http_upstream.c:921
> > > #1  0x0000000000427839 in
> > ngx_event_expire_timers () at
> > > src/event/ngx_event_timer.c:149
> >
> > Could you please test if the attached patch fixes
> > things for you?
> >
> > You may also grab it from here:
> > http://mdounin.ru/files/patch-nginx-ssl-shutdown.t
> > xt
> >
> > Maxim Dounin
> > # HG changeset patch
> > # User Maxim Dounin <mdounin at mdounin.ru>
> > # Date 1314581235 -14400
> > # Node ID 804454a04e22d32c064a3af8e7b24326c7fc2d63
> > # Parent  4b58ea791d9f09b0e2d019fcdea340c0a390af6c
> > Handle quiet ssl shutdown.
> >
> > OpenSSL's SSL_shutdown() may still try to talk to
> > network even if
> > SSL_set_shutdown(SSL_RECEIVED_SHUTDOWN|SSL_SENT_SH
> > UTDOWN) was used.
> > This happens if there are some unsent alerts.
> >
> > Use SSL_set_quiet_shutdown() to actually shutdown
> > quitely if we were asked
> > to.  Note that SSL_set_shutdown() is still
> > required as not setting it will
> > invalidate session.
> >
> > diff --git a/src/event/ngx_event_openssl.c
> > b/src/event/ngx_event_openssl.c
> > --- a/src/event/ngx_event_openssl.c
> > +++ b/src/event/ngx_event_openssl.c
> > @@ -1205,6 +1205,7 @@
> > ngx_ssl_shutdown(ngx_connection_t *c)
> >
> >      if (c->timedout) {
> >          mode =
> > SSL_RECEIVED_SHUTDOWN|SSL_SENT_SHUTDOWN;
> > +
> > SSL_set_quiet_shutdown(c->ssl->connection, 1);
> >
> >      } else {
> >          mode =
> > SSL_get_shutdown(c->ssl->connection);
> > @@ -1216,6 +1217,10 @@
> > ngx_ssl_shutdown(ngx_connection_t *c)
> >          if (c->ssl->no_send_shutdown) {
> >              mode |= SSL_SENT_SHUTDOWN;
> >          }
> > +
> > +        if (c->ssl->no_wait_shutdown &&
> > c->ssl->no_send_shutdown) {
> > +
> > SSL_set_quiet_shutdown(c->ssl->connection, 1);
> > +        }
> >      }
> >
> >      SSL_set_shutdown(c->ssl->connection, mode);
> > _______________________________________________
> > nginx mailing list
> > nginx at nginx.org
> > http://mailman.nginx.org/mailman/listinfo/nginx
>
> Posted at Nginx Forum:
> http://forum.nginx.org/read.php?2,213207,214602#msg-214602
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110901/55ae0443/attachment-0001.html>

From nginx-forum at nginx.us  Thu Sep  1 03:16:42 2011
From: nginx-forum at nginx.us (speedfirst)
Date: Wed, 31 Aug 2011 23:16:42 -0400
Subject: A bug when Host header contain IPv6 address
In-Reply-To: <20110831090804.GO1137@mdounin.ru>
References: <20110831090804.GO1137@mdounin.ru>
Message-ID: <1b1e24f5621e0c7ada55750e10c9367d.NginxMailingListEnglish@forum.nginx.org>

Thanks. the general idea of that patch is similar to mine but better. So
I'll apply it. 

If you can tell me why you feel that patch is not enough, I can help to
improve it.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,214541,214605#msg-214605


From erik at erikwickstrom.com  Thu Sep  1 05:48:54 2011
From: erik at erikwickstrom.com (Erik Wickstrom)
Date: Wed, 31 Aug 2011 22:48:54 -0700
Subject: Serving stale items from cache
Message-ID: <CAHfS1hO0shh0j1tcwHsrJPmUMR8SFbDWhYHF6i9u0sywGPiaBg@mail.gmail.com>

I have a web application where the objects are expensive to generate,
yet each individual object is accessed infrequently (a few times a
week).  The dataset in total is about 24 GB.

Can Nginx be configured to cache these items for a short amount of
time (say 20 minutes), but also have a relatively long (i.e. 7 days)
stale setting so clients will always get a fast response (but at the
same time will trigger an update of the cached item).

proxy_cache_path  /var/lib/nginx/cache  levels=2:2:2
keys_zone=staticfilecache:512m inactive=7d max_size=25000m;
# serve stale cache when the cache is updating, or there is a timeout or error.
proxy_cache_use_stale timeout updating error invalid_header http_500
http_502 http_503 http_504;
proxy_cache_valid 200 20m;

My understanding is that the by setting proxy_cache_path to
inactive=7d, cached items will be kept for 7 days and available to be
served as stale objects.  Basically any request that comes in after 20
minutes will get a stale object from the cache for up to 7 days (and
at the same time refresh the cache with the latest version from the
backend).

However, in testing, it appears that stale objects are not returned,
but instead a fresh request is made to my appservers.  Here is some
sample data from my logs (Notice the $upstream_cache_status and
$request_time at the end of the log line)

71.111.3.11 - - [31/Aug/2011:19:44:26 +0000]  "GET /?p=3093 HTTP/1.1"
200 3849 "-" "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_7_1)
AppleWebKit/535.1 (KHTML, like Gecko) Chrome/13.0.782.215
Safari/535.1" - EXPIRED 15.789
71.111.3.11 - - [31/Aug/2011:19:44:42 +0000]  "GET /?p=3093 HTTP/1.1"
200 3849 "-" "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_7_1)
AppleWebKit/535.1 (KHTML, like Gecko) Chrome/13.0.782.215
Safari/535.1" - HIT 0.000
71.111.3.11 - - [31/Aug/2011:19:51:02 +0000]  "GET /?p=3093 HTTP/1.1"
200 3849 "-" "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_7_1)
AppleWebKit/535.1 (KHTML, like Gecko) Chrome/13.0.782.215
Safari/535.1" - HIT 0.000
71.111.3.11 - - [31/Aug/2011:20:01:50 +0000]  "GET /?p=3093 HTTP/1.1"
200 3849 "-" "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_7_1)
AppleWebKit/535.1 (KHTML, like Gecko) Chrome/13.0.782.215
Safari/535.1" - HIT 0.000
71.111.3.11 - - [31/Aug/2011:21:03:56 +0000]  "GET /?p=3093 HTTP/1.1"
200 3851 "-" "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_7_1)
AppleWebKit/535.1 (KHTML, like Gecko) Chrome/13.0.782.218
Safari/535.1" - EXPIRED 0.725
71.111.3.11 - - [31/Aug/2011:21:14:55 +0000]  "GET /?p=3093 HTTP/1.1"
200 3851 "-" "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_7_1)
AppleWebKit/535.1 (KHTML, like Gecko) Chrome/13.0.782.218
Safari/535.1" - HIT 0.000
71.111.3.11 - - [31/Aug/2011:21:54:18 +0000]  "GET /?p=3093 HTTP/1.1"
200 3833 "-" "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_7_1)
AppleWebKit/535.1 (KHTML, like Gecko) Chrome/13.0.782.218
Safari/535.1" - EXPIRED 0.856
71.111.3.11 - - [31/Aug/2011:22:01:25 +0000]  "GET /?p=3093 HTTP/1.1"
200 3833 "-" "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_7_1)
AppleWebKit/535.1 (KHTML, like Gecko) Chrome/13.0.782.218
Safari/535.1" - HIT 0.000
71.111.3.11 - - [01/Sep/2011:00:02:01 +0000]  "GET /?p=3093 HTTP/1.1"
200 3847 "-" "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_7_1)
AppleWebKit/535.1 (KHTML, like Gecko) Chrome/13.0.782.218
Safari/535.1" - MISS 11.019


Basically I'm trying to keep my cache warm, without overwhelming my
app servers.  Is this doable?

Thanks!
Erik


From nginx-forum at nginx.us  Thu Sep  1 07:30:34 2011
From: nginx-forum at nginx.us (bigplum)
Date: Thu, 01 Sep 2011 03:30:34 -0400
Subject: doubt with proxy busy buffer
Message-ID: <8c559c7af61f18e50b47dd6f606bcca0.NginxMailingListEnglish@forum.nginx.org>

Hi,

I write a module for file download, but sometimes it will alert 
"the http output chain is empty while sending to client".
And I check the code, it seems that if proxy busy buffer run out,
write_filter will always return ngx error.

Is that right? 

In function: ngx_event_pipe_write_to_downstream(ngx_event_pipe_t *p)

line 540:

        out = NULL;

        if (bsize >= (size_t) p->busy_size) {
            flush = 1;
            goto flush;
        }

//If busy size is larger than that value in ngix.conf, 
//out chain will be NULL and goto fulsh 
//and call output_filter with NULL out chain.

line 617: 

    flush:
        ...............
        rc = p->output_filter(p->output_ctx, out);

//So get into ngx_http_write_filter(ngx_http_request_t *r, ngx_chain_t
*in), 
//print the alert log and return error.

line: 186 

    if (size == 0 && !(c->buffered & NGX_LOWLEVEL_BUFFERED)) {
        if (last) {
            r->out = NULL;
            c->buffered &= ~NGX_HTTP_WRITE_BUFFERED;

            return NGX_OK;
        }

        if (flush) {               // flush will never be true, because
of NULL in chain
            do {
                r->out = r->out->next;
            } while (r->out);

            c->buffered &= ~NGX_HTTP_WRITE_BUFFERED;

            return NGX_OK;
        }

        ngx_log_error(NGX_LOG_ALERT, c->log, 0,
                      "the http output chain is empty");

        ngx_debug_point();

        return NGX_ERROR;

--
Best Regard.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,214608,214608#msg-214608


From mdounin at mdounin.ru  Thu Sep  1 08:01:37 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Thu, 1 Sep 2011 12:01:37 +0400
Subject: doubt with proxy busy buffer
In-Reply-To: <8c559c7af61f18e50b47dd6f606bcca0.NginxMailingListEnglish@forum.nginx.org>
References: <8c559c7af61f18e50b47dd6f606bcca0.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20110901080136.GC1137@mdounin.ru>

Hello!

On Thu, Sep 01, 2011 at 03:30:34AM -0400, bigplum wrote:

> Hi,
> 
> I write a module for file download, but sometimes it will alert 
> "the http output chain is empty while sending to client".
> And I check the code, it seems that if proxy busy buffer run out,
> write_filter will always return ngx error.
> 
> Is that right? 

If ngx_http_write_filter() is called with NULL, r->out must not be 
empty.  If it's empty - this means that buffer(s) to flush 
disappeared somewhere, and alert will warn you about this problem.

Maxim Dounin


From nginx-forum at nginx.us  Thu Sep  1 08:17:28 2011
From: nginx-forum at nginx.us (bigplum)
Date: Thu, 01 Sep 2011 04:17:28 -0400
Subject: doubt with proxy busy buffer
In-Reply-To: <20110901080136.GC1137@mdounin.ru>
References: <20110901080136.GC1137@mdounin.ru>
Message-ID: <26c4f70c115aecb734dac57044c1eaa9.NginxMailingListEnglish@forum.nginx.org>

Thanks to Maxim, I make a mistake that do not read the r->out
statements. 
The r->out is NULL, so something goes wrong.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,214608,214610#msg-214610


From john at bitfieldconsulting.com  Thu Sep  1 08:38:58 2011
From: john at bitfieldconsulting.com (John Arundel)
Date: Thu, 1 Sep 2011 09:38:58 +0100
Subject: Connections time out before proxy_connect_timeout
In-Reply-To: <20110831170758.GA1137@mdounin.ru>
References: <CA+KJPf2_T6e356gdR4b5TSeFkLaSBvwNyJpSFb+8Rai8KPRHsA@mail.gmail.com>
	<20110831170758.GA1137@mdounin.ru>
Message-ID: <CA+KJPf2969rZQ63R3AWMSVNnoadbCW8Dfe7jDb+Q4_-BRXgbeA@mail.gmail.com>

On Wed, Aug 31, 2011 at 6:07 PM, Maxim Dounin <mdounin at mdounin.ru> wrote:
> Connect timeout is limited by your OS's SYN retransmit count limit
> and timeout. ?Under FreeBSD it's 75s by default (3 retransmits
> with exponential backoff: 3s + 6s + 12s + 24s), under Linux it's
> looks like something about 20s by default. ?Looks like you are
> hitting this limit.
>
> The only way to enlarge this limit is to tune OS, e.g. on Linux
> try adjusting net.ipv4.tcp_syn_retries sysctl.

Yes, that fixed it. Thanks a lot!

Regards,
John
-- 
Bitfield Consulting: we make software that makes things work
http://bitfieldconsulting.com/


From toplek at polok.pl  Thu Sep  1 08:42:43 2011
From: toplek at polok.pl (Piotr Polok)
Date: Thu, 01 Sep 2011 10:42:43 +0200
Subject: upload file problem (500 Internal Server Error)
Message-ID: <5d1a65521a0d9db8f54c1e9b4031e9e8@polok.pl>

HI,

I'm using version: 'nginx version: nginx/1.1.0'

While uploading file I'm getting an error: '500 Internal Server Error'

This is an nginx.log result:

------
2011/09/01 10:39:49 [alert] 28764#0: accept() failed (24: Too many open 
files)
2011/09/01 10:39:49 [alert] 28764#0: accept() failed (24: Too many open 
files)
2011/09/01 10:39:49 [alert] 28764#0: accept() failed (24: Too many open 
files)
2011/09/01 10:39:49 [alert] 28764#0: accept() failed (24: Too many open 
files)
2011/09/01 10:39:49 [alert] 28764#0: *13013 socket() failed (24: Too 
many open files) while connecting to upstream, client: 127.0.0.1, 
server: 192.168.0.222, request: "POST /upload HTTP/1.0", upstream: 
"http://127.0.0.1:80/upload", host: "localhost", referrer: 
"http://192.168.0.222/CRM/upload4.html"
2011/09/01 10:39:49 [crit] 28764#0: *13013 open() "/var/www/50x.html" 
failed (24: Too many open files), client: 127.0.0.1, server: 
192.168.0.222, request: "POST /upload HTTP/1.0", upstream: 
"http://127.0.0.1:80/upload", host: "localhost", referrer: 
"http://192.168.0.222/CRM/upload4.html"
------

The 'lsof -n | wc -l' command gives a result: '1980' value.
The 'cat /proc/sys/fs/file-max' command gives a result: '203296'.

So it is not a too many open files problem?

The upload4.html containt:
-----
<html>
<head>
<title>Test upload</title>
</head>
<body>
<h2>Select files to upload</h2>
<form enctype="multipart/form-data" action="/upload" method="post">
<input type="file" name="file1"><br>
<input type="submit" name="submit" value="upload">
<input type="hidden" name="test" value="value">
</form>
</body>
</html>
-----


Please help ;)

-- 
best regards
Piotr Polok


From chinix at gmail.com  Thu Sep  1 09:20:33 2011
From: chinix at gmail.com (chinix)
Date: Thu, 01 Sep 2011 17:20:33 +0800
Subject: How to define multiple resolvers?
Message-ID: <4e5f4e5e.6731440a.12df.fffff98f@mx.google.com>

I want to define more than one resolver to failover if one server is down,how to do this?

2011-09-01



chinix
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110901/039281be/attachment.html>

From zjay1987 at gmail.com  Thu Sep  1 09:49:29 2011
From: zjay1987 at gmail.com (li zJay)
Date: Thu, 1 Sep 2011 17:49:29 +0800
Subject: upload file problem (500 Internal Server Error)
In-Reply-To: <5d1a65521a0d9db8f54c1e9b4031e9e8@polok.pl>
References: <5d1a65521a0d9db8f54c1e9b4031e9e8@polok.pl>
Message-ID: <CABPr_Taddf9VdGfRe2bvaBz0A1zqY-_=tHJvc20Obb3aFmCROA@mail.gmail.com>

try this?

> worker_rlimit_nofile 10000;
>

On Thu, Sep 1, 2011 at 4:42 PM, Piotr Polok <toplek at polok.pl> wrote:

> HI,
>
> I'm using version: 'nginx version: nginx/1.1.0'
>
> While uploading file I'm getting an error: '500 Internal Server Error'
>
> This is an nginx.log result:
>
> ------
> 2011/09/01 10:39:49 [alert] 28764#0: accept() failed (24: Too many open
> files)
> 2011/09/01 10:39:49 [alert] 28764#0: accept() failed (24: Too many open
> files)
> 2011/09/01 10:39:49 [alert] 28764#0: accept() failed (24: Too many open
> files)
> 2011/09/01 10:39:49 [alert] 28764#0: accept() failed (24: Too many open
> files)
> 2011/09/01 10:39:49 [alert] 28764#0: *13013 socket() failed (24: Too many
> open files) while connecting to upstream, client: 127.0.0.1, server:
> 192.168.0.222, request: "POST /upload HTTP/1.0", upstream: "
> http://127.0.0.1:80/upload", host: "localhost", referrer: "
> http://192.168.0.222/CRM/**upload4.html<http://192.168.0.222/CRM/upload4.html>
> "
> 2011/09/01 10:39:49 [crit] 28764#0: *13013 open() "/var/www/50x.html"
> failed (24: Too many open files), client: 127.0.0.1, server: 192.168.0.222,
> request: "POST /upload HTTP/1.0", upstream: "http://127.0.0.1:80/upload",
> host: "localhost", referrer: "http://192.168.0.222/CRM/**upload4.html<http://192.168.0.222/CRM/upload4.html>
> "
> ------
>
> The 'lsof -n | wc -l' command gives a result: '1980' value.
> The 'cat /proc/sys/fs/file-max' command gives a result: '203296'.
>
> So it is not a too many open files problem?
>
> The upload4.html containt:
> -----
> <html>
> <head>
> <title>Test upload</title>
> </head>
> <body>
> <h2>Select files to upload</h2>
> <form enctype="multipart/form-data" action="/upload" method="post">
> <input type="file" name="file1"><br>
> <input type="submit" name="submit" value="upload">
> <input type="hidden" name="test" value="value">
> </form>
> </body>
> </html>
> -----
>
>
> Please help ;)
>
> --
> best regards
> Piotr Polok
>
> ______________________________**_________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/**mailman/listinfo/nginx<http://mailman.nginx.org/mailman/listinfo/nginx>
>



-- 
??
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110901/f30fe1f5/attachment-0001.html>

From simone.fumagalli at contactlab.com  Thu Sep  1 09:53:51 2011
From: simone.fumagalli at contactlab.com (Simone Fumagalli)
Date: Thu, 1 Sep 2011 11:53:51 +0200
Subject: Wordpress Setup (Nginx + Apache)
In-Reply-To: <1314803185467-6746900.post@n2.nabble.com>
References: <1314803185467-6746900.post@n2.nabble.com>
Message-ID: <4E5F562F.90407@contactlab.com>

On 08/31/2011 05:06 PM, antituhan wrote:
> Hello fellas, I've been setup wordpress blog using NginX as front end 
> and Apache as a backend. 

If you are using this setup you might be interested in this plugin http://wordpress.org/extend/plugins/nginx-manager/
In this way you can serve directly with Nginx static version of your pages and offload Apache 

Ciao

-- 
Simone Fumagalli


From zjay1987 at gmail.com  Thu Sep  1 10:01:31 2011
From: zjay1987 at gmail.com (li zJay)
Date: Thu, 1 Sep 2011 18:01:31 +0800
Subject: EmbeddedPerlModule question
Message-ID: <CABPr_TZHGToxEFrrKMXkNBzGkfd4oWEU8+-=Qps+9x4y0QnTZw@mail.gmail.com>

I want to access nginx $uri variable in perl, the following is my nginx.conf

worker_processes  1;
>
> events {
>     worker_connections  1024;
> }
>
> http {
>     perl_set $perlvar 'sub {return "ngx_uri:$uri";}';
>
>     server {
>         listen 80;
>         location / {
>             add_header "perlvar" $perlvar;
>             empty_gif;
>         }
>     }
> }*
> *


but it doesn't work, nginx $uri variable is empty in perl, can somebody
figure out what the problem is?

Thanks!
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110901/a8fde7e5/attachment.html>

From toplek at polok.pl  Thu Sep  1 10:02:49 2011
From: toplek at polok.pl (Piotr Polok)
Date: Thu, 01 Sep 2011 12:02:49 +0200
Subject: upload file problem (500 Internal Server Error)
In-Reply-To: <CABPr_Taddf9VdGfRe2bvaBz0A1zqY-_=tHJvc20Obb3aFmCROA@mail.gmail.com>
References: <5d1a65521a0d9db8f54c1e9b4031e9e8@polok.pl>
	<CABPr_Taddf9VdGfRe2bvaBz0A1zqY-_=tHJvc20Obb3aFmCROA@mail.gmail.com>
Message-ID: <87bd7ee8ed3c1a8bc0f33206b4db4a37@polok.pl>

On Thu, 1 Sep 2011 17:49:29 +0800, li zJay wrote:
> try this?
>
>> worker_rlimit_nofile 10000;

HI,

Thank you, The 'worker_rlimit_nofile 10000' is not solving the problem.

I have added a 'fastcgi_pass  127.0.0.1:9000' to a '/upload' location 
and now I'm getting the 'No input file specified.' error on html form 
and the  '/var/log/nginx/error.log' is clear.

This is my '/upload' configuration:

------
location /upload {
  #Pass altered request body to this location
  upload_pass   @test;

  #Store files to this directory
  #The directory is hashed, subdirectories 0 1 2 3 4 5 6 7 8 9 should 
exist
  upload_store /var/www/upload;

  #Allow uploaded files to be read only by user
  #upload_store_access user:r;
  upload_store_access user:rw group:rw all:r;

  #Set specified fields in request body
  upload_set_form_field $upload_field_name.name "$upload_file_name";
  upload_set_form_field $upload_field_name.content_type 
"$upload_content_type";
  upload_set_form_field $upload_field_name.path "$upload_tmp_path";

  #Inform backend about hash and size of a file
  upload_aggregate_form_field "$upload_field_name.md5" 
"$upload_file_md5";
  upload_aggregate_form_field "$upload_field_name.size" 
"$upload_file_size";

  upload_pass_form_field "^submit$|^description$";

  upload_cleanup 400 404 499 500-505;

  fastcgi_pass  127.0.0.1:9000;
}

# Pass altered request body to a backend
location @test {
  proxy_pass   http://localhost:80;
}
------

-- 
best regards
Piotr Polok


From igor at sysoev.ru  Thu Sep  1 10:06:18 2011
From: igor at sysoev.ru (Igor Sysoev)
Date: Thu, 1 Sep 2011 14:06:18 +0400
Subject: EmbeddedPerlModule question
In-Reply-To: <CABPr_TZHGToxEFrrKMXkNBzGkfd4oWEU8+-=Qps+9x4y0QnTZw@mail.gmail.com>
References: <CABPr_TZHGToxEFrrKMXkNBzGkfd4oWEU8+-=Qps+9x4y0QnTZw@mail.gmail.com>
Message-ID: <20110901100617.GB25615@nginx.com>

On Thu, Sep 01, 2011 at 06:01:31PM +0800, li zJay wrote:
> I want to access nginx $uri variable in perl, the following is my nginx.conf
> 
> worker_processes  1;
> >
> > events {
> >     worker_connections  1024;
> > }
> >
> > http {
> >     perl_set $perlvar 'sub {return "ngx_uri:$uri";}';

You should use $r->variable() method to get nginx variable:

'sub {
    my $r = shift;
    return $r->variable("uri");
}';

or special $r->uri() method for URI:

'sub {
    my $r = shift;
    return $r->uri;
}';

> >     server {
> >         listen 80;
> >         location / {
> >             add_header "perlvar" $perlvar;
> >             empty_gif;
> >         }
> >     }
> > }*
> > *
> 
> 
> but it doesn't work, nginx $uri variable is empty in perl, can somebody
> figure out what the problem is?
> 
> Thanks!

> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx


-- 
Igor Sysoev


From igor at sysoev.ru  Thu Sep  1 10:14:56 2011
From: igor at sysoev.ru (Igor Sysoev)
Date: Thu, 1 Sep 2011 14:14:56 +0400
Subject: How to define multiple resolvers?
In-Reply-To: <4e5f4e5e.6731440a.12df.fffff98f@mx.google.com>
References: <4e5f4e5e.6731440a.12df.fffff98f@mx.google.com>
Message-ID: <20110901101456.GC25615@nginx.com>

On Thu, Sep 01, 2011 at 05:20:33PM +0800, chinix wrote:
> I want to define more than one resolver to failover if one server is down,how to do this?

Currently it's not possible.


-- 
Igor Sysoev


From zjay1987 at gmail.com  Thu Sep  1 10:30:36 2011
From: zjay1987 at gmail.com (li zJay)
Date: Thu, 1 Sep 2011 18:30:36 +0800
Subject: EmbeddedPerlModule question
In-Reply-To: <20110901100617.GB25615@nginx.com>
References: <CABPr_TZHGToxEFrrKMXkNBzGkfd4oWEU8+-=Qps+9x4y0QnTZw@mail.gmail.com>
	<20110901100617.GB25615@nginx.com>
Message-ID: <CABPr_Tb2YbyK10VJn9vwKgVHmNA_bKuWw2b7wCTdToE9psymEA@mail.gmail.com>

Thank you, it works!

On Thu, Sep 1, 2011 at 6:06 PM, Igor Sysoev <igor at sysoev.ru> wrote:

> On Thu, Sep 01, 2011 at 06:01:31PM +0800, li zJay wrote:
> > I want to access nginx $uri variable in perl, the following is my
> nginx.conf
> >
> > worker_processes  1;
> > >
> > > events {
> > >     worker_connections  1024;
> > > }
> > >
> > > http {
> > >     perl_set $perlvar 'sub {return "ngx_uri:$uri";}';
>
> You should use $r->variable() method to get nginx variable:
>
> 'sub {
>    my $r = shift;
>    return $r->variable("uri");
> }';
>
> or special $r->uri() method for URI:
>
> 'sub {
>    my $r = shift;
>    return $r->uri;
> }';
>
> > >     server {
> > >         listen 80;
> > >         location / {
> > >             add_header "perlvar" $perlvar;
> > >             empty_gif;
> > >         }
> > >     }
> > > }*
> > > *
> >
> >
> > but it doesn't work, nginx $uri variable is empty in perl, can somebody
> > figure out what the problem is?
> >
> > Thanks!
>
> > _______________________________________________
> > nginx mailing list
> > nginx at nginx.org
> > http://mailman.nginx.org/mailman/listinfo/nginx
>
>
> --
> Igor Sysoev
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110901/f59242cc/attachment.html>

From jim at ohlste.in  Thu Sep  1 12:00:04 2011
From: jim at ohlste.in (Jim Ohlstein)
Date: Thu, 01 Sep 2011 08:00:04 -0400
Subject: nginx and Apache killer
In-Reply-To: <20110827081104.GB47641@nginx.com>
References: <20110827081104.GB47641@nginx.com>
Message-ID: <4E5F73C4.2050100@ohlste.in>

On 8/27/11 4:11 AM, Igor Sysoev wrote:
> Following "Apache Killer" discussions and the advisory from 2011-08-24
> (Advisory: Range header DoS vulnerability Apache HTTPD 2.x CVE-2011-3192)
> we'd like to clarify a couple of things in regards to nginx behavior
> either in standalone or "combo" (nginx+apache) modes.
> 
> First of all, nginx doesn't favor HEAD requests with compression,
> so the exact mentioned attack doesn't work against a standalone
> nginx installation.
> 
> If you're using nginx in combination with proxying to apache backend,
> please check your configuration to see if nginx actually passes range
> requests to the backend:
> 
> 1) If you're using proxying WITH caching then range requests are not
> sent to backend and your apache should be safe.
> 
> 2) If you're NOT using caching then you might be vulnerable to the attack.
> 
> In order to mitigate this attack when your installation includes
> apache behind nginx we recommend you the following:
> 
> 1. Refer to the above mentioned security advisory CVE-2011-3192 for apache
> and implement described measures accordingly.

Apache 2.2.20 has been released to address this issue. Please see
http://www.apache.org/dist/httpd/Announcement2.2.html.


> 
> 2. Consider using nginx configuration below (in server{} section of
> configuration). This particular example filters 5 and more ranges
> in the request:
> 
>   if ($http_range ~ "(?:\d*\s*-\s*\d*\s*,\s*){5,}") {
>       return 416;
>   }
> 
> We'd also like to notify you that for standalone nginx installations
> we've produced the attached patch. This patch prevents handling
> malicious range requests at all, instead outputting just the entire file
> if the total size of all ranges is greater than the expected response.
> 
> 
> 
> 
> 
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx


-- 
Jim Ohlstein


From nginx-forum at nginx.us  Thu Sep  1 13:38:00 2011
From: nginx-forum at nginx.us (magicbear)
Date: Thu, 01 Sep 2011 09:38:00 -0400
Subject: upstream keepalive - call for testing
In-Reply-To: <CALFfGYa0wKpusQZMjwsZ=uW1jZgRCibQkXSM37FRfBtAkYjpiw@mail.gmail.com>
References: <CALFfGYa0wKpusQZMjwsZ=uW1jZgRCibQkXSM37FRfBtAkYjpiw@mail.gmail.com>
Message-ID: <a3f7dbf329ea6723ef23e19a4a012ebd.NginxMailingListEnglish@forum.nginx.org>

I use at 2x100Mbps server (40% and 10% utilisation), when stable
enought, then I will continue to test at 1000Mbps server.

splitice Wrote:
-------------------------------------------------------
> I havent had any crashes on 3x10mbit servers (90%
> utilisation) or 2x100mbit
> servers (20% utilisation).
> 
> Seems very stable.
> 
> On Thu, Sep 1, 2011 at 6:04 AM, magicbear
> <nginx-forum at nginx.us> wrote:
> 
> > OK, It works for 3 days with 40 million request
> haven't coredump.
> > Thanks.
> >
> > Maxim Dounin Wrote:
> >
> --------------------------------------------------
> -----
> > > Hello!
> > >
> > > On Sun, Aug 28, 2011 at 01:07:25PM -0400,
> > > magicbear wrote:
> > >
> > > > Hello, tested for 2 days, the segmentation
> > > failure frequence is
> > > > decreased, but sometime still will dead.
> > > >
> > > > Here is the coredump information. It seems
> > > appear too at official
> > > > without patch version.
> > >
> > > [...]
> > >
> > > > (gdb) bt
> > > > #0  0x000000000045bd99 in
> > > ngx_http_upstream_handler (ev=0x7f99438901e8)
> > > > at src/http/ngx_http_upstream.c:921
> > > > #1  0x0000000000427839 in
> > > ngx_event_expire_timers () at
> > > > src/event/ngx_event_timer.c:149
> > >
> > > Could you please test if the attached patch
> fixes
> > > things for you?
> > >
> > > You may also grab it from here:
> > >
> http://mdounin.ru/files/patch-nginx-ssl-shutdown.t
> > > xt
> > >
> > > Maxim Dounin
> > > # HG changeset patch
> > > # User Maxim Dounin <mdounin at mdounin.ru>
> > > # Date 1314581235 -14400
> > > # Node ID
> 804454a04e22d32c064a3af8e7b24326c7fc2d63
> > > # Parent 
> 4b58ea791d9f09b0e2d019fcdea340c0a390af6c
> > > Handle quiet ssl shutdown.
> > >
> > > OpenSSL's SSL_shutdown() may still try to talk
> to
> > > network even if
> > >
> SSL_set_shutdown(SSL_RECEIVED_SHUTDOWN|SSL_SENT_SH
> > > UTDOWN) was used.
> > > This happens if there are some unsent alerts.
> > >
> > > Use SSL_set_quiet_shutdown() to actually
> shutdown
> > > quitely if we were asked
> > > to.  Note that SSL_set_shutdown() is still
> > > required as not setting it will
> > > invalidate session.
> > >
> > > diff --git a/src/event/ngx_event_openssl.c
> > > b/src/event/ngx_event_openssl.c
> > > --- a/src/event/ngx_event_openssl.c
> > > +++ b/src/event/ngx_event_openssl.c
> > > @@ -1205,6 +1205,7 @@
> > > ngx_ssl_shutdown(ngx_connection_t *c)
> > >
> > >      if (c->timedout) {
> > >          mode =
> > > SSL_RECEIVED_SHUTDOWN|SSL_SENT_SHUTDOWN;
> > > +
> > > SSL_set_quiet_shutdown(c->ssl->connection, 1);
> > >
> > >      } else {
> > >          mode =
> > > SSL_get_shutdown(c->ssl->connection);
> > > @@ -1216,6 +1217,10 @@
> > > ngx_ssl_shutdown(ngx_connection_t *c)
> > >          if (c->ssl->no_send_shutdown) {
> > >              mode |= SSL_SENT_SHUTDOWN;
> > >          }
> > > +
> > > +        if (c->ssl->no_wait_shutdown &&
> > > c->ssl->no_send_shutdown) {
> > > +
> > > SSL_set_quiet_shutdown(c->ssl->connection, 1);
> > > +        }
> > >      }
> > >
> > >      SSL_set_shutdown(c->ssl->connection,
> mode);
> > >
> _______________________________________________
> > > nginx mailing list
> > > nginx at nginx.org
> > >
> http://mailman.nginx.org/mailman/listinfo/nginx
> >
> > Posted at Nginx Forum:
> >
> http://forum.nginx.org/read.php?2,213207,214602#ms
> g-214602
> >
> > _______________________________________________
> > nginx mailing list
> > nginx at nginx.org
> > http://mailman.nginx.org/mailman/listinfo/nginx
> >
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,213207,214622#msg-214622


From nginx-forum at nginx.us  Thu Sep  1 14:19:57 2011
From: nginx-forum at nginx.us (ssaravalli)
Date: Thu, 01 Sep 2011 10:19:57 -0400
Subject: Issue with NGINX as reverse proxy with basic auth in front of
	Apache with basic auth
In-Reply-To: <20110831122554.GD1468@nginx.com>
References: <20110831122554.GD1468@nginx.com>
Message-ID: <b0be4100db8d80129a3ec469ff3fa944.NginxMailingListEnglish@forum.nginx.org>

Igor Sysoev Wrote:
> 
> You should
> 1) use either the same user/password on both nginx
> and Apache,

It's not possible in my situation. NGINX uses basic auth based on a
file, while Apache uses basic auth through LDAP.

> 2) or enable authentication only on one host: on
> nginx OR Apache,

I can't because I need two layer of authentication. The first one on
NGINX and the second one on the web apps with authentication based on
our LDAP server.

> 3) or enable authentication on nginx and go to
> Apache as some
>    backend user:
> 
>     proxy_set_header   Authorization  
> base64_encoding_of_"user:password";
> 

This is not my case.

So, is there a way to tell NGINX to manage only the first authentication
level and when the user provides username/password (LDAP) for the
backend web app, these credentials must be managed by the backend web
server?

Another solution may be to keep on the backend the LDAP auth and on the
frontend (NGINX) a form based authentication?

Thanks, Simone Saravalli

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,214550,214624#msg-214624


From mdounin at mdounin.ru  Thu Sep  1 14:50:49 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Thu, 1 Sep 2011 18:50:49 +0400
Subject: Issue with NGINX as reverse proxy with basic auth in front of
	Apache with basic auth
In-Reply-To: <b0be4100db8d80129a3ec469ff3fa944.NginxMailingListEnglish@forum.nginx.org>
References: <20110831122554.GD1468@nginx.com>
	<b0be4100db8d80129a3ec469ff3fa944.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20110901145049.GD1137@mdounin.ru>

Hello!

On Thu, Sep 01, 2011 at 10:19:57AM -0400, ssaravalli wrote:

[...]

> So, is there a way to tell NGINX to manage only the first authentication
> level and when the user provides username/password (LDAP) for the
> backend web app, these credentials must be managed by the backend web
> server?

There is no levels of authentication possible in HTTP 
authentication.

More strictly, there are two levels: proxy and client.  The 
proxy level is reserved to working with proxies, and it's not 
generally possible to use it anywhere else (additionally, I 
believe it's not supported by majority of software).

Please refer to RFC 2616 and RFC 2617 for more details.

> Another solution may be to keep on the backend the LDAP auth and on the
> frontend (NGINX) a form based authentication?

This should work.

Maxim Dounin


From lists at ruby-forum.com  Thu Sep  1 16:13:47 2011
From: lists at ruby-forum.com (Hongli Lai)
Date: Thu, 01 Sep 2011 18:13:47 +0200
Subject: Please open the Nginx SVN repository
In-Reply-To: <b8c370517b7a0a34302823509a8e96d9@ruby-forum.com>
References: <2116ea892041ed6fae726223de1a154c@ruby-forum.com>
	<b8c370517b7a0a34302823509a8e96d9@ruby-forum.com>
Message-ID: <d3f903431afd882cdc4f38730da27808@ruby-forum.com>

The SVN repository seems to be down.

  $ svn list svn://svn.nginx.org
  svn: No repository found in 'svn://svn.nginx.org'

-- 
Posted via http://www.ruby-forum.com/.


From mdounin at mdounin.ru  Thu Sep  1 16:16:36 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Thu, 1 Sep 2011 20:16:36 +0400
Subject: Please open the Nginx SVN repository
In-Reply-To: <d3f903431afd882cdc4f38730da27808@ruby-forum.com>
References: <2116ea892041ed6fae726223de1a154c@ruby-forum.com>
	<b8c370517b7a0a34302823509a8e96d9@ruby-forum.com>
	<d3f903431afd882cdc4f38730da27808@ruby-forum.com>
Message-ID: <20110901161635.GE1137@mdounin.ru>

Hello!

On Thu, Sep 01, 2011 at 06:13:47PM +0200, Hongli Lai wrote:

> The SVN repository seems to be down.
> 
>   $ svn list svn://svn.nginx.org
>   svn: No repository found in 'svn://svn.nginx.org'

http://mailman.nginx.org/pipermail/nginx-devel/2011-August/001115.html

Maxim Dounin


From nginx-forum at nginx.us  Thu Sep  1 23:17:06 2011
From: nginx-forum at nginx.us (magicbear)
Date: Thu, 01 Sep 2011 19:17:06 -0400
Subject: upload file problem (500 Internal Server Error)
In-Reply-To: <87bd7ee8ed3c1a8bc0f33206b4db4a37@polok.pl>
References: <87bd7ee8ed3c1a8bc0f33206b4db4a37@polok.pl>
Message-ID: <d238b81ff2a4d42c51acfa6f500996b7.NginxMailingListEnglish@forum.nginx.org>

you can try to add such line to /etc/security/limits.conf 
nginx       soft    nofile   131072
nginx       hard    nofile  1048576


and fastcgi_pass to run PHP
you need to set SCRIPT_FILENAME, you may first include fastcgi_params at
/etc/nginx (if you install nginx with binary), and then add such line
before fastcgi_pass: fastcgi_param SCRIPT_FILENAME
$document_root$fastcgi_script_name;
Piotr Polok Wrote:
-------------------------------------------------------
> On Thu, 1 Sep 2011 17:49:29 +0800, li zJay wrote:
> > try this?
> >
> >> worker_rlimit_nofile 10000;
> 
> HI,
> 
> Thank you, The 'worker_rlimit_nofile 10000' is not
> solving the problem.
> 
> I have added a 'fastcgi_pass  127.0.0.1:9000' to a
> '/upload' location 
> and now I'm getting the 'No input file specified.'
> error on html form 
> and the  '/var/log/nginx/error.log' is clear.
> 
> This is my '/upload' configuration:
> 
> ------
> location /upload {
>   #Pass altered request body to this location
>   upload_pass   @test;
> 
>   #Store files to this directory
>   #The directory is hashed, subdirectories 0 1 2 3
> 4 5 6 7 8 9 should 
> exist
>   upload_store /var/www/upload;
> 
>   #Allow uploaded files to be read only by user
>   #upload_store_access user:r;
>   upload_store_access user:rw group:rw all:r;
> 
>   #Set specified fields in request body
>   upload_set_form_field $upload_field_name.name
> "$upload_file_name";
>   upload_set_form_field
> $upload_field_name.content_type 
> "$upload_content_type";
>   upload_set_form_field $upload_field_name.path
> "$upload_tmp_path";
> 
>   #Inform backend about hash and size of a file
>   upload_aggregate_form_field
> "$upload_field_name.md5" 
> "$upload_file_md5";
>   upload_aggregate_form_field
> "$upload_field_name.size" 
> "$upload_file_size";
> 
>   upload_pass_form_field "^submit$|^description$";
> 
>   upload_cleanup 400 404 499 500-505;
> 
>   fastcgi_pass  127.0.0.1:9000;
> }
> 
> # Pass altered request body to a backend
> location @test {
>   proxy_pass   http://localhost:80;
> }
> ------
> 
> -- 
> best regards
> Piotr Polok
> 
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,214612,214639#msg-214639


From nginx-forum at nginx.us  Fri Sep  2 06:03:03 2011
From: nginx-forum at nginx.us (James_Lee)
Date: Fri, 02 Sep 2011 02:03:03 -0400
Subject: The patch of Nginx SSL: PEM pass phrase problem
Message-ID: <8c1d65b918b52351b2382a2fa6c5d4ae.NginxMailingListEnglish@forum.nginx.org>

Hi,

If we configured SSL in Nginx and the Private Key files are encrypted,
then the following dialog occurs at Nginx startup time:
         Enter PEM pass phrase:

It maybe difficulty for management. Sometimes it's needed to avoid the
interactive dialogue at startup time.
So I develop the patch for Nginx ssl module. Wish it helpful!



Directive:  ssl_pass_phrase_dialog
--------------------------------------------------------------------
Description: Type of pass phrase dialog for encrypted private keys
Syntax: ssl_pass_phrase_dialog type
Default: ssl_pass_phrase_dialog builtin
Context: http, server

Usage:
--------------------------------------------------------------------
When Nginx starts up it has to read the various Certificate (see
ssl_certificate) and Private Key (see ssl_certificate_key) files of the
SSL-enabled virtual servers. Because for security reasons the Private
Key files are usually encrypted, ngx_ssl module needs to query the
administrator for a Pass Phrase in order to decrypt those files. This
query can be done in two ways which can be configured by type:

 *  builtin

    This is the default where an interactive terminal dialog occurs at
startup time. Here the administrator has to manually enter the Pass
Phrase for each encrypted Private Key file.
   
 *  exec:/path/to/program

    Here an external program is configured which is called at startup
for each encrypted Private Key file.

Example:
--------------------------------------------------------------------
(1) ssl_pass_phrase_dialog builtin;
The "Enter PEM pass phrase:" will occurs at the Nginx startup time.

(2) ssl_pass_phrase_dialog "exec:/home/ssl_files/ssl_pass_phrase.sh";
The code of ssl_pass_phrase.sh:
#!/bin/sh
echo "password"

The relevant configuration in Apache is:
http://httpd.apache.org/docs/2.0/mod/mod_ssl.html#sslpassphrasedialog

If more information is needed, please refer to the Nginx HttpSslModule:
http://wiki.nginx.org/HttpSslModule


Patch?
--------------------------------------------------------------------
This patch has been tested in nginx-0.8.54.

The download url is :
http://www.cx.com/dl/?sn=f3c7b79133b7


Author:
--------------------------------------------------------------------
beagem#163.com  (Here # is @ in fact)

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,214641,214641#msg-214641


From nginx-forum at nginx.us  Fri Sep  2 06:09:06 2011
From: nginx-forum at nginx.us (ssaravalli)
Date: Fri, 02 Sep 2011 02:09:06 -0400
Subject: Issue with NGINX as reverse proxy with basic auth in front of
	Apache with basic auth
In-Reply-To: <20110901145049.GD1137@mdounin.ru>
References: <20110901145049.GD1137@mdounin.ru>
Message-ID: <9ae4ed579af1b1731379d0ed94d15745.NginxMailingListEnglish@forum.nginx.org>

Hello Maxim

Maxim Dounin Wrote:
-------------------------------------------------------
> Hello!
> 
> On Thu, Sep 01, 2011 at 10:19:57AM -0400,
> ssaravalli wrote:
> 
> [...]
> 
> > So, is there a way to tell NGINX to manage only
> the first authentication
> > level and when the user provides
> username/password (LDAP) for the
> > backend web app, these credentials must be
> managed by the backend web
> > server?
> 
> There is no levels of authentication possible in
> HTTP 
> authentication.
> 
> More strictly, there are two levels: proxy and
> client.  The 
> proxy level is reserved to working with proxies,
> and it's not 
> generally possible to use it anywhere else
> (additionally, I 
> believe it's not supported by majority of
> software).
> 
> Please refer to RFC 2616 and RFC 2617 for more
> details.

ok, trank you

> > Another solution may be to keep on the backend
> the LDAP auth and on the
> > frontend (NGINX) a form based authentication?
> 
> This should work.
> 

I'll try with this solution. Thank you for your reply!

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,214550,214642#msg-214642


From zblut at cerego.com  Fri Sep  2 07:27:08 2011
From: zblut at cerego.com (Zev Blut)
Date: Fri, 2 Sep 2011 16:27:08 +0900
Subject: Is there a method to allow a particular user agent access to a server
	rule that uses the access and auth basic module?
Message-ID: <4E60854C.3040001@cerego.com>

Hello,

I have a configuration in a server rule that typically only allows 
access by either an auth basic request or by certain ip addresses.

Something along the lines of this:

     server {
         listen       80;

         charset off;
         server_name authsite;

         satisfy any;

         auth_basic "Auth Message";
         auth_basic_user_file xyz.passwd;

         # Allow Internal Network
         allow 192.168.1.0/24;
         deny all;

         # many includes and location directives below
     }


We have a use case were we need to allow an external agent to have
access to this site.
I'd rather not play whack-a-mole and keep adding ip addresses for this
agent.
At the same time I cannot give the agent an login and password, because
we can't control the URLs.

So I was wondering if there is a way to also allow access to this site 
based on the user agent?

I tried using an if directive but that is not working.

Thanks for any ideas


-- 

Zev Blut
Cerego Japan Inc
Developer
Development / Technology

?? ????
????????????
????
???

Office : +81-3-3463-7266
Email : zblut at cerego.com

Corporate website: http://www.cerego.com
Core service website: http://iknow.jp


From agentzh at gmail.com  Fri Sep  2 08:24:50 2011
From: agentzh at gmail.com (agentzh)
Date: Fri, 2 Sep 2011 16:24:50 +0800
Subject: [ANN] ngx_lua v0.3.0 released: new regex API, request args API,
	header filter hooks, and more
In-Reply-To: <CAB4Tn6PoWWvTSKtYevCK9qe0Av6pkHeB0rM4i9oDamwfgi_sZg@mail.gmail.com>
References: <CAB4Tn6PoWWvTSKtYevCK9qe0Av6pkHeB0rM4i9oDamwfgi_sZg@mail.gmail.com>
Message-ID: <CAB4Tn6Oz=4M4j=O7hOyYnMc3C0Niy-d2cY=dQkOwwoani_3dJg@mail.gmail.com>

Hello, folks!

After two months' active development of our contributors, I'm happy to
announce the v0.3.0 release of our ngx_lua module.

You can get the release tarball from the download page:

??? https://github.com/chaoslawful/lua-nginx-module/downloads

This release contains a new PCRE regex API for Lua, an API for
retrieving parsed URI query arguments and POST query arguments, as
well as the new hooks that allow you to define nginx header filters
directly in Lua.

Also, the special "ngx.ctx" table is implemented so that you can store
arbitrary Lua values into this data which will have a lifetime
identical to the current nginx request.

There's many more new features and bug fixes that you can find out in
the following complete change log for this release:

??? http://wiki.nginx.org/HttpLuaModule#v0.3.0

Special thanks go to our contributors and users, Liseen Wan, Lance, Ou
Yuanning, Bertrand Mansion (golgote), James Hurst, Patrick Crosby,
Bill Donahue, Jiang Dapao, and others.

This module embeds the Lua interpreter or LuaJIT into the nginx core
and integrates the powerful Lua threads (aka Lua coroutines) into the
nginx event model by means of nginx subrequests.

Unlike Apache's mod_lua and Lighttpd's mod_magnet, Lua code written
atop this module can be 100% non-blocking on network traffic as long
as you use the ngx.location.capture or ngx.location.capture_multi
interfaces to let the nginx core do all your requests to mysql,
postgresql, memcached, redis, upstream http web services, and etc etc
etc.

This module is also included and enabled by default in our
ngx_openresty bundle: http://openresty.org/

You can find the complete documentation for this module on the
following wiki page:

??? http://wiki.nginx.org/HttpLuaModule

And you always get the latest source code from the git repository here:

??? https://github.com/chaoslawful/lua-nginx-module

Have fun!
-agentzh


From mdounin at mdounin.ru  Fri Sep  2 08:31:05 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Fri, 2 Sep 2011 12:31:05 +0400
Subject: Is there a method to allow a particular user agent access to a
	server rule that uses the access and auth basic module?
In-Reply-To: <4E60854C.3040001@cerego.com>
References: <4E60854C.3040001@cerego.com>
Message-ID: <20110902083105.GH1137@mdounin.ru>

Hello!

On Fri, Sep 02, 2011 at 04:27:08PM +0900, Zev Blut wrote:

> Hello,
> 
> I have a configuration in a server rule that typically only allows
> access by either an auth basic request or by certain ip addresses.
> 
> Something along the lines of this:
> 
>     server {
>         listen       80;
> 
>         charset off;
>         server_name authsite;
> 
>         satisfy any;
> 
>         auth_basic "Auth Message";
>         auth_basic_user_file xyz.passwd;
> 
>         # Allow Internal Network
>         allow 192.168.1.0/24;
>         deny all;
> 
>         # many includes and location directives below
>     }
> 
> 
> We have a use case were we need to allow an external agent to have
> access to this site.
> I'd rather not play whack-a-mole and keep adding ip addresses for this
> agent.
> At the same time I cannot give the agent an login and password, because
> we can't control the URLs.
> 
> So I was wondering if there is a way to also allow access to this
> site based on the user agent?
> 
> I tried using an if directive but that is not working.

Something like this should work:

    server {
        ...

        location / {
            error_page 418 = @allowed;

            if ($http_user_agent ~ something) {
                return 418;
            }

            satisfy any;

            allow ...
            auth_basic ...
        }

        location @allowed {
            # ...
        }
    }

Alternatively, you may use auth request module[1] and write something 
like this:


    server {
        ...

        location / {
            satisfy any;

            allow ...
            auth_basic ...
            auth_request /auth;
        }

        location = /auth {
            if ($http_user_agent ~ something) {
                return 200;
            }
            return 403;
        }
    }

[1] http://mdounin.ru/hg/ngx_http_auth_request_module/

Maxim Dounin


From zblut at cerego.com  Fri Sep  2 09:34:14 2011
From: zblut at cerego.com (Zev Blut)
Date: Fri, 2 Sep 2011 18:34:14 +0900
Subject: Is there a method to allow a particular user agent access to
	a	server rule that uses the access and auth basic module?
In-Reply-To: <20110902083105.GH1137@mdounin.ru>
References: <4E60854C.3040001@cerego.com> <20110902083105.GH1137@mdounin.ru>
Message-ID: <4E60A316.4050701@cerego.com>

Hello,


On 09/02/2011 05:31 PM, Maxim Dounin wrote:
> On Fri, Sep 02, 2011 at 04:27:08PM +0900, Zev Blut wrote:

>> We have a use case were we need to allow an external agent to have
>> access to this site.
>> I'd rather not play whack-a-mole and keep adding ip addresses for this
>> agent.
>> At the same time I cannot give the agent an login and password, because
>> we can't control the URLs.
>>
>> So I was wondering if there is a way to also allow access to this
>> site based on the user agent?
>>
>> I tried using an if directive but that is not working.


> Alternatively, you may use auth request module[1] and write something
> like this:
>
>
>      server {
>          ...
>
>          location / {
>              satisfy any;
>
>              allow ...
>              auth_basic ...
>              auth_request /auth;
>          }
>
>          location = /auth {
>              if ($http_user_agent ~ something) {
>                  return 200;
>              }
>              return 403;
>          }
>      }
>
> [1] http://mdounin.ru/hg/ngx_http_auth_request_module/

Thanks for the link to the auth request module
That might serve my purposes best without have to refactor and duplicate 
my configs as much as the extra location will.

Thanks,
Zev


From nginx-forum at nginx.us  Fri Sep  2 11:50:32 2011
From: nginx-forum at nginx.us (ykse)
Date: Fri, 02 Sep 2011 07:50:32 -0400
Subject: nginx and php header + flush
Message-ID: <fbde961fc87c41fa4f38d05f7693f78e.NginxMailingListEnglish@forum.nginx.org>

Hello guys,
I've got strange problem with my webserver. I've got nginx and apache
with php as backend www server, and now when I tryin to run this code
->

<?php

    header('HTTP/1.x 404 Not Found');
    echo "sometext";
    flush();
?>

( it's on apache )

nginx returns something like this ->


8 sometext 0


As you can see it's returns some numbers, but why ? When I'm openning
same script directly from apache it's returns only "sometext" without
any numbers.


Do you have any idea why nginx returns this numbers ?

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,214654,214654#msg-214654


From mdounin at mdounin.ru  Fri Sep  2 11:57:16 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Fri, 2 Sep 2011 15:57:16 +0400
Subject: nginx and php header + flush
In-Reply-To: <fbde961fc87c41fa4f38d05f7693f78e.NginxMailingListEnglish@forum.nginx.org>
References: <fbde961fc87c41fa4f38d05f7693f78e.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20110902115716.GL1137@mdounin.ru>

Hello!

On Fri, Sep 02, 2011 at 07:50:32AM -0400, ykse wrote:

> Hello guys,
> I've got strange problem with my webserver. I've got nginx and apache
> with php as backend www server, and now when I tryin to run this code
> ->
> 
> <?php
> 
>     header('HTTP/1.x 404 Not Found');
>     echo "sometext";
>     flush();
> ?>
> 
> ( it's on apache )
> 
> nginx returns something like this ->
> 
> 
> 8 sometext 0
> 
> 
> As you can see it's returns some numbers, but why ? When I'm openning
> same script directly from apache it's returns only "sometext" without
> any numbers.
> 
> 
> Do you have any idea why nginx returns this numbers ?

These numbers are from chunked encoding incorrectly sent by apache 
to nginx.  This behaviour is the result of a bug in apache's 
mod_php which forces HTTP/1.1 reply despite the fact client 
(nginx) uses HTTP/1.0 in request.

To fix this, just change

    header('HTTP/1.1 ...');

to

    header('HTTP/1.0 ...');

in your php code, or use "SetEnv downgrade-1.0" in apache config 
as a workaround if changing code is problematic for some reason.  
See [1] for more information.

[1] http://mailman.nginx.org/pipermail/nginx/2011-August/028508.html

Maxim Dounin


From nginx-forum at nginx.us  Fri Sep  2 12:00:21 2011
From: nginx-forum at nginx.us (ykse)
Date: Fri, 02 Sep 2011 08:00:21 -0400
Subject: nginx and php header + flush
In-Reply-To: <20110902115716.GL1137@mdounin.ru>
References: <20110902115716.GL1137@mdounin.ru>
Message-ID: <f4308072370b16f5b2704ac7c4e09a8e.NginxMailingListEnglish@forum.nginx.org>

Maxim, now it's works fine!

Thanks a lot!

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,214654,214656#msg-214656


From andrew at nginx.com  Fri Sep  2 15:59:50 2011
From: andrew at nginx.com (Andrew Alexeev)
Date: Fri, 2 Sep 2011 19:59:50 +0400
Subject: Serving stale items from cache
In-Reply-To: <CAHfS1hO0shh0j1tcwHsrJPmUMR8SFbDWhYHF6i9u0sywGPiaBg@mail.gmail.com>
References: <CAHfS1hO0shh0j1tcwHsrJPmUMR8SFbDWhYHF6i9u0sywGPiaBg@mail.gmail.com>
Message-ID: <4262924B-D501-46F1-95D8-C0DA85D975D3@nginx.com>

Erik,

That's not currently doable with nginx. Proxy_cache_use_stale directive works when you've got your backend down.

What you want is more like doing a triggered update of the content in background and it's not possible at this time.

Hope this helps

On Sep 1, 2011, at 9:48 AM, Erik Wickstrom wrote:

> I have a web application where the objects are expensive to generate,
> yet each individual object is accessed infrequently (a few times a
> week).  The dataset in total is about 24 GB.
> 
> Can Nginx be configured to cache these items for a short amount of
> time (say 20 minutes), but also have a relatively long (i.e. 7 days)
> stale setting so clients will always get a fast response (but at the
> same time will trigger an update of the cached item).
> 
> proxy_cache_path  /var/lib/nginx/cache  levels=2:2:2
> keys_zone=staticfilecache:512m inactive=7d max_size=25000m;
> # serve stale cache when the cache is updating, or there is a timeout or error.
> proxy_cache_use_stale timeout updating error invalid_header http_500
> http_502 http_503 http_504;
> proxy_cache_valid 200 20m;
> 
> My understanding is that the by setting proxy_cache_path to
> inactive=7d, cached items will be kept for 7 days and available to be
> served as stale objects.  Basically any request that comes in after 20
> minutes will get a stale object from the cache for up to 7 days (and
> at the same time refresh the cache with the latest version from the
> backend).
> 
> However, in testing, it appears that stale objects are not returned,
> but instead a fresh request is made to my appservers.  Here is some
> sample data from my logs (Notice the $upstream_cache_status and
> $request_time at the end of the log line)
> 
> 71.111.3.11 - - [31/Aug/2011:19:44:26 +0000]  "GET /?p=3093 HTTP/1.1"
> 200 3849 "-" "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_7_1)
> AppleWebKit/535.1 (KHTML, like Gecko) Chrome/13.0.782.215
> Safari/535.1" - EXPIRED 15.789
> 71.111.3.11 - - [31/Aug/2011:19:44:42 +0000]  "GET /?p=3093 HTTP/1.1"
> 200 3849 "-" "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_7_1)
> AppleWebKit/535.1 (KHTML, like Gecko) Chrome/13.0.782.215
> Safari/535.1" - HIT 0.000
> 71.111.3.11 - - [31/Aug/2011:19:51:02 +0000]  "GET /?p=3093 HTTP/1.1"
> 200 3849 "-" "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_7_1)
> AppleWebKit/535.1 (KHTML, like Gecko) Chrome/13.0.782.215
> Safari/535.1" - HIT 0.000
> 71.111.3.11 - - [31/Aug/2011:20:01:50 +0000]  "GET /?p=3093 HTTP/1.1"
> 200 3849 "-" "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_7_1)
> AppleWebKit/535.1 (KHTML, like Gecko) Chrome/13.0.782.215
> Safari/535.1" - HIT 0.000
> 71.111.3.11 - - [31/Aug/2011:21:03:56 +0000]  "GET /?p=3093 HTTP/1.1"
> 200 3851 "-" "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_7_1)
> AppleWebKit/535.1 (KHTML, like Gecko) Chrome/13.0.782.218
> Safari/535.1" - EXPIRED 0.725
> 71.111.3.11 - - [31/Aug/2011:21:14:55 +0000]  "GET /?p=3093 HTTP/1.1"
> 200 3851 "-" "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_7_1)
> AppleWebKit/535.1 (KHTML, like Gecko) Chrome/13.0.782.218
> Safari/535.1" - HIT 0.000
> 71.111.3.11 - - [31/Aug/2011:21:54:18 +0000]  "GET /?p=3093 HTTP/1.1"
> 200 3833 "-" "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_7_1)
> AppleWebKit/535.1 (KHTML, like Gecko) Chrome/13.0.782.218
> Safari/535.1" - EXPIRED 0.856
> 71.111.3.11 - - [31/Aug/2011:22:01:25 +0000]  "GET /?p=3093 HTTP/1.1"
> 200 3833 "-" "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_7_1)
> AppleWebKit/535.1 (KHTML, like Gecko) Chrome/13.0.782.218
> Safari/535.1" - HIT 0.000
> 71.111.3.11 - - [01/Sep/2011:00:02:01 +0000]  "GET /?p=3093 HTTP/1.1"
> 200 3847 "-" "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_7_1)
> AppleWebKit/535.1 (KHTML, like Gecko) Chrome/13.0.782.218
> Safari/535.1" - MISS 11.019
> 
> 
> Basically I'm trying to keep my cache warm, without overwhelming my
> app servers.  Is this doable?
> 
> Thanks!
> Erik
> 
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx
> 


From nginx-forum at nginx.us  Fri Sep  2 20:32:32 2011
From: nginx-forum at nginx.us (Xan)
Date: Fri, 02 Sep 2011 16:32:32 -0400
Subject: Strange directory behaviour: rename causes 404 error
Message-ID: <86f14be754becd175c07a61f1db101ba.NginxMailingListEnglish@forum.nginx.org>

Hi,

I have strange behaviour. I have a directory "@docencia". All is ok:

172.26.0.6 - - [02/Sep/2011:22:26:10 +0200] "GET /@docencia/ HTTP/1.1"
200 218 "https://172.26.0.3/" "Mozilla/5.0 (X11; Linux i686; rv:6.0.1)
Gecko/20100101 Firefox/6.0.1"

but when I rename to "docencia", I have:

172.26.0.6 - - [02/Sep/2011:22:27:59 +0200] "GET /docencia/ HTTP/1.1"
404 244 "https://172.26.0.3/" "Mozilla/5.0 (X11; Linux i686; rv:6.0.1)
Gecko/20100101 Firefox/6.0.1"


I don't know why. Any hint.
Please, help me.

Xan.


My nginx.conf is:
user www-data;
worker_processes  1;

pid        /var/run/nginx.pid;

events {
    worker_connections  1024;
    # multi_accept on;
}

http {
    include       /etc/nginx/mime.types;

    # =logs:
    access_log|_/var/log/nginx/access.log;
    error_log  /var/log/nginx/error.log;

    sendfile        on;
    #tcp_nopush     on;

    #keepalive_timeout  0;
    keepalive_timeout  65;
    tcp_nodelay        on;

    # =compressi?:
    gzip  on;
    gzip_min_length 100;
    gzip_types text/plain text/css text/xml application/x-javascript
application/atom+xml;

    # =inclusi? dels servidors virtuals
    include /etc/nginx/conf.d/*.conf;
    include /etc/nginx/sites-enabled/*;
}

and /etc/nginx/sites-enabled/ only has default:
# You may add here your
# server {
#	...
# }
# statements for each of your virtual hosts

# Servidor pla (http).
server {
	listen   80 default;
	server_name  localhost;

	location / {
		 rewrite ^(.*)$ https://$host$1 permanent;
	}

	# Per a qu? no me doni error si el navegador me demana favicon.ico i
no el
	# tenc. De [http://wiki.nginx.org/Piwik]
	# location = /favicon.ico {
        #         try_files /favicon.ico =204;
        #}

	# Seguint http://wiki.nginx.org/ThttpdCGI
	#location ~ \.pl$ {
	#	#include proxy.conf;
	#	proxy_intercept_errors on;
	#	proxy_pass http://127.0.0.1:8082;
	#	proxy_redirect     off;
	#	proxy_set_header  X-Real-IP  $remote_addr;
	#	proxy_set_header        Host            $host;
	#	proxy_set_header        X-Forwarded-For $proxy_add_x_forwarded_for;

	#	#proxy_redirect  default;
	#	#real_ip_header     X-Real-IP;
	#	#proxy_set_header Connection Close;
        #   	#proxy_intercept_errors on;
	#}

        #location ~ \.php$ {
        #        #include proxy.conf;
        #        proxy_intercept_errors on;
        #        proxy_pass http://127.0.0.1:8082; #8082 --> lighttpd
        #        proxy_redirect     off;
        #        proxy_set_header  X-Real-IP  $remote_addr;
        #        proxy_set_header        Host            $host;
        #        proxy_set_header        X-Forwarded-For
$proxy_add_x_forwarded_for;

         #       #proxy_redirect  default;
         #       #real_ip_header     X-Real-IP;
         #       #proxy_set_header Connection Close;
         #       #proxy_intercept_errors on;
        #}


	# Estat de les connexions del nginx.
	# De [http://kovyrin.net/2006/04/29/monitoring-nginx-with-rrdtool/]
	#location /status {
	#	stub_status on;
	#	access_log   off;
	#	allow 127.0.0.1;
        #        allow 172.26.0.0/24;
	#	deny all;
        #}


	# proxy the PHP scripts to Apache listening on 127.0.0.1:80
	#
	#location ~ \.php$ {
		#proxy_pass   http://127.0.0.1;
	#}

	# pass the PHP scripts to FastCGI server listening on 127.0.0.1:9000
	#
	#location ~ \.php$ {
		#fastcgi_pass   127.0.0.1:9000;
		#fastcgi_index  index.php;
		#fastcgi_param  SCRIPT_FILENAME  /scripts$fastcgi_script_name;
		#includefastcgi_params;
	#}

	# deny access to .htaccess files, if Apache's document root
	# concurs with nginx's one
	#
	#location ~ /\.ht {
		#deny  all;
	#}
}


# Servidor SSL (https)
server {
    server_name localhost;
    listen 443;

    ssl on;
    ssl_certificate /usr/local/nginx/conf/server.crt;
    ssl_certificate_key /usr/local/nginx/conf/server.key;


    location / {
    root   /var/www;
    index  index.html index.htm index.php;
    autoindex on;
    autoindex_exact_size off;
    }


    location /doc {
                root   /usr/share;
                autoindex on;
		autoindex_exact_size off;
                allow 127.0.0.1;
                allow 172.26.0.0/24;
                deny all;
    }

	# 400 = Bad request
        error_page  400  /400.html;
        location = /400.html {
                root /var/www/code-pages;
        }

        # 401 = Unauthorized
        error_page  401  /401.html;
        location = /401.html {
                root /var/www/code-pages;
        }

        # 403 = Forbidden page
        error_page  403  /403.html;
        location = /403.html {
                root /var/www/code-pages;
        }

        # 404 = Not found
        error_page  404  /404.html;
        location = /404.html {
                root /var/www/code-pages;
        }

        # 500 = Internal Server Error
        #
        error_page   500 /500.html;
        location = /500.html {
                root   /var/www/code-pages;
        }

        # 501 = Not Implemented
        #
        error_page   501 /501.html;
        location = /501.html {
                root   /var/www/code-pages;
        }


        # 502 = Bad Gateway
        #
        error_page  502 /502.html;
        location = /502.html {
                root   /var/www/code-pages;
        }

        # 503 = Service Unavaliable
        #
        error_page   503 /503.html;
        location = /503.html {
                root   /var/www/code-pages;
        }

        # 504 = Gateway Timeout
        #
        error_page   504  /504.html;
        location = /504.html {
                root   /var/www/code-pages;
        }

        # 505 = HTTP Version Not Supported
        #
        error_page   505  /505.html;
        location = /505.html {
                root   /var/www/code-pages;
        }


        # Per a qu? no me doni error si el navegador me demana
favicon.ico i no el
        # tenc. De [http://wiki.nginx.org/Piwik]
         location = /favicon.ico {
                 try_files /favicon.ico =204;
        }

	# Execuci? de scripts de perl per proxy i lighttpd
	# Informaci? original http://wiki.nginx.org/ThttpdCGI
        location ~ \.pl$ {
                #include proxy.conf;
                proxy_intercept_errors on;
                proxy_pass http://127.0.0.1:8082;
		#En teoria 8443; per ser segur entre servidors
                proxy_redirect     off;
                proxy_set_header  X-Real-IP  $remote_addr;
                proxy_set_header        Host            $host;
                proxy_set_header        X-Forwarded-For
$proxy_add_x_forwarded_for;

                #proxy_redirect  default;
                #real_ip_header     X-Real-IP;
                #proxy_set_header Connection Close;
                #proxy_intercept_errors on;
        }

#	# Per executar php via fastcgi (php-fpm).
#	# NO descomentar.
#        location ~ \.php$ {
#                include /etc/nginx/fastcgi_params;
#		#fastcgi_pass   unix:/var/run/php-fpm/php-fpm.sock;
#		#fastcgi_param  SCRIPT_FILENAME   
$document_root$fastcgi_script_name;
#		fastcgi_pass  127.0.0.1:9000;
#        }

	# Execuci? de php per proxy a lighttpd
        location ~ \.php$ {
                #include proxy.conf;
                #proxy_intercept_errors on;
                proxy_pass http://127.0.0.1:8082; #8082 --> lighttpd
		##proxy_set_header X-Forwarded-Host $host;
    		##proxy_set_header X-Forwarded-Server $host;
    		##proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for; 

                #proxy_redirect    off;
		#proxy_buffering   off;
                proxy_set_header  X-Real-IP  $remote_addr;
                proxy_set_header  Host            $host;
                proxy_set_header  X-Forwarded-For $remote_addr;
#$proxy_add_x_forwarded_for;
		#proxy_set_header  X-Forwarded-Ssl on;

                ##proxy_redirect  default;
                ##real_ip_header     X-Real-IP;
                ##proxy_set_header Connection Close;
                ##proxy_intercept_errors on;
        }

	location  /segur/ {
  		auth_basic            "Zona restringida";
		auth_basic_user_file  conf.d/contrasenyes;
                root   /var/www/;
		autoindex on;
	}

        # Estat de les connexions del nginx.
        # De
[http://kovyrin.net/2006/04/29/monitoring-nginx-with-rrdtool/]
        location /status {
               stub_status on;
               access_log   off;
               allow 127.0.0.1;
               allow 172.26.0.0/24;
               deny all;
        }



}




# another virtual host using mix of IP-, name-, and port-based
configuration
#
#server {
#listen   8000;
#listen   somename:8080;
#server_name  somename  alias  another.alias;

#location / {
#root   html;
#index  index.html index.htm;
#}
#}


# HTTPS server
#
#server {
#listen   443;
#server_name  localhost;

#ssl  on;
#ssl_certificate  cert.pem;
#ssl_certificate_key  cert.key;

#ssl_session_timeout  5m;

#ssl_protocols  SSLv2 SSLv3 TLSv1;
#ssl_ciphers 
ALL:!ADH:!EXPORT56:RC4+RSA:+HIGH:+MEDIUM:+LOW:+SSLv2:+EXP;
#ssl_prefer_server_ciphers   on;

#location / {
#root   html;
#index  index.html index.htm;
#}
#}

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,214663,214663#msg-214663


From nginx-forum at nginx.us  Fri Sep  2 20:36:00 2011
From: nginx-forum at nginx.us (Xan)
Date: Fri, 02 Sep 2011 16:36:00 -0400
Subject: SOLVED Strange directory behaviour: rename causes 404 error
In-Reply-To: <86f14be754becd175c07a61f1db101ba.NginxMailingListEnglish@forum.nginx.org>
References: <86f14be754becd175c07a61f1db101ba.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <43d7cb0d9aef0ebb36575cfad40c285c.NginxMailingListEnglish@forum.nginx.org>

Solved:

simply change: "location /doc" to "location /doc/"

That's all. :-)

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,214663,214664#msg-214664


From rami.essaid at gmail.com  Fri Sep  2 20:42:39 2011
From: rami.essaid at gmail.com (Rami Essaid)
Date: Fri, 2 Sep 2011 16:42:39 -0400
Subject: Storing URI in a cookie
Message-ID: <CAPMF7wg+TN80KDkzq4JMCcG+Y=8eTxT+L6bUcGb1bqxN49GNsw@mail.gmail.com>

I dont know why I am struggling with this so much but I need a bit of help.
 I am trying to redirect users to a verification page and then back to their
original requested URI but because of the rewrite I lose both the $uri and
$request_uri variables.  I figure the easiest way to solve this problem is
to store the original URI in a cookie and read it from that after the
verification page.  I cant seem to store the cookie correctly and read it
back.  Can anyone provide some direction?  Also, is there any risk that by
doing this I disrupt other cookies being stored by my application?

Thanks!

Rami
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110902/cc39d985/attachment.html>

From nginx-forum at nginx.us  Sat Sep  3 00:36:31 2011
From: nginx-forum at nginx.us (allegro)
Date: Fri, 02 Sep 2011 20:36:31 -0400
Subject: How to take full advantage system memory?
Message-ID: <3ece42c4e2934ac2dd6110e6c3b00432.NginxMailingListEnglish@forum.nginx.org>

Hi,All:

    My host have a very large physical memory, how to get nginx take
full advantage of memory as the file system cache?

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,214666,214666#msg-214666


From ianevans at digitalhit.com  Sat Sep  3 00:51:11 2011
From: ianevans at digitalhit.com (Ian Evans)
Date: Fri, 02 Sep 2011 20:51:11 -0400
Subject: SSL location keep redirecting to non-SSL
Message-ID: <4E6179FF.7060209@digitalhit.com>

Ok...my brain's been hurting on this one tonight.

My site's been running fine on nginx for years. have SSL and non-SSL 
locations. PHP runs through .shtml files

Working on a Facebook app and right now need to have it work with both 
the SSL and non-SSL URL. Testing the coding on my server right now.

If I go to http://www.example.com/myappsname the page loads.
If I go to https://www.example.com/myappsname/index.shtml the page 
loads. <-note https
If I go to https://www.example.com/myappsname/ the page redirects to 
http://www.example.com/myappsname. <-note https on first URL

Here are the relevant locations in the SSL server section:

location / {
root /usr/local/apache/htdocs;
rewrite ^(.+) http://www.digitalhit.com$1 permanent;
index index.shtml index.php;
}
				
location ~ \.(shtml|php|inc)$ {
root /usr/local/apache/htdocs;
include /usr/local/nginx/conf/fastcgi-php.conf;
fastcgi_pass 127.0.0.1:10004;
}

location ^~ /myappsname/ {
index index.shtml
root /usr/local/apache/htdocs/;
fastcgi_intercept_errors on;
include /usr/local/nginx/conf/fastcgi-php.conf;
fastcgi_pass 127.0.0.1:10004;
fastcgi_param HTTPS on;
}

I'll probably smack my head when this is pointed out to me. All my other 
SSL locations work fine. Any idea why including the filename stays SSL 
but dropping it redirects?

Thanks.


From nginx-forum at nginx.us  Sat Sep  3 01:21:33 2011
From: nginx-forum at nginx.us (signe)
Date: Fri, 02 Sep 2011 21:21:33 -0400
Subject: URL decode rewrite handling: + vs. %20
Message-ID: <2f879159200e58c32a5b1eb6b3d01cef.NginxMailingListEnglish@forum.nginx.org>

Ubuntu Natty
nginx/0.8.54
PHP 5.3.5 / FPM / FastCGI

I'm just beginning to work with nginx for the first time.  Converting my
home server (very few hits) as an experiment for possible broader-scale
testing and implementation on work servers (billions of hits per
month).

Everything has gone beautifully... up until I begin working on an older
app: Gallery 2.  Some of the photos in my Gallery contain spaces, so
they're being encoded in the old-but-still-accepted + format.  However,
when I make the requests in nginx, it's passing the pathinfo through to
fastcgi as an escaped + (%2b) instead of a space (+ or %20).

The path is something like this:

http://www.example.com/gallery/foo+bar.jpg.html
which gets rewritten to
http://www.example.com/gallery/main.php?g2_path=foo+bar.jpg.html

In a path, + should be handled literally per RFC.  In a query string,
it's supposed to be interpreted as a space per x-www-form-urlencoded. 
This is the disparity - what once was path is now query string.  Because
of the conversion, Apache translates the + from literal string to
encoded space during rewrite.  nginx simply encodes the literal string.

I don't know if there's a solid answer as to which one is strictly
"correct"... but what I do need to know is if there's a way to achieve
the same behavior in nginx?

This is an interesting edge case...

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,214668,214668#msg-214668


From ianevans at digitalhit.com  Sat Sep  3 08:17:27 2011
From: ianevans at digitalhit.com (Ian Evans)
Date: Sat, 03 Sep 2011 04:17:27 -0400
Subject: SSL location keep redirecting to non-SSL
In-Reply-To: <4E6179FF.7060209@digitalhit.com>
References: <4E6179FF.7060209@digitalhit.com>
Message-ID: <4E61E297.1070506@digitalhit.com>

Been scratching my head on this one all night.

Been working on it a bit and now if I go to:
https://www.example.com/myfacebookapp/index.shtml the page loads but 
https://www.example.com/myfacebookapp is tossing a 404

I'm really boggled here because the other subdirs are woring and the 
only difference is they're password protected.

Here's the full ssl server conf section:

server {
server_name www.example.com;
listen 443;
root /usr/local/apache/htdocs/;

ssl on;
ssl_certificate /usr/local/apache/conf/server.pem;
ssl_certificate_key /usr/local/apache/conf/server.key;
ssl_session_timeout 5m;
error_page 404 /dhe404.shtml;

location / {
rewrite ^(.+) http://www.example.com$1 permanent;
index index.shtml index.php;
}
		
location ~ \.(shtml|php|inc)$ {
include /usr/local/nginx/conf/fastcgi-php.conf;
fastcgi_pass 127.0.0.1:10004;
}
	
location ^~ /myfacebookapp/ {
fastcgi_intercept_errors on;
include /usr/local/nginx/conf/fastcgi-php.conf;
fastcgi_pass 127.0.0.1:10004;
fastcgi_param HTTPS on;
}

location ^~ /subdir2/ {
index index.php;
fastcgi_intercept_errors on;
include /usr/local/nginx/conf/fastcgi-php.conf;
fastcgi_pass 127.0.0.1:10004;
fastcgi_index index.php;
fastcgi_param HTTPS on;
}

location ^~ /subdir3/ {
fastcgi_intercept_errors on;
include /usr/local/nginx/conf/fastcgi-php.conf;
fastcgi_pass 127.0.0.1:10004;
fastcgi_param HTTPS on;
auth_basic "example";
auth_basic_user_file /usr/local/apache/passwd/passwords;
}

location /subdir4/ {
index index.shtml index.php index.html;
auth_basic "example";
auth_basic_user_file /usr/local/apache/passwd/passwords;
}

location /subdir5/ {
index index.html;
auth_basic "example";
auth_basic_user_file /usr/local/apache/passwd/passwords;
}

location ~* ^.+\.(png|gif)$ {
expires off;
}

# serve static files directly
location ~* ^.+.(css|gif|jpe?g|bmp|js|png|ico)$ {
expires 30d;
}

}



From mdounin at mdounin.ru  Sat Sep  3 08:44:45 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Sat, 3 Sep 2011 12:44:45 +0400
Subject: How to take full advantage system memory?
In-Reply-To: <3ece42c4e2934ac2dd6110e6c3b00432.NginxMailingListEnglish@forum.nginx.org>
References: <3ece42c4e2934ac2dd6110e6c3b00432.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20110903084444.GN1137@mdounin.ru>

Hello!

On Fri, Sep 02, 2011 at 08:36:31PM -0400, allegro wrote:

> Hi,All:
> 
>     My host have a very large physical memory, how to get nginx take
> full advantage of memory as the file system cache?

It's up to your OS to use memory as file system cache.  Depending 
on your OS you may need to tune something, see your OS tuning 
guides for details.

In nginx you may want to tune various buffer sizes if you have 
enough memory, e.g. output_buffers, proxy_buffers, fastcgi_buffers 
and so on.  This is not about file system cache, but may reduce 
disk load if properly tuned for your typical use pattern.

Maxim Dounin


From nginx-forum at nginx.us  Sat Sep  3 08:49:56 2011
From: nginx-forum at nginx.us (magicbear)
Date: Sat, 03 Sep 2011 04:49:56 -0400
Subject: Storing URI in a cookie
In-Reply-To: <CAPMF7wg+TN80KDkzq4JMCcG+Y=8eTxT+L6bUcGb1bqxN49GNsw@mail.gmail.com>
References: <CAPMF7wg+TN80KDkzq4JMCcG+Y=8eTxT+L6bUcGb1bqxN49GNsw@mail.gmail.com>
Message-ID: <212564c9ef34b2933c9f8d7c91b1bc83.NginxMailingListEnglish@forum.nginx.org>

may try this:
add_header Set-Cookie "uri=$uri; path=/";

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,214665,214675#msg-214675


From nginx-forum at nginx.us  Sat Sep  3 09:02:39 2011
From: nginx-forum at nginx.us (Samael)
Date: Sat, 03 Sep 2011 05:02:39 -0400
Subject: Strange directory behaviour: rename causes 404 error
In-Reply-To: <86f14be754becd175c07a61f1db101ba.NginxMailingListEnglish@forum.nginx.org>
References: <86f14be754becd175c07a61f1db101ba.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <6f364f206a797a344011329e8ca022fd.NginxMailingListEnglish@forum.nginx.org>

Xan, I suggest you take a look at this link -
https://nealpoole.com/blog/2011/04/setting-up-php-fastcgi-and-nginx-dont-trust-the-tutorials-check-your-configuration/

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,214663,214676#msg-214676


From calin.don at gmail.com  Sat Sep  3 09:04:50 2011
From: calin.don at gmail.com (Calin Don)
Date: Sat, 3 Sep 2011 12:04:50 +0300
Subject: Storing URI in a cookie
In-Reply-To: <CAPMF7wg+TN80KDkzq4JMCcG+Y=8eTxT+L6bUcGb1bqxN49GNsw@mail.gmail.com>
References: <CAPMF7wg+TN80KDkzq4JMCcG+Y=8eTxT+L6bUcGb1bqxN49GNsw@mail.gmail.com>
Message-ID: <CAEOe2Jyi=sHbqN_iGix0DfEJ9cr2vhbQtNnQgiX1mvEwC4nk9w@mail.gmail.com>

A better way to redirect users is to store the uri in the redirection url.
Storing it in a cookie causes problems if user opens two different tabs with
two different uris for example.
On Sep 2, 2011 11:42 PM, "Rami Essaid" <rami.essaid at gmail.com> wrote:
> I dont know why I am struggling with this so much but I need a bit of
help.
> I am trying to redirect users to a verification page and then back to
their
> original requested URI but because of the rewrite I lose both the $uri and
> $request_uri variables. I figure the easiest way to solve this problem is
> to store the original URI in a cookie and read it from that after the
> verification page. I cant seem to store the cookie correctly and read it
> back. Can anyone provide some direction? Also, is there any risk that by
> doing this I disrupt other cookies being stored by my application?
>
> Thanks!
>
> Rami
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110903/d9374d66/attachment.html>

From ianevans at digitalhit.com  Sat Sep  3 09:20:30 2011
From: ianevans at digitalhit.com (Ian Evans)
Date: Sat, 03 Sep 2011 05:20:30 -0400
Subject: SSL location keep redirecting to non-SSL
In-Reply-To: <4E61E297.1070506@digitalhit.com>
References: <4E6179FF.7060209@digitalhit.com> <4E61E297.1070506@digitalhit.com>
Message-ID: <4E61F15E.7040808@digitalhit.com>

still expermineting...just looked at the debug log. Even if I change the 
location to location = /myfacebookapp a call to myfacebookapp still 
isn't matching the location.

Really don't get what's happening here. Why are my other ssl locations 
working?


From calin.don at gmail.com  Sat Sep  3 09:23:57 2011
From: calin.don at gmail.com (Calin Don)
Date: Sat, 3 Sep 2011 12:23:57 +0300
Subject: SSL location keep redirecting to non-SSL
In-Reply-To: <4E61E297.1070506@digitalhit.com>
References: <4E6179FF.7060209@digitalhit.com> <4E61E297.1070506@digitalhit.com>
Message-ID: <CAEOe2JxORb5vBxBaOAq8bVTUX-w-dzscCmFu1xj5ppGuYNuA+A@mail.gmail.com>

You should set a serverwide index directive. You should move index
index.shtml index.php; from location / {} to server {}

Also your rewrite rule rewrite ^(.+) http://www.example.com$1 permanent;
sould be rewritten into rewrite ^ http://www.example.com$request_uri?
permanent;

Take a look at http://wiki.nginx.org/Pitfalls


On Sat, Sep 3, 2011 at 11:17, Ian Evans <ianevans at digitalhit.com> wrote:

> Been scratching my head on this one all night.
>
> Been working on it a bit and now if I go to:
> https://www.example.com/**myfacebookapp/index.shtml<https://www.example.com/myfacebookapp/index.shtml>the page loads but
> https://www.example.com/**myfacebookapp<https://www.example.com/myfacebookapp>is tossing a 404
>
> I'm really boggled here because the other subdirs are woring and the only
> difference is they're password protected.
>
> Here's the full ssl server conf section:
>
> server {
> server_name www.example.com;
> listen 443;
> root /usr/local/apache/htdocs/;
>
> ssl on;
> ssl_certificate /usr/local/apache/conf/server.**pem;
> ssl_certificate_key /usr/local/apache/conf/server.**key;
> ssl_session_timeout 5m;
> error_page 404 /dhe404.shtml;
>
> location / {
> rewrite ^(.+) http://www.example.com$1 permanent;
>
> index index.shtml index.php;
> }
>
> location ~ \.(shtml|php|inc)$ {
> include /usr/local/nginx/conf/fastcgi-**php.conf;
> fastcgi_pass 127.0.0.1:10004;
> }
>
> location ^~ /myfacebookapp/ {
>
> fastcgi_intercept_errors on;
> include /usr/local/nginx/conf/fastcgi-**php.conf;
> fastcgi_pass 127.0.0.1:10004;
> fastcgi_param HTTPS on;
> }
>
> location ^~ /subdir2/ {
> index index.php;
>
> fastcgi_intercept_errors on;
> include /usr/local/nginx/conf/fastcgi-**php.conf;
> fastcgi_pass 127.0.0.1:10004;
> fastcgi_index index.php;
> fastcgi_param HTTPS on;
> }
>
> location ^~ /subdir3/ {
>
> fastcgi_intercept_errors on;
> include /usr/local/nginx/conf/fastcgi-**php.conf;
> fastcgi_pass 127.0.0.1:10004;
> fastcgi_param HTTPS on;
> auth_basic "example";
> auth_basic_user_file /usr/local/apache/passwd/**passwords;
> }
>
> location /subdir4/ {
> index index.shtml index.php index.html;
> auth_basic "example";
> auth_basic_user_file /usr/local/apache/passwd/**passwords;
> }
>
> location /subdir5/ {
> index index.html;
> auth_basic "example";
> auth_basic_user_file /usr/local/apache/passwd/**passwords;
> }
>
> location ~* ^.+\.(png|gif)$ {
> expires off;
> }
>
> # serve static files directly
> location ~* ^.+.(css|gif|jpe?g|bmp|js|png|**ico)$ {
> expires 30d;
>
> }
>
> }
>
>
> ______________________________**_________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/**mailman/listinfo/nginx<http://mailman.nginx.org/mailman/listinfo/nginx>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110903/72c74d71/attachment.html>

From ianevans at digitalhit.com  Sat Sep  3 09:34:57 2011
From: ianevans at digitalhit.com (Ian Evans)
Date: Sat, 03 Sep 2011 05:34:57 -0400
Subject: SSL location keep redirecting to non-SSL
In-Reply-To: <CAEOe2JxORb5vBxBaOAq8bVTUX-w-dzscCmFu1xj5ppGuYNuA+A@mail.gmail.com>
References: <4E6179FF.7060209@digitalhit.com> <4E61E297.1070506@digitalhit.com>
	<CAEOe2JxORb5vBxBaOAq8bVTUX-w-dzscCmFu1xj5ppGuYNuA+A@mail.gmail.com>
Message-ID: <4E61F4C1.6000103@digitalhit.com>

On 03/09/2011 5:23 AM, Calin Don wrote:
> You should set a serverwide index directive. You should move index
> index.shtml index.php; from location / {} to server {}
>
> Also your rewrite rule rewrite ^(.+) http://www.example.com
> <http://www.example.com/>$1 permanent; sould be rewritten into rewrite ^
> http://www.example.com <http://www.example.com/>$request_uri? permanent;

Made your suggested changes.

After doing so tested location = /myfacebookapp which redirected to 
http, not what I want and location ~ /myfacebookapp which tossed a 
404...also not what I want.

5:34 am...argh. :-)


From drdr.xp at gmail.com  Sat Sep  3 09:46:39 2011
From: drdr.xp at gmail.com (dr-dr xp)
Date: Sat, 3 Sep 2011 17:46:39 +0800
Subject: nginx 1.0.2 AIO problem with open_file_cache
Message-ID: <CAMcFNRJv_KSkj+DLQzXsCT02Ktd-W5uFXRAOz3MQJ+vErKOsuQ@mail.gmail.com>

Hello everybody. I recently have had some problem with AIO support and need
some help or advice on it.

With AIO on, everything was OK. But when I enabled open_file_cache, nginx
frequently closed client connection directly without sending back any data.


My platform is linux 2.6.30. CentOS 5.4;

The nginx server that has this issue serves static file at backend. Another
nginx serves as proxy in front of it.
Backend nginx config is like this:
...
aio            on;


directio       512;


output_buffers 1 128k;
...
    location /files {



        internal;


        alias /;
        open_file_cache max=50000 inactive=20s;


        open_file_cache_valid    30s;


        open_file_cache_min_uses 1;


        open_file_cache_errors   on;


With this config, on the frontend proxy server there are error logs like
this:

2011/09/03 17:01:19 [error] 12498#0: *78679759 *upstream prematurely closed
connection while reading response header from upstream*, client:
221.193.212.77, server: , request: "GET .......... HTTP/1.1", upstream:
"http://**/file/871586e763bf4417bf8d475858deee2b/260110/0b47fc7f4d84e4bcb01a6d6d1fcacc94-3581749",
host.....


And I saw in tcpdump output the connection was closed very quickly by
backend nginx:

22:02:22.892327 IP 172.16.113.32.58271 > 172.16.228.63.irdmi: S
1788188296:1788188296(0) win 5840 <mss 1460,sackOK,timestamp 666482608
0,nop,wscale 7>
22:02:22.895683 IP 172.16.228.63.irdmi > 172.16.113.32.58271: S
350810413:350810413(0) ack 1788188297 win 5792 <mss 1460,sackOK,timestamp
1948226601 66<
22:02:22.895695 IP 172.16.113.32.58271 > 172.16.228.63.irdmi: . ack 1 win 46
<nop,nop,timestamp 666482612 1948226601>
22:02:22.895703 IP 172.16.113.32.58271 > 172.16.228.63.irdmi: P 1:181(180)
ack 1 win 46 <nop,nop,timestamp 666482612 1948226601>
22:02:22.897680 IP 172.16.228.63.irdmi > 172.16.113.32.58271: . ack 181 win
54 <nop,nop,timestamp 1948226603 666482612>
22:02:22.898348 IP 172.16.228.63.irdmi > 172.16.113.32.58271: F 1:1(0) ack
181 win 54 <nop,nop,timestamp 1948226603 666482612>
22:02:22.898473 IP 172.16.113.32.58271 > 172.16.228.63.irdmi: F 181:181(0)
ack 2 win 46 <nop,nop,timestamp 666482615 1948226603>
22:02:22.899804 IP 172.16.228.63.irdmi > 172.16.113.32.58271: . ack 182 win
54 <nop,nop,timestamp 1948226605 666482615>


On the backend nginx, strace shows that io_getevents got an error *"-22"
EINVAL*.

  [pid  5275] recvfrom(48, 0x7fffe4400ae7, 1, 2, 0, 0) = -1 EAGAIN (Resource
temporarily unavailable)
  [pid  5275] epoll_wait(6, {{EPOLLIN|EPOLLOUT, {u32=2445415233,
u64=139850875540289}}}, 512, 5672) = 1
  [pid  5275] recvfrom(201, "\1\6\0\1\0\337\1\0Status: 301 Moved Perman"...,
4096, 0, NULL, NULL) = 256
  [pid  5275] close(201)                  = 0
  [pid  5275]
stat("//data1/871586e763bf4417bf8d475858deee2b/g268315/252/995f20f76d0d3871fdb5d71bbc92956c-27769564",
{st_mode=S_IFREG|0600, st_size=122507, ...<
  [pid  5275] io_submit(139850942226432, 1, {{0xf43718, 0, 0, 0, 40}}) = 1


  [pid  5275] epoll_wait(6, {{EPOLLIN, {u32=7173248, u64=7173248}}}, 512,
5671) = 1

  [pid  5275] read(7, "\1\0\0\0\0\0\0\0", 8) = 8
  [pid  5275] *io_getevents(139850942226432, 1, 64, {{0xf43718, 0xf436d8,
-22, 0}}, {0, 0}) = 1*
  [pid  5275] write(24, "172.16.164.30 - - 172.16.228.63:"..., 338) = 338
  [pid  5275] close(48)

And access.log shows that response length is ZERO:

[03/Sep/2011:06:28:48 +0800] "GET
/file/871586e763bf4417bf8d475858deee2b/260284/ed31933859d27e33b8398deaea1d2ade-3969549
HTTP/1.0" 206 0 0.001 "-" "-" "-" Range:bytes=27459105-


After I removed those *open_file_cache* directives, No such error occurred.

And the shorter *inactive* is,  the more often I get such errors.


Anybody has idea about avoiding this or fixing it?



-- 
????????????????
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110903/187158b4/attachment-0001.html>

From mdounin at mdounin.ru  Sat Sep  3 10:03:36 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Sat, 3 Sep 2011 14:03:36 +0400
Subject: nginx 1.0.2 AIO problem with open_file_cache
In-Reply-To: <CAMcFNRJv_KSkj+DLQzXsCT02Ktd-W5uFXRAOz3MQJ+vErKOsuQ@mail.gmail.com>
References: <CAMcFNRJv_KSkj+DLQzXsCT02Ktd-W5uFXRAOz3MQJ+vErKOsuQ@mail.gmail.com>
Message-ID: <20110903100335.GO1137@mdounin.ru>

Hello!

On Sat, Sep 03, 2011 at 05:46:39PM +0800, dr-dr xp wrote:

> Hello everybody. I recently have had some problem with AIO support and need
> some help or advice on it.
> 
> With AIO on, everything was OK. But when I enabled open_file_cache, nginx
> frequently closed client connection directly without sending back any data.
> 
> 
> My platform is linux 2.6.30. CentOS 5.4;
> 
> The nginx server that has this issue serves static file at backend. Another
> nginx serves as proxy in front of it.
> Backend nginx config is like this:
> ...
> aio            on;
> 
> 
> directio       512;
> 
> 
> output_buffers 1 128k;
> ...
>     location /files {
> 
> 
> 
>         internal;
> 
> 
>         alias /;
>         open_file_cache max=50000 inactive=20s;
> 
> 
>         open_file_cache_valid    30s;
> 
> 
>         open_file_cache_min_uses 1;
> 
> 
>         open_file_cache_errors   on;
> 
> 
> With this config, on the frontend proxy server there are error logs like
> this:
> 
> 2011/09/03 17:01:19 [error] 12498#0: *78679759 *upstream prematurely closed
> connection while reading response header from upstream*, client:
> 221.193.212.77, server: , request: "GET .......... HTTP/1.1", upstream:
> "http://**/file/871586e763bf4417bf8d475858deee2b/260110/0b47fc7f4d84e4bcb01a6d6d1fcacc94-3581749",
> host.....
> 
> 
> And I saw in tcpdump output the connection was closed very quickly by
> backend nginx:
> 
> 22:02:22.892327 IP 172.16.113.32.58271 > 172.16.228.63.irdmi: S
> 1788188296:1788188296(0) win 5840 <mss 1460,sackOK,timestamp 666482608
> 0,nop,wscale 7>
> 22:02:22.895683 IP 172.16.228.63.irdmi > 172.16.113.32.58271: S
> 350810413:350810413(0) ack 1788188297 win 5792 <mss 1460,sackOK,timestamp
> 1948226601 66<
> 22:02:22.895695 IP 172.16.113.32.58271 > 172.16.228.63.irdmi: . ack 1 win 46
> <nop,nop,timestamp 666482612 1948226601>
> 22:02:22.895703 IP 172.16.113.32.58271 > 172.16.228.63.irdmi: P 1:181(180)
> ack 1 win 46 <nop,nop,timestamp 666482612 1948226601>
> 22:02:22.897680 IP 172.16.228.63.irdmi > 172.16.113.32.58271: . ack 181 win
> 54 <nop,nop,timestamp 1948226603 666482612>
> 22:02:22.898348 IP 172.16.228.63.irdmi > 172.16.113.32.58271: F 1:1(0) ack
> 181 win 54 <nop,nop,timestamp 1948226603 666482612>
> 22:02:22.898473 IP 172.16.113.32.58271 > 172.16.228.63.irdmi: F 181:181(0)
> ack 2 win 46 <nop,nop,timestamp 666482615 1948226603>
> 22:02:22.899804 IP 172.16.228.63.irdmi > 172.16.113.32.58271: . ack 182 win
> 54 <nop,nop,timestamp 1948226605 666482615>
> 
> 
> On the backend nginx, strace shows that io_getevents got an error *"-22"
> EINVAL*.
> 
>   [pid  5275] recvfrom(48, 0x7fffe4400ae7, 1, 2, 0, 0) = -1 EAGAIN (Resource
> temporarily unavailable)
>   [pid  5275] epoll_wait(6, {{EPOLLIN|EPOLLOUT, {u32=2445415233,
> u64=139850875540289}}}, 512, 5672) = 1
>   [pid  5275] recvfrom(201, "\1\6\0\1\0\337\1\0Status: 301 Moved Perman"...,
> 4096, 0, NULL, NULL) = 256
>   [pid  5275] close(201)                  = 0
>   [pid  5275]
> stat("//data1/871586e763bf4417bf8d475858deee2b/g268315/252/995f20f76d0d3871fdb5d71bbc92956c-27769564",
> {st_mode=S_IFREG|0600, st_size=122507, ...<
>   [pid  5275] io_submit(139850942226432, 1, {{0xf43718, 0, 0, 0, 40}}) = 1
> 
> 
>   [pid  5275] epoll_wait(6, {{EPOLLIN, {u32=7173248, u64=7173248}}}, 512,
> 5671) = 1
> 
>   [pid  5275] read(7, "\1\0\0\0\0\0\0\0", 8) = 8
>   [pid  5275] *io_getevents(139850942226432, 1, 64, {{0xf43718, 0xf436d8,
> -22, 0}}, {0, 0}) = 1*
>   [pid  5275] write(24, "172.16.164.30 - - 172.16.228.63:"..., 338) = 338
>   [pid  5275] close(48)
> 
> And access.log shows that response length is ZERO:
> 
> [03/Sep/2011:06:28:48 +0800] "GET
> /file/871586e763bf4417bf8d475858deee2b/260284/ed31933859d27e33b8398deaea1d2ade-3969549
> HTTP/1.0" 206 0 0.001 "-" "-" "-" Range:bytes=27459105-
> 
> 
> After I removed those *open_file_cache* directives, No such error occurred.
> 
> And the shorter *inactive* is,  the more often I get such errors.
> 
> 
> Anybody has idea about avoiding this or fixing it?

Could you please provide debug log?  See 
http://wiki.nginx.org/Debugging for more information.

Additionally, some more info about file requested might be 
helpful: how big it is? whether it changed on backend during 
requests? if yes, is it changed atomically?

Maxim Dounin


From stef at caunter.ca  Sat Sep  3 10:14:07 2011
From: stef at caunter.ca (Stefan Caunter)
Date: Sat, 3 Sep 2011 06:14:07 -0400
Subject: SSL location keep redirecting to non-SSL
In-Reply-To: <4E61E297.1070506@digitalhit.com>
References: <4E6179FF.7060209@digitalhit.com> <4E61E297.1070506@digitalhit.com>
Message-ID: <CALdpnrz2DP0OR_mVOREqbmjg6=3OzuBdOyZ8Yh2WZeH4772NTw@mail.gmail.com>

On Sat, Sep 3, 2011 at 4:17 AM, Ian Evans <ianevans at digitalhit.com> wrote:
> Been scratching my head on this one all night.
>
> Been working on it a bit and now if I go to:
> https://www.example.com/myfacebookapp/index.shtml the page loads but
> https://www.example.com/myfacebookapp is tossing a 404
>
> I'm really boggled here because the other subdirs are woring and the only
> difference is they're password protected.
>
> Here's the full ssl server conf section:
>
> server {
> server_name www.example.com;
> listen 443;
> root /usr/local/apache/htdocs/;
>
> ssl on;
> ssl_certificate /usr/local/apache/conf/server.pem;
> ssl_certificate_key /usr/local/apache/conf/server.key;
> ssl_session_timeout 5m;
> error_page 404 /dhe404.shtml;
>
> location / {
> rewrite ^(.+) http://www.example.com$1 permanent;
> index index.shtml index.php;
> }

This looks odd to me, I've never seen a rewrite in an ssl config. It
will complicate things.

Leave the index to the server config, and take out the location / rewrite.

Simplify.

Get the general app working, and then apply specialized rules for
directories and urls. Remember, if you have told nginx to do something
that it *can* do, the logs won't show much except that it's doing what
it's been told.


>
> location ~ \.(shtml|php|inc)$ {
> include /usr/local/nginx/conf/fastcgi-php.conf;
> fastcgi_pass 127.0.0.1:10004;
> }
>
> location ^~ /myfacebookapp/ {
> fastcgi_intercept_errors on;
> include /usr/local/nginx/conf/fastcgi-php.conf;
> fastcgi_pass 127.0.0.1:10004;
> fastcgi_param HTTPS on;
> }
>
> location ^~ /subdir2/ {
> index index.php;
> fastcgi_intercept_errors on;
> include /usr/local/nginx/conf/fastcgi-php.conf;
> fastcgi_pass 127.0.0.1:10004;
> fastcgi_index index.php;
> fastcgi_param HTTPS on;
> }
>
> location ^~ /subdir3/ {
> fastcgi_intercept_errors on;
> include /usr/local/nginx/conf/fastcgi-php.conf;
> fastcgi_pass 127.0.0.1:10004;
> fastcgi_param HTTPS on;
> auth_basic "example";
> auth_basic_user_file /usr/local/apache/passwd/passwords;
> }
>
> location /subdir4/ {
> index index.shtml index.php index.html;
> auth_basic "example";
> auth_basic_user_file /usr/local/apache/passwd/passwords;
> }
>
> location /subdir5/ {
> index index.html;
> auth_basic "example";
> auth_basic_user_file /usr/local/apache/passwd/passwords;
> }
>
> location ~* ^.+\.(png|gif)$ {
> expires off;
> }
>
> # serve static files directly
> location ~* ^.+.(css|gif|jpe?g|bmp|js|png|ico)$ {
> expires 30d;
> }
>
> }
>
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx
>


From ianevans at digitalhit.com  Sat Sep  3 10:18:23 2011
From: ianevans at digitalhit.com (Ian Evans)
Date: Sat, 03 Sep 2011 06:18:23 -0400
Subject: SSL location keep redirecting to non-SSL
In-Reply-To: <CAEOe2JxORb5vBxBaOAq8bVTUX-w-dzscCmFu1xj5ppGuYNuA+A@mail.gmail.com>
References: <4E6179FF.7060209@digitalhit.com> <4E61E297.1070506@digitalhit.com>
	<CAEOe2JxORb5vBxBaOAq8bVTUX-w-dzscCmFu1xj5ppGuYNuA+A@mail.gmail.com>
Message-ID: <4E61FEEF.6020206@digitalhit.com>

On 03/09/2011 5:23 AM, Calin Don wrote:
> Also your rewrite rule rewrite ^(.+) http://www.example.com
> <http://www.example.com/>$1 permanent; sould be rewritten into rewrite ^
> http://www.example.com <http://www.example.com/>$request_uri? permanent;

Just wondering...instead of a blanket redirect and then location that 
try to overide it, would it be better to have a rewrite that says "if 
not subdir1|subdir2|myfacebookapp rewrite to http". Then locations for 
those subdirs if they require passwords, etc.

How would I do the if not subdir1|subdir2|myfacebookapp rewrite to http 
check?


From mat999 at gmail.com  Sat Sep  3 10:22:06 2011
From: mat999 at gmail.com (SplitIce)
Date: Sat, 3 Sep 2011 20:22:06 +1000
Subject: URL decode rewrite handling: + vs. %20
In-Reply-To: <2f879159200e58c32a5b1eb6b3d01cef.NginxMailingListEnglish@forum.nginx.org>
References: <2f879159200e58c32a5b1eb6b3d01cef.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <CALFfGYYHxsvL7q6x057eH2O3QP7qRPDAr+UcobgT=x82AGpf=Q@mail.gmail.com>

edit the php script to do something like this before the rest of the script

$_GET['g2_path'] = urldecode($_GET['g2_path'])

?

On Sat, Sep 3, 2011 at 11:21 AM, signe <nginx-forum at nginx.us> wrote:

> Ubuntu Natty
> nginx/0.8.54
> PHP 5.3.5 / FPM / FastCGI
>
> I'm just beginning to work with nginx for the first time.  Converting my
> home server (very few hits) as an experiment for possible broader-scale
> testing and implementation on work servers (billions of hits per
> month).
>
> Everything has gone beautifully... up until I begin working on an older
> app: Gallery 2.  Some of the photos in my Gallery contain spaces, so
> they're being encoded in the old-but-still-accepted + format.  However,
> when I make the requests in nginx, it's passing the pathinfo through to
> fastcgi as an escaped + (%2b) instead of a space (+ or %20).
>
> The path is something like this:
>
> http://www.example.com/gallery/foo+bar.jpg.html
> which gets rewritten to
> http://www.example.com/gallery/main.php?g2_path=foo+bar.jpg.html
>
> In a path, + should be handled literally per RFC.  In a query string,
> it's supposed to be interpreted as a space per x-www-form-urlencoded.
> This is the disparity - what once was path is now query string.  Because
> of the conversion, Apache translates the + from literal string to
> encoded space during rewrite.  nginx simply encodes the literal string.
>
> I don't know if there's a solid answer as to which one is strictly
> "correct"... but what I do need to know is if there's a way to achieve
> the same behavior in nginx?
>
> This is an interesting edge case...
>
> Posted at Nginx Forum:
> http://forum.nginx.org/read.php?2,214668,214668#msg-214668
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110903/d5b1a960/attachment.html>

From ianevans at digitalhit.com  Sat Sep  3 10:24:59 2011
From: ianevans at digitalhit.com (Ian Evans)
Date: Sat, 03 Sep 2011 06:24:59 -0400
Subject: SSL location keep redirecting to non-SSL
In-Reply-To: <CALdpnrz2DP0OR_mVOREqbmjg6=3OzuBdOyZ8Yh2WZeH4772NTw@mail.gmail.com>
References: <4E6179FF.7060209@digitalhit.com> <4E61E297.1070506@digitalhit.com>
	<CALdpnrz2DP0OR_mVOREqbmjg6=3OzuBdOyZ8Yh2WZeH4772NTw@mail.gmail.com>
Message-ID: <4E62007B.6010601@digitalhit.com>

On 03/09/2011 6:14 AM, Stefan Caunter wrote:
>
> This looks odd to me, I've never seen a rewrite in an ssl config. It
> will complicate things.
>
> Leave the index to the server config, and take out the location / rewrite.
>
> Simplify.
Stefan,

The only reason I had the rewrite in there (and I think it was 
recommended to me years ago) was to prevent people or bots from calling 
the non-ssl directories of the site with https and creating duplicates 
in Google, etc. e.g. calling http://www.example.com/ and 
https://www.example.com/ when my homepage doesn't need https.

I only needs https calls on my phpMyAdmin, webmail, facebookapp and site 
admin subdirectories. I just wrote another email in the thread asking if 
I should only do a rewrite if it's NOT those directories.


From nginx-forum at nginx.us  Sat Sep  3 10:32:29 2011
From: nginx-forum at nginx.us (signe)
Date: Sat, 03 Sep 2011 06:32:29 -0400
Subject: URL decode rewrite handling: + vs. %20
In-Reply-To: <CALFfGYYHxsvL7q6x057eH2O3QP7qRPDAr+UcobgT=x82AGpf=Q@mail.gmail.com>
References: <CALFfGYYHxsvL7q6x057eH2O3QP7qRPDAr+UcobgT=x82AGpf=Q@mail.gmail.com>
Message-ID: <5e2ad7736ec3f420b7f1992a428e6969.NginxMailingListEnglish@forum.nginx.org>

splitice Wrote:
-------------------------------------------------------
> edit the php script to do something like this
> before the rest of the script
> 
> $_GET['g2_path'] = urldecode($_GET['g2_path'])
> 
> ?

That's not the problem - query parameters are automatically decoded by
PHP before processing even begins.  nginx is turning a + (space) into a
%2b (plus).  Once it's handed off to the application, it can't tell the
difference between a filename that actually had a +, and one that was
converted by nginx.  It needs to happen before it's handed off to
fastcgi.

I've already modified G2 to present URLs which are more accurately
encoded, but that doesn't resolve the issue since these URLs have been
published for years.  They're in search engines, blog posts, etc.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,214668,214687#msg-214687


From nginx-forum at nginx.us  Sat Sep  3 14:10:10 2011
From: nginx-forum at nginx.us (Xan)
Date: Sat, 03 Sep 2011 10:10:10 -0400
Subject: SOLVED Strange directory behaviour: rename causes 404 error
In-Reply-To: <43d7cb0d9aef0ebb36575cfad40c285c.NginxMailingListEnglish@forum.nginx.org>
References: <86f14be754becd175c07a61f1db101ba.NginxMailingListEnglish@forum.nginx.org>
	<43d7cb0d9aef0ebb36575cfad40c285c.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <d8e2ac58390ae37a97e353b7b5438352.NginxMailingListEnglish@forum.nginx.org>

Report as bug
[https://bugs.launchpad.net/ubuntu/+source/nginx/+bug/840358] in
ubuntu.

@ Samael: thanks.

Xan.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,214663,214688#msg-214688


From rami.essaid at gmail.com  Sat Sep  3 14:43:03 2011
From: rami.essaid at gmail.com (Rami Essaid)
Date: Sat, 3 Sep 2011 10:43:03 -0400
Subject: Storing URI in a cookie
In-Reply-To: <CAEOe2Jyi=sHbqN_iGix0DfEJ9cr2vhbQtNnQgiX1mvEwC4nk9w@mail.gmail.com>
References: <CAPMF7wg+TN80KDkzq4JMCcG+Y=8eTxT+L6bUcGb1bqxN49GNsw@mail.gmail.com>
	<CAEOe2Jyi=sHbqN_iGix0DfEJ9cr2vhbQtNnQgiX1mvEwC4nk9w@mail.gmail.com>
Message-ID: <CAPMF7wgoWPc59iOdR0x32zxdc5TA8OH=pD1bkDx0-zHedoOEDg@mail.gmail.com>

Thanks guys for the suggestions.

@calin - I have tried doing that but it messes with the verification page
and am not sure how to implement it correctly.

@Magicbear - I have put that code in there before the rewrite but then after
the rewrite when I try to access that cookie it is not there.  I should have
been a little more clear on what I have tried and my config.  Here is the
flow of the rewirte.

location / {
            error_page 503  = @recaptcha;
            if ($verifyvariable)
             {
             return 503;
              }
}

        location @recaptcha {

            add_header Set-Cookie  "originrequest=$request_uri path=/
expires=2m";
            secure_cookie $cookie_CAPTCHA_SESSION,$cookie_CAPTCHA_EXPIRES;
            rewrite ^ /captcha.html redirect;
    }

location = /captcha.html {
                #this has a form that submits information and goes to
/verify
}

location = /verify {
eval_inherit_body on;
eval_override_content_type 'text/plain';

eval $verify_content {
            proxy_method POST;
proxy_set_header  Accept-Encoding  "";
proxy_set_body ;
rewrite .* /recaptcha/api/verify break;
proxy_pass 'http://www.google.com';
}

if ($verify_content ~* ^true[\s\R]*(.*)) {
set $error_code $1;
rewrite .* /set_secure_cookie last;
}

}

location = /set_secure_cookie {
internal;
## SETTING
rewrite ^ /$cookie_ originrequest redirect;
}



On Sat, Sep 3, 2011 at 5:04 AM, Calin Don <calin.don at gmail.com> wrote:

> A better way to redirect users is to store the uri in the redirection url.
> Storing it in a cookie causes problems if user opens two different tabs with
> two different uris for example.
>  On Sep 2, 2011 11:42 PM, "Rami Essaid" <rami.essaid at gmail.com> wrote:
> > I dont know why I am struggling with this so much but I need a bit of
> help.
> > I am trying to redirect users to a verification page and then back to
> their
> > original requested URI but because of the rewrite I lose both the $uri
> and
> > $request_uri variables. I figure the easiest way to solve this problem is
> > to store the original URI in a cookie and read it from that after the
> > verification page. I cant seem to store the cookie correctly and read it
> > back. Can anyone provide some direction? Also, is there any risk that by
> > doing this I disrupt other cookies being stored by my application?
> >
> > Thanks!
> >
> > Rami
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110903/c1e597a6/attachment.html>

From drdr.xp at gmail.com  Sat Sep  3 15:06:24 2011
From: drdr.xp at gmail.com (dr-dr xp)
Date: Sat, 3 Sep 2011 23:06:24 +0800
Subject: nginx 1.0.2 AIO problem with open_file_cache
In-Reply-To: <20110903100335.GO1137@mdounin.ru>
References: <CAMcFNRJv_KSkj+DLQzXsCT02Ktd-W5uFXRAOz3MQJ+vErKOsuQ@mail.gmail.com>
	<20110903100335.GO1137@mdounin.ru>
Message-ID: <CAMcFNRJ_CbX-7OsonG20BBNF_59vKwFw1Y=34CjSxUg+0=1i0g@mail.gmail.com>

The attachment is the error.log with only the failed request. And stat
output of this file:

stat
//data2/0d97e01d7fbc48e498897492f0863929/g260674/257/26fd3732cd9f92f2413d10206cd8b0c9-5157235
  File:
`//data2/0d97e01d7fbc48e498897492f0863929/g260674/257/26fd3732cd9f92f2413d10206cd8b0c9-5157235'
  Size: 51163499        Blocks: 100048     IO Block: 4096   regular file
Device: 821h/2081d      Inode: 464404482   Links: 1
Access: (0664/-rw-rw-r--)  Uid: (    1003/    www)   Gid: (    1003/    www)
Access: 2011-09-03 22:31:22.000000000 +0800
Modify: 2011-07-30 01:45:56.000000000 +0800
Change: 2011-07-30 01:45:56.000000000 +0800


Such error occurs randomly on random file. The only relavent thing seems
like to be open_file_cache directive.
File sizes are almost 5M to 20M.
Files had not been changed. I checked the modify time of the files involved
in these errors. At first I had thought it was caused by a file change too.
But not.

In my cluster there are also some other similar nodes running with the same
open_file_cache parameters but without aio enabled(sendfile is used). They
have no such issue.




On Sat, Sep 3, 2011 at 6:03 PM, Maxim Dounin <mdounin at mdounin.ru> wrote:

> Hello!
>
> On Sat, Sep 03, 2011 at 05:46:39PM +0800, dr-dr xp wrote:
>
> > Hello everybody. I recently have had some problem with AIO support and
> need
> > some help or advice on it.
> >
> > With AIO on, everything was OK. But when I enabled open_file_cache, nginx
> > frequently closed client connection directly without sending back any
> data.
> >
> >
> > My platform is linux 2.6.30. CentOS 5.4;
> >
> > The nginx server that has this issue serves static file at backend.
> Another
> > nginx serves as proxy in front of it.
> > Backend nginx config is like this:
> > ...
> > aio            on;
> >
> >
> > directio       512;
> >
> >
> > output_buffers 1 128k;
> > ...
> >     location /files {
> >
> >
> >
> >         internal;
> >
> >
> >         alias /;
> >         open_file_cache max=50000 inactive=20s;
> >
> >
> >         open_file_cache_valid    30s;
> >
> >
> >         open_file_cache_min_uses 1;
> >
> >
> >         open_file_cache_errors   on;
> >
> >
> > With this config, on the frontend proxy server there are error logs like
> > this:
> >
> > 2011/09/03 17:01:19 [error] 12498#0: *78679759 *upstream prematurely
> closed
> > connection while reading response header from upstream*, client:
> > 221.193.212.77, server: , request: "GET .......... HTTP/1.1", upstream:
> > "http://
> **/file/871586e763bf4417bf8d475858deee2b/260110/0b47fc7f4d84e4bcb01a6d6d1fcacc94-3581749",
> > host.....
> >
> >
> > And I saw in tcpdump output the connection was closed very quickly by
> > backend nginx:
> >
> > 22:02:22.892327 IP 172.16.113.32.58271 > 172.16.228.63.irdmi: S
> > 1788188296:1788188296(0) win 5840 <mss 1460,sackOK,timestamp 666482608
> > 0,nop,wscale 7>
> > 22:02:22.895683 IP 172.16.228.63.irdmi > 172.16.113.32.58271: S
> > 350810413:350810413(0) ack 1788188297 win 5792 <mss 1460,sackOK,timestamp
> > 1948226601 66<
> > 22:02:22.895695 IP 172.16.113.32.58271 > 172.16.228.63.irdmi: . ack 1 win
> 46
> > <nop,nop,timestamp 666482612 1948226601>
> > 22:02:22.895703 IP 172.16.113.32.58271 > 172.16.228.63.irdmi: P
> 1:181(180)
> > ack 1 win 46 <nop,nop,timestamp 666482612 1948226601>
> > 22:02:22.897680 IP 172.16.228.63.irdmi > 172.16.113.32.58271: . ack 181
> win
> > 54 <nop,nop,timestamp 1948226603 666482612>
> > 22:02:22.898348 IP 172.16.228.63.irdmi > 172.16.113.32.58271: F 1:1(0)
> ack
> > 181 win 54 <nop,nop,timestamp 1948226603 666482612>
> > 22:02:22.898473 IP 172.16.113.32.58271 > 172.16.228.63.irdmi: F
> 181:181(0)
> > ack 2 win 46 <nop,nop,timestamp 666482615 1948226603>
> > 22:02:22.899804 IP 172.16.228.63.irdmi > 172.16.113.32.58271: . ack 182
> win
> > 54 <nop,nop,timestamp 1948226605 666482615>
> >
> >
> > On the backend nginx, strace shows that io_getevents got an error *"-22"
> > EINVAL*.
> >
> >   [pid  5275] recvfrom(48, 0x7fffe4400ae7, 1, 2, 0, 0) = -1 EAGAIN
> (Resource
> > temporarily unavailable)
> >   [pid  5275] epoll_wait(6, {{EPOLLIN|EPOLLOUT, {u32=2445415233,
> > u64=139850875540289}}}, 512, 5672) = 1
> >   [pid  5275] recvfrom(201, "\1\6\0\1\0\337\1\0Status: 301 Moved
> Perman"...,
> > 4096, 0, NULL, NULL) = 256
> >   [pid  5275] close(201)                  = 0
> >   [pid  5275]
> >
> stat("//data1/871586e763bf4417bf8d475858deee2b/g268315/252/995f20f76d0d3871fdb5d71bbc92956c-27769564",
> > {st_mode=S_IFREG|0600, st_size=122507, ...<
> >   [pid  5275] io_submit(139850942226432, 1, {{0xf43718, 0, 0, 0, 40}}) =
> 1
> >
> >
> >   [pid  5275] epoll_wait(6, {{EPOLLIN, {u32=7173248, u64=7173248}}}, 512,
> > 5671) = 1
> >
> >   [pid  5275] read(7, "\1\0\0\0\0\0\0\0", 8) = 8
> >   [pid  5275] *io_getevents(139850942226432, 1, 64, {{0xf43718, 0xf436d8,
> > -22, 0}}, {0, 0}) = 1*
> >   [pid  5275] write(24, "172.16.164.30 - - 172.16.228.63:"..., 338) =
> 338
> >   [pid  5275] close(48)
> >
> > And access.log shows that response length is ZERO:
> >
> > [03/Sep/2011:06:28:48 +0800] "GET
> >
> /file/871586e763bf4417bf8d475858deee2b/260284/ed31933859d27e33b8398deaea1d2ade-3969549
> > HTTP/1.0" 206 0 0.001 "-" "-" "-" Range:bytes=27459105-
> >
> >
> > After I removed those *open_file_cache* directives, No such error
> occurred.
> >
> > And the shorter *inactive* is,  the more often I get such errors.
> >
> >
> > Anybody has idea about avoiding this or fixing it?
>
> Could you please provide debug log?  See
> http://wiki.nginx.org/Debugging for more information.
>
> Additionally, some more info about file requested might be
> helpful: how big it is? whether it changed on backend during
> requests? if yes, is it changed atomically?
>
> Maxim Dounin
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx
>



-- 
????????????????
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110903/4b9cc5c0/attachment-0001.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: error.log
Type: application/octet-stream
Size: 18946 bytes
Desc: not available
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110903/4b9cc5c0/attachment-0001.obj>

From drdr.xp at gmail.com  Sat Sep  3 15:10:03 2011
From: drdr.xp at gmail.com (dr-dr xp)
Date: Sat, 3 Sep 2011 23:10:03 +0800
Subject: nginx 1.0.2 AIO problem with open_file_cache
In-Reply-To: <CAMcFNRJ_CbX-7OsonG20BBNF_59vKwFw1Y=34CjSxUg+0=1i0g@mail.gmail.com>
References: <CAMcFNRJv_KSkj+DLQzXsCT02Ktd-W5uFXRAOz3MQJ+vErKOsuQ@mail.gmail.com>
	<20110903100335.GO1137@mdounin.ru>
	<CAMcFNRJ_CbX-7OsonG20BBNF_59vKwFw1Y=34CjSxUg+0=1i0g@mail.gmail.com>
Message-ID: <CAMcFNRJvOSr7MODEPfPPkMf1E+tJXvqMK0eL5CB1aD3nAEY=OQ@mail.gmail.com>

And this is the access log entry:
172.16.113.32 - - 172.16.228.63:8000 [03/Sep/2011:22:43:44 +0800] "GET
/file/0d97e01d7fbc48e498897492f0863929/260674/26fd3732cd9f92f2413d10206cd8b0c9-5157235
HTTP/1.0" 206 0 0.002 "-" "-" "-" Range:bytes=1059980- usTime:0.001
us:unix:/tmp/storage_fcgi_sock RID:0000052a-1109-0322-4344-842b2b0c0467
ERR:- ReqLength:171 ReqBody:- COOKIE:"-"

0000052a is connection id in hex.


2011/9/3 dr-dr xp <drdr.xp at gmail.com>

> The attachment is the error.log with only the failed request. And stat
> output of this file:
>
> stat
> //data2/0d97e01d7fbc48e498897492f0863929/g260674/257/26fd3732cd9f92f2413d10206cd8b0c9-5157235
>   File:
> `//data2/0d97e01d7fbc48e498897492f0863929/g260674/257/26fd3732cd9f92f2413d10206cd8b0c9-5157235'
>   Size: 51163499        Blocks: 100048     IO Block: 4096   regular file
> Device: 821h/2081d      Inode: 464404482   Links: 1
> Access: (0664/-rw-rw-r--)  Uid: (    1003/    www)   Gid: (    1003/
>  www)
> Access: 2011-09-03 22:31:22.000000000 +0800
> Modify: 2011-07-30 01:45:56.000000000 +0800
> Change: 2011-07-30 01:45:56.000000000 +0800
>
>
> Such error occurs randomly on random file. The only relavent thing seems
> like to be open_file_cache directive.
> File sizes are almost 5M to 20M.
> Files had not been changed. I checked the modify time of the files involved
> in these errors. At first I had thought it was caused by a file change too.
> But not.
>
> In my cluster there are also some other similar nodes running with the same
> open_file_cache parameters but without aio enabled(sendfile is used). They
> have no such issue.
>
>
>
>
> On Sat, Sep 3, 2011 at 6:03 PM, Maxim Dounin <mdounin at mdounin.ru> wrote:
>
>> Hello!
>>
>> On Sat, Sep 03, 2011 at 05:46:39PM +0800, dr-dr xp wrote:
>>
>> > Hello everybody. I recently have had some problem with AIO support and
>> need
>> > some help or advice on it.
>> >
>> > With AIO on, everything was OK. But when I enabled open_file_cache,
>> nginx
>> > frequently closed client connection directly without sending back any
>> data.
>> >
>> >
>> > My platform is linux 2.6.30. CentOS 5.4;
>> >
>> > The nginx server that has this issue serves static file at backend.
>> Another
>> > nginx serves as proxy in front of it.
>> > Backend nginx config is like this:
>> > ...
>> > aio            on;
>> >
>> >
>> > directio       512;
>> >
>> >
>> > output_buffers 1 128k;
>> > ...
>> >     location /files {
>> >
>> >
>> >
>> >         internal;
>> >
>> >
>> >         alias /;
>> >         open_file_cache max=50000 inactive=20s;
>> >
>> >
>> >         open_file_cache_valid    30s;
>> >
>> >
>> >         open_file_cache_min_uses 1;
>> >
>> >
>> >         open_file_cache_errors   on;
>> >
>> >
>> > With this config, on the frontend proxy server there are error logs like
>> > this:
>> >
>> > 2011/09/03 17:01:19 [error] 12498#0: *78679759 *upstream prematurely
>> closed
>> > connection while reading response header from upstream*, client:
>> > 221.193.212.77, server: , request: "GET .......... HTTP/1.1", upstream:
>> > "http://
>> **/file/871586e763bf4417bf8d475858deee2b/260110/0b47fc7f4d84e4bcb01a6d6d1fcacc94-3581749",
>> > host.....
>> >
>> >
>> > And I saw in tcpdump output the connection was closed very quickly by
>> > backend nginx:
>> >
>> > 22:02:22.892327 IP 172.16.113.32.58271 > 172.16.228.63.irdmi: S
>> > 1788188296:1788188296(0) win 5840 <mss 1460,sackOK,timestamp 666482608
>> > 0,nop,wscale 7>
>> > 22:02:22.895683 IP 172.16.228.63.irdmi > 172.16.113.32.58271: S
>> > 350810413:350810413(0) ack 1788188297 win 5792 <mss
>> 1460,sackOK,timestamp
>> > 1948226601 66<
>> > 22:02:22.895695 IP 172.16.113.32.58271 > 172.16.228.63.irdmi: . ack 1
>> win 46
>> > <nop,nop,timestamp 666482612 1948226601>
>> > 22:02:22.895703 IP 172.16.113.32.58271 > 172.16.228.63.irdmi: P
>> 1:181(180)
>> > ack 1 win 46 <nop,nop,timestamp 666482612 1948226601>
>> > 22:02:22.897680 IP 172.16.228.63.irdmi > 172.16.113.32.58271: . ack 181
>> win
>> > 54 <nop,nop,timestamp 1948226603 666482612>
>> > 22:02:22.898348 IP 172.16.228.63.irdmi > 172.16.113.32.58271: F 1:1(0)
>> ack
>> > 181 win 54 <nop,nop,timestamp 1948226603 666482612>
>> > 22:02:22.898473 IP 172.16.113.32.58271 > 172.16.228.63.irdmi: F
>> 181:181(0)
>> > ack 2 win 46 <nop,nop,timestamp 666482615 1948226603>
>> > 22:02:22.899804 IP 172.16.228.63.irdmi > 172.16.113.32.58271: . ack 182
>> win
>> > 54 <nop,nop,timestamp 1948226605 666482615>
>> >
>> >
>> > On the backend nginx, strace shows that io_getevents got an error *"-22"
>> > EINVAL*.
>> >
>> >   [pid  5275] recvfrom(48, 0x7fffe4400ae7, 1, 2, 0, 0) = -1 EAGAIN
>> (Resource
>> > temporarily unavailable)
>> >   [pid  5275] epoll_wait(6, {{EPOLLIN|EPOLLOUT, {u32=2445415233,
>> > u64=139850875540289}}}, 512, 5672) = 1
>> >   [pid  5275] recvfrom(201, "\1\6\0\1\0\337\1\0Status: 301 Moved
>> Perman"...,
>> > 4096, 0, NULL, NULL) = 256
>> >   [pid  5275] close(201)                  = 0
>> >   [pid  5275]
>> >
>> stat("//data1/871586e763bf4417bf8d475858deee2b/g268315/252/995f20f76d0d3871fdb5d71bbc92956c-27769564",
>> > {st_mode=S_IFREG|0600, st_size=122507, ...<
>> >   [pid  5275] io_submit(139850942226432, 1, {{0xf43718, 0, 0, 0, 40}}) =
>> 1
>> >
>> >
>> >   [pid  5275] epoll_wait(6, {{EPOLLIN, {u32=7173248, u64=7173248}}},
>> 512,
>> > 5671) = 1
>> >
>> >   [pid  5275] read(7, "\1\0\0\0\0\0\0\0", 8) = 8
>> >   [pid  5275] *io_getevents(139850942226432, 1, 64, {{0xf43718,
>> 0xf436d8,
>> > -22, 0}}, {0, 0}) = 1*
>> >   [pid  5275] write(24, "172.16.164.30 - - 172.16.228.63:"..., 338) =
>> 338
>> >   [pid  5275] close(48)
>> >
>> > And access.log shows that response length is ZERO:
>> >
>> > [03/Sep/2011:06:28:48 +0800] "GET
>> >
>> /file/871586e763bf4417bf8d475858deee2b/260284/ed31933859d27e33b8398deaea1d2ade-3969549
>> > HTTP/1.0" 206 0 0.001 "-" "-" "-" Range:bytes=27459105-
>> >
>> >
>> > After I removed those *open_file_cache* directives, No such error
>> occurred.
>> >
>> > And the shorter *inactive* is,  the more often I get such errors.
>> >
>> >
>> > Anybody has idea about avoiding this or fixing it?
>>
>> Could you please provide debug log?  See
>> http://wiki.nginx.org/Debugging for more information.
>>
>> Additionally, some more info about file requested might be
>> helpful: how big it is? whether it changed on backend during
>> requests? if yes, is it changed atomically?
>>
>> Maxim Dounin
>>
>> _______________________________________________
>> nginx mailing list
>> nginx at nginx.org
>> http://mailman.nginx.org/mailman/listinfo/nginx
>>
>
>
>
> --
> ????????????????
>



-- 
????????????????
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110903/358fb907/attachment.html>

From calin at presslabs.com  Sat Sep  3 14:11:07 2011
From: calin at presslabs.com (Calin Don)
Date: Sat, 3 Sep 2011 17:11:07 +0300
Subject: Performance of map directive
Message-ID: <CAEOe2JxJkLvicxZ2WbV8id=0gkbW2ej1aBeZ6SR8Ux3tGbkSLA@mail.gmail.com>

Hi,

I understood that if you have only strings in a map directive the access
time is O(1).
What if you have the hostnames directive enabled and some of the items are
in the format *.example.com?
What about if you mix with regular expressions?

Thanks,
Calin
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110903/f5b1004c/attachment.html>

From igor at sysoev.ru  Sat Sep  3 15:50:20 2011
From: igor at sysoev.ru (Igor Sysoev)
Date: Sat, 3 Sep 2011 19:50:20 +0400
Subject: Performance of map directive
In-Reply-To: <CAEOe2JxJkLvicxZ2WbV8id=0gkbW2ej1aBeZ6SR8Ux3tGbkSLA@mail.gmail.com>
References: <CAEOe2JxJkLvicxZ2WbV8id=0gkbW2ej1aBeZ6SR8Ux3tGbkSLA@mail.gmail.com>
Message-ID: <20110903155020.GB86123@nginx.com>

On Sat, Sep 03, 2011 at 05:11:07PM +0300, Calin Don wrote:
> Hi,
> 
> I understood that if you have only strings in a map directive the access
> time is O(1).

Yes.

> What if you have the hostnames directive enabled and some of the items are
> in the format *.example.com?

nginx tests address by its parts, first "com", then "example".
So access of the operation varies from O(1) to O(N) where N is number of
parts of longest name. For example, if you test example.net only against

*.example.com
*.sub.domain.com

then this will be O(1). If you test www.sub.domain.com, it will be O(3).

> What about if you mix with regular expressions?

The regular expressions tested sequentially.

nginx tests map in the following order:

1) exact names,
2) *.names,
3) names.*,
4) regexes.

The first match stops the tesing.


-- 
Igor Sysoev


From ianevans at digitalhit.com  Sat Sep  3 19:51:00 2011
From: ianevans at digitalhit.com (Ian Evans)
Date: Sat, 03 Sep 2011 15:51:00 -0400
Subject: SSL location keep redirecting to non-SSL
In-Reply-To: <4E62007B.6010601@digitalhit.com>
References: <4E6179FF.7060209@digitalhit.com> <4E61E297.1070506@digitalhit.com>
	<CALdpnrz2DP0OR_mVOREqbmjg6=3OzuBdOyZ8Yh2WZeH4772NTw@mail.gmail.com>
	<4E62007B.6010601@digitalhit.com>
Message-ID: <4E628524.1070409@digitalhit.com>

After a few hours sleep, it appears that the location was missing:

fastcgi_index index.shtml;

Seems to work now.


From nginx at nginxuser.net  Sat Sep  3 21:02:34 2011
From: nginx at nginxuser.net (Nginx User)
Date: Sun, 4 Sep 2011 00:02:34 +0300
Subject: http_auth_request_module Instructions?
Message-ID: <CAN=WzV2e8_OVk24OyUcB=3JOXUWXfONNV-7b9VW-NmhGu5b4cw@mail.gmail.com>

Is the any documentation for the http_auth_request_module anywhere? Trying
to find out what the configuration parameters are if any.

Thanks
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110904/a5f9621c/attachment.html>

From mdounin at mdounin.ru  Sat Sep  3 22:25:15 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Sun, 4 Sep 2011 02:25:15 +0400
Subject: http_auth_request_module Instructions?
In-Reply-To: <CAN=WzV2e8_OVk24OyUcB=3JOXUWXfONNV-7b9VW-NmhGu5b4cw@mail.gmail.com>
References: <CAN=WzV2e8_OVk24OyUcB=3JOXUWXfONNV-7b9VW-NmhGu5b4cw@mail.gmail.com>
Message-ID: <20110903222515.GP1137@mdounin.ru>

Hello!

On Sun, Sep 04, 2011 at 12:02:34AM +0300, Nginx User wrote:

> Is the any documentation for the http_auth_request_module anywhere? Trying
> to find out what the configuration parameters are if any.

Try README in the module tarball.  Alternatively, you may find it 
here:

http://mdounin.ru/hg/ngx_http_auth_request_module/file/tip/README

Maxim Dounin


From agentzh at gmail.com  Sun Sep  4 05:22:24 2011
From: agentzh at gmail.com (agentzh)
Date: Sun, 4 Sep 2011 13:22:24 +0800
Subject: [ANN] The lua-rds-parser Lua library and ngx_rds_csv nginx module
In-Reply-To: <CAB4Tn6P-LyhkUz9NQYYvm6aF-mcD=mcQRapoB-EHDV3QZz1mhA@mail.gmail.com>
References: <CAB4Tn6P-LyhkUz9NQYYvm6aF-mcD=mcQRapoB-EHDV3QZz1mhA@mail.gmail.com>
Message-ID: <CAB4Tn6OyN7xwsqveh_jAp07xro3gnyKUJ0eo3yF232H8YdZDgQ@mail.gmail.com>

Hello!

Recently I wrote a Lua C library named lua-rds-parser for the
ngx_openresty [1] world that we can now efficiently parse Resty DBD
Stream (RDS) formatted data generated by ngx_drizzle [2] and
ngx_postgres [3] into Lua data structures without using JSON as the
intermediate data format. This can save both CPU time and memory for
web applications running atop ngx_lua [4].

Here is the project homepage for this Lua library:

? ?https://github.com/agentzh/lua-rds-parser

Also, there's a new Nginx filter module named ngx_rds_csv that can
convert RDS output into the Comma-Separated Values (CSV) format in a
streaming fashion:

??? https://github.com/agentzh/rds-csv-nginx-module

Both of these two components have already been included and enabled by
default in the latest ngx_openresty stable release [5] :)

Enjoy!
-agentzh

[1] http://openresty.org
[2] http://wiki.nginx.org/HttpDrizzleModule
[3] http://github.com/FRiCKLE/ngx_postgres/
[4] http://wiki.nginx.org/HttpLuaModule
[5] http://openresty.org/#Download


From sy.meteor at msn.com  Sun Sep  4 04:33:14 2011
From: sy.meteor at msn.com (Xin Liu)
Date: Sun, 4 Sep 2011 12:33:14 +0800
Subject: php-fpm is very very slow...
Message-ID: <SNT123-DS1859DE5E8065DAA750AC3EE51A0@phx.gbl>

Hi all,

After upgrade from php 5.3.6 to php 5.3.8, now php-fpm is very very slow. Even I tried to roll back to php 5.3.6, it still doen't work.

And there are lots of errors in php-fpm.log,

  [04-Sep-2011 12:28:27] WARNING: [pool www] child 18793, script '/web/www/forum.php' (request: "GET /forum.php") executing too slow (60.073197 sec), logging
  [04-Sep-2011 12:28:27] WARNING: [pool www] child 18784, script '/web/www/forum.php' (request: "GET /forum.php") executing too slow (60.080807 sec), logging
  [04-Sep-2011 12:28:27] NOTICE: child 18793 stopped for tracing
  [04-Sep-2011 12:28:27] NOTICE: about to trace 18793
  [04-Sep-2011 12:28:27] ERROR: ptrace(PT_IO) failed: Bad address (14)
  [04-Sep-2011 12:28:27] NOTICE: finished trace of 18793
  [04-Sep-2011 12:28:27] NOTICE: child 18784 stopped for tracing
  [04-Sep-2011 12:28:27] NOTICE: about to trace 18784
  [04-Sep-2011 12:28:27] ERROR: ptrace(PT_IO) failed: Bad address (14)
  [04-Sep-2011 12:28:27] NOTICE: finished trace of 18784
And in php-fpm.slow

  [04-Sep-2011 12:27:30]  [pool www] pid 18622
  script_filename = /web/www/forum.php
  [0x000000080455aaa8] mysql_query() /web/www/source/class/class_core.php:815
  [0x000000080455a958] query() /web/www/source/class/class_core.php:1003
  [0x000000080455a858] _execute() /web/www/source/class/class_core.php:977
  [0x00000008045591a0] query() /web/www/source/module/forum/forum_viewthread.php:514
  [0x0000000804559058] +++ dump failed


And one can help me about this issue?
My server is running Freebsd 8.1 + nginx + php 5.3.8
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110904/7495716e/attachment.html>

From ben at benweblife.fr  Sun Sep  4 09:32:56 2011
From: ben at benweblife.fr (Ben)
Date: Sun, 04 Sep 2011 11:32:56 +0200
Subject: php-fpm is very very slow...
In-Reply-To: <SNT123-DS1859DE5E8065DAA750AC3EE51A0@phx.gbl>
References: <SNT123-DS1859DE5E8065DAA750AC3EE51A0@phx.gbl>
Message-ID: <4E6345C8.7060800@benweblife.fr>

Le 04/09/2011 06:33, Xin Liu a ?crit :
> Hi all,
Hi,
> After upgrade from php 5.3.6 to php 5.3.8, now php-fpm is very very 
> slow. Even I tried to roll back to php 5.3.6, it still doen't work.
> And one can help me about this issue?
PHP (and PHP-FPM) is not part of nginx, you should post your issue on 
PHP lists.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110904/ba83c4c4/attachment.html>

From rainer at ultra-secure.de  Sun Sep  4 11:37:06 2011
From: rainer at ultra-secure.de (Rainer Duffner)
Date: Sun, 4 Sep 2011 13:37:06 +0200
Subject: php-fpm is very very slow...
In-Reply-To: <SNT123-DS1859DE5E8065DAA750AC3EE51A0@phx.gbl>
References: <SNT123-DS1859DE5E8065DAA750AC3EE51A0@phx.gbl>
Message-ID: <20110904133706.31a42c4a@linux-0n9l.site.local>

Am Sun, 4 Sep 2011 12:33:14 +0800
schrieb "Xin Liu" <sy.meteor at msn.com>:

> Hi all,
> 
> After upgrade from php 5.3.6 to php 5.3.8, now php-fpm is very very
> slow. Even I tried to roll back to php 5.3.6, it still doen't work.
...
> And one can help me about this issue?
> My server is running Freebsd 8.1 + nginx + php 5.3.8



Haven't had that one.
If this is a test-server, upgrade to 8.2 and recompile everything.

(make a pkg_info list, then pkg_delete everything,
even /usr/local/lib/compat/, then install the pkgs again, as per your
list.)


If this is a production-server, maybe it's time to have a
test-server....


But I agree - it's most likely not a problem with NGINX. Would still be
slow with apache...



From edho at myconan.net  Sun Sep  4 12:23:28 2011
From: edho at myconan.net (Edho P Arief)
Date: Sun, 4 Sep 2011 19:23:28 +0700
Subject: php-fpm is very very slow...
In-Reply-To: <SNT123-DS1859DE5E8065DAA750AC3EE51A0@phx.gbl>
References: <SNT123-DS1859DE5E8065DAA750AC3EE51A0@phx.gbl>
Message-ID: <CABTjkKk2wd+PNAYx9gf=uwax9oYmwdfwW3iZYfKyV+XKJc1CBw@mail.gmail.com>

2011/9/4 Xin Liu <sy.meteor at msn.com>:
> Hi all,
>
> After upgrade from php 5.3.6 to php 5.3.8, now php-fpm is very very slow.
> Even I tried to roll back to php 5.3.6, it still doen't work.
>
> [0x000000080455aaa8] mysql_query() /web/www/source/class/class_core.php:815
> [0x000000080455a958] query() /web/www/source/class/class_core.php:1003
> [0x000000080455a858] _execute() /web/www/source/class/class_core.php:977
> [0x00000008045591a0] query()
> /web/www/source/module/forum/forum_viewthread.php:514
> [0x0000000804559058] +++ dump failed

looks like db issue. Try turning on your db's slow query logging to
see which query slowed the page down.


From quan.nexthop at gmail.com  Sun Sep  4 15:51:31 2011
From: quan.nexthop at gmail.com (Geoge.Q)
Date: Sun, 4 Sep 2011 23:51:31 +0800
Subject: =?UTF-8?Q?upstream_configuration_is_changed=EF=BC=9FThe_keyword_=28weight/?=
	=?UTF-8?Q?max=5Ffail/fail=5Ftimeout=29_is_not_supported=3F?=
Message-ID: <CAP_76Mbvn6-MhH0Kmj2-jp0-mJDvYhggLBTuqP1BUUc95qwmjA@mail.gmail.com>

Hi All?

I met a configuation issue. the configuration is as following:?NGINX is
V0.7.68) and upstream is enabled when compiled.
I also test it in V1.0.5, the issue happens again.

----------------------------------------------------------------------------------------------------------------------------------------------
                server {
                        listen 80;
                        server_name aaa.com;
                        server_name 192.168.2.140;
                        location / {
                                proxy_pass http://haha;
                                proxy_redirect off;
                                proxy_buffering off;
                                proxy_set_header Host $host;
                                proxy_set_header X-Real-IP  $remote_addr;
                                proxy_set_header X-Forwarded-For
$proxy_add_x_forwarded_for;
                                client_max_body_size 1000m;
                                client_body_buffer_size 256k;
                                proxy_connect_timeout 600;
                                proxy_temp_file_write_size  64k;
                        }
                }
                upstream haha {
                        server 1.1.1.2 max_fails=3  fail_timeout=30s;
<---------max_fails/fail_timeouts/weight is not support ?
                        server 1.1.1.1 weight=10;
                }
----------------------------------------------------------------------------------------------------------------------------------------------

When I run nginx, the error message print out as following:
./nginx

Root# /etc/init.d/nginx restart
[emerg]: invalid parameter "max_fails=3" in
/var/log/nginx/conf/nginx.conf:19
~
~
~
~
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110904/e348fb2d/attachment.html>

From quan.nexthop at gmail.com  Sun Sep  4 15:52:39 2011
From: quan.nexthop at gmail.com (Geoge.Q)
Date: Sun, 4 Sep 2011 23:52:39 +0800
Subject: [ANN] The lua-rds-parser Lua library and ngx_rds_csv nginx module
In-Reply-To: <CAB4Tn6OyN7xwsqveh_jAp07xro3gnyKUJ0eo3yF232H8YdZDgQ@mail.gmail.com>
References: <CAB4Tn6P-LyhkUz9NQYYvm6aF-mcD=mcQRapoB-EHDV3QZz1mhA@mail.gmail.com>
	<CAB4Tn6OyN7xwsqveh_jAp07xro3gnyKUJ0eo3yF232H8YdZDgQ@mail.gmail.com>
Message-ID: <CAP_76Mb4__E=BgGZAtGd2wRk0cPPq2K5LYv8CuGrRH6Zy9xBtw@mail.gmail.com>

Thanks your share with us.

george

On Sun, Sep 4, 2011 at 1:22 PM, agentzh <agentzh at gmail.com> wrote:

> Hello!
>
> Recently I wrote a Lua C library named lua-rds-parser for the
> ngx_openresty [1] world that we can now efficiently parse Resty DBD
> Stream (RDS) formatted data generated by ngx_drizzle [2] and
> ngx_postgres [3] into Lua data structures without using JSON as the
> intermediate data format. This can save both CPU time and memory for
> web applications running atop ngx_lua [4].
>
> Here is the project homepage for this Lua library:
>
>    https://github.com/agentzh/lua-rds-parser
>
> Also, there's a new Nginx filter module named ngx_rds_csv that can
> convert RDS output into the Comma-Separated Values (CSV) format in a
> streaming fashion:
>
>     https://github.com/agentzh/rds-csv-nginx-module
>
> Both of these two components have already been included and enabled by
> default in the latest ngx_openresty stable release [5] :)
>
> Enjoy!
> -agentzh
>
> [1] http://openresty.org
> [2] http://wiki.nginx.org/HttpDrizzleModule
> [3] http://github.com/FRiCKLE/ngx_postgres/
> [4] http://wiki.nginx.org/HttpLuaModule
> [5] http://openresty.org/#Download
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110904/794d82bf/attachment.html>

From mdounin at mdounin.ru  Sun Sep  4 16:34:13 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Sun, 4 Sep 2011 20:34:13 +0400
Subject: nginx 1.0.2 AIO problem with open_file_cache
In-Reply-To: <CAMcFNRJ_CbX-7OsonG20BBNF_59vKwFw1Y=34CjSxUg+0=1i0g@mail.gmail.com>
References: <CAMcFNRJv_KSkj+DLQzXsCT02Ktd-W5uFXRAOz3MQJ+vErKOsuQ@mail.gmail.com>
	<20110903100335.GO1137@mdounin.ru>
	<CAMcFNRJ_CbX-7OsonG20BBNF_59vKwFw1Y=34CjSxUg+0=1i0g@mail.gmail.com>
Message-ID: <20110904163413.GR1137@mdounin.ru>

Hello!

On Sat, Sep 03, 2011 at 11:06:24PM +0800, dr-dr xp wrote:

> The attachment is the error.log with only the failed request. And stat
> output of this file:
> 
> stat
> //data2/0d97e01d7fbc48e498897492f0863929/g260674/257/26fd3732cd9f92f2413d10206cd8b0c9-5157235
>   File:
> `//data2/0d97e01d7fbc48e498897492f0863929/g260674/257/26fd3732cd9f92f2413d10206cd8b0c9-5157235'
>   Size: 51163499        Blocks: 100048     IO Block: 4096   regular file
> Device: 821h/2081d      Inode: 464404482   Links: 1
> Access: (0664/-rw-rw-r--)  Uid: (    1003/    www)   Gid: (    1003/    www)
> Access: 2011-09-03 22:31:22.000000000 +0800
> Modify: 2011-07-30 01:45:56.000000000 +0800
> Change: 2011-07-30 01:45:56.000000000 +0800
> 
> 
> Such error occurs randomly on random file. The only relavent thing seems
> like to be open_file_cache directive.
> File sizes are almost 5M to 20M.
> Files had not been changed. I checked the modify time of the files involved
> in these errors. At first I had thought it was caused by a file change too.
> But not.
> 
> In my cluster there are also some other similar nodes running with the same
> open_file_cache parameters but without aio enabled(sendfile is used). They
> have no such issue.

Please try the attached patch.

Maxim Dounin

> 
> 
> 
> 
> On Sat, Sep 3, 2011 at 6:03 PM, Maxim Dounin <mdounin at mdounin.ru> wrote:
> 
> > Hello!
> >
> > On Sat, Sep 03, 2011 at 05:46:39PM +0800, dr-dr xp wrote:
> >
> > > Hello everybody. I recently have had some problem with AIO support and
> > need
> > > some help or advice on it.
> > >
> > > With AIO on, everything was OK. But when I enabled open_file_cache, nginx
> > > frequently closed client connection directly without sending back any
> > data.
> > >
> > >
> > > My platform is linux 2.6.30. CentOS 5.4;
> > >
> > > The nginx server that has this issue serves static file at backend.
> > Another
> > > nginx serves as proxy in front of it.
> > > Backend nginx config is like this:
> > > ...
> > > aio            on;
> > >
> > >
> > > directio       512;
> > >
> > >
> > > output_buffers 1 128k;
> > > ...
> > >     location /files {
> > >
> > >
> > >
> > >         internal;
> > >
> > >
> > >         alias /;
> > >         open_file_cache max=50000 inactive=20s;
> > >
> > >
> > >         open_file_cache_valid    30s;
> > >
> > >
> > >         open_file_cache_min_uses 1;
> > >
> > >
> > >         open_file_cache_errors   on;
> > >
> > >
> > > With this config, on the frontend proxy server there are error logs like
> > > this:
> > >
> > > 2011/09/03 17:01:19 [error] 12498#0: *78679759 *upstream prematurely
> > closed
> > > connection while reading response header from upstream*, client:
> > > 221.193.212.77, server: , request: "GET .......... HTTP/1.1", upstream:
> > > "http://
> > **/file/871586e763bf4417bf8d475858deee2b/260110/0b47fc7f4d84e4bcb01a6d6d1fcacc94-3581749",
> > > host.....
> > >
> > >
> > > And I saw in tcpdump output the connection was closed very quickly by
> > > backend nginx:
> > >
> > > 22:02:22.892327 IP 172.16.113.32.58271 > 172.16.228.63.irdmi: S
> > > 1788188296:1788188296(0) win 5840 <mss 1460,sackOK,timestamp 666482608
> > > 0,nop,wscale 7>
> > > 22:02:22.895683 IP 172.16.228.63.irdmi > 172.16.113.32.58271: S
> > > 350810413:350810413(0) ack 1788188297 win 5792 <mss 1460,sackOK,timestamp
> > > 1948226601 66<
> > > 22:02:22.895695 IP 172.16.113.32.58271 > 172.16.228.63.irdmi: . ack 1 win
> > 46
> > > <nop,nop,timestamp 666482612 1948226601>
> > > 22:02:22.895703 IP 172.16.113.32.58271 > 172.16.228.63.irdmi: P
> > 1:181(180)
> > > ack 1 win 46 <nop,nop,timestamp 666482612 1948226601>
> > > 22:02:22.897680 IP 172.16.228.63.irdmi > 172.16.113.32.58271: . ack 181
> > win
> > > 54 <nop,nop,timestamp 1948226603 666482612>
> > > 22:02:22.898348 IP 172.16.228.63.irdmi > 172.16.113.32.58271: F 1:1(0)
> > ack
> > > 181 win 54 <nop,nop,timestamp 1948226603 666482612>
> > > 22:02:22.898473 IP 172.16.113.32.58271 > 172.16.228.63.irdmi: F
> > 181:181(0)
> > > ack 2 win 46 <nop,nop,timestamp 666482615 1948226603>
> > > 22:02:22.899804 IP 172.16.228.63.irdmi > 172.16.113.32.58271: . ack 182
> > win
> > > 54 <nop,nop,timestamp 1948226605 666482615>
> > >
> > >
> > > On the backend nginx, strace shows that io_getevents got an error *"-22"
> > > EINVAL*.
> > >
> > >   [pid  5275] recvfrom(48, 0x7fffe4400ae7, 1, 2, 0, 0) = -1 EAGAIN
> > (Resource
> > > temporarily unavailable)
> > >   [pid  5275] epoll_wait(6, {{EPOLLIN|EPOLLOUT, {u32=2445415233,
> > > u64=139850875540289}}}, 512, 5672) = 1
> > >   [pid  5275] recvfrom(201, "\1\6\0\1\0\337\1\0Status: 301 Moved
> > Perman"...,
> > > 4096, 0, NULL, NULL) = 256
> > >   [pid  5275] close(201)                  = 0
> > >   [pid  5275]
> > >
> > stat("//data1/871586e763bf4417bf8d475858deee2b/g268315/252/995f20f76d0d3871fdb5d71bbc92956c-27769564",
> > > {st_mode=S_IFREG|0600, st_size=122507, ...<
> > >   [pid  5275] io_submit(139850942226432, 1, {{0xf43718, 0, 0, 0, 40}}) =
> > 1
> > >
> > >
> > >   [pid  5275] epoll_wait(6, {{EPOLLIN, {u32=7173248, u64=7173248}}}, 512,
> > > 5671) = 1
> > >
> > >   [pid  5275] read(7, "\1\0\0\0\0\0\0\0", 8) = 8
> > >   [pid  5275] *io_getevents(139850942226432, 1, 64, {{0xf43718, 0xf436d8,
> > > -22, 0}}, {0, 0}) = 1*
> > >   [pid  5275] write(24, "172.16.164.30 - - 172.16.228.63:"..., 338) =
> > 338
> > >   [pid  5275] close(48)
> > >
> > > And access.log shows that response length is ZERO:
> > >
> > > [03/Sep/2011:06:28:48 +0800] "GET
> > >
> > /file/871586e763bf4417bf8d475858deee2b/260284/ed31933859d27e33b8398deaea1d2ade-3969549
> > > HTTP/1.0" 206 0 0.001 "-" "-" "-" Range:bytes=27459105-
> > >
> > >
> > > After I removed those *open_file_cache* directives, No such error
> > occurred.
> > >
> > > And the shorter *inactive* is,  the more often I get such errors.
> > >
> > >
> > > Anybody has idea about avoiding this or fixing it?
> >
> > Could you please provide debug log?  See
> > http://wiki.nginx.org/Debugging for more information.
> >
> > Additionally, some more info about file requested might be
> > helpful: how big it is? whether it changed on backend during
> > requests? if yes, is it changed atomically?
> >
> > Maxim Dounin
> >
> > _______________________________________________
> > nginx mailing list
> > nginx at nginx.org
> > http://mailman.nginx.org/mailman/listinfo/nginx
> >
> 
> 
> 
> -- 
> ????????????????


> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx

-------------- next part --------------
# HG changeset patch
# User Maxim Dounin <mdounin at mdounin.ru>
# Date 1315153882 0
# Node ID 16eaef9fbf7335a8ab02ce308799647b26bd11f9
# Parent  5d94f8b3e01d74ec6bd5bdcae176a8d3b998237d
Bugfix: open_file_cache lost is_directio flag.

On file retest open_file_cache lost is_directio if file wasn't changed.
This caused unaligned operations under Linux to fail with EINVAL.  It
wasn't noticiable with AIO though, as errors wasn't properly logged.

diff --git a/src/core/ngx_open_file_cache.c b/src/core/ngx_open_file_cache.c
--- a/src/core/ngx_open_file_cache.c
+++ b/src/core/ngx_open_file_cache.c
@@ -290,6 +290,8 @@ ngx_open_cached_file(ngx_open_file_cache
                     file->use_event = 1;
                 }
 
+                of->is_directio = file->is_directio;
+
                 goto renew;
             }
 
diff --git a/src/os/unix/ngx_file_aio_read.c b/src/os/unix/ngx_file_aio_read.c
--- a/src/os/unix/ngx_file_aio_read.c
+++ b/src/os/unix/ngx_file_aio_read.c
@@ -86,6 +86,9 @@ ngx_file_aio_read(ngx_file_t *file, u_ch
             return aio->nbytes;
         }
 
+        ngx_log_error(NGX_LOG_CRIT, file->log, ngx_errno,
+                      "aio read \"%s\" failed", file->name.data);
+
         return NGX_ERROR;
     }
 
diff --git a/src/os/unix/ngx_linux_aio_read.c b/src/os/unix/ngx_linux_aio_read.c
--- a/src/os/unix/ngx_linux_aio_read.c
+++ b/src/os/unix/ngx_linux_aio_read.c
@@ -74,6 +74,10 @@ ngx_file_aio_read(ngx_file_t *file, u_ch
         }
 
         ngx_set_errno(-aio->res);
+
+        ngx_log_error(NGX_LOG_CRIT, file->log, ngx_errno,
+                      "aio read \"%s\" failed", file->name.data);
+
         return NGX_ERROR;
     }
 

From nginx at nginxuser.net  Sun Sep  4 16:40:44 2011
From: nginx at nginxuser.net (Nginx User)
Date: Sun, 4 Sep 2011 19:40:44 +0300
Subject: http_auth_request_module Instructions?
In-Reply-To: <20110903222515.GP1137@mdounin.ru>
References: <CAN=WzV2e8_OVk24OyUcB=3JOXUWXfONNV-7b9VW-NmhGu5b4cw@mail.gmail.com>
	<20110903222515.GP1137@mdounin.ru>
Message-ID: <CAN=WzV3F=x3K6_L3SGzw6D8e-3gBuUz1oqyv7R+cFHJc=s0hiQ@mail.gmail.com>

On Sun, Sep 4, 2011 at 1:25 AM, Maxim Dounin <mdounin at mdounin.ru> wrote:

> Hello!
>
> On Sun, Sep 04, 2011 at 12:02:34AM +0300, Nginx User wrote:
>
> > Is the any documentation for the http_auth_request_module anywhere?
> Trying
> > to find out what the configuration parameters are if any.
>
> Try README in the module tarball.  Alternatively, you may find it
> here:
>


Hi Thanks for that.

To clarify, My understanding of how this works is that when a request from a
client (I'll call this "Client Request") hits Nginx, the module handler will
spin off a request (I'll call this "Module Request") to a location where I
would have arranged for authentication to occur. This can be auth basic etc
or my own custom process. Assuming my own custom process, I should arrange
for it to return status code "200" to allow access, status code "403" to
deny access or status code "401" to ask for username and
password (responding by 200, 403 or 401 as required). When the module
receives a "200" code for the Module Request, it will pass the Client
Request on through to the next normal stage of Nginx processing. If a "403"
code is received, The user is sent the same and processing stops.

Four queries:

1. Is my understanding of the process correct?
2. When the README says "it is not currently possible to use
proxy_cache/proxy_store (and fastcgi_cache/fastcgi_store) for requests
initiated by auth request module", does this apply to the Module Request
only as it suggests and that the Client Request will proceed as normal or is
there a twist to it?
3. Does the module cover "post" requests as well
4. I notice "proxy_set_header X-Original-URI $request_uri;" in the README
example. Is this a requirement?


Thanks again
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110904/1d44e637/attachment-0001.html>

From mdounin at mdounin.ru  Sun Sep  4 17:17:48 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Sun, 4 Sep 2011 21:17:48 +0400
Subject: =?UTF-8?Q?Re=3A_upstream_configuration_is_changed=EF=BC=9FThe_keyword_=28w?=
	=?UTF-8?Q?eight/max=5Ffail/fail=5Ftimeout=29_is_not_supported=3F?=
In-Reply-To: <CAP_76Mbvn6-MhH0Kmj2-jp0-mJDvYhggLBTuqP1BUUc95qwmjA@mail.gmail.com>
References: <CAP_76Mbvn6-MhH0Kmj2-jp0-mJDvYhggLBTuqP1BUUc95qwmjA@mail.gmail.com>
Message-ID: <20110904171748.GU1137@mdounin.ru>

Hello!

On Sun, Sep 04, 2011 at 11:51:31PM +0800, Geoge.Q wrote:

> Hi All?
> 
> I met a configuation issue. the configuration is as following:?NGINX is
> V0.7.68) and upstream is enabled when compiled.
> I also test it in V1.0.5, the issue happens again.
> 
> ----------------------------------------------------------------------------------------------------------------------------------------------
>                 server {
>                         listen 80;
>                         server_name aaa.com;
>                         server_name 192.168.2.140;
>                         location / {
>                                 proxy_pass http://haha;
>                                 proxy_redirect off;
>                                 proxy_buffering off;
>                                 proxy_set_header Host $host;
>                                 proxy_set_header X-Real-IP  $remote_addr;
>                                 proxy_set_header X-Forwarded-For
> $proxy_add_x_forwarded_for;
>                                 client_max_body_size 1000m;
>                                 client_body_buffer_size 256k;
>                                 proxy_connect_timeout 600;
>                                 proxy_temp_file_write_size  64k;
>                         }
>                 }
>                 upstream haha {
>                         server 1.1.1.2 max_fails=3  fail_timeout=30s;
> <---------max_fails/fail_timeouts/weight is not support ?
>                         server 1.1.1.1 weight=10;
>                 }
> ----------------------------------------------------------------------------------------------------------------------------------------------
> 
> When I run nginx, the error message print out as following:
> ./nginx
> 
> Root# /etc/init.d/nginx restart
> [emerg]: invalid parameter "max_fails=3" in
> /var/log/nginx/conf/nginx.conf:19

The problem is that you've tried to define upstream after it was 
already implicitly created by proxy_pass directive.  Move upstream 
definition before it's first use.

Maxim Dounin


From nginx-forum at nginx.us  Sun Sep  4 17:33:59 2011
From: nginx-forum at nginx.us (magicbear)
Date: Sun, 04 Sep 2011 13:33:59 -0400
Subject: upstream keepalive - call for testing
In-Reply-To: <20110801160725.GH1137@mdounin.ru>
References: <20110801160725.GH1137@mdounin.ru>
Message-ID: <0d5cc130a417bb62098701656fd44e42.NginxMailingListEnglish@forum.nginx.org>

Hello , when run for 5 days, today have 2 segment failure.

2011/09/04 12:00:47 [alert] 32147#0: worker process 8496 exited on
signal 11 (core dumped)
2011/09/04 12:17:32 [alert] 8497#0: ignore long locked inactive cache
entry 57c7c6c612a651727b880a1c9a2c7d2d, count:1
2011/09/04 12:21:31 [alert] 32147#0: worker process 19185 exited on
signal 11 (core dumped)




Core was generated by `nginx:'.
Program terminated with signal 11, Segmentation fault.
#0  ngx_http_upstream_handler (ev=0x7fc45735f8a8) at
src/http/ngx_http_upstream.c:915
915	    ctx->current_request = r;
(gdb) bt
#0  ngx_http_upstream_handler (ev=0x7fc45735f8a8) at
src/http/ngx_http_upstream.c:915
#1  0x0000000000419e95 in ngx_event_expire_timers () at
src/event/ngx_event_timer.c:149
#2  0x0000000000419ac7 in ngx_process_events_and_timers
(cycle=0x21f1990) at src/event/ngx_event.c:261
#3  0x000000000041f76e in ngx_worker_process_cycle (cycle=0x21f1990,
data=<value optimized out>) at src/os/unix/ngx_process_cycle.c:800
#4  0x000000000041e16b in ngx_spawn_process (cycle=0x21f1990,
proc=0x41f69d <ngx_worker_process_cycle>, data=0x0, name=0x463f3d
"worker process", respawn=7) at src/os/unix/ngx_process.c:196
#5  0x000000000042008d in ngx_reap_children (cycle=0x21f1990) at
src/os/unix/ngx_process_cycle.c:617
#6  ngx_master_process_cycle (cycle=0x21f1990) at
src/os/unix/ngx_process_cycle.c:180
#7  0x0000000000406061 in main (argc=<value optimized out>, argv=<value
optimized out>) at src/core/nginx.c:405





(gdb) bt full
#0  ngx_http_upstream_handler (ev=0x7fc45735f8a8) at
src/http/ngx_http_upstream.c:915
        c = 0x23b4160
        r = 0x22a2780
        ctx = <error reading variable ctx (Cannot access memory at
address 0x20)>
        u = 0x0
#1  0x0000000000419e95 in ngx_event_expire_timers () at
src/event/ngx_event_timer.c:149
        ev = 0x7fc45735f8a8
        root = 0x7fc4574ffa30
        sentinel = 0x689560
#2  0x0000000000419ac7 in ngx_process_events_and_timers
(cycle=0x21f1990) at src/event/ngx_event.c:261
        flags = 1
        timer = <value optimized out>
        delta = 21
#3  0x000000000041f76e in ngx_worker_process_cycle (cycle=0x21f1990,
data=<value optimized out>) at src/os/unix/ngx_process_cycle.c:800
        i = <value optimized out>
        c = <value optimized out>
#4  0x000000000041e16b in ngx_spawn_process (cycle=0x21f1990,
proc=0x41f69d <ngx_worker_process_cycle>, data=0x0, name=0x463f3d
"worker process", respawn=7) at src/os/unix/ngx_process.c:196
        on = 1
        pid = 0
        s = 7
#5  0x000000000042008d in ngx_reap_children (cycle=0x21f1990) at
src/os/unix/ngx_process_cycle.c:617
        i = <value optimized out>
        live = <value optimized out>
        n = <value optimized out>
        ch = {command = 2, pid = 8496, slot = 7, fd = -1}
        ccf = <value optimized out>
#6  ngx_master_process_cycle (cycle=0x21f1990) at
src/os/unix/ngx_process_cycle.c:180
        title = <value optimized out>
        p = <value optimized out>
        size = <value optimized out>
        i = <value optimized out>
        n = <value optimized out>
        sigio = 0
        set = {__val = {0 <repeats 16 times>}}
        itv = {it_interval = {tv_sec = 0, tv_usec = 35114632}, it_value
= {tv_sec = 0, tv_usec = 32146}}
        live = <value optimized out>
        delay = 0
        ls = <value optimized out>
        ccf = 0x21f2188
#7  0x0000000000406061 in main (argc=<value optimized out>, argv=<value
optimized out>) at src/core/nginx.c:405
        i = <value optimized out>
        log = 0x687380
        cycle = 0x217ce70
        init_cycle = {conf_ctx = 0x0, pool = 0x217c240, log = 0x687380,
new_log = {log_level = 0, file = 0x0, connection = 0, handler = 0, data
= 0x0, action = 0x0}, files = 0x0, 
          free_connections = 0x0, free_connection_n = 0,
reusable_connections_queue = {prev = 0x0, next = 0x0}, listening = {elts
= 0x0, nelts = 0, size = 0, nalloc = 0, pool = 0x0}, 
          pathes = {elts = 0x0, nelts = 0, size = 0, nalloc = 0, pool =
0x0}, open_files = {last = 0x0, part = {elts = 0x0, nelts = 0, next =
0x0}, size = 0, nalloc = 0, pool = 0x0}, 
          shared_memory = {last = 0x0, part = {elts = 0x0, nelts = 0,
next = 0x0}, size = 0, nalloc = 0, pool = 0x0}, connection_n = 0,
files_n = 0, connections = 0x0, read_events = 0x0, 
          write_events = 0x0, old_cycle = 0x0, conf_file = {len = 21,
data = 0x4606b6 "/etc/nginx/nginx.conf"}, conf_param = {len = 0, data =
0x0}, conf_prefix = {len = 11, 
            data = 0x4606b6 "/etc/nginx/nginx.conf"}, prefix = {len =
17, data = 0x4606a4 "/usr/local/nginx/"}, lock_file = {len = 0, data =
0x0}, hostname = {len = 0, data = 0x0}}
---Type <return> to continue, or q <return> to quit---
        ccf = <value optimized out>

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,213207,214739#msg-214739


From nginx at nginxuser.net  Sun Sep  4 17:38:04 2011
From: nginx at nginxuser.net (Nginx User)
Date: Sun, 4 Sep 2011 20:38:04 +0300
Subject: OpenResty Queries
Message-ID: <CAN=WzV2BqGZdAVzeo4_y=QEPyxEM9M-u1Cue7EbVmcSbRBmrHA@mail.gmail.com>

Some quick queries about some things missing from available documentation
about OpenResty

1. Notice current version is bundled with Nginx 1.0.4. Is it possible for an
end user to upgrade Nginx? I.E. can I download OpenResty and slip in Nginx
1.0.6?
2. The bundled modules are awesome but what is the process wants an
extra third party module?

Thanks
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110904/24e4bf0b/attachment.html>

From mdounin at mdounin.ru  Sun Sep  4 18:00:43 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Sun, 4 Sep 2011 22:00:43 +0400
Subject: http_auth_request_module Instructions?
In-Reply-To: <CAN=WzV3F=x3K6_L3SGzw6D8e-3gBuUz1oqyv7R+cFHJc=s0hiQ@mail.gmail.com>
References: <CAN=WzV2e8_OVk24OyUcB=3JOXUWXfONNV-7b9VW-NmhGu5b4cw@mail.gmail.com>
	<20110903222515.GP1137@mdounin.ru>
	<CAN=WzV3F=x3K6_L3SGzw6D8e-3gBuUz1oqyv7R+cFHJc=s0hiQ@mail.gmail.com>
Message-ID: <20110904180042.GV1137@mdounin.ru>

Hello!

On Sun, Sep 04, 2011 at 07:40:44PM +0300, Nginx User wrote:

> On Sun, Sep 4, 2011 at 1:25 AM, Maxim Dounin <mdounin at mdounin.ru> wrote:
> 
> > Hello!
> >
> > On Sun, Sep 04, 2011 at 12:02:34AM +0300, Nginx User wrote:
> >
> > > Is the any documentation for the http_auth_request_module anywhere?
> > Trying
> > > to find out what the configuration parameters are if any.
> >
> > Try README in the module tarball.  Alternatively, you may find it
> > here:
> >
> 
> 
> Hi Thanks for that.
> 
> To clarify, My understanding of how this works is that when a request from a
> client (I'll call this "Client Request") hits Nginx, the module handler will
> spin off a request (I'll call this "Module Request") to a location where I
> would have arranged for authentication to occur. This can be auth basic etc
> or my own custom process. Assuming my own custom process, I should arrange
> for it to return status code "200" to allow access, status code "403" to
> deny access or status code "401" to ask for username and
> password (responding by 200, 403 or 401 as required). When the module
> receives a "200" code for the Module Request, it will pass the Client
> Request on through to the next normal stage of Nginx processing. If a "403"
> code is received, The user is sent the same and processing stops.
> 
> Four queries:
> 
> 1. Is my understanding of the process correct?

Yes.

> 2. When the README says "it is not currently possible to use
> proxy_cache/proxy_store (and fastcgi_cache/fastcgi_store) for requests
> initiated by auth request module", does this apply to the Module Request
> only as it suggests and that the Client Request will proceed as normal or is
> there a twist to it?

It applies only to auth subrequests.  Client requests will proceed 
as usual.

> 3. Does the module cover "post" requests as well

Yes.  Note though, request body won't be read at auth_request (and 
won't be passed to auth subrequest).

> 4. I notice "proxy_set_header X-Original-URI $request_uri;" in the README
> example. Is this a requirement?

It's just an example how to pass original request uri to your 
custom auth script.

Note though, that

    proxy_pass_request_body off;
    proxy_set_header Content-Length "";

is actually required when proxy_pass'ing auth subrequests, as 
request body won't read (see above).

Maxim Dounin


From ggarron at alketech.com  Sun Sep  4 18:01:50 2011
From: ggarron at alketech.com (Guillermo Garron)
Date: Sun, 4 Sep 2011 14:01:50 -0400
Subject: Archives / or question
Message-ID: <5B05EBB5-52E1-42EC-9AA6-EC754649A4F7@alketech.com>

Hi,

I'm new to the list, and I do not want to bother you, as I'm sure this has been posted in past please tell me where to find the archives.

If for some reason someone has the answer here is my question:

I've this setup.

Gentoo Linux

Nginx with this USE flags:

www-servers/nginx  http http-cache ipv6 pcre vim-syntax -aio -debug -libatomic -ssl nginx_modules_http_access nginx_modules_http_auth_basic anginx_modules_http_utoindex nginx_modules_http_browser nginx_modules_http_charset nginx_modules_http_empty_gif -nginx_modules_http_fastcgi nginx_modules_http_geo nginx_modules_http_gzip nginx_modules_http_limit_req nginx_modules_http_limit_zone map -nginx_modules_http_memcached nginx_modules_http_proxy nginx_modules_http_referer nginx_modules_http_rewrite -nginx_modules_http_scgi nginx_modules_http_split_clients -nginx_modules_http_ssi nginx_modules_http_upstream_ip_hash nginx_modules_http_userid nginx_modules_http_uwsgi -nginx_modules_http_addition -nginx_modules_http_cache_purge -nginx_modules_http_dav -nginx_modules_http_degradation -nginx_modules_http_ey_balancer -nginx_modules_http_flv -nginx_modules_http_geoip nginx_modules_http_gzip_static -nginx_modules_http_headers_more -nginx_modules_http_image_filter -nginx_modules_http_passenger -nginx_modules_http_perl -nginx_modules_http_push -nginx_modules_http_random_index -nginx_modules_http_realip -nginx_modules_http_secure_link -nginx_modules_http_slowfs_cache -nginx_modules_http_stub_status -nginx_modules_http_sub -nginx_modules_http_upload -nginx_modules_http_xslt -nginx_modules_mail_imap -nginx_modules_mail_pop3 -nginx_modules_mail_smtp

Serving only static files (My blog is Jekyll generated, so only html/css/png files).

Now I want to know how to achieve the best possible performance, leave my server as it is? or implement (on the same machine) Varnish in front of Nginx to serve from RAM instead of disk.

I am using a Linode server with 512 Megs or RAM.

thanks for your help, either answering me or pointing me in the right direction to find an answer.

best regards.


Guillermo Garron





From mdounin at mdounin.ru  Sun Sep  4 18:21:37 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Sun, 4 Sep 2011 22:21:37 +0400
Subject: upstream keepalive - call for testing
In-Reply-To: <0d5cc130a417bb62098701656fd44e42.NginxMailingListEnglish@forum.nginx.org>
References: <20110801160725.GH1137@mdounin.ru>
	<0d5cc130a417bb62098701656fd44e42.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20110904182137.GW1137@mdounin.ru>

Hello!

On Sun, Sep 04, 2011 at 01:33:59PM -0400, magicbear wrote:

> Hello , when run for 5 days, today have 2 segment failure.
> 
> 2011/09/04 12:00:47 [alert] 32147#0: worker process 8496 exited on
> signal 11 (core dumped)
> 2011/09/04 12:17:32 [alert] 8497#0: ignore long locked inactive cache
> entry 57c7c6c612a651727b880a1c9a2c7d2d, count:1
> 2011/09/04 12:21:31 [alert] 32147#0: worker process 19185 exited on
> signal 11 (core dumped)
> 
> 
> 
> 
> Core was generated by `nginx:'.
> Program terminated with signal 11, Segmentation fault.
> #0  ngx_http_upstream_handler (ev=0x7fc45735f8a8) at
> src/http/ngx_http_upstream.c:915
> 915	    ctx->current_request = r;
> (gdb) bt
> #0  ngx_http_upstream_handler (ev=0x7fc45735f8a8) at
> src/http/ngx_http_upstream.c:915
> #1  0x0000000000419e95 in ngx_event_expire_timers () at
> src/event/ngx_event_timer.c:149
> #2  0x0000000000419ac7 in ngx_process_events_and_timers
> (cycle=0x21f1990) at src/event/ngx_event.c:261

Hm, this looks identical to one you've reported previously.  Do 
you have patch from

http://mailman.nginx.org/pipermail/nginx/2011-August/028809.html

applied?

> #3  0x000000000041f76e in ngx_worker_process_cycle (cycle=0x21f1990,
> data=<value optimized out>) at src/os/unix/ngx_process_cycle.c:800
> #4  0x000000000041e16b in ngx_spawn_process (cycle=0x21f1990,
> proc=0x41f69d <ngx_worker_process_cycle>, data=0x0, name=0x463f3d
> "worker process", respawn=7) at src/os/unix/ngx_process.c:196
> #5  0x000000000042008d in ngx_reap_children (cycle=0x21f1990) at
> src/os/unix/ngx_process_cycle.c:617
> #6  ngx_master_process_cycle (cycle=0x21f1990) at
> src/os/unix/ngx_process_cycle.c:180
> #7  0x0000000000406061 in main (argc=<value optimized out>, argv=<value
> optimized out>) at src/core/nginx.c:405
> 
> 
> (gdb) bt full
> #0  ngx_http_upstream_handler (ev=0x7fc45735f8a8) at
> src/http/ngx_http_upstream.c:915
>         c = 0x23b4160
>         r = 0x22a2780
>         ctx = <error reading variable ctx (Cannot access memory at
> address 0x20)>
>         u = 0x0

Could you please show results of the following gdb commands:

fr 0
p *ev
p *((ngx_connection_t *) ev->data)


Maxim Dounin


From magicbearmo at gmail.com  Sun Sep  4 18:31:54 2011
From: magicbearmo at gmail.com (MagicBear)
Date: Mon, 5 Sep 2011 02:31:54 +0800
Subject: upstream keepalive - call for testing
In-Reply-To: <20110904182137.GW1137@mdounin.ru>
References: <20110801160725.GH1137@mdounin.ru>
	<0d5cc130a417bb62098701656fd44e42.NginxMailingListEnglish@forum.nginx.org>
	<20110904182137.GW1137@mdounin.ru>
Message-ID: <CAENo=s3G3XBrYN7K9G6HzHZkBWYHEEi4DLCH8G-7_yC8dJyu-w@mail.gmail.com>

(gdb) fr 0
#0  ngx_http_upstream_handler (ev=0x7fc45735f8a8)
    at src/http/ngx_http_upstream.c:915
915    ctx->current_request = r;
(gdb) p *ev
$1 = {data = 0x7fc4576aa750, write = 1, accept = 0, instance = 1, active =
1,
  disabled = 0, ready = 1, oneshot = 0, complete = 0, eof = 0, error = 0,
  timedout = 1, timer_set = 0, delayed = 0, read_discarded = 0,
  unexpected_eof = 0, deferred_accept = 0, pending_eof = 0, posted_ready =
0,
  available = 0, handler = 0x43bf9c <ngx_http_upstream_handler>,
  index = 3503345872, log = 0x21f1770, timer = {key = 0, left = 0x0,
    right = 0x0, parent = 0x0, color = 0 '\000', data = 0 '\000'},
  closed = 0, channel = 0, resolver = 0, next = 0x0, prev = 0x0}
(gdb) p *((ngx_connection_t *) ev->data)
$2 = {data = 0x22a2780, read = 0x7fc4575008a8, write = 0x7fc45735f8a8,
  fd = 66, recv = 0x41cc50 <ngx_unix_recv>, send = 0x41cf08
<ngx_unix_send>,
  recv_chain = 0x41cd00 <ngx_readv_chain>,
  send_chain = 0x421050 <ngx_linux_sendfile_chain>, listening = 0x0,
  sent = 0, log = 0x21f1770, pool = 0x237ff70, sockaddr = 0x0, socklen = 0,
  addr_text = {len = 0, data = 0x0}, ssl = 0x0, local_sockaddr = 0x0,
  buffer = 0x0, queue = {prev = 0x0, next = 0x0}, number = 63713716,
  requests = 0, buffered = 0, log_error = 1, single_connection = 0,
  unexpected_eof = 0, timedout = 0, error = 0, destroyed = 0, idle = 0,
  reusable = 0, close = 0, sendfile = 1, sndlowat = 0, tcp_nodelay = 0,
  tcp_nopush = 0}


2011/9/5 Maxim Dounin <mdounin at mdounin.ru>

> Hello!
>
> On Sun, Sep 04, 2011 at 01:33:59PM -0400, magicbear wrote:
>
> > Hello , when run for 5 days, today have 2 segment failure.
> >
> > 2011/09/04 12:00:47 [alert] 32147#0: worker process 8496 exited on
> > signal 11 (core dumped)
> > 2011/09/04 12:17:32 [alert] 8497#0: ignore long locked inactive cache
> > entry 57c7c6c612a651727b880a1c9a2c7d2d, count:1
> > 2011/09/04 12:21:31 [alert] 32147#0: worker process 19185 exited on
> > signal 11 (core dumped)
> >
> >
> >
> >
> > Core was generated by `nginx:'.
> > Program terminated with signal 11, Segmentation fault.
> > #0  ngx_http_upstream_handler (ev=0x7fc45735f8a8) at
> > src/http/ngx_http_upstream.c:915
> > 915       ctx->current_request = r;
> > (gdb) bt
> > #0  ngx_http_upstream_handler (ev=0x7fc45735f8a8) at
> > src/http/ngx_http_upstream.c:915
> > #1  0x0000000000419e95 in ngx_event_expire_timers () at
> > src/event/ngx_event_timer.c:149
> > #2  0x0000000000419ac7 in ngx_process_events_and_timers
> > (cycle=0x21f1990) at src/event/ngx_event.c:261
>
> Hm, this looks identical to one you've reported previously.  Do
> you have patch from
>
> http://mailman.nginx.org/pipermail/nginx/2011-August/028809.html
>
> applied?
>
> > #3  0x000000000041f76e in ngx_worker_process_cycle (cycle=0x21f1990,
> > data=<value optimized out>) at src/os/unix/ngx_process_cycle.c:800
> > #4  0x000000000041e16b in ngx_spawn_process (cycle=0x21f1990,
> > proc=0x41f69d <ngx_worker_process_cycle>, data=0x0, name=0x463f3d
> > "worker process", respawn=7) at src/os/unix/ngx_process.c:196
> > #5  0x000000000042008d in ngx_reap_children (cycle=0x21f1990) at
> > src/os/unix/ngx_process_cycle.c:617
> > #6  ngx_master_process_cycle (cycle=0x21f1990) at
> > src/os/unix/ngx_process_cycle.c:180
> > #7  0x0000000000406061 in main (argc=<value optimized out>, argv=<value
> > optimized out>) at src/core/nginx.c:405
> >
> >
> > (gdb) bt full
> > #0  ngx_http_upstream_handler (ev=0x7fc45735f8a8) at
> > src/http/ngx_http_upstream.c:915
> >         c = 0x23b4160
> >         r = 0x22a2780
> >         ctx = <error reading variable ctx (Cannot access memory at
> > address 0x20)>
> >         u = 0x0
>
> Could you please show results of the following gdb commands:
>
> fr 0
> p *ev
> p *((ngx_connection_t *) ev->data)
>
>
> Maxim Dounin
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110905/fed3533b/attachment-0001.html>

From nginx-forum at nginx.us  Sun Sep  4 18:37:17 2011
From: nginx-forum at nginx.us (loganbest)
Date: Sun, 04 Sep 2011 14:37:17 -0400
Subject: Can't fix "No input file specified." with php-fastcgi
Message-ID: <acb3097a23a38c8bd741974b9ba5c158.NginxMailingListEnglish@forum.nginx.org>

I've been trying to fix this pesky error for a couple days and no forums
or anything is suggesting anything that is helping. 

I simply just need someone to login to my EC2 instance and figure out
what's going on. I've tried reinstalling php to all the default configs
and everything but nothing is working.

I don't have any money to pay for this right now but you can invoice my
paypal and I will get to it as soon as I have the money to give.

My timeline for this is pretty quick as I have something that needs to
be launched on tuesday.

Please let me know if someone can fix this asap. Thanks in advance.

Logan

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,214746,214746#msg-214746


From ggarron at alketech.com  Sun Sep  4 18:40:45 2011
From: ggarron at alketech.com (Guillermo Garron)
Date: Sun, 4 Sep 2011 14:40:45 -0400
Subject: Can't fix "No input file specified." with php-fastcgi
In-Reply-To: <acb3097a23a38c8bd741974b9ba5c158.NginxMailingListEnglish@forum.nginx.org>
References: <acb3097a23a38c8bd741974b9ba5c158.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <CAHMf-MvyL6KN--U2zaZ+i8RGKxXF6kFkxsCc+DaUGoxJ9QsFkw@mail.gmail.com>

Do you really need it to be fastcgi now?
I mean you can start with Apache behind nginx and switch to fastcgi later,
after your launch.

Regards

Guillermo.

On Sep 4, 2011 2:37 PM, "loganbest" <nginx-forum at nginx.us> wrote:

I've been trying to fix this pesky error for a couple days and no forums
or anything is suggesting anything that is helping.

I simply just need someone to login to my EC2 instance and figure out
what's going on. I've tried reinstalling php to all the default configs
and everything but nothing is working.

I don't have any money to pay for this right now but you can invoice my
paypal and I will get to it as soon as I have the money to give.

My timeline for this is pretty quick as I have something that needs to
be launched on tuesday.

Please let me know if someone can fix this asap. Thanks in advance.

Logan

Posted at Nginx Forum:
http://forum.nginx.org/read.php?2,214746,214746#msg-214746

_______________________________________________
nginx mailing list
nginx at nginx.org
http://mailman.nginx.org/mailman/listinfo/nginx
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110904/96501cbb/attachment.html>

From nginx-forum at nginx.us  Sun Sep  4 18:46:07 2011
From: nginx-forum at nginx.us (loganbest)
Date: Sun, 04 Sep 2011 14:46:07 -0400
Subject: Can't fix "No input file specified." with php-fastcgi
In-Reply-To: <CAHMf-MvyL6KN--U2zaZ+i8RGKxXF6kFkxsCc+DaUGoxJ9QsFkw@mail.gmail.com>
References: <CAHMf-MvyL6KN--U2zaZ+i8RGKxXF6kFkxsCc+DaUGoxJ9QsFkw@mail.gmail.com>
Message-ID: <83b7a11505d9b23a1afd7deff640edf1.NginxMailingListEnglish@forum.nginx.org>

I'm not sure what the standard practices of nginx+php are. This is my
first time working with nginx and I didn't install nginx, I just
installed fastcgi based on an article I found at the top of google. I've
seen php-fpm and php-fastcgi and I don't know the differences or
pro/cons of each. I'm really just looking for some quick help to get
this running. I'd also rather not have both nginx and apache running on
the same server since this isn't a very large ec2 instance I can't take
the memory footprint of apache.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,214746,214748#msg-214748


From ggarron at alketech.com  Sun Sep  4 19:03:29 2011
From: ggarron at alketech.com (Guillermo Garron)
Date: Sun, 4 Sep 2011 15:03:29 -0400
Subject: Can't fix "No input file specified." with php-fastcgi
In-Reply-To: <CAHMf-MsRPrtA-4uPveGQffry6sKoWNdSi40xzGRnri-_gDhvxg@mail.gmail.com>
References: <CAHMf-MvyL6KN--U2zaZ+i8RGKxXF6kFkxsCc+DaUGoxJ9QsFkw@mail.gmail.com>
	<83b7a11505d9b23a1afd7deff640edf1.NginxMailingListEnglish@forum.nginx.org>
	<CAHMf-MsRPrtA-4uPveGQffry6sKoWNdSi40xzGRnri-_gDhvxg@mail.gmail.com>
Message-ID: <CAHMf-Mvyn2EPeVBmq_x-bd4=QPt__UucwN5-HUZp8t13ktRD2w@mail.gmail.com>

Well
I've only worked in the past with nginx serving static files and sending php
to apache behind.

If you decide to go that way or not find help in the current scenario, I can
help you then.

Let me know, I'll be checking my blackberry, in 5 hours I will be available

On Sep 4, 2011 2:46 PM, "loganbest" <nginx-forum at nginx.us> wrote:

I'm not sure what the standard practices of nginx+php are. This is my
first time working with nginx and I didn't install nginx, I just
installed fastcgi based on an article I found at the top of google. I've
seen php-fpm and php-fastcgi and I don't know the differences or
pro/cons of each. I'm really just looking for some quick help to get
this running. I'd also rather not have both nginx and apache running on
the same server since this isn't a very large ec2 instance I can't take
the memory footprint of apache.

Posted at Nginx Forum:
http://forum.nginx.org/read.php?2,214746,214748#msg-214748


_______________________________________________
nginx mailing list
nginx at nginx.org
http://mailman.n...
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110904/e45c0f50/attachment.html>

From nginx at nginxuser.net  Sun Sep  4 19:08:30 2011
From: nginx at nginxuser.net (Nginx User)
Date: Sun, 4 Sep 2011 22:08:30 +0300
Subject: http_auth_request_module Instructions?
In-Reply-To: <20110904180042.GV1137@mdounin.ru>
References: <CAN=WzV2e8_OVk24OyUcB=3JOXUWXfONNV-7b9VW-NmhGu5b4cw@mail.gmail.com>
	<20110903222515.GP1137@mdounin.ru>
	<CAN=WzV3F=x3K6_L3SGzw6D8e-3gBuUz1oqyv7R+cFHJc=s0hiQ@mail.gmail.com>
	<20110904180042.GV1137@mdounin.ru>
Message-ID: <CAN=WzV18TiODF8r_5m6WW17V9JtFnBpcPcqzGbvc_Gq0YaZb4Q@mail.gmail.com>

On Sun, Sep 4, 2011 at 9:00 PM, Maxim Dounin <mdounin at mdounin.ru> wrote:

> Hello!
>
> On Sun, Sep 04, 2011 at 07:40:44PM +0300, Nginx User wrote:
>
> > On Sun, Sep 4, 2011 at 1:25 AM, Maxim Dounin <mdounin at mdounin.ru> wrote:
> >
> > > Hello!
> > >
> > > On Sun, Sep 04, 2011 at 12:02:34AM +0300, Nginx User wrote:
> > >
> > > > Is the any documentation for the http_auth_request_module anywhere?
> > > Trying
> > > > to find out what the configuration parameters are if any.
> > >
> > > Try README in the module tarball.  Alternatively, you may find it
> > > here:
> > >
> >
> >
> > Hi Thanks for that.
> >
> > To clarify, My understanding of how this works is that when a request
> from a
> > client (I'll call this "Client Request") hits Nginx, the module handler
> will
> > spin off a request (I'll call this "Module Request") to a location where
> I
> > would have arranged for authentication to occur. This can be auth basic
> etc
> > or my own custom process. Assuming my own custom process, I should
> arrange
> > for it to return status code "200" to allow access, status code "403" to
> > deny access or status code "401" to ask for username and
> > password (responding by 200, 403 or 401 as required). When the module
> > receives a "200" code for the Module Request, it will pass the Client
> > Request on through to the next normal stage of Nginx processing. If a
> "403"
> > code is received, The user is sent the same and processing stops.
> >
> > Four queries:
> >
> > 1. Is my understanding of the process correct?
>
> Yes.
>
> > 2. When the README says "it is not currently possible to use
> > proxy_cache/proxy_store (and fastcgi_cache/fastcgi_store) for requests
> > initiated by auth request module", does this apply to the Module Request
> > only as it suggests and that the Client Request will proceed as normal or
> is
> > there a twist to it?
>
> It applies only to auth subrequests.  Client requests will proceed
> as usual.
>
> > 3. Does the module cover "post" requests as well
>
> Yes.  Note though, request body won't be read at auth_request (and
> won't be passed to auth subrequest).
>
> > 4. I notice "proxy_set_header X-Original-URI $request_uri;" in the README
> > example. Is this a requirement?
>
> It's just an example how to pass original request uri to your
> custom auth script.
>
> Note though, that
>
>    proxy_pass_request_body off;
>    proxy_set_header Content-Length "";
>
> is actually required when proxy_pass'ing auth subrequests, as
> request body won't read (see above).


Brilliant.

I think a bigger deal should be made of this module as it makes a nonsense
of the "lack of a WAF" concern about Nginx. I'll suggest pushing it to the
core as many are not aware of it. I think the low profile is also because it
is not documented on the wiki. I had clicked on the link in the early days
but on being taken to the sparse website with a few log items, I beat a
hurried retreat.

Since you have Igor's ear, I'll suggest you whisper the following into it:

"Hey ?????! ? ????? ??? ?? ?????? ??????? ???? ?????? ???????? ?? ?????????.
??? ??????? ????? ??? FUD ?????? Nginx ? ????? ?????? WAF ? ?????? ?????
????."

On second thoughts, don't ... He might think you have had too much vodka
since that is machine translated.


Last query on this.

The information on the "satisfy" directive in the wiki states that:

* all - Both Access *and* Auth Basic directives must grant access to the
context
* any - Either Access *or* Auth Basic directives grant access to the context
Is this miss leading? Should it be something like:

* all - All access phase handlers must grant access to the context
* any - Any access phase handler can grant access to the context

Thanks
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110904/bad62225/attachment.html>

From ano at bestmx.ru  Sun Sep  4 19:33:51 2011
From: ano at bestmx.ru (Andrey N. Oktyabrski)
Date: Sun, 04 Sep 2011 23:33:51 +0400
Subject: http_auth_request_module Instructions?
In-Reply-To: <CAN=WzV18TiODF8r_5m6WW17V9JtFnBpcPcqzGbvc_Gq0YaZb4Q@mail.gmail.com>
References: <CAN=WzV2e8_OVk24OyUcB=3JOXUWXfONNV-7b9VW-NmhGu5b4cw@mail.gmail.com>
	<20110903222515.GP1137@mdounin.ru>
	<CAN=WzV3F=x3K6_L3SGzw6D8e-3gBuUz1oqyv7R+cFHJc=s0hiQ@mail.gmail.com>
	<20110904180042.GV1137@mdounin.ru>
	<CAN=WzV18TiODF8r_5m6WW17V9JtFnBpcPcqzGbvc_Gq0YaZb4Q@mail.gmail.com>
Message-ID: <4E63D29F.1030209@bestmx.ru>

On 04.09.11 23:08, Nginx User wrote:
> "Hey ?????! ? ????? ??? ?? ?????? ??????? ???? ?????? ???????? ??
> ?????????. ??? ??????? ????? ??? FUD ?????? Nginx ? ????? ?????? WAF ?
> ?????? ????? ????."
Excuse me, your english much better than russian, please use it :-)


From nginx at nginxuser.net  Sun Sep  4 19:50:04 2011
From: nginx at nginxuser.net (Nginx User)
Date: Sun, 4 Sep 2011 22:50:04 +0300
Subject: http_auth_request_module Instructions?
In-Reply-To: <4E63D29F.1030209@bestmx.ru>
References: <CAN=WzV2e8_OVk24OyUcB=3JOXUWXfONNV-7b9VW-NmhGu5b4cw@mail.gmail.com>
	<20110903222515.GP1137@mdounin.ru>
	<CAN=WzV3F=x3K6_L3SGzw6D8e-3gBuUz1oqyv7R+cFHJc=s0hiQ@mail.gmail.com>
	<20110904180042.GV1137@mdounin.ru>
	<CAN=WzV18TiODF8r_5m6WW17V9JtFnBpcPcqzGbvc_Gq0YaZb4Q@mail.gmail.com>
	<4E63D29F.1030209@bestmx.ru>
Message-ID: <CAN=WzV3EqjWkAeNn+hH+Mgj8v6+n5ZGfWPBDjUBq3UL9-trpXg@mail.gmail.com>

2011/9/4 Andrey N. Oktyabrski <ano at bestmx.ru>

> On 04.09.11 23:08, Nginx User wrote:
>
>> "Hey ?????! ? ????? ??? ?? ?????? ??????? ???? ?????? ???????? ??
>> ?????????. ??? ??????? ????? ??? FUD ?????? Nginx ? ????? ?????? WAF ?
>> ?????? ????? ????."
>>
> Excuse me, your english much better than russian, please use it :-)


You obviously type faster than you read since you managed to send your email
before you finished reading the next line which says ...

On second thoughts, don't ... He might think you have had too much vodka
> since that is machine translated.


You should learn to read a bit faster but we digress  :-)
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110904/a101270d/attachment-0001.html>

From nginx-forum at nginx.us  Sun Sep  4 20:24:00 2011
From: nginx-forum at nginx.us (loganbest)
Date: Sun, 04 Sep 2011 16:24:00 -0400
Subject: Can't fix "No input file specified." with php-fastcgi
In-Reply-To: <83b7a11505d9b23a1afd7deff640edf1.NginxMailingListEnglish@forum.nginx.org>
References: <CAHMf-MvyL6KN--U2zaZ+i8RGKxXF6kFkxsCc+DaUGoxJ9QsFkw@mail.gmail.com>
	<83b7a11505d9b23a1afd7deff640edf1.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <f830a61564502bd1d97cf8c959bf2de6.NginxMailingListEnglish@forum.nginx.org>

Somehow, between purging all php configs and packages and reinstalling
them and altering the php location definition in the nginx.conf it now
works as it did before.... I"m confused but I'm not going to touch it
now.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,214746,214754#msg-214754


From nginx-forum at nginx.us  Sun Sep  4 21:14:27 2011
From: nginx-forum at nginx.us (loganbest)
Date: Sun, 04 Sep 2011 17:14:27 -0400
Subject: Can't fix "No input file specified." with php-fastcgi
In-Reply-To: <acb3097a23a38c8bd741974b9ba5c158.NginxMailingListEnglish@forum.nginx.org>
References: <acb3097a23a38c8bd741974b9ba5c158.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <369c4c70e6b9f1a77e70a62ea93d8aa0.NginxMailingListEnglish@forum.nginx.org>

I take that back.... it started again.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,214746,214756#msg-214756


From ggarron at alketech.com  Sun Sep  4 21:19:03 2011
From: ggarron at alketech.com (Guillermo Garron)
Date: Sun, 4 Sep 2011 17:19:03 -0400
Subject: Can't fix "No input file specified." with php-fastcgi
Message-ID: <2a1a1ceedd33352155268607634ce01f@mail.gmail.com>

What aplication are you using? I mean with php?
------Original Message------
From: loganbest
To: nginx at nginx.org
ReplyTo: nginx at nginx.org
Subject: Re: Can't fix "No input file specified." with php-fastcgi
Sent: Sep 4, 2011 4:24 PM

Somehow, between purging all php configs and packages and reinstalling
them and altering the php location definition in the nginx.conf it now
works as it did before.... I"m confused but I'm not going to touch it
now.

Posted at Nginx Forum:
http://forum.nginx.org/read.php?2,214746,214754#msg-214754

_______________________________________________
nginx mailing list
nginx at nginx.org
http://mailman.nginx.org/mailman/listinfo/nginx


From nginx-forum at nginx.us  Sun Sep  4 21:21:21 2011
From: nginx-forum at nginx.us (loganbest)
Date: Sun, 04 Sep 2011 17:21:21 -0400
Subject: Can't fix "No input file specified." with php-fastcgi
In-Reply-To: <2a1a1ceedd33352155268607634ce01f@mail.gmail.com>
References: <2a1a1ceedd33352155268607634ce01f@mail.gmail.com>
Message-ID: <110b6d83f8fc0d1ab9b8cd761506b946.NginxMailingListEnglish@forum.nginx.org>

I only have 2 php files at this time because the rest of the server is
using ruby. It's just standard php and not even very much php.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,214746,214758#msg-214758


From mdounin at mdounin.ru  Sun Sep  4 22:01:25 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Mon, 5 Sep 2011 02:01:25 +0400
Subject: http_auth_request_module Instructions?
In-Reply-To: <CAN=WzV18TiODF8r_5m6WW17V9JtFnBpcPcqzGbvc_Gq0YaZb4Q@mail.gmail.com>
References: <CAN=WzV2e8_OVk24OyUcB=3JOXUWXfONNV-7b9VW-NmhGu5b4cw@mail.gmail.com>
	<20110903222515.GP1137@mdounin.ru>
	<CAN=WzV3F=x3K6_L3SGzw6D8e-3gBuUz1oqyv7R+cFHJc=s0hiQ@mail.gmail.com>
	<20110904180042.GV1137@mdounin.ru>
	<CAN=WzV18TiODF8r_5m6WW17V9JtFnBpcPcqzGbvc_Gq0YaZb4Q@mail.gmail.com>
Message-ID: <20110904220125.GX1137@mdounin.ru>

Hello!

On Sun, Sep 04, 2011 at 10:08:30PM +0300, Nginx User wrote:

[...]

> Last query on this.
> 
> The information on the "satisfy" directive in the wiki states that:
> 
> * all - Both Access *and* Auth Basic directives must grant access to the
> context
> * any - Either Access *or* Auth Basic directives grant access to the context
> Is this miss leading? Should it be something like:
> 
> * all - All access phase handlers must grant access to the context
> * any - Any access phase handler can grant access to the context

Both versions are right, though first one implicitly assumes you 
are using only standard access modules.

Maxim Dounin


From nginx at nginxuser.net  Sun Sep  4 22:36:09 2011
From: nginx at nginxuser.net (Nginx User)
Date: Mon, 5 Sep 2011 01:36:09 +0300
Subject: http_auth_request_module Instructions?
In-Reply-To: <20110904220125.GX1137@mdounin.ru>
References: <CAN=WzV2e8_OVk24OyUcB=3JOXUWXfONNV-7b9VW-NmhGu5b4cw@mail.gmail.com>
	<20110903222515.GP1137@mdounin.ru>
	<CAN=WzV3F=x3K6_L3SGzw6D8e-3gBuUz1oqyv7R+cFHJc=s0hiQ@mail.gmail.com>
	<20110904180042.GV1137@mdounin.ru>
	<CAN=WzV18TiODF8r_5m6WW17V9JtFnBpcPcqzGbvc_Gq0YaZb4Q@mail.gmail.com>
	<20110904220125.GX1137@mdounin.ru>
Message-ID: <CAN=WzV0PG7prqRUjqHbbUKJikhLaeOcFMDYc5c3ELdq6FPsYOA@mail.gmail.com>

On Mon, Sep 5, 2011 at 1:01 AM, Maxim Dounin <mdounin at mdounin.ru> wrote:

> Hello!
>
> On Sun, Sep 04, 2011 at 10:08:30PM +0300, Nginx User wrote:
>
> [...]
>
> > Last query on this.
> >
> > The information on the "satisfy" directive in the wiki states that:
> >
> > * all - Both Access *and* Auth Basic directives must grant access to the
> > context
> > * any - Either Access *or* Auth Basic directives grant access to the
> context
> > Is this miss leading? Should it be something like:
> >
> > * all - All access phase handlers must grant access to the context
> > * any - Any access phase handler can grant access to the context
>
> Both versions are right, though first one implicitly assumes you
> are using only standard access modules.



I'll say the first set is technically wrong as it *implies* the directive
only applies to those two modules. The key is in the use of the words "both"
and "either" which imply a sum total of two possibilities.

I edited the wiki to read:

Access policy if directives from multiple access phase handlers, such as the
Access and Auth Basic modules, are defined in a context:
* all - All access phase handlers must grant access to the context
* any - Any access phase handler can grant access to the context

Thanks!
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110905/82f352b6/attachment.html>

From nginx-forum at nginx.us  Sun Sep  4 22:46:18 2011
From: nginx-forum at nginx.us (Mickie D)
Date: Sun, 04 Sep 2011 18:46:18 -0400
Subject: Help with rewrite
Message-ID: <461adae45a9d3ba0d9d4ed211a89df13.NginxMailingListEnglish@forum.nginx.org>

Hi everyone,

I am using a plugin for my site called VBSEO

They use the following code:

--------------------------------------------------------
 if ($request_filename ~ "\.php$" ) {
 rewrite ^/(.*)$ /vbseo.php break;
 }
 if (!-e $request_filename) {
 rewrite ^/(.*)$ /vbseo.php break;
 }
--------------------------------------------------------

I read the article about IF IS EVIL and thought it would be better to
make this code better.

Could someone who knows about the rewrites and nginx coding please help
me.

I will post the results (if it's possible of course) to VBSEO so they
can update there rules.

Kind Regards
Mick

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,214761,214761#msg-214761


From ianevans at digitalhit.com  Sun Sep  4 23:46:56 2011
From: ianevans at digitalhit.com (Ian Evans)
Date: Sun, 04 Sep 2011 19:46:56 -0400
Subject: PositiveSSL certificate keeps giving the untrusted issuer warnings in
	browser
Message-ID: <4E640DF0.1020000@digitalhit.com>

Since I'm working on a Facebook app, the self-signed certificate I 
created for SSL access to my admin area is no longer useful.

I grabbed a PositiveSSL cert from Comodo, but after following the 
instructions here ( http://kbeezie.com/view/free-ssl-with-nginx/2/ ) but 
still get the untrusted or self-signed warnings.

This page 
(http://www.digicert.com/ssl-certificate-installation-nginx.htm) seems 
to suggest I have to concatenate several crt files together.

This is what Comodo sent me:

AddTrustExternalCARoot.crt
PositiveSSLCA.crt
UTNAddTrustServerCA.crt and
www_example_com.crt

Do I need to concatenate all of these?

Also, just curious why do some example sites I've seen say Ngix needs a 
pem file and others say a crt?

Thanks!


From ianevans at digitalhit.com  Mon Sep  5 00:04:30 2011
From: ianevans at digitalhit.com (Ian Evans)
Date: Sun, 04 Sep 2011 20:04:30 -0400
Subject: PositiveSSL certificate keeps giving the untrusted issuer
	warnings in browser
In-Reply-To: <4E640DF0.1020000@digitalhit.com>
References: <4E640DF0.1020000@digitalhit.com>
Message-ID: <4E64120E.3090304@digitalhit.com>

On 04/09/2011 7:46 PM, Ian Evans wrote:
> Since I'm working on a Facebook app, the self-signed certificate I
> created for SSL access to my admin area is no longer useful.
>
> I grabbed a PositiveSSL cert from Comodo, but after following the
> instructions here ( http://kbeezie.com/view/free-ssl-with-nginx/2/ ) but
> still get the untrusted or self-signed warnings.

Upon further testing I see that it's still using the old cert despite 
issuing a kill -HUP.

How do I get nginx to start using the new cert?


From quan.nexthop at gmail.com  Mon Sep  5 00:35:58 2011
From: quan.nexthop at gmail.com (Geoge.Q)
Date: Mon, 5 Sep 2011 08:35:58 +0800
Subject: =?UTF-8?Q?Re=3A_upstream_configuration_is_changed=EF=BC=9FThe_keyword_=28w?=
	=?UTF-8?Q?eight/max=5Ffail/fail=5Ftimeout=29_is_not_supported=3F?=
In-Reply-To: <20110904171748.GU1137@mdounin.ru>
References: <CAP_76Mbvn6-MhH0Kmj2-jp0-mJDvYhggLBTuqP1BUUc95qwmjA@mail.gmail.com>
	<20110904171748.GU1137@mdounin.ru>
Message-ID: <CAP_76MYK3TOBvFzogaFxRWeTpB4q3RxwHWbhDOt3gygPbi7eTw@mail.gmail.com>

Thanks Max, it works now.

George

On Mon, Sep 5, 2011 at 1:17 AM, Maxim Dounin <mdounin at mdounin.ru> wrote:

> Hello!
>
> On Sun, Sep 04, 2011 at 11:51:31PM +0800, Geoge.Q wrote:
>
> > Hi All?
> >
> > I met a configuation issue. the configuration is as following:?NGINX is
> > V0.7.68) and upstream is enabled when compiled.
> > I also test it in V1.0.5, the issue happens again.
> >
> >
> ----------------------------------------------------------------------------------------------------------------------------------------------
> >                 server {
> >                         listen 80;
> >                         server_name aaa.com;
> >                         server_name 192.168.2.140;
> >                         location / {
> >                                 proxy_pass http://haha;
> >                                 proxy_redirect off;
> >                                 proxy_buffering off;
> >                                 proxy_set_header Host $host;
> >                                 proxy_set_header X-Real-IP  $remote_addr;
> >                                 proxy_set_header X-Forwarded-For
> > $proxy_add_x_forwarded_for;
> >                                 client_max_body_size 1000m;
> >                                 client_body_buffer_size 256k;
> >                                 proxy_connect_timeout 600;
> >                                 proxy_temp_file_write_size  64k;
> >                         }
> >                 }
> >                 upstream haha {
> >                         server 1.1.1.2 max_fails=3  fail_timeout=30s;
> > <---------max_fails/fail_timeouts/weight is not support ?
> >                         server 1.1.1.1 weight=10;
> >                 }
> >
> ----------------------------------------------------------------------------------------------------------------------------------------------
> >
> > When I run nginx, the error message print out as following:
> > ./nginx
> >
> > Root# /etc/init.d/nginx restart
> > [emerg]: invalid parameter "max_fails=3" in
> > /var/log/nginx/conf/nginx.conf:19
>
> The problem is that you've tried to define upstream after it was
> already implicitly created by proxy_pass directive.  Move upstream
> definition before it's first use.
>
> Maxim Dounin
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110905/0aa75cc2/attachment-0001.html>

From ianevans at digitalhit.com  Mon Sep  5 01:06:07 2011
From: ianevans at digitalhit.com (Ian Evans)
Date: Sun, 04 Sep 2011 21:06:07 -0400
Subject: PositiveSSL certificate keeps giving the untrusted issuer
	warnings in browser
In-Reply-To: <4E64120E.3090304@digitalhit.com>
References: <4E640DF0.1020000@digitalhit.com> <4E64120E.3090304@digitalhit.com>
Message-ID: <4E64207F.6040101@digitalhit.com>

On 04/09/2011 8:04 PM, Ian Evans wrote:
> Upon further testing I see that it's still using the old cert despite
> issuing a kill -HUP.

Sigh...okay, never mind. Seems I had one typo in the conf change so the 
kill -HUP was still using the old conf. So it's working okay now.

Still curious why some examples use pem and others use crt.

Also any best usage additions I should add to the ssl section?


From sy.meteor at msn.com  Mon Sep  5 02:41:45 2011
From: sy.meteor at msn.com (LiuXin)
Date: Mon, 5 Sep 2011 10:41:45 +0800
Subject: php-fpm is very very slow...
Message-ID: <SNT123-DS2007A8B078FDF142BEA67EE51D0@phx.gbl>

Hi all,
After upgrade from php 5.3.6 to php 5.3.8, now php-fpm is very very slow. Even I tried to roll back to php 5.3.6, it still doen't work.
And there are lots of errors in php-fpm.log,
  [04-Sep-2011 12:28:27] WARNING: [pool www] child 18793, script '/web/www/forum.php' (request: "GET /forum.php") executing too slow (60.073197 sec), logging
  [04-Sep-2011 12:28:27] WARNING: [pool www] child 18784, script '/web/www/forum.php' (request: "GET /forum.php") executing too slow (60.080807 sec), logging
  [04-Sep-2011 12:28:27] NOTICE: child 18793 stopped for tracing
  [04-Sep-2011 12:28:27] NOTICE: about to trace 18793
  [04-Sep-2011 12:28:27] ERROR: ptrace(PT_IO) failed: Bad address (14)
  [04-Sep-2011 12:28:27] NOTICE: finished trace of 18793
  [04-Sep-2011 12:28:27] NOTICE: child 18784 stopped for tracing
  [04-Sep-2011 12:28:27] NOTICE: about to trace 18784
  [04-Sep-2011 12:28:27] ERROR: ptrace(PT_IO) failed: Bad address (14)
  [04-Sep-2011 12:28:27] NOTICE: finished trace of 18784
And in php-fpm.slow
  [04-Sep-2011 12:27:30] [pool www] pid 18622
  script_filename = /web/www/forum.php
  [0x000000080455aaa8] mysql_query() /web/www/source/class/class_core.php:815
  [0x000000080455a958] query() /web/www/source/class/class_core.php:1003
  [0x000000080455a858] _execute() /web/www/source/class/class_core.php:977
  [0x00000008045591a0] query() /web/www/source/module/forum/forum_viewthread.php:514
  [0x0000000804559058] +++ dump failed
And one can help me about this issue?
My server is running Freebsd 8.1 + nginx + php 5.3.8
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110905/70eb9329/attachment.html>

From mat999 at gmail.com  Mon Sep  5 02:55:53 2011
From: mat999 at gmail.com (SplitIce)
Date: Mon, 5 Sep 2011 12:55:53 +1000
Subject: php-fpm is very very slow...
In-Reply-To: <SNT123-DS2007A8B078FDF142BEA67EE51D0@phx.gbl>
References: <SNT123-DS2007A8B078FDF142BEA67EE51D0@phx.gbl>
Message-ID: <CALFfGYbo2EC4w=d8_XZDYQvFscTRmn8WU3EX8+zR-RnFcSCFDg@mail.gmail.com>

a) not nginx related
b) not even php-fpm related

its a slow mysql query.

2011/9/5 LiuXin <sy.meteor at msn.com>

>   Hi all,
>  After upgrade from php 5.3.6 to php 5.3.8, now php-fpm is very very slow.
> Even I tried to roll back to php 5.3.6, it still doen't work.
>  And there are lots of errors in php-fpm.log,
>
> [04-Sep-2011 12:28:27] WARNING: [pool www] child 18793, script
> '/web/www/forum.php' (request: "GET /forum.php") executing too slow
> (60.073197 sec), logging
> [04-Sep-2011 12:28:27] WARNING: [pool www] child 18784, script
> '/web/www/forum.php' (request: "GET /forum.php") executing too slow
> (60.080807 sec), logging
> [04-Sep-2011 12:28:27] NOTICE: child 18793 stopped for tracing
> [04-Sep-2011 12:28:27] NOTICE: about to trace 18793
> [04-Sep-2011 12:28:27] ERROR: ptrace(PT_IO) failed: Bad address (14)
> [04-Sep-2011 12:28:27] NOTICE: finished trace of 18793
> [04-Sep-2011 12:28:27] NOTICE: child 18784 stopped for tracing
> [04-Sep-2011 12:28:27] NOTICE: about to trace 18784
> [04-Sep-2011 12:28:27] ERROR: ptrace(PT_IO) failed: Bad address (14)
> [04-Sep-2011 12:28:27] NOTICE: finished trace of 18784
>
> And in php-fpm.slow
>
> [04-Sep-2011 12:27:30] [pool www] pid 18622
> script_filename = /web/www/forum.php
> [0x000000080455aaa8] mysql_query() /web/www/source/class/class_core.php:815
> [0x000000080455a958] query() /web/www/source/class/class_core.php:1003
> [0x000000080455a858] _execute() /web/www/source/class/class_core.php:977
> [0x00000008045591a0] query()
> /web/www/source/module/forum/forum_viewthread.php:514
> [0x0000000804559058] +++ dump failed
>
>  And one can help me about this issue?
> My server is running Freebsd 8.1 + nginx + php 5.3.8
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110905/3a09f7f5/attachment.html>

From sy.meteor at msn.com  Mon Sep  5 03:52:02 2011
From: sy.meteor at msn.com (LiuXin)
Date: Mon, 5 Sep 2011 11:52:02 +0800
Subject: php-fpm is very very slow...
In-Reply-To: <CALFfGYbo2EC4w=d8_XZDYQvFscTRmn8WU3EX8+zR-RnFcSCFDg@mail.gmail.com>
References: <SNT123-DS2007A8B078FDF142BEA67EE51D0@phx.gbl>
	<CALFfGYbo2EC4w=d8_XZDYQvFscTRmn8WU3EX8+zR-RnFcSCFDg@mail.gmail.com>
Message-ID: <SNT123-DS199232D2FC868CDA19F800E51D0@phx.gbl>

Hi SplitIce,

Do you mean this problem is casued by mysql? Thanks

[04-Sep-2011 12:28:27] ERROR: ptrace(PT_IO) failed: Bad address (14)


From: SplitIce 
Sent: Monday, September 05, 2011 10:55 AM
To: nginx at nginx.org 
Subject: Re: php-fpm is very very slow...

a) not nginx related
b) not even php-fpm related

its a slow mysql query.


2011/9/5 LiuXin <sy.meteor at msn.com>

  Hi all,
  After upgrade from php 5.3.6 to php 5.3.8, now php-fpm is very very slow. Even I tried to roll back to php 5.3.6, it still doen't work.
  And there are lots of errors in php-fpm.log,
    [04-Sep-2011 12:28:27] WARNING: [pool www] child 18793, script '/web/www/forum.php' (request: "GET /forum.php") executing too slow (60.073197 sec), logging
    [04-Sep-2011 12:28:27] WARNING: [pool www] child 18784, script '/web/www/forum.php' (request: "GET /forum.php") executing too slow (60.080807 sec), logging
    [04-Sep-2011 12:28:27] NOTICE: child 18793 stopped for tracing
    [04-Sep-2011 12:28:27] NOTICE: about to trace 18793
    [04-Sep-2011 12:28:27] ERROR: ptrace(PT_IO) failed: Bad address (14)
    [04-Sep-2011 12:28:27] NOTICE: finished trace of 18793
    [04-Sep-2011 12:28:27] NOTICE: child 18784 stopped for tracing
    [04-Sep-2011 12:28:27] NOTICE: about to trace 18784
    [04-Sep-2011 12:28:27] ERROR: ptrace(PT_IO) failed: Bad address (14)
    [04-Sep-2011 12:28:27] NOTICE: finished trace of 18784
  And in php-fpm.slow
    [04-Sep-2011 12:27:30] [pool www] pid 18622
    script_filename = /web/www/forum.php
    [0x000000080455aaa8] mysql_query() /web/www/source/class/class_core.php:815
    [0x000000080455a958] query() /web/www/source/class/class_core.php:1003
    [0x000000080455a858] _execute() /web/www/source/class/class_core.php:977
    [0x00000008045591a0] query() /web/www/source/module/forum/forum_viewthread.php:514
    [0x0000000804559058] +++ dump failed
  And one can help me about this issue?
  My server is running Freebsd 8.1 + nginx + php 5.3.8

  _______________________________________________
  nginx mailing list
  nginx at nginx.org
  http://mailman.nginx.org/mailman/listinfo/nginx





--------------------------------------------------------------------------------
_______________________________________________
nginx mailing list
nginx at nginx.org
http://mailman.nginx.org/mailman/listinfo/nginx
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110905/d4671212/attachment.html>

From edho at myconan.net  Mon Sep  5 03:57:55 2011
From: edho at myconan.net (Edho P Arief)
Date: Mon, 5 Sep 2011 10:57:55 +0700
Subject: php-fpm is very very slow...
In-Reply-To: <SNT123-DS199232D2FC868CDA19F800E51D0@phx.gbl>
References: <SNT123-DS2007A8B078FDF142BEA67EE51D0@phx.gbl>
	<CALFfGYbo2EC4w=d8_XZDYQvFscTRmn8WU3EX8+zR-RnFcSCFDg@mail.gmail.com>
	<SNT123-DS199232D2FC868CDA19F800E51D0@phx.gbl>
Message-ID: <CABTjkK=Z0a+gmtot9YoMDsCrK=XiB5wDXO=0Bx0F0i6uNWyE8g@mail.gmail.com>

On Mon, Sep 5, 2011 at 10:52 AM, LiuXin <sy.meteor at msn.com> wrote:
> Hi SplitIce,
>
> Do you mean this problem is casued by mysql? Thanks
>

[04-Sep-2011 12:27:30] [pool www] pid 18622
script_filename = /web/www/forum.php
[0x000000080455aaa8] mysql_query() /web/www/source/class/class_core.php:815
[0x000000080455a958] query() /web/www/source/class/class_core.php:1003
[0x000000080455a858] _execute() /web/www/source/class/class_core.php:977
[0x00000008045591a0] query()
/web/www/source/module/forum/forum_viewthread.php:514


From igor at sysoev.ru  Mon Sep  5 04:55:16 2011
From: igor at sysoev.ru (Igor Sysoev)
Date: Mon, 5 Sep 2011 08:55:16 +0400
Subject: PositiveSSL certificate keeps giving the untrusted issuer
	warnings in browser
In-Reply-To: <4E64207F.6040101@digitalhit.com>
References: <4E640DF0.1020000@digitalhit.com> <4E64120E.3090304@digitalhit.com>
	<4E64207F.6040101@digitalhit.com>
Message-ID: <20110905045516.GA25408@nginx.com>

On Sun, Sep 04, 2011 at 09:06:07PM -0400, Ian Evans wrote:
> On 04/09/2011 8:04 PM, Ian Evans wrote:
> > Upon further testing I see that it's still using the old cert despite
> > issuing a kill -HUP.
> 
> Sigh...okay, never mind. Seems I had one typo in the conf change so the 
> kill -HUP was still using the old conf. So it's working okay now.
> 
> Still curious why some examples use pem and others use crt.

This is just a file extention. Both use the same PEM formay inside.

> Also any best usage additions I should add to the ssl section?

Please read this:
http://nginx.org/en/docs/http/configuring_https_servers.html


-- 
Igor Sysoev


From dewanggaba at gmail.com  Mon Sep  5 04:59:35 2011
From: dewanggaba at gmail.com (antituhan)
Date: Sun, 4 Sep 2011 21:59:35 -0700 (PDT)
Subject: Wordpress Setup (Nginx + Apache)
In-Reply-To: <4E5F562F.90407@contactlab.com>
References: <1314803185467-6746900.post@n2.nabble.com>
	<4E5F562F.90407@contactlab.com>
Message-ID: <1315198775336-6759814.post@n2.nabble.com>


Simone fumagalli wrote:
> 
> On 08/31/2011 05:06 PM, antituhan wrote:
>> Hello fellas, I've been setup wordpress blog using NginX as front end 
>> and Apache as a backend. 
> 
> If you are using this setup you might be interested in this plugin
> http://wordpress.org/extend/plugins/nginx-manager/
> In this way you can serve directly with Nginx static version of your pages
> and offload Apache 
> 
> Ciao
> 
> -- 
> Simone Fumagalli
> 
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx
> 
Ok, I'll try it ;)

-----
[daemon at antituhan.com ~]# 
--
View this message in context: http://nginx.2469901.n2.nabble.com/Wordpress-Setup-Nginx-Apache-tp6746900p6759814.html
Sent from the nginx mailing list archive at Nabble.com.


From nginx-forum at nginx.us  Mon Sep  5 05:33:41 2011
From: nginx-forum at nginx.us (magicbear)
Date: Mon, 05 Sep 2011 01:33:41 -0400
Subject: php-fpm is very very slow...
In-Reply-To: <SNT123-DS2007A8B078FDF142BEA67EE51D0@phx.gbl>
References: <SNT123-DS2007A8B078FDF142BEA67EE51D0@phx.gbl>
Message-ID: <f7e8ad090acfd2a22d3209a9afe7b288.NginxMailingListEnglish@forum.nginx.org>

You can check iowait, Discuz when have many of data, suggest you using
MySQL Cluster to process.

MagicBear

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,214766,214773#msg-214773


From nginx-forum at nginx.us  Mon Sep  5 05:40:22 2011
From: nginx-forum at nginx.us (jeff14)
Date: Mon, 05 Sep 2011 01:40:22 -0400
Subject: can i pass $time_local to backend server?
Message-ID: <42a94fd4a531ed9786ef1c5ee73f1a87.NginxMailingListEnglish@forum.nginx.org>

hi, i want set the variable $time_local in log to the header , and then
the backend server can got it
like: proxy_set_header MY_TIME $time_local
but i got error : unknown "time_local" variable
if i can set a variable about time to the header?  many thanks!

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,214774,214774#msg-214774


From toplek at polok.pl  Mon Sep  5 06:28:22 2011
From: toplek at polok.pl (Piotr Polok)
Date: Mon, 05 Sep 2011 08:28:22 +0200
Subject: upload file problem (500 Internal Server Error)
In-Reply-To: <d238b81ff2a4d42c51acfa6f500996b7.NginxMailingListEnglish@forum.nginx.org>
References: <87bd7ee8ed3c1a8bc0f33206b4db4a37@polok.pl>
	<d238b81ff2a4d42c51acfa6f500996b7.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <5fb9f62c9818d7337af57460493714f4@polok.pl>

On Thu, 01 Sep 2011 19:17:06 -0400, magicbear wrote:
> you can try to add such line to /etc/security/limits.conf
> nginx       soft    nofile   131072
> nginx       hard    nofile  1048576
>
>
> and fastcgi_pass to run PHP
> you need to set SCRIPT_FILENAME, you may first include fastcgi_params 
> at
> /etc/nginx (if you install nginx with binary), and then add such line
> before fastcgi_pass: fastcgi_param SCRIPT_FILENAME
> $document_root$fastcgi_script_name;

HI,

Thanks, this is also not solving the problem. I will start from the 
beginning in new thread.

-- 
best regards
Piotr Polok

---------------------------
POLOK WELDING
44-121 Gliwice
ul. Sowi?skiego 3
NIP: 631-112-45-46

tel: (32) 725 77 77
fax: (32) 444 64 50
http://www.polok.pl


From toplek at polok.pl  Mon Sep  5 06:59:58 2011
From: toplek at polok.pl (Piotr Polok)
Date: Mon, 05 Sep 2011 08:59:58 +0200
Subject: Nginx upload module (v 2.2.0) :: 502 Bad Gateway
Message-ID: <8d3997b08745b16189742395b9544d85@polok.pl>

HI,

I have compiled newest version of nginx (nginx/1.1.1) with nginx upload 
module (v 2.2.0), also the example config (nginx.conf) was copied from 
https://github.com/vkholodkov/nginx-upload-module into nginx direcoty. 
The html form is also copied from this location. So the installation and 
configuration is oryginal now without any changes.

I'm getting some errors while uploading the file;

The html form: '502 Bad Gateway';

The nginx error.log (/usr/local/nginx/logs/error.log):

------
2011/09/05 08:35:15 [error] 21161#0: *6 connect() failed (111: 
Connection refused) while connecting to upstream, client: 192.168.0.165, 
server: , request: "POST /upload HTTP/1.1", upstream: 
"http://127.0.0.1:8080/upload", host: "192.168.0.127", referrer: 
"http://192.168.0.127/http/upload.html"
------

Maybe it's a problem of my wrong understanding how does it works, I do 
not understand why in this example config the proxy_pass is going 
through local 8080 port but the server in this configuration is not 
listening on such port.

Please help.

-- 
best regards
Piotr Polok


From mdounin at mdounin.ru  Mon Sep  5 07:08:13 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Mon, 5 Sep 2011 11:08:13 +0400
Subject: upstream keepalive - call for testing
In-Reply-To: <CAENo=s3G3XBrYN7K9G6HzHZkBWYHEEi4DLCH8G-7_yC8dJyu-w@mail.gmail.com>
References: <20110801160725.GH1137@mdounin.ru>
	<0d5cc130a417bb62098701656fd44e42.NginxMailingListEnglish@forum.nginx.org>
	<20110904182137.GW1137@mdounin.ru>
	<CAENo=s3G3XBrYN7K9G6HzHZkBWYHEEi4DLCH8G-7_yC8dJyu-w@mail.gmail.com>
Message-ID: <20110905070813.GY1137@mdounin.ru>

Hello!

On Mon, Sep 05, 2011 at 02:31:54AM +0800, MagicBear wrote:

> (gdb) fr 0
> #0  ngx_http_upstream_handler (ev=0x7fc45735f8a8)
>     at src/http/ngx_http_upstream.c:915
> 915    ctx->current_request = r;
> (gdb) p *ev
> $1 = {data = 0x7fc4576aa750, write = 1, accept = 0, instance = 1, active =
> 1,
>   disabled = 0, ready = 1, oneshot = 0, complete = 0, eof = 0, error = 0,
>   timedout = 1, timer_set = 0, delayed = 0, read_discarded = 0,
>   unexpected_eof = 0, deferred_accept = 0, pending_eof = 0, posted_ready =
> 0,
>   available = 0, handler = 0x43bf9c <ngx_http_upstream_handler>,
>   index = 3503345872, log = 0x21f1770, timer = {key = 0, left = 0x0,
>     right = 0x0, parent = 0x0, color = 0 '\000', data = 0 '\000'},
>   closed = 0, channel = 0, resolver = 0, next = 0x0, prev = 0x0}
> (gdb) p *((ngx_connection_t *) ev->data)
> $2 = {data = 0x22a2780, read = 0x7fc4575008a8, write = 0x7fc45735f8a8,
>   fd = 66, recv = 0x41cc50 <ngx_unix_recv>, send = 0x41cf08
> <ngx_unix_send>,
>   recv_chain = 0x41cd00 <ngx_readv_chain>,
>   send_chain = 0x421050 <ngx_linux_sendfile_chain>, listening = 0x0,
>   sent = 0, log = 0x21f1770, pool = 0x237ff70, sockaddr = 0x0, socklen = 0,
>   addr_text = {len = 0, data = 0x0}, ssl = 0x0, local_sockaddr = 0x0,
>   buffer = 0x0, queue = {prev = 0x0, next = 0x0}, number = 63713716,
>   requests = 0, buffered = 0, log_error = 1, single_connection = 0,
>   unexpected_eof = 0, timedout = 0, error = 0, destroyed = 0, idle = 0,
>   reusable = 0, close = 0, sendfile = 1, sndlowat = 0, tcp_nodelay = 0,
>   tcp_nopush = 0}

Could you please also show the following (in the same frame, i.e. 
after "fr 0"):

p ngx_cycle->log
p *r
p *r->connection

Maxim Dounin

> 
> 
> 2011/9/5 Maxim Dounin <mdounin at mdounin.ru>
> 
> > Hello!
> >
> > On Sun, Sep 04, 2011 at 01:33:59PM -0400, magicbear wrote:
> >
> > > Hello , when run for 5 days, today have 2 segment failure.
> > >
> > > 2011/09/04 12:00:47 [alert] 32147#0: worker process 8496 exited on
> > > signal 11 (core dumped)
> > > 2011/09/04 12:17:32 [alert] 8497#0: ignore long locked inactive cache
> > > entry 57c7c6c612a651727b880a1c9a2c7d2d, count:1
> > > 2011/09/04 12:21:31 [alert] 32147#0: worker process 19185 exited on
> > > signal 11 (core dumped)
> > >
> > >
> > >
> > >
> > > Core was generated by `nginx:'.
> > > Program terminated with signal 11, Segmentation fault.
> > > #0  ngx_http_upstream_handler (ev=0x7fc45735f8a8) at
> > > src/http/ngx_http_upstream.c:915
> > > 915       ctx->current_request = r;
> > > (gdb) bt
> > > #0  ngx_http_upstream_handler (ev=0x7fc45735f8a8) at
> > > src/http/ngx_http_upstream.c:915
> > > #1  0x0000000000419e95 in ngx_event_expire_timers () at
> > > src/event/ngx_event_timer.c:149
> > > #2  0x0000000000419ac7 in ngx_process_events_and_timers
> > > (cycle=0x21f1990) at src/event/ngx_event.c:261
> >
> > Hm, this looks identical to one you've reported previously.  Do
> > you have patch from
> >
> > http://mailman.nginx.org/pipermail/nginx/2011-August/028809.html
> >
> > applied?
> >
> > > #3  0x000000000041f76e in ngx_worker_process_cycle (cycle=0x21f1990,
> > > data=<value optimized out>) at src/os/unix/ngx_process_cycle.c:800
> > > #4  0x000000000041e16b in ngx_spawn_process (cycle=0x21f1990,
> > > proc=0x41f69d <ngx_worker_process_cycle>, data=0x0, name=0x463f3d
> > > "worker process", respawn=7) at src/os/unix/ngx_process.c:196
> > > #5  0x000000000042008d in ngx_reap_children (cycle=0x21f1990) at
> > > src/os/unix/ngx_process_cycle.c:617
> > > #6  ngx_master_process_cycle (cycle=0x21f1990) at
> > > src/os/unix/ngx_process_cycle.c:180
> > > #7  0x0000000000406061 in main (argc=<value optimized out>, argv=<value
> > > optimized out>) at src/core/nginx.c:405
> > >
> > >
> > > (gdb) bt full
> > > #0  ngx_http_upstream_handler (ev=0x7fc45735f8a8) at
> > > src/http/ngx_http_upstream.c:915
> > >         c = 0x23b4160
> > >         r = 0x22a2780
> > >         ctx = <error reading variable ctx (Cannot access memory at
> > > address 0x20)>
> > >         u = 0x0
> >
> > Could you please show results of the following gdb commands:
> >
> > fr 0
> > p *ev
> > p *((ngx_connection_t *) ev->data)
> >
> >
> > Maxim Dounin
> >
> > _______________________________________________
> > nginx mailing list
> > nginx at nginx.org
> > http://mailman.nginx.org/mailman/listinfo/nginx
> >

> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx


From agentzh at gmail.com  Mon Sep  5 07:09:44 2011
From: agentzh at gmail.com (agentzh)
Date: Mon, 5 Sep 2011 15:09:44 +0800
Subject: OpenResty Queries
In-Reply-To: <CAN=WzV2BqGZdAVzeo4_y=QEPyxEM9M-u1Cue7EbVmcSbRBmrHA@mail.gmail.com>
References: <CAN=WzV2BqGZdAVzeo4_y=QEPyxEM9M-u1Cue7EbVmcSbRBmrHA@mail.gmail.com>
Message-ID: <CAB4Tn6NmD5VVikPbwUPPVQawZsNXu8ukWv7DUzEc7UKCdiZYFA@mail.gmail.com>

On Mon, Sep 5, 2011 at 1:38 AM, Nginx User <nginx at nginxuser.net> wrote:
> Some quick queries about some things missing from available documentation
> about OpenResty
> 1. Notice current version is bundled with Nginx 1.0.4. Is it possible for an
> end user to upgrade Nginx? I.E. can I download OpenResty and slip in Nginx
> 1.0.6?

The current stable version of OpenResty is 1.0.5.1, which bundles the
1.0.5 Nginx core. The current devel release of OpenResty is
1.0.6.0rc2, which bundles the 1.0.6 core. Check out

    http://openresty.org/#Download

Manually replacing the nginx core is possible, just replace the
directory of OpenResty's bundle/nginx-1.0.5/ with your own, say,
nginx-1.1.1/. OpenResty's configure script should be smart enough to
pick up the new version of the Nginx core.

But be careful with self replacing the Nginx core in OpenResty because
forward compatibility is not guaranteed. A new untested Nginx core can
break things, and that's all the point of bundling the Nginx core in
the OpenResty bundle to ensure that it works well with *all* of the
bundled components.

> 2. The bundled modules are awesome but what is the process wants an
> extra?third?party module?

All of the nginx's configure options are inherited by OpenResty's
configure script, so you can still add extra 3rd-party modules like
this:

    ./configure --add-module=/path/to/your/other/module

Regards,
-agentzh


From ggarron at alketech.com  Mon Sep  5 08:04:48 2011
From: ggarron at alketech.com (Guillermo Garron)
Date: Mon, 5 Sep 2011 04:04:48 -0400
Subject: Nginx upload module (v 2.2.0) :: 502 Bad Gateway
Message-ID: <bb0747d2966eee270635fe7910d8326c@mail.gmail.com>

Bad gateway is a proxy problem
I have not used upload module. But try having something listen to that
port, or modify the proxy directive to point to the right port.
Guillermo
------Original Message------
From: Piotr Polok
To: Nginx
ReplyTo: nginx at nginx.org
Subject: Nginx upload module (v 2.2.0) :: 502 Bad Gateway
Sent: Sep 5, 2011 2:59 AM

HI,

I have compiled newest version of nginx (nginx/1.1.1) with nginx upload
module (v 2.2.0), also the example config (nginx.conf) was copied from
https://github.com/vkholodkov/nginx-upload-module into nginx direcoty.
The html form is also copied from this location. So the installation and
configuration is oryginal now without any changes.

I'm getting some errors while uploading the file;

The html form: '502 Bad Gateway';

The nginx error.log (/usr/local/nginx/logs/error.log):

------
2011/09/05 08:35:15 [error] 21161#0: *6 connect() failed (111:
Connection refused) while connecting to upstream, client: 192.168.0.165,
server: , request: "POST /upload HTTP/1.1", upstream:
"http://127.0.0.1:8080/upload", host: "192.168.0.127", referrer:
"http://192.168.0.127/http/upload.html"
------

Maybe it's a problem of my wrong understanding how does it works, I do
not understand why in this example config the proxy_pass is going
through local 8080 port but the server in this configuration is not
listening on such port.

Please help.

-- 
best regards
Piotr Polok

_______________________________________________
nginx mailing list
nginx at nginx.org
http://mailman.nginx.org/mailman/listinfo/nginx


From toplek at polok.pl  Mon Sep  5 08:16:14 2011
From: toplek at polok.pl (Piotr Polok)
Date: Mon, 05 Sep 2011 10:16:14 +0200
Subject: Nginx upload module (v 2.2.0) :: 502 Bad Gateway
In-Reply-To: <bb0747d2966eee270635fe7910d8326c@mail.gmail.com>
References: <bb0747d2966eee270635fe7910d8326c@mail.gmail.com>
Message-ID: <881e6fdbefb126e2e272aeff53a4f205@polok.pl>

On Mon, 5 Sep 2011 04:04:48 -0400, Guillermo Garron wrote:
> Bad gateway is a proxy problem
> I have not used upload module. But try having something listen to 
> that
> port, or modify the proxy directive to point to the right port.

HI,

I do not use proxy.

-- 
pozdrawiam
Piotr Polok

---------------------------
POLOK WELDING
44-121 Gliwice
ul. Sowi?skiego 3
NIP: 631-112-45-46

tel: (32) 725 77 77
fax: (32) 444 64 50
http://www.polok.pl


From ggarron at alketech.com  Mon Sep  5 08:38:20 2011
From: ggarron at alketech.com (Guillermo Garron)
Date: Mon, 5 Sep 2011 04:38:20 -0400
Subject: Nginx upload module (v 2.2.0) :: 502 Bad Gateway
Message-ID: <f726fb893c0899b06d6f58a6da8e999c@mail.gmail.com>

Your problem is that in your conf file /test location is going to some no
existent back end

Either deal with it inside nginx or have some back end listening to 8080.

That is the directive causing the problem

Guillermo
Ps sorry for top posting (bb issue)
------Original Message------
From: Piotr Polok
To: Nginx
ReplyTo: nginx at nginx.org
Subject: Nginx upload module (v 2.2.0) :: 502 Bad Gateway
Sent: Sep 5, 2011 2:59 AM

HI,

I have compiled newest version of nginx (nginx/1.1.1) with nginx upload
module (v 2.2.0), also the example config (nginx.conf) was copied from
https://github.com/vkholodkov/nginx-upload-module into nginx direcoty.
The html form is also copied from this location. So the installation and
configuration is oryginal now without any changes.

I'm getting some errors while uploading the file;

The html form: '502 Bad Gateway';

The nginx error.log (/usr/local/nginx/logs/error.log):

------
2011/09/05 08:35:15 [error] 21161#0: *6 connect() failed (111:
Connection refused) while connecting to upstream, client: 192.168.0.165,
server: , request: "POST /upload HTTP/1.1", upstream:
"http://127.0.0.1:8080/upload", host: "192.168.0.127", referrer:
"http://192.168.0.127/http/upload.html"
------

Maybe it's a problem of my wrong understanding how does it works, I do
not understand why in this example config the proxy_pass is going
through local 8080 port but the server in this configuration is not
listening on such port.

Please help.

-- 
best regards
Piotr Polok

_______________________________________________
nginx mailing list
nginx at nginx.org
http://mailman.nginx.org/mailman/listinfo/nginx


From toplek at polok.pl  Mon Sep  5 08:54:12 2011
From: toplek at polok.pl (Piotr Polok)
Date: Mon, 05 Sep 2011 10:54:12 +0200
Subject: Nginx upload module (v 2.2.0) :: 502 Bad Gateway
In-Reply-To: <f726fb893c0899b06d6f58a6da8e999c@mail.gmail.com>
References: <f726fb893c0899b06d6f58a6da8e999c@mail.gmail.com>
Message-ID: <36cedc1c4584f3d39d3b4899232b7e9b@polok.pl>

> Your problem is that in your conf file /test location is going to 
> some no
> existent back end
>
> Either deal with it inside nginx or have some back end listening to 
> 8080.
>
> That is the directive causing the problem

xserver:/usr/local/nginx/conf# tshark -i lo port 8080
Running as user "root" and group "root". This could be dangerous.
Capturing on lo
   0.000000    127.0.0.1 -> 127.0.0.1    TCP 50768 > http-alt [SYN] 
Seq=0 Win=32792 Len=0 MSS=16396 TSV=340033941 TSER=0 WS=7
   0.000039    127.0.0.1 -> 127.0.0.1    TCP http-alt > 50768 [RST, ACK] 
Seq=1 Ack=1 Win=0 Len=0

There is no any service on this port.

I'm not getting the idea what should happend on 'location @test'. Maybe 
in this place the file should be saved in the specyfic location? If yes, 
how this should be configured?

-- 
best regards
Piotr Polok


From varia at e-healthexpert.org  Mon Sep  5 09:11:17 2011
From: varia at e-healthexpert.org (Mark Alan)
Date: Mon, 5 Sep 2011 10:11:17 +0100
Subject: Nginx upload module (v 2.2.0) :: 502 Bad Gateway
In-Reply-To: <8d3997b08745b16189742395b9544d85@polok.pl>
References: <8d3997b08745b16189742395b9544d85@polok.pl>
Message-ID: <20110905101117.05d2416a@e-healthexpert.org>

On Mon, 05 Sep 2011 08:59:58 +0200, Piotr Polok <toplek at polok.pl> wrote:
> I'm getting some errors while uploading the file;
> The html form: '502 Bad Gateway';
> The nginx error.log (/usr/local/nginx/logs/error.log):
> ------
> 2011/09/05 08:35:15 [error] 21161#0: *6 connect() failed (111: 
> Connection refused) while connecting to upstream, client:
> 192.168.0.165, server: , request: "POST /upload HTTP/1.1", upstream: 
> "http://127.0.0.1:8080/upload", host: "192.168.0.127", referrer: 
> "http://192.168.0.127/http/upload.html"
> ------

Your setup includes:
client_max_body_size 100m;

Do you also your php.ini configured to accommodate 100m files?
Check:
memory_limit (should be something able to cope with 100M files)
post_max_size (should be >= 110M)
file_uploads (should be On)
upload_max_filesize (should be >= 100M)

Regards,
M.


From toplek at polok.pl  Mon Sep  5 09:33:26 2011
From: toplek at polok.pl (Piotr Polok)
Date: Mon, 05 Sep 2011 11:33:26 +0200
Subject: Nginx upload module (v 2.2.0) :: 502 Bad Gateway
In-Reply-To: <20110905101117.05d2416a@e-healthexpert.org>
References: <8d3997b08745b16189742395b9544d85@polok.pl>
	<20110905101117.05d2416a@e-healthexpert.org>
Message-ID: <95e66db4b6c311ca3b0987a55e9c2b30@polok.pl>

> Your setup includes:
> client_max_body_size 100m;
>
> Do you also your php.ini configured to accommodate 100m files?
> Check:
> memory_limit (should be something able to cope with 100M files)
> post_max_size (should be >= 110M)
> file_uploads (should be On)
> upload_max_filesize (should be >= 100M)

Thank you all for help!

I found the solution on this thread:  
http://www.ruby-forum.com/topic/1432037

The problem was that I did not wrote the form handler script.

This is my working configuration of nginx:

/usr/local/nginx/conf/nginx.conf:
----
worker_processes  20;

error_log  logs/error.log notice;

working_directory /usr/local/nginx/html;

events {
     worker_connections  1024;
}

http {
     include       mime.types;
     default_type  application/octet-stream;

     server {
         listen       80;
         client_max_body_size 100m;

         # Upload form should be submitted to this location
         location /upload {
             # Pass altered request body to this location
             upload_pass   /my-form-handler;

             # Store files to this directory
             # The directory is hashed, subdirectories 0 1 2 3 4 5 6 7 8 
9 should exist
             upload_store /tmp 1;

             # Allow uploaded files to be read only by user
             upload_store_access user:r;

             # Set specified fields in request body
             upload_set_form_field "${upload_field_name}_name" 
$upload_file_name;
             upload_set_form_field "${upload_field_name}_content_type" 
$upload_content_type;
             upload_set_form_field "${upload_field_name}_path" 
$upload_tmp_path;

             # Inform backend about hash and size of a file
             upload_aggregate_form_field "${upload_field_name}_md5" 
$upload_file_md5;
             upload_aggregate_form_field "${upload_field_name}_size" 
$upload_file_size;

             upload_pass_form_field "^submit$|^description$";
         }

        include fastcgi.conf;
        location = /my-form-handler {
           fastcgi_pass  127.0.0.1:9000;
         }

     }
}
----

/usr/local/nginx/html/my-form-handler
----
<?php
header("Content-Type: text/plain");
print("GET: "); print_r($_GET);
print("POST: "); print_r($_POST);
print("FILES: "); print_r($_FILES);
?>
----

-- 
best regards
Piotr Polok


From nginx-forum at nginx.us  Mon Sep  5 10:58:56 2011
From: nginx-forum at nginx.us (locojohn)
Date: Mon, 05 Sep 2011 06:58:56 -0400
Subject: URL decode rewrite handling: + vs. %20
In-Reply-To: <5e2ad7736ec3f420b7f1992a428e6969.NginxMailingListEnglish@forum.nginx.org>
References: <CALFfGYYHxsvL7q6x057eH2O3QP7qRPDAr+UcobgT=x82AGpf=Q@mail.gmail.com>
	<5e2ad7736ec3f420b7f1992a428e6969.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <c21d279e72ae78a1915852a0b3f4311b.NginxMailingListEnglish@forum.nginx.org>

You can also use PCRE capturing inside virtual host configuration, to
capture both SCRIPT_FILENAME and PATH_INFO without URL encoding:

location ~ ^(?<SCRIPT_FILENAME>.+\.php)$ {
                  include fastcgi_params;
                  fastcgi_param  SCRIPT_FILENAME
$document_root$SCRIPT_FILENAME;
                  fastcgi_pass ...;
}

location ~ ^(?<SCRIPT_FILENAME>.+\.php)(?<PATH_INFO>.+)$ {
                  include fastcgi_params;
                  fastcgi_param  SCRIPT_FILENAME
$document_root$SCRIPT_FILENAME;
                  fastcgi_param  PATH_INFO $PATH_INFO;
                  fastcgi_param  PATH_TRANSLATED
$document_root$PATH_INFO;
                  fastcgi_pass phpfarm;
                }

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,214668,214792#msg-214792


From igor at sysoev.ru  Mon Sep  5 13:19:50 2011
From: igor at sysoev.ru (Igor Sysoev)
Date: Mon, 5 Sep 2011 17:19:50 +0400
Subject: nginx-1.1.2
Message-ID: <20110905131950.GG34606@nginx.com>

Changes with nginx 1.1.2                                         05 Sep 2011

    *) Change: now if total size of all ranges is greater than source
       response size, then nginx disables ranges and returns just the source
       response.

    *) Feature: the "max_ranges" directive.

    *) Bugfix: the "ssl_verify_client", "ssl_verify_depth", and
       "ssl_prefer_server_ciphers" directives might work incorrectly if SNI
       was used.

    *) Bugfix: in the "proxy/fastcgi/scgi/uwsgi_ignore_client_abort"
       directives.


-- 
Igor Sysoev


From wendal1985 at gmail.com  Mon Sep  5 13:32:46 2011
From: wendal1985 at gmail.com (Wendal Chen)
Date: Mon, 5 Sep 2011 21:32:46 +0800
Subject: nginx-1.1.2
In-Reply-To: <20110905131950.GG34606@nginx.com>
References: <20110905131950.GG34606@nginx.com>
Message-ID: <CAM79QPW5mVFQZX3_v8nscYqFV_V8_a=4y853sC7iJk1KkmH--A@mail.gmail.com>

Great!!

2011/9/5 Igor Sysoev <igor at sysoev.ru>

> Changes with nginx 1.1.2                                         05 Sep
> 2011
>
>    *) Change: now if total size of all ranges is greater than source
>       response size, then nginx disables ranges and returns just the source
>       response.
>
>    *) Feature: the "max_ranges" directive.
>
>    *) Bugfix: the "ssl_verify_client", "ssl_verify_depth", and
>       "ssl_prefer_server_ciphers" directives might work incorrectly if SNI
>       was used.
>
>    *) Bugfix: in the "proxy/fastcgi/scgi/uwsgi_ignore_client_abort"
>       directives.
>
>
> --
> Igor Sysoev
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx
>



-- 
Wendal Chen
<http://t.sina.com.cn/1068040557?s=6uyXnP>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110905/ee75e5a6/attachment.html>

From mdounin at mdounin.ru  Mon Sep  5 15:16:18 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Mon, 5 Sep 2011 19:16:18 +0400
Subject: openssl-1.0.0d & proxy_buffering
In-Reply-To: <4E64DDF0.4040202@bestmx.ru>
References: <4E64B9AA.9050302@bestmx.ru> <20110905121224.GA34606@nginx.com>
	<4E64BFEE.6030502@bestmx.ru> <20110905122936.GC34606@nginx.com>
	<4E64C36B.6080701@bestmx.ru> <20110905140432.GC1137@mdounin.ru>
	<4E64DDF0.4040202@bestmx.ru>
Message-ID: <20110905151618.GE1137@mdounin.ru>

Hello!

On Mon, Sep 05, 2011 at 06:34:24PM +0400, Andrey N. Oktyabrski wrote:

> On 05.09.11 18:04, Maxim Dounin wrote:
> >>>????? ?? ?
> >>?? ?????????? ??????? dragonfly x86_64 (???? ???? ?????? ???
> >>?????????????), ?? ????????????? - centos-5.2 i386 ?? vps.
> >
> >? ?????:
> >
> >1. ????? ./configure ?????????.
> >2. Debug log.
> ??????? ? ????? ??? ??? ????? ? ?????? ??????.

???? ?? debug log'?, ?????? ?? ????????? ?????????? ????? ???????? 
??????.  ? ?????????? nginx ????????? ?????????? ?? ????????, ? 
??? ?? ???????????? ??????? ?????? (if any) - ????????????.

????????? ??? ???? ?????? - ????????? ????? ?????????? ?????? ?? 
[1], ??? ?????? ?????????? Content-Length.  ??? ??? ?????????? 
???????? ??? SSL ?/??? ?? ?????? ??????? - ??? ???? ???????, 
?????? ??? ?????? ????? ???? ?? ???????.

[1] http://mailman.nginx.org/pipermail/nginx-devel/2011-September/001129.html

Maxim Dounin


From mdounin at mdounin.ru  Mon Sep  5 15:18:06 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Mon, 5 Sep 2011 19:18:06 +0400
Subject: openssl-1.0.0d & proxy_buffering
In-Reply-To: <20110905151618.GE1137@mdounin.ru>
References: <4E64B9AA.9050302@bestmx.ru> <20110905121224.GA34606@nginx.com>
	<4E64BFEE.6030502@bestmx.ru> <20110905122936.GC34606@nginx.com>
	<4E64C36B.6080701@bestmx.ru> <20110905140432.GC1137@mdounin.ru>
	<4E64DDF0.4040202@bestmx.ru> <20110905151618.GE1137@mdounin.ru>
Message-ID: <20110905151806.GF1137@mdounin.ru>

Hello!

On Mon, Sep 05, 2011 at 07:16:18PM +0400, Maxim Dounin wrote:

> Hello!
> 
> On Mon, Sep 05, 2011 at 06:34:24PM +0400, Andrey N. Oktyabrski wrote:
> 
> > On 05.09.11 18:04, Maxim Dounin wrote:
> > >>>????? ?? ?
> > >>?? ?????????? ??????? dragonfly x86_64 (???? ???? ?????? ???
> > >>?????????????), ?? ????????????? - centos-5.2 i386 ?? vps.
> > >
> > >? ?????:
> > >
> > >1. ????? ./configure ?????????.
> > >2. Debug log.
> > ??????? ? ????? ??? ??? ????? ? ?????? ??????.
> 
> ???? ?? debug log'?, ?????? ?? ????????? ?????????? ????? ???????? 
> ??????.  ? ?????????? nginx ????????? ?????????? ?? ????????, ? 
> ??? ?? ???????????? ??????? ?????? (if any) - ????????????.
> 
> ????????? ??? ???? ?????? - ????????? ????? ?????????? ?????? ?? 
> [1], ??? ?????? ?????????? Content-Length.  ??? ??? ?????????? 
> ???????? ??? SSL ?/??? ?? ?????? ??????? - ??? ???? ???????, 
> ?????? ??? ?????? ????? ???? ?? ???????.
> 
> [1] http://mailman.nginx.org/pipermail/nginx-devel/2011-September/001129.html

Ooops, sorry, this was for nginx-ru at .

Maxim Dounin


From orz at loli.my  Mon Sep  5 15:42:31 2011
From: orz at loli.my (=?UTF-8?B?44OT44Oq44OT44Oq4oWk?=)
Date: Mon, 5 Sep 2011 23:42:31 +0800
Subject: upstream keepalive - call for testing
In-Reply-To: <20110905070813.GY1137@mdounin.ru>
References: <20110801160725.GH1137@mdounin.ru>
	<0d5cc130a417bb62098701656fd44e42.NginxMailingListEnglish@forum.nginx.org>
	<20110904182137.GW1137@mdounin.ru>
	<CAENo=s3G3XBrYN7K9G6HzHZkBWYHEEi4DLCH8G-7_yC8dJyu-w@mail.gmail.com>
	<20110905070813.GY1137@mdounin.ru>
Message-ID: <CAENo=s1GJzGJ8OPbNBvQhTEECN=jKFAtgYN=SMx=2o9KS+W02A@mail.gmail.com>

(gdb) fr 0
#0  ngx_http_upstream_handler (ev=0x7fc45735f8a8)
    at src/http/ngx_http_upstream.c:915
915    ctx->current_request = r;
(gdb) p ngx_cycle->log
$1 = (ngx_log_t *) 0x21f19a8
(gdb) p *r
$2 = {signature = 51686928, connection = 0x23b4160, ctx = 0x0,
  main_conf = 0x0, srv_conf = 0x23e6dc8, loc_conf = 0x2216d58,
  read_event_handler = 0x42d5dd <ngx_http_block_reading>,
  write_event_handler = 0x42d2a6 <ngx_http_request_empty_handler>,
  cache = 0x0, upstream = 0x0, upstream_states = 0x0, pool = 0x23e92d0,
  header_in = 0x2309ea0, headers_in = {headers = {last = 0x22a27f0, part = {
        elts = 0x23e9ae8, nelts = 5, next = 0x0}, size = 48, nalloc = 20,
      pool = 0x23e92d0}, host = 0x23e9b78, connection = 0x0,
    if_modified_since = 0x0, if_unmodified_since = 0x0,
    user_agent = 0x23e9b48, referer = 0x0, content_length = 0x0,
    content_type = 0x0, range = 0x0, if_range = 0x0, transfer_encoding =
0x0,
    expect = 0x0, accept_encoding = 0x0, via = 0x0, authorization = 0x0,
    keep_alive = 0x0, x_forwarded_for = 0x0, user = {len = 0, data = 0x0},
    passwd = {len = 0, data = 0x0}, cookies = {elts = 0x23e9ea8, nelts = 0,
      size = 8, nalloc = 2, pool = 0x23e92d0}, server = {len = 14,
      data = 0x23e9edb "static.loli.myconnectioner-ag\360\324,\002"},
    content_length_n = -1, keep_alive_n = -1, connection_type = 1, msie =
1,
    msie6 = 1, opera = 0, gecko = 0, chrome = 0, safari = 0, konqueror =
0},
  headers_out = {headers = {last = 0x22a2920, part = {elts = 0x23e9320,
        nelts = 4, next = 0x0}, size = 48, nalloc = 20, pool = 0x23e92d0},
    status = 200, status_line = {len = 0, data = 0x0}, server = 0x0,
    date = 0x0, content_length = 0x0, content_encoding = 0x0, location =
0x0,
    refresh = 0x0, last_modified = 0x0, content_range = 0x0,
---Type <return> to continue, or q <return> to quit---
    accept_ranges = 0x23e93b0, www_authenticate = 0x0, expires = 0x23e9320,
    etag = 0x0, override_charset = 0x0, content_type_len = 24, content_type
= {
      len = 24, data = 0x23e5bf0 "application/x-javascript"}, charset = {
      len = 0, data = 0x0},
    content_type_lowcase = 0x23ea1e8 "application/x-javascripts\016>\002",
    content_type_hash = 15285649345721276481, cache_control = {
      elts = 0x23ea188, nelts = 1, size = 8, nalloc = 1, pool = 0x23e92d0},
    content_length_n = 19773, date_time = 0, last_modified_time =
1313998502},
  request_body = 0x0, lingering_time = 1315110106, start_sec = 1315110076,
  start_msec = 683, method = 2, http_version = 1001, request_line = {len =
0,
    data = 0x314ae20 "\360\f>\002"}, uri = {len = 15, data = 0x314ae24 ""},
  args = {len = 0, data = 0x0}, exten = {len = 2, data = 0x314ae31 ""},
  unparsed_uri = {len = 15, data = 0x314ae24 ""}, method_name = {len = 3,
    data = 0x314ae20 "\360\f>\002"}, http_protocol = {len = 8,
    data = 0x314ae34 ""}, out = 0x0, main = 0x22a2780, parent = 0x0,
  postponed = 0x0, post_subrequest = 0x0, posted_requests = 0x0,
  virtual_names = 0x2465c20, phase_handler = 11, content_handler = 0,
  access_code = 0, variables = 0x23e9808, ncaptures = 4, captures =
0x23e9ff0,
  captures_data = 0x314ae24 "", limit_rate = 0, header_size = 339,
  request_length = 183, err_status = 0, http_connection = 0x23b4228,
  log_handler = 0x42da14 <ngx_http_log_error_handler>, cleanup = 0x0,
  subrequests = 51, count = 0, blocked = 0, aio = 0, http_state = 2,
  complex_uri = 0, quoted_uri = 0, plus_in_uri = 0, space_in_uri = 0,
---Type <return> to continue, or q <return> to quit---
  invalid_header = 0, add_uri_to_alias = 0, valid_location = 1,
  valid_unparsed_uri = 1, uri_changed = 0, uri_changes = 11,
  request_body_in_single_buf = 0, request_body_in_file_only = 0,
  request_body_in_persistent_file = 0, request_body_in_clean_file = 0,
  request_body_file_group_access = 0, request_body_file_log_level = 5,
  subrequest_in_memory = 0, waited = 0, cached = 0, gzip_tested = 1,
  gzip_ok = 0, gzip_vary = 1, proxy = 0, bypass_cache = 0, no_cache = 0,
  limit_zone_set = 0, limit_req_set = 0, pipeline = 0, plain_http = 0,
  chunked = 0, header_only = 0, keepalive = 0, lingering_close = 0,
  discard_body = 0, internal = 0, error_page = 0, ignore_content_encoding =
0,
  filter_finalize = 0, post_action = 0, request_complete = 1,
  request_output = 1, header_sent = 1, expect_tested = 0, root_tested = 1,
  done = 1, logged = 0, buffered = 0, main_filter_need_in_memory = 0,
  filter_need_in_memory = 0, filter_need_temporary = 0, allow_ranges = 1,
  stat_reading = 0, stat_writing = 1, state = 0,
  header_hash = 2715320498552542, lowcase_index = 10,
  lowcase_header = "connectionguage", '\000' <repeats 16 times>,
  header_name_start = 0x314aed5 "\r\nible; MSIE 8.0; Windows NT 6.1;
Trident/4.0; GTB7.0; SLCC2; .NET CLR 2.0.50727; .NET CLR 3.5.30729; .NET CLR
3.0.30729; MALN; InfoPath.1; Media Center PC 6.0; Tablet PC 2.0; .NET4.0C;
.NET4.0E)",
  header_name_end = 0x314aecc "", header_start = 0x314aece "close",
  header_end = 0x314aed5 "\r\nible; MSIE 8.0; Windows NT 6.1; Trident/4.0;
GTB7.0; SLCC2; .NET CLR 2.0.50727; .NET CLR 3.5.30729; .NET CLR 3.0.30729;
MALN; Info---Type <return> to continue, or q <return> to quit---
Path.1; Media Center PC 6.0; Tablet PC 2.0; .NET4.0C; .NET4.0E)",
  uri_start = 0x314ae24 "", uri_end = 0x314ae33 "", uri_ext = 0x314ae31 "",
  args_start = 0x0, request_start = 0x314ae20 "\360\f>\002",
  request_end = 0x314ae3c "", method_end = 0x314ae22 ">\002",
  schema_start = 0x0, schema_end = 0x0, host_start = 0x0, host_end = 0x0,
  port_start = 0x0, port_end = 0x0, http_minor = 1264, http_major = 0}
(gdb) p *r->connection
$3 = {data = 0x30, read = 0x111, write = 0x22a2770, fd = 36740720,
  recv = 0x2309e80, send = 0, recv_chain = 0xb0, send_chain = 0x23b4170,
  listening = 0x0, sent = 36740848, log = 0x0, pool = 0x23b41d0,
  sockaddr = 0xf088e714b9a0002, socklen = 0, addr_text = {len = 4,
    data = 0x21f1c18 "S"}, ssl = 0x3cc35a0, local_sockaddr = 0x42d957,
  buffer = 0x23b4210, queue = {prev = 0x465bfc, next = 0x2e3234312e333131},
  number = 3256446919138946616, requests = 140481256921360, buffered = 0,
  log_error = 0, single_connection = 0, unexpected_eof = 0, timedout = 0,
  error = 0, destroyed = 0, idle = 0, reusable = 0, close = 0, sendfile =
0,
  sndlowat = 0, tcp_nodelay = 0, tcp_nopush = 0}

2011/9/5 Maxim Dounin <mdounin at mdounin.ru>

> Hello!
>
> On Mon, Sep 05, 2011 at 02:31:54AM +0800, MagicBear wrote:
>
> > (gdb) fr 0
> > #0  ngx_http_upstream_handler (ev=0x7fc45735f8a8)
> >     at src/http/ngx_http_upstream.c:915
> > 915    ctx->current_request = r;
> > (gdb) p *ev
> > $1 = {data = 0x7fc4576aa750, write = 1, accept = 0, instance = 1, active
> =
> > 1,
> >   disabled = 0, ready = 1, oneshot = 0, complete = 0, eof = 0, error = 0,
> >   timedout = 1, timer_set = 0, delayed = 0, read_discarded = 0,
> >   unexpected_eof = 0, deferred_accept = 0, pending_eof = 0, posted_ready
> =
> > 0,
> >   available = 0, handler = 0x43bf9c <ngx_http_upstream_handler>,
> >   index = 3503345872, log = 0x21f1770, timer = {key = 0, left = 0x0,
> >     right = 0x0, parent = 0x0, color = 0 '\000', data = 0 '\000'},
> >   closed = 0, channel = 0, resolver = 0, next = 0x0, prev = 0x0}
> > (gdb) p *((ngx_connection_t *) ev->data)
> > $2 = {data = 0x22a2780, read = 0x7fc4575008a8, write = 0x7fc45735f8a8,
> >   fd = 66, recv = 0x41cc50 <ngx_unix_recv>, send = 0x41cf08
> > <ngx_unix_send>,
> >   recv_chain = 0x41cd00 <ngx_readv_chain>,
> >   send_chain = 0x421050 <ngx_linux_sendfile_chain>, listening = 0x0,
> >   sent = 0, log = 0x21f1770, pool = 0x237ff70, sockaddr = 0x0, socklen =
> 0,
> >   addr_text = {len = 0, data = 0x0}, ssl = 0x0, local_sockaddr = 0x0,
> >   buffer = 0x0, queue = {prev = 0x0, next = 0x0}, number = 63713716,
> >   requests = 0, buffered = 0, log_error = 1, single_connection = 0,
> >   unexpected_eof = 0, timedout = 0, error = 0, destroyed = 0, idle = 0,
> >   reusable = 0, close = 0, sendfile = 1, sndlowat = 0, tcp_nodelay = 0,
> >   tcp_nopush = 0}
>
> Could you please also show the following (in the same frame, i.e.
> after "fr 0"):
>
> p ngx_cycle->log
> p *r
> p *r->connection
>
> Maxim Dounin
>
> >
> >
> > 2011/9/5 Maxim Dounin <mdounin at mdounin.ru>
> >
> > > Hello!
> > >
> > > On Sun, Sep 04, 2011 at 01:33:59PM -0400, magicbear wrote:
> > >
> > > > Hello , when run for 5 days, today have 2 segment failure.
> > > >
> > > > 2011/09/04 12:00:47 [alert] 32147#0: worker process 8496 exited on
> > > > signal 11 (core dumped)
> > > > 2011/09/04 12:17:32 [alert] 8497#0: ignore long locked inactive cache
> > > > entry 57c7c6c612a651727b880a1c9a2c7d2d, count:1
> > > > 2011/09/04 12:21:31 [alert] 32147#0: worker process 19185 exited on
> > > > signal 11 (core dumped)
> > > >
> > > >
> > > >
> > > >
> > > > Core was generated by `nginx:'.
> > > > Program terminated with signal 11, Segmentation fault.
> > > > #0  ngx_http_upstream_handler (ev=0x7fc45735f8a8) at
> > > > src/http/ngx_http_upstream.c:915
> > > > 915       ctx->current_request = r;
> > > > (gdb) bt
> > > > #0  ngx_http_upstream_handler (ev=0x7fc45735f8a8) at
> > > > src/http/ngx_http_upstream.c:915
> > > > #1  0x0000000000419e95 in ngx_event_expire_timers () at
> > > > src/event/ngx_event_timer.c:149
> > > > #2  0x0000000000419ac7 in ngx_process_events_and_timers
> > > > (cycle=0x21f1990) at src/event/ngx_event.c:261
> > >
> > > Hm, this looks identical to one you've reported previously.  Do
> > > you have patch from
> > >
> > > http://mailman.nginx.org/pipermail/nginx/2011-August/028809.html
> > >
> > > applied?
> > >
> > > > #3  0x000000000041f76e in ngx_worker_process_cycle (cycle=0x21f1990,
> > > > data=<value optimized out>) at src/os/unix/ngx_process_cycle.c:800
> > > > #4  0x000000000041e16b in ngx_spawn_process (cycle=0x21f1990,
> > > > proc=0x41f69d <ngx_worker_process_cycle>, data=0x0, name=0x463f3d
> > > > "worker process", respawn=7) at src/os/unix/ngx_process.c:196
> > > > #5  0x000000000042008d in ngx_reap_children (cycle=0x21f1990) at
> > > > src/os/unix/ngx_process_cycle.c:617
> > > > #6  ngx_master_process_cycle (cycle=0x21f1990) at
> > > > src/os/unix/ngx_process_cycle.c:180
> > > > #7  0x0000000000406061 in main (argc=<value optimized out>,
> argv=<value
> > > > optimized out>) at src/core/nginx.c:405
> > > >
> > > >
> > > > (gdb) bt full
> > > > #0  ngx_http_upstream_handler (ev=0x7fc45735f8a8) at
> > > > src/http/ngx_http_upstream.c:915
> > > >         c = 0x23b4160
> > > >         r = 0x22a2780
> > > >         ctx = <error reading variable ctx (Cannot access memory at
> > > > address 0x20)>
> > > >         u = 0x0
> > >
> > > Could you please show results of the following gdb commands:
> > >
> > > fr 0
> > > p *ev
> > > p *((ngx_connection_t *) ev->data)
> > >
> > >
> > > Maxim Dounin
> > >
> > > _______________________________________________
> > > nginx mailing list
> > > nginx at nginx.org
> > > http://mailman.nginx.org/mailman/listinfo/nginx
> > >
>
> > _______________________________________________
> > nginx mailing list
> > nginx at nginx.org
> > http://mailman.nginx.org/mailman/listinfo/nginx
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110905/4122706c/attachment-0001.html>

From drdr.xp at gmail.com  Mon Sep  5 16:16:19 2011
From: drdr.xp at gmail.com (dr-dr xp)
Date: Tue, 6 Sep 2011 00:16:19 +0800
Subject: nginx 1.0.2 AIO problem with open_file_cache
In-Reply-To: <20110904163413.GR1137@mdounin.ru>
References: <CAMcFNRJv_KSkj+DLQzXsCT02Ktd-W5uFXRAOz3MQJ+vErKOsuQ@mail.gmail.com>
	<20110903100335.GO1137@mdounin.ru>
	<CAMcFNRJ_CbX-7OsonG20BBNF_59vKwFw1Y=34CjSxUg+0=1i0g@mail.gmail.com>
	<20110904163413.GR1137@mdounin.ru>
Message-ID: <CAMcFNRJdNyvEPTNLXNXinVT70q62vUR8a_HLn0VKBNFCcAhWwg@mail.gmail.com>

Thank you very much Dounin. With the patch it works fine now.




On Mon, Sep 5, 2011 at 12:34 AM, Maxim Dounin <mdounin at mdounin.ru> wrote:

> Hello!
>
> On Sat, Sep 03, 2011 at 11:06:24PM +0800, dr-dr xp wrote:
>
> > The attachment is the error.log with only the failed request. And stat
> > output of this file:
> >
> > stat
> >
> //data2/0d97e01d7fbc48e498897492f0863929/g260674/257/26fd3732cd9f92f2413d10206cd8b0c9-5157235
> >   File:
> >
> `//data2/0d97e01d7fbc48e498897492f0863929/g260674/257/26fd3732cd9f92f2413d10206cd8b0c9-5157235'
> >   Size: 51163499        Blocks: 100048     IO Block: 4096   regular file
> > Device: 821h/2081d      Inode: 464404482   Links: 1
> > Access: (0664/-rw-rw-r--)  Uid: (    1003/    www)   Gid: (    1003/
>  www)
> > Access: 2011-09-03 22:31:22.000000000 +0800
> > Modify: 2011-07-30 01:45:56.000000000 +0800
> > Change: 2011-07-30 01:45:56.000000000 +0800
> >
> >
> > Such error occurs randomly on random file. The only relavent thing seems
> > like to be open_file_cache directive.
> > File sizes are almost 5M to 20M.
> > Files had not been changed. I checked the modify time of the files
> involved
> > in these errors. At first I had thought it was caused by a file change
> too.
> > But not.
> >
> > In my cluster there are also some other similar nodes running with the
> same
> > open_file_cache parameters but without aio enabled(sendfile is used).
> They
> > have no such issue.
>
> Please try the attached patch.
>
> Maxim Dounin
>
> >
> >
> >
> >
> > On Sat, Sep 3, 2011 at 6:03 PM, Maxim Dounin <mdounin at mdounin.ru> wrote:
> >
> > > Hello!
> > >
> > > On Sat, Sep 03, 2011 at 05:46:39PM +0800, dr-dr xp wrote:
> > >
> > > > Hello everybody. I recently have had some problem with AIO support
> and
> > > need
> > > > some help or advice on it.
> > > >
> > > > With AIO on, everything was OK. But when I enabled open_file_cache,
> nginx
> > > > frequently closed client connection directly without sending back any
> > > data.
> > > >
> > > >
> > > > My platform is linux 2.6.30. CentOS 5.4;
> > > >
> > > > The nginx server that has this issue serves static file at backend.
> > > Another
> > > > nginx serves as proxy in front of it.
> > > > Backend nginx config is like this:
> > > > ...
> > > > aio            on;
> > > >
> > > >
> > > > directio       512;
> > > >
> > > >
> > > > output_buffers 1 128k;
> > > > ...
> > > >     location /files {
> > > >
> > > >
> > > >
> > > >         internal;
> > > >
> > > >
> > > >         alias /;
> > > >         open_file_cache max=50000 inactive=20s;
> > > >
> > > >
> > > >         open_file_cache_valid    30s;
> > > >
> > > >
> > > >         open_file_cache_min_uses 1;
> > > >
> > > >
> > > >         open_file_cache_errors   on;
> > > >
> > > >
> > > > With this config, on the frontend proxy server there are error logs
> like
> > > > this:
> > > >
> > > > 2011/09/03 17:01:19 [error] 12498#0: *78679759 *upstream prematurely
> > > closed
> > > > connection while reading response header from upstream*, client:
> > > > 221.193.212.77, server: , request: "GET .......... HTTP/1.1",
> upstream:
> > > > "http://
> > >
> **/file/871586e763bf4417bf8d475858deee2b/260110/0b47fc7f4d84e4bcb01a6d6d1fcacc94-3581749",
> > > > host.....
> > > >
> > > >
> > > > And I saw in tcpdump output the connection was closed very quickly by
> > > > backend nginx:
> > > >
> > > > 22:02:22.892327 IP 172.16.113.32.58271 > 172.16.228.63.irdmi: S
> > > > 1788188296:1788188296(0) win 5840 <mss 1460,sackOK,timestamp
> 666482608
> > > > 0,nop,wscale 7>
> > > > 22:02:22.895683 IP 172.16.228.63.irdmi > 172.16.113.32.58271: S
> > > > 350810413:350810413(0) ack 1788188297 win 5792 <mss
> 1460,sackOK,timestamp
> > > > 1948226601 66<
> > > > 22:02:22.895695 IP 172.16.113.32.58271 > 172.16.228.63.irdmi: . ack 1
> win
> > > 46
> > > > <nop,nop,timestamp 666482612 1948226601>
> > > > 22:02:22.895703 IP 172.16.113.32.58271 > 172.16.228.63.irdmi: P
> > > 1:181(180)
> > > > ack 1 win 46 <nop,nop,timestamp 666482612 1948226601>
> > > > 22:02:22.897680 IP 172.16.228.63.irdmi > 172.16.113.32.58271: . ack
> 181
> > > win
> > > > 54 <nop,nop,timestamp 1948226603 666482612>
> > > > 22:02:22.898348 IP 172.16.228.63.irdmi > 172.16.113.32.58271: F
> 1:1(0)
> > > ack
> > > > 181 win 54 <nop,nop,timestamp 1948226603 666482612>
> > > > 22:02:22.898473 IP 172.16.113.32.58271 > 172.16.228.63.irdmi: F
> > > 181:181(0)
> > > > ack 2 win 46 <nop,nop,timestamp 666482615 1948226603>
> > > > 22:02:22.899804 IP 172.16.228.63.irdmi > 172.16.113.32.58271: . ack
> 182
> > > win
> > > > 54 <nop,nop,timestamp 1948226605 666482615>
> > > >
> > > >
> > > > On the backend nginx, strace shows that io_getevents got an error
> *"-22"
> > > > EINVAL*.
> > > >
> > > >   [pid  5275] recvfrom(48, 0x7fffe4400ae7, 1, 2, 0, 0) = -1 EAGAIN
> > > (Resource
> > > > temporarily unavailable)
> > > >   [pid  5275] epoll_wait(6, {{EPOLLIN|EPOLLOUT, {u32=2445415233,
> > > > u64=139850875540289}}}, 512, 5672) = 1
> > > >   [pid  5275] recvfrom(201, "\1\6\0\1\0\337\1\0Status: 301 Moved
> > > Perman"...,
> > > > 4096, 0, NULL, NULL) = 256
> > > >   [pid  5275] close(201)                  = 0
> > > >   [pid  5275]
> > > >
> > >
> stat("//data1/871586e763bf4417bf8d475858deee2b/g268315/252/995f20f76d0d3871fdb5d71bbc92956c-27769564",
> > > > {st_mode=S_IFREG|0600, st_size=122507, ...<
> > > >   [pid  5275] io_submit(139850942226432, 1, {{0xf43718, 0, 0, 0,
> 40}}) =
> > > 1
> > > >
> > > >
> > > >   [pid  5275] epoll_wait(6, {{EPOLLIN, {u32=7173248, u64=7173248}}},
> 512,
> > > > 5671) = 1
> > > >
> > > >   [pid  5275] read(7, "\1\0\0\0\0\0\0\0", 8) = 8
> > > >   [pid  5275] *io_getevents(139850942226432, 1, 64, {{0xf43718,
> 0xf436d8,
> > > > -22, 0}}, {0, 0}) = 1*
> > > >   [pid  5275] write(24, "172.16.164.30 - - 172.16.228.63:"..., 338)
> =
> > > 338
> > > >   [pid  5275] close(48)
> > > >
> > > > And access.log shows that response length is ZERO:
> > > >
> > > > [03/Sep/2011:06:28:48 +0800] "GET
> > > >
> > >
> /file/871586e763bf4417bf8d475858deee2b/260284/ed31933859d27e33b8398deaea1d2ade-3969549
> > > > HTTP/1.0" 206 0 0.001 "-" "-" "-" Range:bytes=27459105-
> > > >
> > > >
> > > > After I removed those *open_file_cache* directives, No such error
> > > occurred.
> > > >
> > > > And the shorter *inactive* is,  the more often I get such errors.
> > > >
> > > >
> > > > Anybody has idea about avoiding this or fixing it?
> > >
> > > Could you please provide debug log?  See
> > > http://wiki.nginx.org/Debugging for more information.
> > >
> > > Additionally, some more info about file requested might be
> > > helpful: how big it is? whether it changed on backend during
> > > requests? if yes, is it changed atomically?
> > >
> > > Maxim Dounin
> > >
> > > _______________________________________________
> > > nginx mailing list
> > > nginx at nginx.org
> > > http://mailman.nginx.org/mailman/listinfo/nginx
> > >
> >
> >
> >
> > --
> > ????????????????
>
>
> > _______________________________________________
> > nginx mailing list
> > nginx at nginx.org
> > http://mailman.nginx.org/mailman/listinfo/nginx
>
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx
>



-- 
????????????????
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110906/ef0f00ea/attachment.html>

From mdounin at mdounin.ru  Mon Sep  5 16:38:07 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Mon, 5 Sep 2011 20:38:07 +0400
Subject: nginx 1.0.2 AIO problem with open_file_cache
In-Reply-To: <CAMcFNRJdNyvEPTNLXNXinVT70q62vUR8a_HLn0VKBNFCcAhWwg@mail.gmail.com>
References: <CAMcFNRJv_KSkj+DLQzXsCT02Ktd-W5uFXRAOz3MQJ+vErKOsuQ@mail.gmail.com>
	<20110903100335.GO1137@mdounin.ru>
	<CAMcFNRJ_CbX-7OsonG20BBNF_59vKwFw1Y=34CjSxUg+0=1i0g@mail.gmail.com>
	<20110904163413.GR1137@mdounin.ru>
	<CAMcFNRJdNyvEPTNLXNXinVT70q62vUR8a_HLn0VKBNFCcAhWwg@mail.gmail.com>
Message-ID: <20110905163807.GI1137@mdounin.ru>

Hello!

On Tue, Sep 06, 2011 at 12:16:19AM +0800, dr-dr xp wrote:

> Thank you very much Dounin. With the patch it works fine now.

Thanks for testing, the patch was committed and will be included 
in next release.

Maxim Dounin

> 
> 
> 
> 
> On Mon, Sep 5, 2011 at 12:34 AM, Maxim Dounin <mdounin at mdounin.ru> wrote:
> 
> > Hello!
> >
> > On Sat, Sep 03, 2011 at 11:06:24PM +0800, dr-dr xp wrote:
> >
> > > The attachment is the error.log with only the failed request. And stat
> > > output of this file:
> > >
> > > stat
> > >
> > //data2/0d97e01d7fbc48e498897492f0863929/g260674/257/26fd3732cd9f92f2413d10206cd8b0c9-5157235
> > >   File:
> > >
> > `//data2/0d97e01d7fbc48e498897492f0863929/g260674/257/26fd3732cd9f92f2413d10206cd8b0c9-5157235'
> > >   Size: 51163499        Blocks: 100048     IO Block: 4096   regular file
> > > Device: 821h/2081d      Inode: 464404482   Links: 1
> > > Access: (0664/-rw-rw-r--)  Uid: (    1003/    www)   Gid: (    1003/
> >  www)
> > > Access: 2011-09-03 22:31:22.000000000 +0800
> > > Modify: 2011-07-30 01:45:56.000000000 +0800
> > > Change: 2011-07-30 01:45:56.000000000 +0800
> > >
> > >
> > > Such error occurs randomly on random file. The only relavent thing seems
> > > like to be open_file_cache directive.
> > > File sizes are almost 5M to 20M.
> > > Files had not been changed. I checked the modify time of the files
> > involved
> > > in these errors. At first I had thought it was caused by a file change
> > too.
> > > But not.
> > >
> > > In my cluster there are also some other similar nodes running with the
> > same
> > > open_file_cache parameters but without aio enabled(sendfile is used).
> > They
> > > have no such issue.
> >
> > Please try the attached patch.
> >
> > Maxim Dounin
> >
> > >
> > >
> > >
> > >
> > > On Sat, Sep 3, 2011 at 6:03 PM, Maxim Dounin <mdounin at mdounin.ru> wrote:
> > >
> > > > Hello!
> > > >
> > > > On Sat, Sep 03, 2011 at 05:46:39PM +0800, dr-dr xp wrote:
> > > >
> > > > > Hello everybody. I recently have had some problem with AIO support
> > and
> > > > need
> > > > > some help or advice on it.
> > > > >
> > > > > With AIO on, everything was OK. But when I enabled open_file_cache,
> > nginx
> > > > > frequently closed client connection directly without sending back any
> > > > data.
> > > > >
> > > > >
> > > > > My platform is linux 2.6.30. CentOS 5.4;
> > > > >
> > > > > The nginx server that has this issue serves static file at backend.
> > > > Another
> > > > > nginx serves as proxy in front of it.
> > > > > Backend nginx config is like this:
> > > > > ...
> > > > > aio            on;
> > > > >
> > > > >
> > > > > directio       512;
> > > > >
> > > > >
> > > > > output_buffers 1 128k;
> > > > > ...
> > > > >     location /files {
> > > > >
> > > > >
> > > > >
> > > > >         internal;
> > > > >
> > > > >
> > > > >         alias /;
> > > > >         open_file_cache max=50000 inactive=20s;
> > > > >
> > > > >
> > > > >         open_file_cache_valid    30s;
> > > > >
> > > > >
> > > > >         open_file_cache_min_uses 1;
> > > > >
> > > > >
> > > > >         open_file_cache_errors   on;
> > > > >
> > > > >
> > > > > With this config, on the frontend proxy server there are error logs
> > like
> > > > > this:
> > > > >
> > > > > 2011/09/03 17:01:19 [error] 12498#0: *78679759 *upstream prematurely
> > > > closed
> > > > > connection while reading response header from upstream*, client:
> > > > > 221.193.212.77, server: , request: "GET .......... HTTP/1.1",
> > upstream:
> > > > > "http://
> > > >
> > **/file/871586e763bf4417bf8d475858deee2b/260110/0b47fc7f4d84e4bcb01a6d6d1fcacc94-3581749",
> > > > > host.....
> > > > >
> > > > >
> > > > > And I saw in tcpdump output the connection was closed very quickly by
> > > > > backend nginx:
> > > > >
> > > > > 22:02:22.892327 IP 172.16.113.32.58271 > 172.16.228.63.irdmi: S
> > > > > 1788188296:1788188296(0) win 5840 <mss 1460,sackOK,timestamp
> > 666482608
> > > > > 0,nop,wscale 7>
> > > > > 22:02:22.895683 IP 172.16.228.63.irdmi > 172.16.113.32.58271: S
> > > > > 350810413:350810413(0) ack 1788188297 win 5792 <mss
> > 1460,sackOK,timestamp
> > > > > 1948226601 66<
> > > > > 22:02:22.895695 IP 172.16.113.32.58271 > 172.16.228.63.irdmi: . ack 1
> > win
> > > > 46
> > > > > <nop,nop,timestamp 666482612 1948226601>
> > > > > 22:02:22.895703 IP 172.16.113.32.58271 > 172.16.228.63.irdmi: P
> > > > 1:181(180)
> > > > > ack 1 win 46 <nop,nop,timestamp 666482612 1948226601>
> > > > > 22:02:22.897680 IP 172.16.228.63.irdmi > 172.16.113.32.58271: . ack
> > 181
> > > > win
> > > > > 54 <nop,nop,timestamp 1948226603 666482612>
> > > > > 22:02:22.898348 IP 172.16.228.63.irdmi > 172.16.113.32.58271: F
> > 1:1(0)
> > > > ack
> > > > > 181 win 54 <nop,nop,timestamp 1948226603 666482612>
> > > > > 22:02:22.898473 IP 172.16.113.32.58271 > 172.16.228.63.irdmi: F
> > > > 181:181(0)
> > > > > ack 2 win 46 <nop,nop,timestamp 666482615 1948226603>
> > > > > 22:02:22.899804 IP 172.16.228.63.irdmi > 172.16.113.32.58271: . ack
> > 182
> > > > win
> > > > > 54 <nop,nop,timestamp 1948226605 666482615>
> > > > >
> > > > >
> > > > > On the backend nginx, strace shows that io_getevents got an error
> > *"-22"
> > > > > EINVAL*.
> > > > >
> > > > >   [pid  5275] recvfrom(48, 0x7fffe4400ae7, 1, 2, 0, 0) = -1 EAGAIN
> > > > (Resource
> > > > > temporarily unavailable)
> > > > >   [pid  5275] epoll_wait(6, {{EPOLLIN|EPOLLOUT, {u32=2445415233,
> > > > > u64=139850875540289}}}, 512, 5672) = 1
> > > > >   [pid  5275] recvfrom(201, "\1\6\0\1\0\337\1\0Status: 301 Moved
> > > > Perman"...,
> > > > > 4096, 0, NULL, NULL) = 256
> > > > >   [pid  5275] close(201)                  = 0
> > > > >   [pid  5275]
> > > > >
> > > >
> > stat("//data1/871586e763bf4417bf8d475858deee2b/g268315/252/995f20f76d0d3871fdb5d71bbc92956c-27769564",
> > > > > {st_mode=S_IFREG|0600, st_size=122507, ...<
> > > > >   [pid  5275] io_submit(139850942226432, 1, {{0xf43718, 0, 0, 0,
> > 40}}) =
> > > > 1
> > > > >
> > > > >
> > > > >   [pid  5275] epoll_wait(6, {{EPOLLIN, {u32=7173248, u64=7173248}}},
> > 512,
> > > > > 5671) = 1
> > > > >
> > > > >   [pid  5275] read(7, "\1\0\0\0\0\0\0\0", 8) = 8
> > > > >   [pid  5275] *io_getevents(139850942226432, 1, 64, {{0xf43718,
> > 0xf436d8,
> > > > > -22, 0}}, {0, 0}) = 1*
> > > > >   [pid  5275] write(24, "172.16.164.30 - - 172.16.228.63:"..., 338)
> > =
> > > > 338
> > > > >   [pid  5275] close(48)
> > > > >
> > > > > And access.log shows that response length is ZERO:
> > > > >
> > > > > [03/Sep/2011:06:28:48 +0800] "GET
> > > > >
> > > >
> > /file/871586e763bf4417bf8d475858deee2b/260284/ed31933859d27e33b8398deaea1d2ade-3969549
> > > > > HTTP/1.0" 206 0 0.001 "-" "-" "-" Range:bytes=27459105-
> > > > >
> > > > >
> > > > > After I removed those *open_file_cache* directives, No such error
> > > > occurred.
> > > > >
> > > > > And the shorter *inactive* is,  the more often I get such errors.
> > > > >
> > > > >
> > > > > Anybody has idea about avoiding this or fixing it?
> > > >
> > > > Could you please provide debug log?  See
> > > > http://wiki.nginx.org/Debugging for more information.
> > > >
> > > > Additionally, some more info about file requested might be
> > > > helpful: how big it is? whether it changed on backend during
> > > > requests? if yes, is it changed atomically?
> > > >
> > > > Maxim Dounin
> > > >
> > > > _______________________________________________
> > > > nginx mailing list
> > > > nginx at nginx.org
> > > > http://mailman.nginx.org/mailman/listinfo/nginx
> > > >
> > >
> > >
> > >
> > > --
> > > ????????????????
> >
> >
> > > _______________________________________________
> > > nginx mailing list
> > > nginx at nginx.org
> > > http://mailman.nginx.org/mailman/listinfo/nginx
> >
> >
> > _______________________________________________
> > nginx mailing list
> > nginx at nginx.org
> > http://mailman.nginx.org/mailman/listinfo/nginx
> >
> 
> 
> 
> -- 
> ????????????????

> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx


From mdounin at mdounin.ru  Mon Sep  5 18:02:10 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Mon, 5 Sep 2011 22:02:10 +0400
Subject: upstream keepalive - call for testing
In-Reply-To: <CAENo=s1GJzGJ8OPbNBvQhTEECN=jKFAtgYN=SMx=2o9KS+W02A@mail.gmail.com>
References: <20110801160725.GH1137@mdounin.ru>
	<0d5cc130a417bb62098701656fd44e42.NginxMailingListEnglish@forum.nginx.org>
	<20110904182137.GW1137@mdounin.ru>
	<CAENo=s3G3XBrYN7K9G6HzHZkBWYHEEi4DLCH8G-7_yC8dJyu-w@mail.gmail.com>
	<20110905070813.GY1137@mdounin.ru>
	<CAENo=s1GJzGJ8OPbNBvQhTEECN=jKFAtgYN=SMx=2o9KS+W02A@mail.gmail.com>
Message-ID: <20110905180210.GL1137@mdounin.ru>

Hello!

On Mon, Sep 05, 2011 at 11:42:31PM +0800, ????? wrote:

> (gdb) fr 0
> #0  ngx_http_upstream_handler (ev=0x7fc45735f8a8)
>     at src/http/ngx_http_upstream.c:915
> 915    ctx->current_request = r;
> (gdb) p ngx_cycle->log
> $1 = (ngx_log_t *) 0x21f19a8
> (gdb) p *r
> $2 = {signature = 51686928, connection = 0x23b4160, ctx = 0x0,

[...]

This looks like memory corruption, but unfortunately I don't see 
any traces of the real cause.  My best quess is improper handling 
of proxy_ignore_client_abort as fixed in 1.1.2.  Please try 1.1.2 
with patches from

http://nginx.org/patches/patch-nginx-keepalive-full-5.txt

It already includes upstream keepalive module, as well as all 
other upstream-keepalive related fixes.  See here for details:

http://mailman.nginx.org/pipermail/nginx-devel/2011-September/001147.html

Maxim Dounin


From nginx-forum at nginx.us  Mon Sep  5 18:13:03 2011
From: nginx-forum at nginx.us (locojohn)
Date: Mon, 05 Sep 2011 14:13:03 -0400
Subject: Help with rewrite
In-Reply-To: <461adae45a9d3ba0d9d4ed211a89df13.NginxMailingListEnglish@forum.nginx.org>
References: <461adae45a9d3ba0d9d4ed211a89df13.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <2daf92f694ebb063350c5d57b27c59a1.NginxMailingListEnglish@forum.nginx.org>

the first "if" can be replaced with the following:


location ^.+\.php$ {
  include fastcgi_params;
  fastcgi_param SCRIPT_FILENAME $document_root/vbseo.php;
  fastcgi_pass ....;  # your fastcgi handler
}

not sure that the second "if" is needed, unless they meant to handle all
non-existing files (404 errors) with vbseo.php.  this can be done with
the following:

location / {
    error_page    404 = /vbseo.php;
}

Andrejs

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,214761,214833#msg-214833


From nginx-forum at nginx.us  Mon Sep  5 18:17:25 2011
From: nginx-forum at nginx.us (locojohn)
Date: Mon, 05 Sep 2011 14:17:25 -0400
Subject: Can't fix "No input file specified." with php-fastcgi
In-Reply-To: <110b6d83f8fc0d1ab9b8cd761506b946.NginxMailingListEnglish@forum.nginx.org>
References: <2a1a1ceedd33352155268607634ce01f@mail.gmail.com>
	<110b6d83f8fc0d1ab9b8cd761506b946.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <4a7c97edb72696de67eabdbd166f82b1.NginxMailingListEnglish@forum.nginx.org>

Please post your nginx.conf here, because the error can be in the nginx
configuration.  Also make sure to have cgi.fix_pathinfo=1 in your
php.ini.

Andrejs

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,214746,214834#msg-214834


From nginx-forum at nginx.us  Mon Sep  5 18:21:09 2011
From: nginx-forum at nginx.us (loganbest)
Date: Mon, 05 Sep 2011 14:21:09 -0400
Subject: Can't fix "No input file specified." with php-fastcgi
In-Reply-To: <4a7c97edb72696de67eabdbd166f82b1.NginxMailingListEnglish@forum.nginx.org>
References: <2a1a1ceedd33352155268607634ce01f@mail.gmail.com>
	<110b6d83f8fc0d1ab9b8cd761506b946.NginxMailingListEnglish@forum.nginx.org>
	<4a7c97edb72696de67eabdbd166f82b1.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <c0356159b6e539ebf095ee4b6f8b76b2.NginxMailingListEnglish@forum.nginx.org>

cgi.fix is already set in php.ini


#user  nobody;
worker_processes  3;

#error_log  logs/error.log;
#error_log  logs/error.log  notice;
#error_log  logs/error.log  info;

#pid        logs/nginx.pid;

events {
    worker_connections  1024;
}

http {
    passenger_root /usr/local/lib/ruby/gems/1.8/gems/passenger-3.0.8;
    passenger_ruby /usr/local/bin/ruby;

    include       mime.types;
    default_type  application/octet-stream;

    #log_format  main  '$remote_addr - $remote_user [$time_local]
"$request" '
    #                  '$status $body_bytes_sent "$http_referer" '
    #                  '"$http_user_agent" "$http_x_forwarded_for"';

    #access_log  logs/access.log  main;

    sendfile        on;
    #tcp_nopush     on;

    #keepalive_timeout  0;
    keepalive_timeout  65;

    #gzip  on;

    server {
        listen       80;
        server_name  localhost;

        #charset koi8-r;

        #access_log  logs/host.access.log  main;
        location / {
                root   /var/www/contest;
                index  index.php index.html index.htm;
        }

        location /staging {
                root  
/var/www/FreeFallHighScoreWebStuff/current/public;
                passenger_enabled on;
                passenger_min_instances 3;
        }

        location /staging_assets {
                root  
/var/www/FreeFallHighScoreWebStuff/current/public;
        }

        #error_page  404              /404.html;

        # redirect server error pages to the static page /50x.html
        #
        error_page   500 502 503 504  /50x.html;
        location = /50x.html {
            root   html;
        }

        location /gitphp {
                root /var/www;
                index index.php index.html index.htm;
        }


        ## Parse all .php file in the /var/www directory
        location ~ .php$ {
                root /var/www;
                fastcgi_split_path_info ^(.+\.php)(.*)$;
                fastcgi_pass   backend;
                fastcgi_index  index.php;
                fastcgi_param  SCRIPT_FILENAME 
/var/www$fastcgi_script_name;
                include fastcgi_params;
                fastcgi_param  QUERY_STRING     $query_string;
                fastcgi_param  REQUEST_METHOD   $request_method;
                fastcgi_param  CONTENT_TYPE     $content_type;
                fastcgi_param  CONTENT_LENGTH   $content_length;
                fastcgi_intercept_errors        on;
                fastcgi_ignore_client_abort     off;
                fastcgi_connect_timeout 60;
                fastcgi_send_timeout 180;
                fastcgi_read_timeout 180;
                fastcgi_buffer_size 128k;
                fastcgi_buffers 4 256k;
                fastcgi_busy_buffers_size 256k;
                fastcgi_temp_file_write_size 256k;
        }

        location ~ /\.ht {
            deny  all;
        }
    }

upstream backend {
        server 127.0.0.1:9000;
}
}

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,214746,214835#msg-214835


From kworthington at gmail.com  Mon Sep  5 20:44:25 2011
From: kworthington at gmail.com (Kevin Worthington)
Date: Mon, 5 Sep 2011 16:44:25 -0400
Subject: nginx-1.0.6
In-Reply-To: <CAPihCJUqakUvDDiOhnnFv-yEARtSg=Q=wqFaOt-T0Z_gJx=ejQ@mail.gmail.com>
References: <20110829144251.GE21595@nginx.com>
	<CAPihCJUqakUvDDiOhnnFv-yEARtSg=Q=wqFaOt-T0Z_gJx=ejQ@mail.gmail.com>
Message-ID: <CAGo79UWH=uw03z6WhNimyOnjC-C0xrariKBkWAgdShUmhJS28w@mail.gmail.com>

Hello!

Now available for free download: Nginx 1.0.6 For Windows http://goo.gl/3H1cJ

These builds were built using Cygwin in order to support legacy users.
Official native Windows builds are available at nginx.org

Thanks,
Kevin
--
Kevin Worthington
kworthington (AT} gmail (DoT] com
http://www.kevinworthington.com/



On Mon, Aug 29, 2011 at 11:26 AM, Joe <frumentius at gmail.com> wrote:
> Thanks Igor.
>
>
> Regards,
> Joe
>
>
> On Mon, Aug 29, 2011 at 9:42 PM, Igor Sysoev <igor at sysoev.ru> wrote:
>>
>> Changes with nginx 1.0.6 ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? 29 Aug
>> 2011
>>
>> ? ?*) Feature: cache loader run time decrease.
>>
>> ? ?*) Feature: loading time decrease of configuration with large number of
>> ? ? ? HTTPS sites.
>>
>> ? ?*) Feature: now nginx supports ECDHE key exchange ciphers.
>> ? ? ? Thanks to Adrian Kotelba.
>>
>> ? ?*) Feature: the "lingering_close" directive.
>>
>> ? ?*) Feature: now shared zones and caches use POSIX semaphores on
>> Solaris.
>> ? ? ? Thanks to Den Ivanov.
>>
>> ? ?*) Bugfix: nginx could not be built on Linux 3.0.
>>
>> ? ?*) Bugfix: a segmentation fault might occur in a worker process if
>> ? ? ? "fastcgi/scgi/uwsgi_param" directives were used with values starting
>> ? ? ? with "HTTP_"; the bug had appeared in 0.8.40.
>>
>> ? ?*) Bugfix: in closing connection for pipelined requests.
>>
>> ? ?*) Bugfix: nginx did not disable gzipping if client sent "gzip;q=0" in
>> ? ? ? "Accept-Encoding" request header line.
>>
>> ? ?*) Bugfix: in timeout in unbuffered proxied mode.
>>
>> ? ?*) Bugfix: memory leaks when a "proxy_pass" directive contains
>> variables
>> ? ? ? and proxies to an HTTPS backend.
>>
>> ? ?*) Bugfix: in parameter validaiton of a "proxy_pass" directive with
>> ? ? ? variables.
>> ? ? ? Thanks to Lanshun Zhou.
>>
>> ? ?*) Bugfix: SSL did not work on QNX.
>>
>> ? ?*) Bugfix: SSL modules could not be built by gcc 4.6 without
>> ? ? ? --with-debug option.
>>
>> --
>> Igor Sysoev
>>
>> _______________________________________________
>> nginx mailing list
>> nginx at nginx.org
>> http://mailman.nginx.org/mailman/listinfo/nginx
>
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx
>
>


From hagaia at qwilt.com  Mon Sep  5 22:21:00 2011
From: hagaia at qwilt.com (Hagai Avrahami)
Date: Tue, 6 Sep 2011 01:21:00 +0300
Subject: Help with AIO
Message-ID: <CAN-ZXpuTdDMmQ7rgo_QP=1W53+j1qc-Bqa=f4=QLNskOubY2dw@mail.gmail.com>

Hi

I configured Nginx with *aio *mode *on *and 1 *worker process*
In benchmarking I saw limit of 1024 concurrent sessions even configured to
4096 *worker connections*
When I looked in the source code I could see that  io_setup is called with
fix value of 1024 max events
can anyone please explain me the hard coded limit
Is it OK to change it or it is derived from other limit I am not aware of
any help will be appreciated

thanks
Hagai






-- 

*Hagai Avrahami*
Qwilt | Work: +972-72-2221644| Mobile: +972-54-4895656 |
hagaia at qwilt.com<yoav at qwilt.com>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110906/20bc66c9/attachment.html>

From lsd at wootangent.net  Tue Sep  6 02:45:46 2011
From: lsd at wootangent.net (Leigh Dyer)
Date: Tue, 06 Sep 2011 12:45:46 +1000
Subject: Issue with If-Range header and X-Accel-Redirect requests with
	upstream Last-Modified header
Message-ID: <4E65895A.4020506@wootangent.net>

Hi,

I have nginx 1.0.5 set up as a proxy in front of a Ruby/Ramaze 
application, using X-Accel-Redirect to efficiently handle sending large 
binary files by handing those transfers over to nginx. Range requests 
against this setup work just fine, but "If-Range" headers from the 
client are ignored -- the server always returns the requested range, 
even when the If-Range date doesn't match the file's Last-Modified date.

The problem seems to be due to this line in our config:

add_header Last-Modified $upstream_http_last_modified;

This sets the Last-Modified date on file sent to the client to the value 
sent to nginx from Ruby. With this line in the config, I get the 
If-Range issue described above. If I comment this out, Last-Modified is 
taken from the file's mtime instead, but If-Range works.

Is there any chance of getting this fixed?

Thanks
Leigh


From amoiz.shine at gmail.com  Tue Sep  6 02:54:32 2011
From: amoiz.shine at gmail.com (Sharl.Jimh.Tsin)
Date: Tue, 06 Sep 2011 10:54:32 +0800
Subject: nginx-1.0.6
In-Reply-To: <CAGo79UWH=uw03z6WhNimyOnjC-C0xrariKBkWAgdShUmhJS28w@mail.gmail.com>
References: <20110829144251.GE21595@nginx.com>
	<CAPihCJUqakUvDDiOhnnFv-yEARtSg=Q=wqFaOt-T0Z_gJx=ejQ@mail.gmail.com>
	<CAGo79UWH=uw03z6WhNimyOnjC-C0xrariKBkWAgdShUmhJS28w@mail.gmail.com>
Message-ID: <1315277672.1958.3.camel@sharl-desktop>

? 2011-09-05?? 16:44 -0400?Kevin Worthington???
> Hello!
> 
> Now available for free download: Nginx 1.0.6 For Windows http://goo.gl/3H1cJ
> 
> These builds were built using Cygwin in order to support legacy users.
> Official native Windows builds are available at nginx.org
> 
> Thanks,
> Kevin
> --
> Kevin Worthington
> kworthington (AT} gmail (DoT] com
> http://www.kevinworthington.com/
> 
> 
> 
> On Mon, Aug 29, 2011 at 11:26 AM, Joe <frumentius at gmail.com> wrote:
> > Thanks Igor.
> >
> >
> > Regards,
> > Joe
> >
> >
> > On Mon, Aug 29, 2011 at 9:42 PM, Igor Sysoev <igor at sysoev.ru> wrote:
> >>
> >> Changes with nginx 1.0.6                                         29 Aug
> >> 2011
> >>
> >>    *) Feature: cache loader run time decrease.
> >>
> >>    *) Feature: loading time decrease of configuration with large number of
> >>       HTTPS sites.
> >>
> >>    *) Feature: now nginx supports ECDHE key exchange ciphers.
> >>       Thanks to Adrian Kotelba.
> >>
> >>    *) Feature: the "lingering_close" directive.
> >>
> >>    *) Feature: now shared zones and caches use POSIX semaphores on
> >> Solaris.
> >>       Thanks to Den Ivanov.
> >>
> >>    *) Bugfix: nginx could not be built on Linux 3.0.
> >>
> >>    *) Bugfix: a segmentation fault might occur in a worker process if
> >>       "fastcgi/scgi/uwsgi_param" directives were used with values starting
> >>       with "HTTP_"; the bug had appeared in 0.8.40.
> >>
> >>    *) Bugfix: in closing connection for pipelined requests.
> >>
> >>    *) Bugfix: nginx did not disable gzipping if client sent "gzip;q=0" in
> >>       "Accept-Encoding" request header line.
> >>
> >>    *) Bugfix: in timeout in unbuffered proxied mode.
> >>
> >>    *) Bugfix: memory leaks when a "proxy_pass" directive contains
> >> variables
> >>       and proxies to an HTTPS backend.
> >>
> >>    *) Bugfix: in parameter validaiton of a "proxy_pass" directive with
> >>       variables.
> >>       Thanks to Lanshun Zhou.
> >>
> >>    *) Bugfix: SSL did not work on QNX.
> >>
> >>    *) Bugfix: SSL modules could not be built by gcc 4.6 without
> >>       --with-debug option.
> >>
> >> --
> >> Igor Sysoev
> >>
> >> _______________________________________________
> >> nginx mailing list
> >> nginx at nginx.org
> >> http://mailman.nginx.org/mailman/listinfo/nginx
> >
> >
> > _______________________________________________
> > nginx mailing list
> > nginx at nginx.org
> > http://mailman.nginx.org/mailman/listinfo/nginx
> >
> >
> 
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx

thanks for your work!

-- 
Best regards,
Sharl.Jimh.Tsin (From China **Obviously Taiwan INCLUDED**)

Using Gmail? Please read this important notice:
http://www.fsf.org/campaigns/jstrap/gmail?10073.


From nginx-forum at nginx.us  Tue Sep  6 06:39:41 2011
From: nginx-forum at nginx.us (magicbear)
Date: Tue, 06 Sep 2011 02:39:41 -0400
Subject: upstream keepalive - call for testing
In-Reply-To: <20110905180210.GL1137@mdounin.ru>
References: <20110905180210.GL1137@mdounin.ru>
Message-ID: <870c8a757b4e464870698eefa90f49d0.NginxMailingListEnglish@forum.nginx.org>

ok, I will try that at another dedicated server, thanks for your Helps.

MagicBear

Maxim Dounin Wrote:
-------------------------------------------------------
> Hello!
> 
> On Mon, Sep 05, 2011 at 11:42:31PM +0800,
> ????? wrote:
> 
> > (gdb) fr 0
> > #0  ngx_http_upstream_handler
> (ev=0x7fc45735f8a8)
> >     at src/http/ngx_http_upstream.c:915
> > 915    ctx->current_request = r;
> > (gdb) p ngx_cycle->log
> > $1 = (ngx_log_t *) 0x21f19a8
> > (gdb) p *r
> > $2 = {signature = 51686928, connection =
> 0x23b4160, ctx = 0x0,
> 
> [...]
> 
> This looks like memory corruption, but
> unfortunately I don't see 
> any traces of the real cause.  My best quess is
> improper handling 
> of proxy_ignore_client_abort as fixed in 1.1.2. 
> Please try 1.1.2 
> with patches from
> 
> http://nginx.org/patches/patch-nginx-keepalive-ful
> l-5.txt
> 
> It already includes upstream keepalive module, as
> well as all 
> other upstream-keepalive related fixes.  See here
> for details:
> 
> http://mailman.nginx.org/pipermail/nginx-devel/201
> 1-September/001147.html
> 
> Maxim Dounin
> 
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,213207,214853#msg-214853


From mdounin at mdounin.ru  Tue Sep  6 06:50:18 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Tue, 6 Sep 2011 10:50:18 +0400
Subject: Issue with If-Range header and X-Accel-Redirect requests with
	upstream Last-Modified header
In-Reply-To: <4E65895A.4020506@wootangent.net>
References: <4E65895A.4020506@wootangent.net>
Message-ID: <20110906065018.GQ1137@mdounin.ru>

Hello!

On Tue, Sep 06, 2011 at 12:45:46PM +1000, Leigh Dyer wrote:

> Hi,
> 
> I have nginx 1.0.5 set up as a proxy in front of a Ruby/Ramaze
> application, using X-Accel-Redirect to efficiently handle sending
> large binary files by handing those transfers over to nginx. Range
> requests against this setup work just fine, but "If-Range" headers
> from the client are ignored -- the server always returns the
> requested range, even when the If-Range date doesn't match the
> file's Last-Modified date.
> 
> The problem seems to be due to this line in our config:
> 
> add_header Last-Modified $upstream_http_last_modified;
> 
> This sets the Last-Modified date on file sent to the client to the
> value sent to nginx from Ruby. With this line in the config, I get
> the If-Range issue described above. If I comment this out,
> Last-Modified is taken from the file's mtime instead, but If-Range
> works.
> 
> Is there any chance of getting this fixed?

Please try the attached patch.

Maxim Dounin
-------------- next part --------------
# HG changeset patch
# User Maxim Dounin <mdounin at mdounin.ru>
# Date 1315291748 -14400
# Node ID e51619385db9694030b9614c833a2b2504b377c9
# Parent  014764a85840606c90317e9f44f2b9fa139cbc8b
Handling of If-Range with add_header Last-Modified.

diff --git a/src/http/modules/ngx_http_headers_filter_module.c b/src/http/modules/ngx_http_headers_filter_module.c
--- a/src/http/modules/ngx_http_headers_filter_module.c
+++ b/src/http/modules/ngx_http_headers_filter_module.c
@@ -369,7 +369,8 @@ ngx_http_set_last_modified(ngx_http_requ
         old = NULL;
     }
 
-    r->headers_out.last_modified_time = -1;
+    r->headers_out.last_modified_time = ngx_http_parse_time(value->data,
+                                                            value->len);
 
     if (old == NULL || *old == NULL) {
 
@@ -382,6 +383,8 @@ ngx_http_set_last_modified(ngx_http_requ
             return NGX_ERROR;
         }
 
+        *old = h;
+
     } else {
         h = *old;
 
diff --git a/src/http/modules/ngx_http_range_filter_module.c b/src/http/modules/ngx_http_range_filter_module.c
--- a/src/http/modules/ngx_http_range_filter_module.c
+++ b/src/http/modules/ngx_http_range_filter_module.c
@@ -173,7 +173,11 @@ ngx_http_range_header_filter(ngx_http_re
         goto next_filter;
     }
 
-    if (r->headers_in.if_range && r->headers_out.last_modified_time != -1) {
+    if (r->headers_in.if_range) {
+
+        if (r->headers_out.last_modified_time == (time_t) -1) {
+            goto next_filter;
+        }
 
         if_range = ngx_http_parse_time(r->headers_in.if_range->value.data,
                                        r->headers_in.if_range->value.len);

From lsd at wootangent.net  Tue Sep  6 06:52:41 2011
From: lsd at wootangent.net (Leigh Dyer)
Date: Tue, 06 Sep 2011 16:52:41 +1000
Subject: Issue with If-Range header and X-Accel-Redirect requests with
	upstream Last-Modified header
In-Reply-To: <4E65895A.4020506@wootangent.net>
References: <4E65895A.4020506@wootangent.net>
Message-ID: <4E65C339.8090801@wootangent.net>

On 6/09/11 12:45 PM, Leigh Dyer wrote:
> Hi,
>
> I have nginx 1.0.5 set up as a proxy in front of a Ruby/Ramaze
> application, using X-Accel-Redirect to efficiently handle sending large
> binary files by handing those transfers over to nginx. Range requests
> against this setup work just fine, but "If-Range" headers from the
> client are ignored -- the server always returns the requested range,
> even when the If-Range date doesn't match the file's Last-Modified date.

Hi again,

I spent some time looking in to this and created a patch that seems to 
fix the problem for me. The issue was in "ngx_http_set_last_modified" in 
http/modules/ngx_http_headers_filter_module.c -- last_modified_time gets 
set to -1 and is then never updated with the new value.

I have no idea how correct this patch is, so I'd greatly appreciate it 
if someone could review this for me.

Thanks
Leigh

-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: nginx-1.0.5-ifrange-add_header-lastmodified-fix.patch
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110906/42885a33/attachment.ksh>

From lsd at wootangent.net  Tue Sep  6 07:00:47 2011
From: lsd at wootangent.net (Leigh Dyer)
Date: Tue, 06 Sep 2011 17:00:47 +1000
Subject: Issue with If-Range header and X-Accel-Redirect requests with
	upstream Last-Modified header
In-Reply-To: <20110906065018.GQ1137@mdounin.ru>
References: <4E65895A.4020506@wootangent.net>
	<20110906065018.GQ1137@mdounin.ru>
Message-ID: <4E65C51F.4090900@wootangent.net>

On 6/09/11 4:50 PM, Maxim Dounin wrote:
> Hello!
>
> Please try the attached patch.

Thanks Maxim -- your patch has fixed the problem for me. I feel a little 
silly for spending the time to come up with my own patch, but at least I 
know more about nginx now than I did this morning :)

Thanks
Leigh

>
> Maxim Dounin
>
>
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx


From andrew at nginx.com  Tue Sep  6 07:02:39 2011
From: andrew at nginx.com (Andrew Alexeev)
Date: Tue, 6 Sep 2011 11:02:39 +0400
Subject: Issue with If-Range header and X-Accel-Redirect requests with
	upstream Last-Modified header
In-Reply-To: <4E65C51F.4090900@wootangent.net>
References: <4E65895A.4020506@wootangent.net>
	<20110906065018.GQ1137@mdounin.ru>
	<4E65C51F.4090900@wootangent.net>
Message-ID: <46444432-D6C1-49FE-A1E0-98ED91544375@nginx.com>

Leigh,

You don't need to be silly at all. Thanks a lot for spotting this one.

-- 
Andrew Alexeev
w. http://nginx.org

On Sep 6, 2011, at 11:00 AM, Leigh Dyer wrote:

> On 6/09/11 4:50 PM, Maxim Dounin wrote:
>> Hello!
>> 
>> Please try the attached patch.
> 
> Thanks Maxim -- your patch has fixed the problem for me. I feel a little silly for spending the time to come up with my own patch, but at least I know more about nginx now than I did this morning :)
> 
> Thanks
> Leigh
> 
>> 
>> Maxim Dounin
>> 
>> 
>> 
>> _______________________________________________
>> nginx mailing list
>> nginx at nginx.org
>> http://mailman.nginx.org/mailman/listinfo/nginx
> 
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx
> 


From mdounin at mdounin.ru  Tue Sep  6 07:33:52 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Tue, 6 Sep 2011 11:33:52 +0400
Subject: Help with AIO
In-Reply-To: <CAN-ZXpuTdDMmQ7rgo_QP=1W53+j1qc-Bqa=f4=QLNskOubY2dw@mail.gmail.com>
References: <CAN-ZXpuTdDMmQ7rgo_QP=1W53+j1qc-Bqa=f4=QLNskOubY2dw@mail.gmail.com>
Message-ID: <20110906073352.GR1137@mdounin.ru>

Hello!

On Tue, Sep 06, 2011 at 01:21:00AM +0300, Hagai Avrahami wrote:

> Hi
> 
> I configured Nginx with *aio *mode *on *and 1 *worker process*
> In benchmarking I saw limit of 1024 concurrent sessions even configured to
> 4096 *worker connections*

Sure it's not limitation of the benchmarking tool you use?  Most 
tools out there use select() and hence limited to 1024 
concurrent connections.

> When I looked in the source code I could see that  io_setup is called with
> fix value of 1024 max events
> can anyone please explain me the hard coded limit
> Is it OK to change it or it is derived from other limit I am not aware of
> any help will be appreciated

io_setup() maxevents is a limit on number of aio requests allowed 
to be submitted simulteneously; if nginx hits this limit, it will 
fallback to normal blocking read.

If you are going to change it, make sure to keep total for all 
processes lower than /proc/sys/fs/aio-max-nr.

Maxim Dounin


From hagaia at qwilt.com  Tue Sep  6 07:46:56 2011
From: hagaia at qwilt.com (Hagai Avrahami)
Date: Tue, 6 Sep 2011 09:46:56 +0200
Subject: Help with AIO
In-Reply-To: <20110906073352.GR1137@mdounin.ru>
References: <CAN-ZXpuTdDMmQ7rgo_QP=1W53+j1qc-Bqa=f4=QLNskOubY2dw@mail.gmail.com>
	<20110906073352.GR1137@mdounin.ru>
Message-ID: <CAN-ZXpuK-Eoe52Q-Jmv2zjsezMM36WuB7Pkacona+SXoHEk8xA@mail.gmail.com>

Thanks a lot
I will try it

On Tue, Sep 6, 2011 at 9:33 AM, Maxim Dounin <mdounin at mdounin.ru> wrote:

> Hello!
>
> On Tue, Sep 06, 2011 at 01:21:00AM +0300, Hagai Avrahami wrote:
>
> > Hi
> >
> > I configured Nginx with *aio *mode *on *and 1 *worker process*
> > In benchmarking I saw limit of 1024 concurrent sessions even configured
> to
> > 4096 *worker connections*
>
> Sure it's not limitation of the benchmarking tool you use?  Most
> tools out there use select() and hence limited to 1024
> concurrent connections.
>
> > When I looked in the source code I could see that  io_setup is called
> with
> > fix value of 1024 max events
> > can anyone please explain me the hard coded limit
> > Is it OK to change it or it is derived from other limit I am not aware of
> > any help will be appreciated
>
> io_setup() maxevents is a limit on number of aio requests allowed
> to be submitted simulteneously; if nginx hits this limit, it will
> fallback to normal blocking read.
>
> If you are going to change it, make sure to keep total for all
> processes lower than /proc/sys/fs/aio-max-nr.
>
> Maxim Dounin
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx
>



-- 

*Hagai Avrahami*
Qwilt | Work: +972-72-2221644| Mobile: +972-54-4895656 |
hagaia at qwilt.com<yoav at qwilt.com>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110906/6f899194/attachment.html>

From nginx-forum at nginx.us  Tue Sep  6 13:12:32 2011
From: nginx-forum at nginx.us (djeyewater)
Date: Tue, 06 Sep 2011 09:12:32 -0400
Subject: nginx not gzipping php output
Message-ID: <bb8f1a7eafb07d6a961dc080faf18ffe.NginxMailingListEnglish@forum.nginx.org>

Nginx isn't gzipping php pages, but I don't know why? I'm using the
Audits tab in Google Chrome Developer tools to determine that the pages
aren't being gzipped.

My config:
nginx version: nginx/0.8.52
built by gcc 4.4.3 (Ubuntu 4.4.3-4ubuntu5) 
configure arguments: --prefix=/home/djeyewater/apps/nginx
--with-pcre=/home/djeyewater/tarballs/pcre-8.10
--without-http_autoindex_module --without-http_charset_module
--without-http_empty_gif_module --without-http_geo_module
--without-http_ssi_module --with-http_gzip_static_module
--with-http_realip_module --with-http_geoip_module --with-cc-opt='-I
/home/djeyewater/apps/GeoIP/include'
--with-ld-opt='-Wl,-R,/home/djeyewater/apps/GeoIP/lib -L
/home/djeyewater/apps/GeoIP/lib'


And http section from the .conf file:
http {
	include       mime.types;
	default_type  application/octet-stream;

	server_names_hash_bucket_size 128;

	sendfile        on;
	keepalive_timeout  10 10;
	
	port_in_redirect off;
	
	#Fix IP address
	set_real_ip_from   127.0.0.1;
	real_ip_header     X-Forwarded-For;

	gzip  on;
	gzip_comp_level 3;
	gzip_proxied any;
	gzip_types text/plain text/css application/x-javascript text/xml
application/xml application/xhtml+xml application/xml+rss
text/javascript;
	gzip_disable "MSIE [1-6]\.";
	
	geoip_country /home/djeyewater/apps/GeoIP/share/GeoIP/GeoIP.dat;
	
	log_format main '$remote_addr - $remote_user [$time_local] '
	             '"$request" $status $body_bytes_sent "$http_referer" '
	             '"$http_user_agent"';

	include sites/*;
}

Thanks

Dave

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,214879,214879#msg-214879


From nginx-forum at nginx.us  Tue Sep  6 13:33:49 2011
From: nginx-forum at nginx.us (magicbear)
Date: Tue, 06 Sep 2011 09:33:49 -0400
Subject: nginx not gzipping php output
In-Reply-To: <bb8f1a7eafb07d6a961dc080faf18ffe.NginxMailingListEnglish@forum.nginx.org>
References: <bb8f1a7eafb07d6a961dc080faf18ffe.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <b2d9045ae5240216105a88d18c8cfa2c.NginxMailingListEnglish@forum.nginx.org>

Check your PHP output header for "Content-Type", if it is not text/html
and one of the gzip_types, it won't be compress. Your config I have
tested on my server, it will compress normally.

MagicBear

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,214879,214881#msg-214881


From david at digitalronin.com  Tue Sep  6 13:45:28 2011
From: david at digitalronin.com (David Salgado)
Date: Tue, 6 Sep 2011 14:45:28 +0100
Subject: Old versions of Nginx
Message-ID: <CAOxCrNb4EM0QdhPNmYKH-DeGON-FFYHGANyuEEhTQV-s+oHCQA@mail.gmail.com>

Hi there

I'm trying to work with some third-party modules, and it seems they
need very specific point releases of nginx in order to work correctly.

e.g. the handlersocket module here;

    https://github.com/akalend/ngx_http_handlersocket_json_module

...apparently works with nginx 0.8.64.

But, that version is not available on the download page here;

    http://nginx.org/en/download.html

I can't even find it in the SVN repository.

    trac.nginx.org/nginx/browser#nginx/tags

Is there a place to get old releases of the nginx source tarballs?

Thanks

David


From nginx-forum at nginx.us  Tue Sep  6 13:55:23 2011
From: nginx-forum at nginx.us (djeyewater)
Date: Tue, 06 Sep 2011 09:55:23 -0400
Subject: nginx not gzipping php output
In-Reply-To: <b2d9045ae5240216105a88d18c8cfa2c.NginxMailingListEnglish@forum.nginx.org>
References: <bb8f1a7eafb07d6a961dc080faf18ffe.NginxMailingListEnglish@forum.nginx.org>
	<b2d9045ae5240216105a88d18c8cfa2c.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <c653439440a34e7837caabecaf37666c.NginxMailingListEnglish@forum.nginx.org>

Thanks for testing it MagicBear.

Google Chrome reports content type header as:
Content-Type:text/html; charset=UTF-8

So still not sure why it is not working.

I checked nginx changelog and couldn't see anything that looked to be
pointing out that this was an issue in older versions of nginx. Still, I
will try latest stable version and see if that makes any difference.

Dave

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,214879,214885#msg-214885


From mdounin at mdounin.ru  Tue Sep  6 14:01:23 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Tue, 6 Sep 2011 18:01:23 +0400
Subject: Old versions of Nginx
In-Reply-To: <CAOxCrNb4EM0QdhPNmYKH-DeGON-FFYHGANyuEEhTQV-s+oHCQA@mail.gmail.com>
References: <CAOxCrNb4EM0QdhPNmYKH-DeGON-FFYHGANyuEEhTQV-s+oHCQA@mail.gmail.com>
Message-ID: <20110906140123.GX1137@mdounin.ru>

Hello!

On Tue, Sep 06, 2011 at 02:45:28PM +0100, David Salgado wrote:

> Hi there
> 
> I'm trying to work with some third-party modules, and it seems they
> need very specific point releases of nginx in order to work correctly.
> 
> e.g. the handlersocket module here;
> 
>     https://github.com/akalend/ngx_http_handlersocket_json_module
> 
> ...apparently works with nginx 0.8.64.
> 
> But, that version is not available on the download page here;
> 
>     http://nginx.org/en/download.html
> 
> I can't even find it in the SVN repository.
> 
>     trac.nginx.org/nginx/browser#nginx/tags

This is because there was no such version.  The last one in 0.8.* 
branch is 0.8.55.

> Is there a place to get old releases of the nginx source tarballs?

The http://nginx.org/download/ should be a proper one.  Currently 
it only contains tarballs since 2007, probably older tarballs 
should be placed there too, just for history.

Maxim Dounin


From lists at ruby-forum.com  Tue Sep  6 14:04:21 2011
From: lists at ruby-forum.com (Matthias R. W.)
Date: Tue, 06 Sep 2011 16:04:21 +0200
Subject: Content-Length header missing after proxying
In-Reply-To: <4E53F504.2020402@cloud9.co.uk>
References: <4E53F504.2020402@cloud9.co.uk>
Message-ID: <767401579bb79c93e252e939b9285059@ruby-forum.com>

Hello outside there,

I'm working on a corresponding issue
To get cloud9 at node.js through an ssl-offload proxy I've set the 
following nginx configuration:

upstream nodejs3001 {
    server 127.0.0.1:3001;
}

server {
    listen 127.0.0.1:84;
    server_name portal.domain.tld;
    access_log /var/log/nginx/customlogfile.log;

    location / {
        proxy_set_header X-Real-IP $remote_addr;
        proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
        proxy_set_header Host $http_host;
        proxy_set_header X-NginX-Proxy true;
        proxy_pass http://nodejs3001;
        proxy_redirect off;
    }
}

this is the last log-entry:
192.168.1.123 - admin [06/Sep/2011:15:28:37 +0200] "GET 
/socket.io/1/xhr-polling/1502126590706860217?t1315315712123 HTTP/1.1" 
200 0 "http://127.0.0.1:84/" "Mozilla/5.0 (Macintosh; Intel Mac OS X 
10_7_1) AppleWebKit/535.1 (KHTML, like Gecko) Chrome/13.0.782.220 
Safari/535.1"

and this is the content of the header regarding to get this last step:
Request 
URL:http://127.0.01:84/socket.io/1/xhr-polling/1603386210380646057?t1315315788760
Request Headersview source
Referer:http://127.0.0.1:84/
User-Agent:Mozilla/5.0 (Macintosh; Intel Mac OS X 10_7_1) 
AppleWebKit/535.1 (KHTML, like Gecko) Chrome/13.0.782.220 Safari/535.1
Query String Parametersview URL encoded

I think we'll have the most success here:
https://github.com/ajaxorg/cloud9/issues/267

Tell me... Are you trying to do the same?
?atthias

-- 
Posted via http://www.ruby-forum.com/.

From david at digitalronin.com  Tue Sep  6 14:05:40 2011
From: david at digitalronin.com (David Salgado)
Date: Tue, 6 Sep 2011 15:05:40 +0100
Subject: Old versions of Nginx
In-Reply-To: <20110906140123.GX1137@mdounin.ru>
References: <CAOxCrNb4EM0QdhPNmYKH-DeGON-FFYHGANyuEEhTQV-s+oHCQA@mail.gmail.com>
	<20110906140123.GX1137@mdounin.ru>
Message-ID: <CAOxCrNYXGvVjNFUFgJDQvkJRMpwbG+MY+LiZ9hb9SJCkzbOTYA@mail.gmail.com>

> This is because there was no such version. ?The last one in 0.8.*
> branch is 0.8.55.

Ah, that would explain it! The module's author must have gotten confused.

Many thanks

David


From orz at loli.my  Tue Sep  6 14:26:57 2011
From: orz at loli.my (=?UTF-8?B?44OT44Oq44OT44Oq4oWk?=)
Date: Tue, 6 Sep 2011 22:26:57 +0800
Subject: nginx not gzipping php output
In-Reply-To: <c653439440a34e7837caabecaf37666c.NginxMailingListEnglish@forum.nginx.org>
References: <bb8f1a7eafb07d6a961dc080faf18ffe.NginxMailingListEnglish@forum.nginx.org>
	<b2d9045ae5240216105a88d18c8cfa2c.NginxMailingListEnglish@forum.nginx.org>
	<c653439440a34e7837caabecaf37666c.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <CAENo=s0Rr4Uc0=c+Z8UOkGoV++H6GQ66FBeR40SA_OAHByCD4w@mail.gmail.com>

In fact, you can try to open php internal gzip compress function, you may
find it at php.ini, change this to On
zlib.output_compression = Off
and php will compress it auto before send to nginx.

MagicBear

2011/9/6 djeyewater <nginx-forum at nginx.us>

> Thanks for testing it MagicBear.
>
> Google Chrome reports content type header as:
> Content-Type:text/html; charset=UTF-8
>
> So still not sure why it is not working.
>
> I checked nginx changelog and couldn't see anything that looked to be
> pointing out that this was an issue in older versions of nginx. Still, I
> will try latest stable version and see if that makes any difference.
>
> Dave
>
> Posted at Nginx Forum:
> http://forum.nginx.org/read.php?2,214879,214885#msg-214885
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110906/23b90793/attachment.html>

From nginx-forum at nginx.us  Tue Sep  6 14:31:14 2011
From: nginx-forum at nginx.us (djeyewater)
Date: Tue, 06 Sep 2011 10:31:14 -0400
Subject: nginx not gzipping php output
In-Reply-To: <c653439440a34e7837caabecaf37666c.NginxMailingListEnglish@forum.nginx.org>
References: <bb8f1a7eafb07d6a961dc080faf18ffe.NginxMailingListEnglish@forum.nginx.org>
	<b2d9045ae5240216105a88d18c8cfa2c.NginxMailingListEnglish@forum.nginx.org>
	<c653439440a34e7837caabecaf37666c.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <ed569b2b80942913f6b2a7d89c16f0be.NginxMailingListEnglish@forum.nginx.org>

Hmm.. well I installed the latest version and found that actually
nothing is being zipped. Was probably the same thing before, but most of
the static files have gzip versions and so were included by gzip_static,
making it look like only PHP wasn't being zipped.

Any ideas?

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,214879,214890#msg-214890


From chinix at gmail.com  Tue Sep  6 14:39:56 2011
From: chinix at gmail.com (chinix)
Date: Tue, 06 Sep 2011 22:39:56 +0800
Subject: proxy with header If-Modified-Since
Message-ID: <4e6630c1.8e42e70a.1b59.4caa@mx.google.com>

hi all,

I find that when the cache is expired,nginx request completely to backend for fresh cache even the file wasn't modified.Could nginx request to backend with header "If-Modified-Since" when the cache is expired? If this,backend will response with code 304, nginx should deal with the 304 response.

2011-09-06



chinix
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110906/01ba541f/attachment-0001.html>

From mdounin at mdounin.ru  Tue Sep  6 15:27:09 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Tue, 6 Sep 2011 19:27:09 +0400
Subject: proxy with header If-Modified-Since
In-Reply-To: <4e6630c1.8e42e70a.1b59.4caa@mx.google.com>
References: <4e6630c1.8e42e70a.1b59.4caa@mx.google.com>
Message-ID: <20110906152709.GB1137@mdounin.ru>

Hello!

On Tue, Sep 06, 2011 at 10:39:56PM +0800, chinix wrote:

> hi all,
> 
> I find that when the cache is expired,nginx request completely 
> to backend for fresh cache even the file wasn't modified.Could 
> nginx request to backend with header "If-Modified-Since" when 
> the cache is expired? If this,backend will response with code 
> 304, nginx should deal with the 304 response.

This is not currently supported.

Maxim Dounin


From lists at ruby-forum.com  Tue Sep  6 17:06:57 2011
From: lists at ruby-forum.com (Ed Tiller)
Date: Tue, 06 Sep 2011 19:06:57 +0200
Subject: nginx-1.1.2
In-Reply-To: <20110905131950.GG34606@nginx.com>
References: <20110905131950.GG34606@nginx.com>
Message-ID: <829a8decd69485ea5f7095cfd62a60f1@ruby-forum.com>

Thank you Igor!!!
Congrats on the growth of Nginx marketshare on the net.
http://news.netcraft.com/archives/2011/09/06/september-2011-web-server-survey.html

-- 
Posted via http://www.ruby-forum.com/.


From nginx-forum at nginx.us  Tue Sep  6 17:49:02 2011
From: nginx-forum at nginx.us (aries)
Date: Tue, 06 Sep 2011 13:49:02 -0400
Subject: 400 bad request errors - Firefox and Chrome
Message-ID: <577bf662b4e1bf80057dbe1c58731896.NginxMailingListEnglish@forum.nginx.org>

Hello.
The access.log, have multiples lines similar to:
 [06/Sep/2011:11:05:32 +0500] "-" 400 0 "-" "-" "-"

These are only created when browsing with Firefox or Chrome, do not
register these errors if you use IExplorer.
The error only shows in access.log, the page loads properly without
showing any error.
I think that it is not a problem of cookies because these errors are
logged even if disabling the use of cookies on your browser.
"error_log /var/log/nginx/error.log info;" not show me any data about
this error.
I noticed that error is recorded between 3 to 7 seconds after that the
apache access.log recorded the entry with 200 code for this visit.

I also tried using:
client_header_buffer_size 256k;
large_client_header_buffers 4 256;

As additional data indicate that I'm using:
The Apache keepalive is disabled.

The nginx.conf
user  nobody;
worker_processes  1;
error_log  /var/log/nginx/error.log info;
worker_rlimit_nofile 20480;
events {
 worker_connections 5120; # increase for busier servers
 use epoll; # you should use epoll here for Linux kernels 2.6.x
}
http {
 server_name_in_redirect off;
 server_names_hash_max_size 10240;
 server_names_hash_bucket_size 1024;
 include    mime.types;
 default_type  application/octet-stream;
 server_tokens off;
 sendfile on;
 tcp_nopush on;
 tcp_nodelay on;
 keepalive_timeout  5;
 gzip on;
 gzip_vary on;
 gzip_disable "MSIE [1-6]\.";
 gzip_proxied any;
 gzip_http_version 1.1;
 gzip_min_length  1000;
 gzip_comp_level  6;
 gzip_buffers  16 8k;
 gzip_types text/plain text/xml text/css application/x-javascript
application/xml application/xml+rss text/javascript
application/atom+xml;
 ignore_invalid_headers on;
 client_header_timeout  3m;
 client_body_timeout 3m;
 send_timeout     3m;
 reset_timedout_connection on;
 connection_pool_size  256;
 client_header_buffer_size 128k;
 large_client_header_buffers 8 32k;
 client_max_body_size 2M; 
 client_body_buffer_size 256k;
 request_pool_size  4k;
 output_buffers   4 32k;
 postpone_output  1460;
 include "/etc/nginx/vhosts/*";
}

Any suggestions how to fix these errors?.

Thanks.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,214924,214924#msg-214924


From mdounin at mdounin.ru  Tue Sep  6 18:12:58 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Tue, 6 Sep 2011 22:12:58 +0400
Subject: 400 bad request errors - Firefox and Chrome
In-Reply-To: <577bf662b4e1bf80057dbe1c58731896.NginxMailingListEnglish@forum.nginx.org>
References: <577bf662b4e1bf80057dbe1c58731896.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20110906181258.GC1137@mdounin.ru>

Hello!

On Tue, Sep 06, 2011 at 01:49:02PM -0400, aries wrote:

> Hello.
> The access.log, have multiples lines similar to:
>  [06/Sep/2011:11:05:32 +0500] "-" 400 0 "-" "-" "-"
> 
> These are only created when browsing with Firefox or Chrome, do not
> register these errors if you use IExplorer.
> The error only shows in access.log, the page loads properly without
> showing any error.
> I think that it is not a problem of cookies because these errors are
> logged even if disabling the use of cookies on your browser.
> "error_log /var/log/nginx/error.log info;" not show me any data about
> this error.
> I noticed that error is recorded between 3 to 7 seconds after that the
> apache access.log recorded the entry with 200 code for this visit.

With Chrome it's caused by Chrome's behaviour: it opens 2 
connections to requested host, and may close one of them without 
sending any single request though it (if one connection was 
enough).  There is no real problem here, just a sign that 
connection was opened but not used.

No idea about Firefox, but likely it does something similar.

Maxim Dounin


From sheng.zheng at etimestech.jp  Wed Sep  7 01:03:56 2011
From: sheng.zheng at etimestech.jp (Sheng.Zheng)
Date: Wed, 07 Sep 2011 10:03:56 +0900
Subject: add header on custom error page
Message-ID: <4E66C2FC.6020405@etimestech.jp>

Hi,
  i'm trying to add a 'X-ua-compatible' header on a custom 444 error
page  like below. but this  does not work. do you have any suggestions
on how to do this?

Any advise will be appreciated.

Thanks.



  error_page 444 = @err_444;

  location = /xxx {
      return 444;
  }

      location ~ \.php$ {
        try_files $uri =404;
        fastcgi_pass   unix:/var/run/php.sock;
        fastcgi_param  SCRIPT_FILENAME  $document_root$fastcgi_script_name;
        include        /etc/nginx/fastcgi_params;
    }


   location @err_444 {
        add_header  X-ua-compatible 'IE=EmulateIE8';
        rewrite ^ /index.php?url=err/444?;
    }



From nginx-forum at nginx.us  Wed Sep  7 02:07:45 2011
From: nginx-forum at nginx.us (speedfirst)
Date: Tue, 06 Sep 2011 22:07:45 -0400
Subject: add header on custom error page
In-Reply-To: <4E66C2FC.6020405@etimestech.jp>
References: <4E66C2FC.6020405@etimestech.jp>
Message-ID: <c0cc19eedd257a1194db01b0f56b1a0e.NginxMailingListEnglish@forum.nginx.org>

add_header is effective only when the response code is equal to 200,
204, 301, 302 or 304. You should try headers_more module.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,214931,214932#msg-214932


From nginx-forum at nginx.us  Wed Sep  7 14:41:31 2011
From: nginx-forum at nginx.us (arleybls)
Date: Wed, 07 Sep 2011 10:41:31 -0400
Subject: block ip/subnet at socket lvl
Message-ID: <8103f75d46a866ec34f9bc64bec9ab13.NginxMailingListEnglish@forum.nginx.org>

Currently Nginx can block IP on application level and thus send a 403
forbidden message back to the client.

I wonder if there's a way to block any request at the socket level,
preferably not even sending a FIN message to the client but simply
dropping the packet. Right now i'm using iptables but i'm really looking
to have such feature on nginx itself, and if not supported right on
current build, how could I have a hook to the socket and control it on a
home made server module/handler?

Cheers,
__Jr

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,214948,214948#msg-214948


From nginx-forum at nginx.us  Wed Sep  7 20:56:22 2011
From: nginx-forum at nginx.us (pvsw1)
Date: Wed, 07 Sep 2011 16:56:22 -0400
Subject: Problem with updated nginx
In-Reply-To: <20110829004530.GS1137@mdounin.ru>
References: <20110829004530.GS1137@mdounin.ru>
Message-ID: <ac32ee8b7bd2eab5da52c00e3fca019d.NginxMailingListEnglish@forum.nginx.org>

Hi Maxim,

Thanks for the reply. That fixed my problem

Regards,
Paul

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,214407,214958#msg-214958


From matthieu.tourne at gmail.com  Wed Sep  7 23:33:38 2011
From: matthieu.tourne at gmail.com (Matthieu Tourne)
Date: Wed, 7 Sep 2011 16:33:38 -0700
Subject: upstream keepalive - call for testing
In-Reply-To: <20110905180210.GL1137@mdounin.ru>
References: <20110801160725.GH1137@mdounin.ru>
	<0d5cc130a417bb62098701656fd44e42.NginxMailingListEnglish@forum.nginx.org>
	<20110904182137.GW1137@mdounin.ru>
	<CAENo=s3G3XBrYN7K9G6HzHZkBWYHEEi4DLCH8G-7_yC8dJyu-w@mail.gmail.com>
	<20110905070813.GY1137@mdounin.ru>
	<CAENo=s1GJzGJ8OPbNBvQhTEECN=jKFAtgYN=SMx=2o9KS+W02A@mail.gmail.com>
	<20110905180210.GL1137@mdounin.ru>
Message-ID: <CAAGYmFfRzVMmP+fkFa_SKBzPS2HAowVSKYcQVKR68j_vAE3r4Q@mail.gmail.com>

Thanks for the all-included patch from 1.1.2

It seems that keepalive connections aren't working in combination with
proxy_buffering off; (c->read->ready is set to 1)
It worked before this last update, is this the intended behavior ?

Thank you,

Matthieu.

On Mon, Sep 5, 2011 at 11:02 AM, Maxim Dounin <mdounin at mdounin.ru> wrote:

> Hello!
>
> On Mon, Sep 05, 2011 at 11:42:31PM +0800, $B%S%j%S%j-9(B wrote:
>
> > (gdb) fr 0
> > #0  ngx_http_upstream_handler (ev=0x7fc45735f8a8)
> >     at src/http/ngx_http_upstream.c:915
> > 915    ctx->current_request = r;
> > (gdb) p ngx_cycle->log
> > $1 = (ngx_log_t *) 0x21f19a8
> > (gdb) p *r
> > $2 = {signature = 51686928, connection = 0x23b4160, ctx = 0x0,
>
> [...]
>
> This looks like memory corruption, but unfortunately I don't see
> any traces of the real cause.  My best quess is improper handling
> of proxy_ignore_client_abort as fixed in 1.1.2.  Please try 1.1.2
> with patches from
>
> http://nginx.org/patches/patch-nginx-keepalive-full-5.txt
>
> It already includes upstream keepalive module, as well as all
> other upstream-keepalive related fixes.  See here for details:
>
> http://mailman.nginx.org/pipermail/nginx-devel/2011-September/001147.html
>
> Maxim Dounin
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110907/8b74bb67/attachment.html>

From mdounin at mdounin.ru  Thu Sep  8 09:27:34 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Thu, 8 Sep 2011 13:27:34 +0400
Subject: upstream keepalive - call for testing
In-Reply-To: <CAAGYmFfRzVMmP+fkFa_SKBzPS2HAowVSKYcQVKR68j_vAE3r4Q@mail.gmail.com>
References: <20110801160725.GH1137@mdounin.ru>
	<0d5cc130a417bb62098701656fd44e42.NginxMailingListEnglish@forum.nginx.org>
	<20110904182137.GW1137@mdounin.ru>
	<CAENo=s3G3XBrYN7K9G6HzHZkBWYHEEi4DLCH8G-7_yC8dJyu-w@mail.gmail.com>
	<20110905070813.GY1137@mdounin.ru>
	<CAENo=s1GJzGJ8OPbNBvQhTEECN=jKFAtgYN=SMx=2o9KS+W02A@mail.gmail.com>
	<20110905180210.GL1137@mdounin.ru>
	<CAAGYmFfRzVMmP+fkFa_SKBzPS2HAowVSKYcQVKR68j_vAE3r4Q@mail.gmail.com>
Message-ID: <20110908092733.GG1137@mdounin.ru>

Hello!

On Wed, Sep 07, 2011 at 04:33:38PM -0700, Matthieu Tourne wrote:

> Thanks for the all-included patch from 1.1.2
> 
> It seems that keepalive connections aren't working in combination with
> proxy_buffering off; (c->read->ready is set to 1)
> It worked before this last update, is this the intended behavior ?

Linux with epoll, right?  Please try the following patch:

diff --git a/src/http/modules/ngx_http_upstream_keepalive_module.c b/src/http/modules/ngx_http_upstream_keepalive_module.c
--- a/src/http/modules/ngx_http_upstream_keepalive_module.c
+++ b/src/http/modules/ngx_http_upstream_keepalive_module.c
@@ -321,7 +321,6 @@ ngx_http_upstream_free_keepalive_peer(ng
     if (kp->failed
         || c == NULL
         || c->read->eof
-        || c->read->ready
         || c->read->error
         || c->read->timedout
         || c->write->error
@@ -382,6 +381,10 @@ ngx_http_upstream_free_keepalive_peer(ng
     item->socklen = pc->socklen;
     ngx_memcpy(&item->sockaddr, pc->sockaddr, pc->socklen);

+    if (c->read->ready) {
+        ngx_http_upstream_keepalive_close_handler(c->read);
+    }
+
 invalid:

     kp->original_free_peer(pc, kp->data, state);
diff --git a/src/http/ngx_http_upstream.c b/src/http/ngx_http_upstream.c
--- a/src/http/ngx_http_upstream.c
+++ b/src/http/ngx_http_upstream.c
@@ -2142,7 +2142,7 @@ ngx_http_upstream_send_response(ngx_http
                 return;
             }

-            if (u->peer.connection->read->ready) {
+            if (u->peer.connection->read->ready || u->length == 0) {
                 ngx_http_upstream_process_non_buffered_upstream(r, u);
             }
         }

Second part fixes one more problem with handling of responses with 
empty body over keepalive connections, found during testing with 
various event methods.

Maxim Dounin


From nginx-forum at nginx.us  Thu Sep  8 11:06:10 2011
From: nginx-forum at nginx.us (liushaobo)
Date: Thu, 08 Sep 2011 07:06:10 -0400
Subject: closed keepalive connection
Message-ID: <ebc8be3fd6b079b1d76f86cffaf5ff56.NginxMailingListEnglish@forum.nginx.org>

hi all
nginx debug log the following problems,please help me

# nginx config
user  www www;
worker_processes 8;
worker_cpu_affinity 00000001 00000010 00000100 00001000 00010000
00100000 01000000 10000000;
worker_rlimit_nofile 65535;

#error_log /var/log/nginx_error.log  crit;
error_log  /var/log/nginx_error.log  debug;

pid        /var/run/nginx.pid;

events {
    worker_connections  65535;
    use epoll;
}

http {
    include       /etc/nginx/mime.types;
    default_type  application/octet-stream;

    sendfile        on;
    tcp_nopush      on;
    tcp_nodelay     on;
    server_tokens   off;
    charset utf8;

    server_names_hash_bucket_size 128;
    client_header_buffer_size 32k;
    large_client_header_buffers 4 32k;
    client_max_body_size 150m;

    gzip            on;
    gzip_comp_level 2;
    gzip_min_length 1024;
    gzip_buffers    4 16k;
    gzip_http_version 1.0;
    gzip_types      text/plain application/x-javascript text/css
application/xml;
    gzip_vary       on;

    fastcgi_connect_timeout 300;
    fastcgi_send_timeout 300;
    fastcgi_read_timeout 300;
    fastcgi_buffer_size 64k;
    fastcgi_buffers 4 64k;
    fastcgi_busy_buffers_size 128k;
    fastcgi_temp_file_write_size 128k;
    proxy_intercept_errors on;
    proxy_ignore_client_abort on;
    fastcgi_intercept_errors on;

    keepalive_timeout  120;
    limit_zone   one  $binary_remote_addr  10m;

    include /etc/nginx/conf.d/*.conf;

error_log  /var/log/nginx_error.log  debug;

# tail -f /var/log/nginx_error.log

2011/09/08 15:38:10 [info] 19178#0: *5713577 client 112.64.188.183
closed keepalive connection (104: Connection reset by peer)
2011/09/08 15:39:42 [info] 19175#0: *5729670 client 112.64.188.183
closed keepalive connection (104: Connection reset by peer)
2011/09/08 15:44:10 [info] 19257#0: *1382 client closed prematurely
connection while sending response to client, client: 112.64.188.183,
server: static.test.com, request: "GET
/apk/48/48598/com.dolphin.browser.cn.apk HTTP/1.1", host:
"static.test.com"
2011/09/08 15:45:06 [info] 19261#0: *2100 client 112.64.188.183 closed
keepalive connection (104: Connection reset by peer)

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,214974,214974#msg-214974


From mdounin at mdounin.ru  Thu Sep  8 11:34:57 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Thu, 8 Sep 2011 15:34:57 +0400
Subject: closed keepalive connection
In-Reply-To: <ebc8be3fd6b079b1d76f86cffaf5ff56.NginxMailingListEnglish@forum.nginx.org>
References: <ebc8be3fd6b079b1d76f86cffaf5ff56.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20110908113457.GJ1137@mdounin.ru>

Hello!

On Thu, Sep 08, 2011 at 07:06:10AM -0400, liushaobo wrote:

> hi all
> nginx debug log the following problems,please help me

[...]

> # tail -f /var/log/nginx_error.log
> 
> 2011/09/08 15:38:10 [info] 19178#0: *5713577 client 112.64.188.183
> closed keepalive connection (104: Connection reset by peer)
> 2011/09/08 15:39:42 [info] 19175#0: *5729670 client 112.64.188.183
> closed keepalive connection (104: Connection reset by peer)

It's info message reporting that client closed keepalive 
connection, nothing more.  This may and will happen under normal 
use.

> 2011/09/08 15:44:10 [info] 19257#0: *1382 client closed prematurely
> connection while sending response to client, client: 112.64.188.183,
> server: static.test.com, request: "GET
> /apk/48/48598/com.dolphin.browser.cn.apk HTTP/1.1", host:
> "static.test.com"

This is another info message, reporting that client closed 
connection before nginx was able to send full response to it.  
This is normal as well and may happen e.g. if browser is closed or 
user navigated to another page.

If you don't see such mesages - just tune your error_log logging 
level to something less verbose.  Something like "notice" or 
"error" is usually a good choice for production.

Maxim Dounin


From nginx-forum at nginx.us  Thu Sep  8 12:01:12 2011
From: nginx-forum at nginx.us (liushaobo)
Date: Thu, 08 Sep 2011 08:01:12 -0400
Subject: closed keepalive connection
In-Reply-To: <05e801c9cb43$68a23cc0$39e6b640$@forum.nginx.org>
References: <05e801c9cb43$68a23cc0$39e6b640$@forum.nginx.org>
Message-ID: <43ff5f44f752e7163ef9638d8edf84fb.NginxMailingListEnglish@forum.nginx.org>

hi all

Modify nginx.conf

error_log  /data0/logs/nginx_error.log  notice;
#error_log  /data0/logs/nginx_error.log  debug;

log are as follows

web download normal
wifi download normal
mobile 3G (WCDMA) download frequently interrupted

2011/09/08 19:54:49 [info] 20098#0: *1137837 client closed prematurely
connection while reading client request line, client: 59.108.246.154,
server: static.nduoa.com
2011/09/08 19:54:49 [info] 20098#0: *1137838 client closed prematurely
connection while reading client request line, client: 59.108.246.154,
server: static.nduoa.com
2011/09/08 19:54:49 [info] 20099#0: *1130906 client closed prematurely
connection while sending response to client, client: 120.196.170.2,
server: static.nduoa.com, request: "GET /apk/22/22881/com.popcap.pvz.apk
HTTP/1.1", host: "static.nduoa.com"

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,1680,214977#msg-214977


From nginx-forum at nginx.us  Thu Sep  8 14:20:44 2011
From: nginx-forum at nginx.us (Giang)
Date: Thu, 08 Sep 2011 10:20:44 -0400
Subject: Please help me
Message-ID: <2465b534c9c4d1f868de438e1e9e3ded.NginxMailingListEnglish@forum.nginx.org>

Hi guys,

I'm new to nginx. Just knew how to use it few months ago.

I got a QuickSSL SSL, but I would like to install it only on mybb admin
folder only (http://myforum.com/admin will rewrite to
https://myforum.com/admin) and keep the rest of the site uses http.

Only http://myforum.com/admin will redirect to
https://myforum.com/admin. It didn't work with any existing file inside
admin (for example: admin/index.php).

Here is my current nginx config:

server {
    listen  80;
    server_name domain.com www.domain.com;
    root /var/www/domain.com;
    include /etc/nginx/fastcgi_php;
    index index.php index.html;
    location / {

        }
    }
    location /admin {
        rewrite ^/ https://$http_host$request_uri permanent;
    }
}

server {
    listen  443 ssl;
    ssl_certificate         /etc/nginx/ssl/csr.csr;
    ssl_certificate_key     /etc/nginx/ssl/csr.key;
    keepalive_timeout    60;
    ssl_protocols  SSLv2 SSLv3 TLSv1;
    ssl_ciphers 
ALL:!ADH:!EXPORT56:RC4+RSA:+HIGH:+MEDIUM:+LOW:+SSLv2:+EXP;
    ssl_prefer_server_ciphers   on;
    
    server_name domain.com www.domain.com;
    root /var/www/domain.com;
    include /etc/nginx/fastcgi_php;
    index index.php index.html;
    location / {
        rewrite ^ http://$http_host$request_uri permanent;
    }   
    location /admin {
    }
}

I don't know how to fix it, please help.

Thank in advanced,
Giang

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,214981,214981#msg-214981


From nginx-forum at nginx.us  Thu Sep  8 14:41:34 2011
From: nginx-forum at nginx.us (ynasser)
Date: Thu, 08 Sep 2011 10:41:34 -0400
Subject: Beginner questions about proxy configuration and white space
Message-ID: <72ca536398bcc048cbca71aa39cb728e.NginxMailingListEnglish@forum.nginx.org>

Hello,

I have installed nginx on my personal computer and would like to use it
as a proxy to force authentication via modified http headers in various
browsers. There's an add-on in firefox to do this, but none for Chrome,
which is why I am using nginx. I've edited configuration file to include
something like this under the http section:

  server {
	listen some_port;

	location http://some_ip {
		add_header X-h1 h1;
		add_header X-h2 h2 h2;
		add_header X-h3 h3;
		add_header X-h4 h4; 
		add_header X-h5 h5;
		add_header X-h6 h6 h6;
		add_header X-h7 h7;
				}
	   }

I figured that out from reading many examples and the wiki. Now, I do
not know how exactly to configure nginx to act as a proxy. I've read
through part of the wiki on the HttpProxy module, but I don't have much
experience with web servers or proxies. Could someone point me in the
right direction?

Also, when editing the configuration file using the http headers module
(this: http://wiki.nginx.org/HttpHeadersModule), can values have spaces
in them? Or should I put quotations around the whole thing? I read here:
http://library.linode.com/web-servers/nginx/configuration/basic that
whitespace doesn't matter, but I just thought to double check, since I
saw no examples where values had spaces in them.

Thanks.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,214982,214982#msg-214982


From mdounin at mdounin.ru  Thu Sep  8 15:42:15 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Thu, 8 Sep 2011 19:42:15 +0400
Subject: upstream keepalive - call for testing
In-Reply-To: <20110908092733.GG1137@mdounin.ru>
References: <20110801160725.GH1137@mdounin.ru>
	<0d5cc130a417bb62098701656fd44e42.NginxMailingListEnglish@forum.nginx.org>
	<20110904182137.GW1137@mdounin.ru>
	<CAENo=s3G3XBrYN7K9G6HzHZkBWYHEEi4DLCH8G-7_yC8dJyu-w@mail.gmail.com>
	<20110905070813.GY1137@mdounin.ru>
	<CAENo=s1GJzGJ8OPbNBvQhTEECN=jKFAtgYN=SMx=2o9KS+W02A@mail.gmail.com>
	<20110905180210.GL1137@mdounin.ru>
	<CAAGYmFfRzVMmP+fkFa_SKBzPS2HAowVSKYcQVKR68j_vAE3r4Q@mail.gmail.com>
	<20110908092733.GG1137@mdounin.ru>
Message-ID: <20110908154215.GM1137@mdounin.ru>

Hello!

On Thu, Sep 08, 2011 at 01:27:34PM +0400, Maxim Dounin wrote:

> Hello!
> 
> On Wed, Sep 07, 2011 at 04:33:38PM -0700, Matthieu Tourne wrote:
> 
> > Thanks for the all-included patch from 1.1.2
> > 
> > It seems that keepalive connections aren't working in combination with
> > proxy_buffering off; (c->read->ready is set to 1)
> > It worked before this last update, is this the intended behavior ?
> 
> Linux with epoll, right?  Please try the following patch:

Updated full patch may be grabbed here:

http://nginx.org/patches/patch-nginx-keepalive-full-6.txt

Maxim Dounin


From nginx-forum at nginx.us  Thu Sep  8 17:28:12 2011
From: nginx-forum at nginx.us (magicbear)
Date: Thu, 08 Sep 2011 13:28:12 -0400
Subject: Beginner questions about proxy configuration and white space
In-Reply-To: <72ca536398bcc048cbca71aa39cb728e.NginxMailingListEnglish@forum.nginx.org>
References: <72ca536398bcc048cbca71aa39cb728e.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <17696a632c172de23cb235a2b6fd33c4.NginxMailingListEnglish@forum.nginx.org>

proxy:

server {
listen some_ip:some_port default;

location / {
add_header X-h1 h1;
add_header X-h2 "h2 h2";
add_header X-h3 h3;
add_header X-h4 h4;
add_header X-h5 h5;
add_header X-h6 "h6 h6";
add_header X-h7 h7;
proxy_pass http://backend;
}
}




space: add a double quotes
add_header X-h2 "h2 h2";

MagicBear

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,214982,214984#msg-214984


From nginx-forum at nginx.us  Thu Sep  8 17:36:09 2011
From: nginx-forum at nginx.us (magicbear)
Date: Thu, 08 Sep 2011 13:36:09 -0400
Subject: Please help me
In-Reply-To: <2465b534c9c4d1f868de438e1e9e3ded.NginxMailingListEnglish@forum.nginx.org>
References: <2465b534c9c4d1f868de438e1e9e3ded.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <6e854cb19c7c2550cde5346387005244.NginxMailingListEnglish@forum.nginx.org>

server {
listen 80;
server_name domain.com www.domain.com;
root /var/www/domain.com;
include /etc/nginx/fastcgi_php;
index index.php index.html;
rewrite ^/admin https://$http_host$request_uri permanent;
}

server {
listen 443 ssl;
ssl_certificate /etc/nginx/ssl/csr.csr;
ssl_certificate_key /etc/nginx/ssl/csr.key;
keepalive_timeout 60;
ssl_protocols SSLv2 SSLv3 TLSv1;
ssl_ciphers ALL:!ADH:!EXPORT56:RC4+RSA:+HIGH:+MEDIUM:+LOW:+SSLv2:+EXP;
ssl_prefer_server_ciphers on;

server_name domain.com www.domain.com;
root /var/www/domain.com;
include /etc/nginx/fastcgi_php;
index index.php index.html;
location / {
rewrite ^/ http://$http_host$request_uri permanent;
}
location ~ ^/admin {
  break;
}
}

MagicBear

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,214981,214985#msg-214985


From nginx-forum at nginx.us  Thu Sep  8 17:42:17 2011
From: nginx-forum at nginx.us (ynasser)
Date: Thu, 08 Sep 2011 13:42:17 -0400
Subject: Beginner questions about proxy configuration and white space
In-Reply-To: <17696a632c172de23cb235a2b6fd33c4.NginxMailingListEnglish@forum.nginx.org>
References: <72ca536398bcc048cbca71aa39cb728e.NginxMailingListEnglish@forum.nginx.org>
	<17696a632c172de23cb235a2b6fd33c4.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <f29afd355b8ba6d9303df0bd7fa38342.NginxMailingListEnglish@forum.nginx.org>

I'll try that out. 

Thank you!

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,214982,214986#msg-214986


From matthieu.tourne at gmail.com  Thu Sep  8 22:05:01 2011
From: matthieu.tourne at gmail.com (Matthieu Tourne)
Date: Thu, 8 Sep 2011 15:05:01 -0700
Subject: upstream keepalive - call for testing
In-Reply-To: <20110908092733.GG1137@mdounin.ru>
References: <20110801160725.GH1137@mdounin.ru>
	<0d5cc130a417bb62098701656fd44e42.NginxMailingListEnglish@forum.nginx.org>
	<20110904182137.GW1137@mdounin.ru>
	<CAENo=s3G3XBrYN7K9G6HzHZkBWYHEEi4DLCH8G-7_yC8dJyu-w@mail.gmail.com>
	<20110905070813.GY1137@mdounin.ru>
	<CAENo=s1GJzGJ8OPbNBvQhTEECN=jKFAtgYN=SMx=2o9KS+W02A@mail.gmail.com>
	<20110905180210.GL1137@mdounin.ru>
	<CAAGYmFfRzVMmP+fkFa_SKBzPS2HAowVSKYcQVKR68j_vAE3r4Q@mail.gmail.com>
	<20110908092733.GG1137@mdounin.ru>
Message-ID: <CAAGYmFf0BTsKTnozP5fSoyO1mZ=pYfzJpMKi2w7jdZoB5cjQ0Q@mail.gmail.com>

On Thu, Sep 8, 2011 at 2:27 AM, Maxim Dounin <mdounin at mdounin.ru> wrote:

> Hello!
>
> On Wed, Sep 07, 2011 at 04:33:38PM -0700, Matthieu Tourne wrote:
>
> > Thanks for the all-included patch from 1.1.2
> >
> > It seems that keepalive connections aren't working in combination with
> > proxy_buffering off; (c->read->ready is set to 1)
> > It worked before this last update, is this the intended behavior ?
>
> Linux with epoll, right?  Please try the following patch:
>

Yes, Linux with epoll.
The patch below seems to fix the issue. Thank you Maxim!


>
> diff --git a/src/http/modules/ngx_http_upstream_keepalive_module.c
> b/src/http/modules/ngx_http_upstream_keepalive_module.c
> --- a/src/http/modules/ngx_http_upstream_keepalive_module.c
> +++ b/src/http/modules/ngx_http_upstream_keepalive_module.c
> @@ -321,7 +321,6 @@ ngx_http_upstream_free_keepalive_peer(ng
>     if (kp->failed
>         || c == NULL
>         || c->read->eof
> -        || c->read->ready
>         || c->read->error
>         || c->read->timedout
>         || c->write->error
> @@ -382,6 +381,10 @@ ngx_http_upstream_free_keepalive_peer(ng
>     item->socklen = pc->socklen;
>     ngx_memcpy(&item->sockaddr, pc->sockaddr, pc->socklen);
>
> +    if (c->read->ready) {
> +        ngx_http_upstream_keepalive_close_handler(c->read);
> +    }
> +
>  invalid:
>
>     kp->original_free_peer(pc, kp->data, state);
> diff --git a/src/http/ngx_http_upstream.c b/src/http/ngx_http_upstream.c
> --- a/src/http/ngx_http_upstream.c
> +++ b/src/http/ngx_http_upstream.c
> @@ -2142,7 +2142,7 @@ ngx_http_upstream_send_response(ngx_http
>                 return;
>             }
>
> -            if (u->peer.connection->read->ready) {
> +            if (u->peer.connection->read->ready || u->length == 0) {
>                 ngx_http_upstream_process_non_buffered_upstream(r, u);
>             }
>         }
>
> Second part fixes one more problem with handling of responses with
> empty body over keepalive connections, found during testing with
> various event methods.
>
> Maxim Dounin
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110908/42843394/attachment.html>

From nginx-forum at nginx.us  Fri Sep  9 09:05:43 2011
From: nginx-forum at nginx.us (jkriesten)
Date: Fri, 09 Sep 2011 05:05:43 -0400
Subject: seg fault with 1.0.6
Message-ID: <fd1d62d1556c177ffafc198f0ce64138.NginxMailingListEnglish@forum.nginx.org>

Hi,

yesterday evening both servers with nginx 1.0.6 set faulted within 10
minutes, i suspect a modified header or something. 

Sep  8 21:33:37 andromeda-1 kernel: [1223057.855838] nginx[28406]:
segfault at 8 ip 000000000041a423 sp 00007ffffb5b0778 error 4 in
nginx[400000+99000]
Sep  8 21:41:14 andromeda-2 kernel: [1220501.571265] nginx[3195]:
segfault at 8 ip 000000000041a423 sp 00007fff8f6b7a48 error 4 in
nginx[400000+99000]

nginx: nginx version: nginx/1.0.6
nginx: TLS SNI support enabled
nginx: configure arguments: --prefix=/usr/local/www
--conf-path=/usr/local/www/cfg/nginx.conf --pid-path=/var/run/nginx.pid
--error-log-path=/usr/local/www/log/error.log
--http-log-path=/usr/local/www/log/transfer.log
--lock-path=/var/lock/nginx.lock --with-http_flv_module
--with-http_ssl_module --with-http_perl_module
--with-http_gzip_static_module --with-http_stub_status_module
--without-http_scgi_module --without-http_uwsgi_module
--add-module=/home/kriesten/translate/src/memc-nginx-module
--add-module=/home/kriesten/translate/src/srcache-nginx-module
--add-module=/home/kriesten/translate/src/ngx_http_upstream_keepalive-0.6
--http-client-body-temp-path=/usr/local/www/tmp/client
--http-proxy-temp-path=/usr/local/www/proxy/tmp
--http-fastcgi-temp-path=/usr/local/www/tmp/fcgi

I'm going to enable debugging to have something to work with when it
occurs again. Anyone else seen this?

Best regards, --- Jan.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215000,215000#msg-215000


From Pekka.Panula at sofor.fi  Fri Sep  9 10:32:31 2011
From: Pekka.Panula at sofor.fi (Pekka.Panula at sofor.fi)
Date: Fri, 9 Sep 2011 13:32:31 +0300
Subject: nginx SSO service for different backend web servers/apps
Message-ID: <OF68438F6A.9A131553-ONC2257906.00399618-C2257906.0039E906@sofor.fi>

Hi

We have a couple different services running in different machines, web 
servers/wep applications. 
We would like to have a Single Sign-On (SSO) in front of these servers, 
now they have their own accounts + passwords, eg. login pages.
So i am asking is there any good SSO modules for nginx to use for or any 
help how we can make it happen?

Pekka Panula | Jatkuvat palvelut | Sofor Oy | www.sofor.fi
Takakaarre 3 | PL 51 |FIN-62201 KAUHAVA | tel. +358 6 432 3111 | fax. +358 
6 432 3555 
Mob. + 358 50 384 3232 | pekka.panula at sofor.fi


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110909/6c2e651e/attachment.html>

From igor at sysoev.ru  Fri Sep  9 11:51:07 2011
From: igor at sysoev.ru (Igor Sysoev)
Date: Fri, 9 Sep 2011 15:51:07 +0400
Subject: mp4 module
Message-ID: <20110909115107.GH97288@nginx.com>

I've commit ngx_http_mp4_module in trunk branch.
The patch for 1.0 and last 0.8 is also available:
http://nginx.org/patches/patch.mp4.txt

The module was tested against JW Player.

The module is not built by default, it should be enabled with
   --with-http_mp4_module
If you use 3rd-party mp4 module, it should be removed.
Configuration:

location /mp4/ {
  mp4;
  mp4_buffer_size    1m;    # default 512k
  mp4_max_moov_size  10m;   # default 5m
}

mp4_buffer_size - initial size of buffer to read mp4 file.
This buffer is also used to keep whole meta data of mp4 file (so
called "moov" atom). Maximum size of moov atom which nginx is capable
to read is set by mp4_max_moov_size.


-- 
Igor Sysoev


From Richard.Kearsley at m247.com  Fri Sep  9 12:16:35 2011
From: Richard.Kearsley at m247.com (Richard Kearsley)
Date: Fri, 9 Sep 2011 12:16:35 +0000
Subject: mp4 module
In-Reply-To: <20110909115107.GH97288@nginx.com>
References: <20110909115107.GH97288@nginx.com>
Message-ID: <D1389253-040D-4CF2-98A3-317E64D64966@m247.com>

That's a good feature
Can you confirm if it's compatible with proxy cache?



On 9 Sep 2011, at 12:51, "Igor Sysoev" <igor at sysoev.ru> wrote:

> I've commit ngx_http_mp4_module in trunk branch.
> The patch for 1.0 and last 0.8 is also available:
> http://nginx.org/patches/patch.mp4.txt
> 
> The module was tested against JW Player.
> 
> The module is not built by default, it should be enabled with
>   --with-http_mp4_module
> If you use 3rd-party mp4 module, it should be removed.
> Configuration:
> 
> location /mp4/ {
>  mp4;
>  mp4_buffer_size    1m;    # default 512k
>  mp4_max_moov_size  10m;   # default 5m
> }
> 
> mp4_buffer_size - initial size of buffer to read mp4 file.
> This buffer is also used to keep whole meta data of mp4 file (so
> called "moov" atom). Maximum size of moov atom which nginx is capable
> to read is set by mp4_max_moov_size.
> 
> 
> -- 
> Igor Sysoev
> 
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx


From lists at ruby-forum.com  Fri Sep  9 12:34:41 2011
From: lists at ruby-forum.com (Jan Bujk)
Date: Fri, 09 Sep 2011 14:34:41 +0200
Subject: Optimizing worker_processes,
	worker_connections & PHP_FCGI_CHILDREN - Any Good Tutorial?
In-Reply-To: <dc7aa241d6987d6d4d3146b0229cdcf7.NginxMailingListEnglish@forum.nginx.org>
References: <dc7aa241d6987d6d4d3146b0229cdcf7.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <bd3a3d449e2bca572809cb1bfec7d506@ruby-forum.com>

Hi,
what would you guys recommend to set for a website loaded heavily with 
PHP and some Ajax features and around 2000 users online at the same time 
?
Currently I have settings like:

PHP_FCGI_CHILDREN=10
PHP_FCGI_MAX_REQUESTS=1000


    worker_connections  20392;
     multi_accept off;
     use epoll;
    client_body_timeout 200;
    client_header_timeout 100;.
    client_max_body_size 40M;
    client_body_buffer_size 1280k;
    send_timeout 100;
    limit_zone limit_per_ip $binary_remote_addr 180m;
    limit_conn limit_per_ip 150;
    limit_rate 2200k;
    keepalive_timeout 15;

Using nginx/0.7.67 and php 5.2.17.

I cannot migrate to php 5.3.3, so I cannot use native php-fpm.
Should I try to tweak fast-cgi better or try to get PHP-FPM working ?
Would nginx upgrade help with performance ?
Website works but with many users it gets veryyyyyyyyy slow with 
responses.
Thanks guys !

-- 
Posted via http://www.ruby-forum.com/.


From igor at sysoev.ru  Fri Sep  9 13:08:12 2011
From: igor at sysoev.ru (Igor Sysoev)
Date: Fri, 9 Sep 2011 17:08:12 +0400
Subject: mp4 module
In-Reply-To: <D1389253-040D-4CF2-98A3-317E64D64966@m247.com>
References: <20110909115107.GH97288@nginx.com>
	<D1389253-040D-4CF2-98A3-317E64D64966@m247.com>
Message-ID: <20110909130812.GI97288@nginx.com>

On Fri, Sep 09, 2011 at 12:16:35PM +0000, Richard Kearsley wrote:
> That's a good feature
> Can you confirm if it's compatible with proxy cache?

No, currently it can work only with static files.


-- 
Igor Sysoev


From r at roze.lv  Fri Sep  9 13:37:52 2011
From: r at roze.lv (Reinis Rozitis)
Date: Fri, 9 Sep 2011 16:37:52 +0300
Subject: Optimizing worker_processes,
	worker_connections & PHP_FCGI_CHILDREN - Any Good Tutorial?
In-Reply-To: <bd3a3d449e2bca572809cb1bfec7d506@ruby-forum.com>
References: <dc7aa241d6987d6d4d3146b0229cdcf7.NginxMailingListEnglish@forum.nginx.org>
	<bd3a3d449e2bca572809cb1bfec7d506@ruby-forum.com>
Message-ID: <14CFA31DBC9D45B1914C2396711DF8EF@DD21>

> I cannot migrate to php 5.3.3, so I cannot use native php-fpm.
> Should I try to tweak fast-cgi better or try to get PHP-FPM working ?

You can apply FPM patches (it was a quite a long time before it got into core) also to 5.2.x tree and for 5.2.17 ( 
http://php-fpm.org/downloads/php-5.2.17-fpm-0.5.14.diff.gz ).
The dynamic manager will let you adjust best process numbers and will show the slow scripts.

rr 


From francois.beausoleil at gmail.com  Fri Sep  9 14:47:10 2011
From: francois.beausoleil at gmail.com (=?utf-8?Q?Fran=C3=A7ois_Beausoleil?=)
Date: Fri, 9 Sep 2011 10:47:10 -0400
Subject: Encoded slashes in URL with proxy = trouble?
Message-ID: <1B2B5635B15447D38C2EFC4E4536FEFE@gmail.com>

Hi!  

Nginx is in front of the RabbitMQ management extension. Some of the URLs the extension generates contain en embedded slash character (%2F):

 http://somehost/#/queues/%2F/events

The encoded slash represents the vhost I want to get information about. I found an older ServerFault question with no answer[1], and was wondering if any of you had a way to let Nginx pass through the encoded slash?

Thanks!
Fran?ois

[1] http://serverfault.com/questions/289188/nginx-passenger-encoded-slash


From igor at sysoev.ru  Fri Sep  9 15:06:57 2011
From: igor at sysoev.ru (Igor Sysoev)
Date: Fri, 9 Sep 2011 19:06:57 +0400
Subject: Encoded slashes in URL with proxy = trouble?
In-Reply-To: <1B2B5635B15447D38C2EFC4E4536FEFE@gmail.com>
References: <1B2B5635B15447D38C2EFC4E4536FEFE@gmail.com>
Message-ID: <20110909150657.GL97288@nginx.com>

On Fri, Sep 09, 2011 at 10:47:10AM -0400, Fran?ois Beausoleil wrote:
> Hi!  
> 
> Nginx is in front of the RabbitMQ management extension. Some of the URLs the extension generates contain en embedded slash character (%2F):
> 
>  http://somehost/#/queues/%2F/events
> 
> The encoded slash represents the vhost I want to get information about. I found an older ServerFault question with no answer[1], and was wondering if any of you had a way to let Nginx pass through the encoded slash?
> 
> Thanks!
> Fran?ois
> 
> [1] http://serverfault.com/questions/289188/nginx-passenger-encoded-slash

First, I'm not sure that browser sends to a server anything after
hash character "#", since hash mean fragment on page.

As to enconded slash, nginx normalizes URI, it decodes all characters
so "/queues/%2F/events" becames "/queues///events" and then it merges
all slashes, "/./", and "/../" to test URI against locations.
Otherwise, anyone can request something like "/%2E%2E%2E../../etc/passwd"
to get files out of server control. Or to get source text of the script
files instead of executing then.

If you want to pass unchanged request to backend, you can use just
backend name without slash in proxy_pass:

location /queues/ {
    proxy_pass   http://backend;
}

i.e.,

-   proxy_pass   http://backend/;
+   proxy_pass   http://backend;


-- 
Igor Sysoev


From nginx-forum at nginx.us  Fri Sep  9 15:18:17 2011
From: nginx-forum at nginx.us (ynasser)
Date: Fri, 09 Sep 2011 11:18:17 -0400
Subject: "48: Address already in use" error
Message-ID: <be24d1c83248e93a532de39a7fe1f302.NginxMailingListEnglish@forum.nginx.org>

This is the error message I get when I try to run nginx:


nginx: [emerg] bind() to 0.0.0.0:80 failed (48: Address already in use)
nginx: [emerg] bind() to 0.0.0.0:80 failed (48: Address already in use)
nginx: [emerg] bind() to 0.0.0.0:80 failed (48: Address already in use)
nginx: [emerg] bind() to 0.0.0.0:80 failed (48: Address already in use)
nginx: [emerg] bind() to 0.0.0.0:80 failed (48: Address already in use)
nginx: [emerg] still could not bind()


I have disabled apache, so it's not that. Previous to this, I
uninstalled nginx (using mac osx snow leopard), but 0.0.0.0 and
127.0.0.1 were still displaying "Welcome to nginx". I then re-installed
nginx and I have been trying to run it, but I am getting these errors.

I've tried checking what is on port 80 with this:
$ sudo lsof -i:80 
but nothing is returned.

0.0.0.0:80 displays: "404 Not Found nginx/1.0.6"

Output of "ps ax -o pid,ppid,%cpu,vsz,wchan,command|egrep
'(nginx|PID)'"

>  PID  PPID  %CPU      VSZ WCHAN  COMMAND
>35198     1   0.0  2435856 -      nginx: master process nginx
>35199 35198   0.0  2436044 -      nginx: worker process
>41669 32470   0.0  2425524 -      egrep (nginx|PID)

nginx.conf can be found here: http://pastebin.com/Rq8vKHXn

Would anyone happen to know how to resolve this?

Thank you in advance.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215031,215031#msg-215031


From mauro.stettler at gmail.com  Fri Sep  9 15:21:26 2011
From: mauro.stettler at gmail.com (Mauro Stettler)
Date: Fri, 9 Sep 2011 23:21:26 +0800
Subject: "48: Address already in use" error
In-Reply-To: <be24d1c83248e93a532de39a7fe1f302.NginxMailingListEnglish@forum.nginx.org>
References: <be24d1c83248e93a532de39a7fe1f302.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <CAP-8hjov3T_DfwLATunBdgE+BXew=x1ObG618U_E1nKG-O9J7A@mail.gmail.com>

you probably still have an nginx running and listening on localhost
port 80. what do you mean with install/uninstall, how did you do that?
after uninstalling did you make sure the running process is also dead?

try to "sudo killall nginx" and then try to run it again.

On Fri, Sep 9, 2011 at 23:18, ynasser <nginx-forum at nginx.us> wrote:
> This is the error message I get when I try to run nginx:
>
>
> nginx: [emerg] bind() to 0.0.0.0:80 failed (48: Address already in use)
> nginx: [emerg] bind() to 0.0.0.0:80 failed (48: Address already in use)
> nginx: [emerg] bind() to 0.0.0.0:80 failed (48: Address already in use)
> nginx: [emerg] bind() to 0.0.0.0:80 failed (48: Address already in use)
> nginx: [emerg] bind() to 0.0.0.0:80 failed (48: Address already in use)
> nginx: [emerg] still could not bind()
>
>
> I have disabled apache, so it's not that. Previous to this, I
> uninstalled nginx (using mac osx snow leopard), but 0.0.0.0 and
> 127.0.0.1 were still displaying "Welcome to nginx". I then re-installed
> nginx and I have been trying to run it, but I am getting these errors.
>
> I've tried checking what is on port 80 with this:
> $ sudo lsof -i:80
> but nothing is returned.
>
> 0.0.0.0:80 displays: "404 Not Found nginx/1.0.6"
>
> Output of "ps ax -o pid,ppid,%cpu,vsz,wchan,command|egrep
> '(nginx|PID)'"
>
>> ?PID ?PPID ?%CPU ? ? ?VSZ WCHAN ?COMMAND
>>35198 ? ? 1 ? 0.0 ?2435856 - ? ? ?nginx: master process nginx
>>35199 35198 ? 0.0 ?2436044 - ? ? ?nginx: worker process
>>41669 32470 ? 0.0 ?2425524 - ? ? ?egrep (nginx|PID)
>
> nginx.conf can be found here: http://pastebin.com/Rq8vKHXn
>
> Would anyone happen to know how to resolve this?
>
> Thank you in advance.
>
> Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215031,215031#msg-215031
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx
>


From nginx-forum at nginx.us  Fri Sep  9 15:26:24 2011
From: nginx-forum at nginx.us (ynasser)
Date: Fri, 09 Sep 2011 11:26:24 -0400
Subject: "48: Address already in use" error
In-Reply-To: <CAP-8hjov3T_DfwLATunBdgE+BXew=x1ObG618U_E1nKG-O9J7A@mail.gmail.com>
References: <CAP-8hjov3T_DfwLATunBdgE+BXew=x1ObG618U_E1nKG-O9J7A@mail.gmail.com>
Message-ID: <8faaad58bd61dcef2765fdc0795a7699.NginxMailingListEnglish@forum.nginx.org>

$ sudo killall nginx
Password:
No matching processes were found

When I try to run it, the same error occurs.

To uninstall it, since there was no "easy" way to remove it, I just
removed every file or folder which contained any nginx in its title or
anywhere in the contents. There didn't seem to be any other way to
remove it.

The first time I installed it, I did a manual install. The second time I
used: "sudo port install nginx".

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215031,215033#msg-215033


From lists at ruby-forum.com  Fri Sep  9 17:35:20 2011
From: lists at ruby-forum.com (dhaval d.)
Date: Fri, 09 Sep 2011 19:35:20 +0200
Subject: nginx uid decoding using Base64.decode64
Message-ID: <b20e4b8d8cede5843ed5050aa66157c3@ruby-forum.com>

Hi,
    I was facing a problem while decoding a nginx generated uid using
ruby.

The uid which I get from nginx cookie is "CqCx7E5qMNUxrmAUAwMJAg==" and
when I try to decode this using
 Base64.decode64("CqCx7E5qMNUxrmAUAwMJAg==") ,  it gives me
 "\n\xA0\xB1\xECNj0\xD51\xAE`\x14\x03\x03\t\x02" instead of
"ECB1A00AD5306A4E1460AE3102090303" which I see in my access
logs($uid_got value).

can any one give some suggestions on this issue?

-- 
Posted via http://www.ruby-forum.com/.


From lists at ruby-forum.com  Fri Sep  9 17:40:07 2011
From: lists at ruby-forum.com (dhaval d.)
Date: Fri, 09 Sep 2011 19:40:07 +0200
Subject: "48: Address already in use" error
In-Reply-To: <8faaad58bd61dcef2765fdc0795a7699.NginxMailingListEnglish@forum.nginx.org>
References: <be24d1c83248e93a532de39a7fe1f302.NginxMailingListEnglish@forum.nginx.org>
	<CAP-8hjov3T_DfwLATunBdgE+BXew=x1ObG618U_E1nKG-O9J7A@mail.gmail.com>
	<8faaad58bd61dcef2765fdc0795a7699.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <d1698fdd0c3a36d1e1482fd7e3ea1169@ruby-forum.com>

Did you check your process using ps -ef | grep nginx

Does it list any of the nginx process running? If yes then kill each one
of those and re-check it using above command.


ynasser wrote in post #1020988:
> $ sudo killall nginx
> Password:
> No matching processes were found
>
> When I try to run it, the same error occurs.
>
> To uninstall it, since there was no "easy" way to remove it, I just
> removed every file or folder which contained any nginx in its title or
> anywhere in the contents. There didn't seem to be any other way to
> remove it.
>
> The first time I installed it, I did a manual install. The second time I
> used: "sudo port install nginx".
>
> Posted at Nginx Forum:
> http://forum.nginx.org/read.php?2,215031,215033#msg-215033

-- 
Posted via http://www.ruby-forum.com/.


From nginx-forum at nginx.us  Fri Sep  9 17:52:19 2011
From: nginx-forum at nginx.us (ynasser)
Date: Fri, 09 Sep 2011 13:52:19 -0400
Subject: "48: Address already in use" error
In-Reply-To: <d1698fdd0c3a36d1e1482fd7e3ea1169@ruby-forum.com>
References: <d1698fdd0c3a36d1e1482fd7e3ea1169@ruby-forum.com>
Message-ID: <df3edf3f2c8ffc4e66a944d2b423fd44.NginxMailingListEnglish@forum.nginx.org>

Thank you! It works now. It's a little weird that 'ps -A | grep nginx'
wasn't catching the other processes though, since -e is identical to -A.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215031,215037#msg-215037


From mdounin at mdounin.ru  Fri Sep  9 18:11:09 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Fri, 9 Sep 2011 22:11:09 +0400
Subject: nginx uid decoding using Base64.decode64
In-Reply-To: <b20e4b8d8cede5843ed5050aa66157c3@ruby-forum.com>
References: <b20e4b8d8cede5843ed5050aa66157c3@ruby-forum.com>
Message-ID: <20110909181109.GR1137@mdounin.ru>

Hello!

On Fri, Sep 09, 2011 at 07:35:20PM +0200, dhaval d. wrote:

> Hi,
>     I was facing a problem while decoding a nginx generated uid using
> ruby.
> 
> The uid which I get from nginx cookie is "CqCx7E5qMNUxrmAUAwMJAg==" and
> when I try to decode this using
>  Base64.decode64("CqCx7E5qMNUxrmAUAwMJAg==") ,  it gives me
>  "\n\xA0\xB1\xECNj0\xD51\xAE`\x14\x03\x03\t\x02" instead of
> "ECB1A00AD5306A4E1460AE3102090303" which I see in my access
> logs($uid_got value).
> 
> can any one give some suggestions on this issue?

Logged value is a hex representation of the data, using 4 host-order 
uint32_t values.  I.e. "\n\xA0\xB1\xEC" matches "ECB1A00A" and so on.

Maxim Dounin


From lists at ruby-forum.com  Fri Sep  9 19:29:20 2011
From: lists at ruby-forum.com (dhaval d.)
Date: Fri, 09 Sep 2011 21:29:20 +0200
Subject: How to read http_userid's Base64 encoded userid from Ruby?
In-Reply-To: <20100209122047.GB2382@oemcomputer.oerks.de>
References: <20100209122047.GB2382@oemcomputer.oerks.de>
Message-ID: <7acd1e5a1b852dc1f0c91b4b4593f88e@ruby-forum.com>

Hello Sven,
           I came across the same problem. But the solution you 
mentioned does not work with my program. When I run this code
cookie = CqCx7E5qMNUxrmAUAwMJAg==

cc = cookie.unpack('m*').first; rr = cc.split("").map{|c| 
c[0].to_i}.inject([]) {|v,s| v.push sprintf("%02X", s); v; 
}.values_at(3, 2, 1, 0, 7, 6, 5, 4, 11, 10, 9, 8, 15, 14, 13, 
12).join("")

I get back the result as "00000000000000000000000100000000" but actually 
it should be ECB1A00AD5306A4E1460AE3102090303

Do you know what am i doing wrong?

-- 
Posted via http://www.ruby-forum.com/.


From nginx-forum at nginx.us  Fri Sep  9 20:05:00 2011
From: nginx-forum at nginx.us (ynasser)
Date: Fri, 09 Sep 2011 16:05:00 -0400
Subject: Beginner questions about proxy configuration and white space
In-Reply-To: <17696a632c172de23cb235a2b6fd33c4.NginxMailingListEnglish@forum.nginx.org>
References: <72ca536398bcc048cbca71aa39cb728e.NginxMailingListEnglish@forum.nginx.org>
	<17696a632c172de23cb235a2b6fd33c4.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <f2b3e6c431ee58181971eb41227a831d.NginxMailingListEnglish@forum.nginx.org>

I reinstalled nginx and re-added that directive to the new configutation
file. Now this happens:

>> sudo nginx
>> nginx: [emerg] bind() to 10.0.1.187:2525 failed (49: Can't assign
requested address)

In regards to the example given above:

server {
	listen some_ip:some_port default;

	location / {
		add_header X-h1 h1;
		add_header X-h2 "h2 h2";
		add_header X-h3 h3;
		add_header X-h4 h4;
		add_header X-h5 h5;
		add_header X-h6 "h6 h6";
		add_header X-h7 h7;
		proxy_pass http://backend;
		}
	}

some_ip:some_port is the site that I'm trying to pass the headers to.
For http://backend, I'm using "http://127.0.0.1"

Any help would be appreciated.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,214982,215040#msg-215040


From magicbearmo at gmail.com  Fri Sep  9 20:08:27 2011
From: magicbearmo at gmail.com (MagicBear)
Date: Sat, 10 Sep 2011 04:08:27 +0800
Subject: Beginner questions about proxy configuration and white space
In-Reply-To: <f2b3e6c431ee58181971eb41227a831d.NginxMailingListEnglish@forum.nginx.org>
References: <72ca536398bcc048cbca71aa39cb728e.NginxMailingListEnglish@forum.nginx.org>
	<17696a632c172de23cb235a2b6fd33c4.NginxMailingListEnglish@forum.nginx.org>
	<f2b3e6c431ee58181971eb41227a831d.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <CAENo=s37Ydi9Zk+PkSfexic3JtXu0DtCpcXrTHP3W59sm7EHiQ@mail.gmail.com>

if you are running nginx, you may use "nginx -s reload" to reload the
config.
or killall -9 nginx; nginx  to restart nginx

2011/9/10 ynasser <nginx-forum at nginx.us>

> I reinstalled nginx and re-added that directive to the new configutation
> file. Now this happens:
>
> >> sudo nginx
> >> nginx: [emerg] bind() to 10.0.1.187:2525 failed (49: Can't assign
> requested address)
>
> In regards to the example given above:
>
> server {
>        listen some_ip:some_port default;
>
>        location / {
>                add_header X-h1 h1;
>                add_header X-h2 "h2 h2";
>                add_header X-h3 h3;
>                add_header X-h4 h4;
>                add_header X-h5 h5;
>                add_header X-h6 "h6 h6";
>                add_header X-h7 h7;
>                proxy_pass http://backend;
>                }
>        }
>
> some_ip:some_port is the site that I'm trying to pass the headers to.
> For http://backend, I'm using "http://127.0.0.1"
>
> Any help would be appreciated.
>
> Posted at Nginx Forum:
> http://forum.nginx.org/read.php?2,214982,215040#msg-215040
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110910/e72db81c/attachment.html>

From lists at ruby-forum.com  Fri Sep  9 20:33:58 2011
From: lists at ruby-forum.com (dhaval d.)
Date: Fri, 09 Sep 2011 22:33:58 +0200
Subject: nginx uid decoding using Base64.decode64
In-Reply-To: <b20e4b8d8cede5843ed5050aa66157c3@ruby-forum.com>
References: <b20e4b8d8cede5843ed5050aa66157c3@ruby-forum.com>
Message-ID: <ecbf4892c0629be0a9a9843faed6e8bc@ruby-forum.com>

Hello Maxim,
            Thanks a lot for your reply. Because of which I was able to 
figure out the way to decode it. Here is the snippet for those who wants 
to know.

cookie = "CqCx7E5qMNUxrmAUAwMJAg=="

Base64.decode64(cookie).unpack("L*").collect{|num| 
num.to_s(16)}.join("").upcase

-- 
Posted via http://www.ruby-forum.com/.


From nginx-forum at nginx.us  Fri Sep  9 21:00:39 2011
From: nginx-forum at nginx.us (ynasser)
Date: Fri, 09 Sep 2011 17:00:39 -0400
Subject: Beginner questions about proxy configuration and white space
In-Reply-To: <CAENo=s37Ydi9Zk+PkSfexic3JtXu0DtCpcXrTHP3W59sm7EHiQ@mail.gmail.com>
References: <CAENo=s37Ydi9Zk+PkSfexic3JtXu0DtCpcXrTHP3W59sm7EHiQ@mail.gmail.com>
Message-ID: <c4e9db129163a8f93e30810345979741.NginxMailingListEnglish@forum.nginx.org>

I don't think it's a problem with reloading the configuration file,
because I have been reloading it every time after I update and save it.

What is happening is that when I run 'sudo nginx', it doesn't start and
outputs this error:

nginx: [emerg] bind() to some_ip:some_port failed (49: Can't assign
requested address)

I can access some_ip:some_port just fine from a browser and ssh-ing into
it is not a problem.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,214982,215044#msg-215044


From masterkorp at gmail.com  Fri Sep  9 21:02:02 2011
From: masterkorp at gmail.com (Alfredo Palhares)
Date: Fri, 9 Sep 2011 22:02:02 +0100
Subject: block ip/subnet at socket lvl
Message-ID: <CAOhncHG4iA_XbaF8XvdxsH=wjr+JdD90-3sw9r=OsL9BmDP2Tw@mail.gmail.com>

Hello,

I think that out of nginx scope.
On 7 Sep 2011 14:42, "arleybls" <nginx-forum at nginx.us> wrote:
> Currently Nginx can block IP on application level and thus send a 403
> forbidden message back to the client.
>
> I wonder if there's a way to block any request at the socket level,
> preferably not even sending a FIN message to the client but simply
> dropping the packet. Right now i'm using iptables but i'm really looking
> to have such feature on nginx itself, and if not supported right on
> current build, how could I have a hook to the socket and control it on a
> home made server module/handler?
>
> Cheers,
> __Jr
>
> Posted at Nginx Forum:
http://forum.nginx.org/read.php?2,214948,214948#msg-214948
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110909/955c1e54/attachment.html>

From caldcv at gmail.com  Fri Sep  9 21:47:14 2011
From: caldcv at gmail.com (Chris)
Date: Fri, 9 Sep 2011 17:47:14 -0400
Subject: block ip/subnet at socket lvl
In-Reply-To: <CAOhncHG4iA_XbaF8XvdxsH=wjr+JdD90-3sw9r=OsL9BmDP2Tw@mail.gmail.com>
References: <CAOhncHG4iA_XbaF8XvdxsH=wjr+JdD90-3sw9r=OsL9BmDP2Tw@mail.gmail.com>
Message-ID: <CAPF5age9EXeRCYGCwcXZTmoCJPRuX5wq64JyytHTLzt65byyJQ@mail.gmail.com>

You're making it too complicated and just stick with iptables

On Fri, Sep 9, 2011 at 5:02 PM, Alfredo Palhares <masterkorp at gmail.com> wrote:
> Hello,
>
> I think that out of nginx scope.
>
> On 7 Sep 2011 14:42, "arleybls" <nginx-forum at nginx.us> wrote:
>> Currently Nginx can block IP on application level and thus send a 403
>> forbidden message back to the client.
>>
>> I wonder if there's a way to block any request at the socket level,
>> preferably not even sending a FIN message to the client but simply
>> dropping the packet. Right now i'm using iptables but i'm really looking
>> to have such feature on nginx itself, and if not supported right on
>> current build, how could I have a hook to the socket and control it on a
>> home made server module/handler?
>>
>> Cheers,
>> __Jr
>>
>> Posted at Nginx Forum:
>> http://forum.nginx.org/read.php?2,214948,214948#msg-214948
>>
>> _______________________________________________
>> nginx mailing list
>> nginx at nginx.org
>> http://mailman.nginx.org/mailman/listinfo/nginx
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx
>



-- 
--C

"The dumber people think you are, the more surprised they're going to
be when you kill them." - Sir William Clayton


From nginx-forum at nginx.us  Sat Sep 10 00:21:59 2011
From: nginx-forum at nginx.us (pedroaxl)
Date: Fri, 09 Sep 2011 20:21:59 -0400
Subject: Decode url in rewrite
Message-ID: <94b42f69891b4faa8507abd58c54367a.NginxMailingListEnglish@forum.nginx.org>

Hey guys,

I am trying to solve a specific problem I have, but have not succeded.

Users would access a link like this:
http://mysite.com/24234234?key=3123&url=http%3A//mailee.me%3Fname%3Djohn%26code%3D123

This is a normal url with some integer params and a url encoded param. I
want nginx to automatically redirect to the url in the parameters. So I
used the following code in the nginx.conf

          location /go/click {
                  set $new_url $arg_url;
                  set $args '';
                  rewrite ^.*$ $new_url redirect;
          }       

It works beautifully. The problem is that if the encoded url also have
params, the client will receive the params encoded and it may not work
in their system.

I try to use perl code to decode the url before redirecting, but it
didnt work.

This is the code inside http context:
           perl_modules perl/lib;
           perl_require mailee.pm;

This is the code inside location context:
          location /go/click {
                  perl mailee::decode;
                  set $new_url $arg_url;
                  #set $args '';
                  rewrite ^.*$ $args redirect;
          }
And the code in mailee.pm:

package mailee;
use nginx;

sub decode {
        my $r = shift;
        $r->args =~ s/\%3D/=/g;
        $r->args =~ s/\%26/&/g;
        $r->args =~ s/\%3F/?/g;
        $r->args =~ s/\%3A/:/g;
        return OK;
}

1;
__END__

But it didnt work too, because I was not able to modify $args inside the
perl function.

I am not sure if this is the best way to do it and if I am doing it
right. Do you have any other solutions?


Thanks

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215047,215047#msg-215047


From nginx-forum at nginx.us  Sat Sep 10 01:08:08 2011
From: nginx-forum at nginx.us (frostbyte)
Date: Fri, 09 Sep 2011 21:08:08 -0400
Subject: nginx 1.0.6 perl module build failure
Message-ID: <a82f61a72ff66a67df6a9885a13e738b.NginxMailingListEnglish@forum.nginx.org>

Snow Leopard 10.6.7, Xcode 4.0.1:

$ cd nginx-1.0.6
$ ./configure --prefix=/usr/local --with-cc-opt="-I/usr/local/include
-O2 -Wno-deprecated-declarations" --with-ld-opt="-L/usr/local/lib"
--conf-path=/etc/nginx/nginx.conf
--error-log-path=/var/log/nginx/error.log
--http-log-path=/var/log/nginx/access.log
--pid-path=/var/run/nginx/nginx.pid
--lock-path=/var/run/nginx/nginx.lock
--http-client-body-temp-path=/var/run/nginx/client_body_temp
--http-proxy-temp-path=/var/run/nginx/proxy_temp
--http-fastcgi-temp-path=/var/run/nginx/fastcgi_temp
--http-uwsgi-temp-path=/var/run/nginx/uwsgi_temp --with-http_dav_module
--with-http_flv_module --with-mail --with-http_ssl_module
--with-mail_ssl_module --with-http_stub_status_module --with-ipv6
--with-http_perl_module --with-perl=/usr/bin/perl
--with-http_realip_module --with-http_addition_module
--with-http_sub_module --with-http_gzip_static_module

$ make
...
gcc -c -pipe -O -W -Wall -Wpointer-arith -Wno-unused-parameter
-Wunused-function -Wunused-variable -Wno-unused-value -Werror -g
-I/usr/local/include -O2 -Wno-deprecated-declarations -arch x86_64 -arch
i386 -arch ppc -g -pipe -fno-common -DPERL_DARWIN -fno-strict-aliasing
-I/usr/local/include
-I/System/Library/Perl/5.10.0/darwin-thread-multi-2level/CORE  -I
src/core -I src/event -I src/event/modules -I src/os/unix -I
/opt/local/include -I objs -I src/http -I src/http/modules -I
src/http/modules/perl \
		-o objs/src/http/modules/perl/ngx_http_perl_module.o \
		src/http/modules/perl/ngx_http_perl_module.c
In file included from src/http/modules/perl/ngx_http_perl_module.h:17,
                 from src/http/modules/perl/ngx_http_perl_module.c:10:
/System/Library/Perl/5.10.0/darwin-thread-multi-2level/CORE/perl.h:3240:
error: expected specifier-qualifier-list before ?bool?
...

Seems to be a know issue in other contexts.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215048,215048#msg-215048


From orz at loli.my  Sat Sep 10 05:22:44 2011
From: orz at loli.my (=?UTF-8?B?44OT44Oq44OT44Oq4oWk?=)
Date: Sat, 10 Sep 2011 13:22:44 +0800
Subject: Beginner questions about proxy configuration and white space
In-Reply-To: <c4e9db129163a8f93e30810345979741.NginxMailingListEnglish@forum.nginx.org>
References: <CAENo=s37Ydi9Zk+PkSfexic3JtXu0DtCpcXrTHP3W59sm7EHiQ@mail.gmail.com>
	<c4e9db129163a8f93e30810345979741.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <CAENo=s0CuGk0UqzQmomX89vPzECw39WOA5izxaP0BwfFuOHfcQ@mail.gmail.com>

If you only using one ip, you can use   listen 80 default;
The some_ip:some_port is using on my product server, it shows no problem.

2011/9/10 ynasser <nginx-forum at nginx.us>

> I don't think it's a problem with reloading the configuration file,
> because I have been reloading it every time after I update and save it.
>
> What is happening is that when I run 'sudo nginx', it doesn't start and
> outputs this error:
>
> nginx: [emerg] bind() to some_ip:some_port failed (49: Can't assign
> requested address)
>
> I can access some_ip:some_port just fine from a browser and ssh-ing into
> it is not a problem.
>
> Posted at Nginx Forum:
> http://forum.nginx.org/read.php?2,214982,215044#msg-215044
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110910/dbefcd63/attachment.html>

From mdounin at mdounin.ru  Sat Sep 10 11:17:24 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Sat, 10 Sep 2011 15:17:24 +0400
Subject: nginx 1.0.6 perl module build failure
In-Reply-To: <a82f61a72ff66a67df6a9885a13e738b.NginxMailingListEnglish@forum.nginx.org>
References: <a82f61a72ff66a67df6a9885a13e738b.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20110910111724.GU1137@mdounin.ru>

Hello!

On Fri, Sep 09, 2011 at 09:08:08PM -0400, frostbyte wrote:

> Snow Leopard 10.6.7, Xcode 4.0.1:
> 
> $ cd nginx-1.0.6
> $ ./configure --prefix=/usr/local --with-cc-opt="-I/usr/local/include
> -O2 -Wno-deprecated-declarations" --with-ld-opt="-L/usr/local/lib"
> --conf-path=/etc/nginx/nginx.conf
> --error-log-path=/var/log/nginx/error.log
> --http-log-path=/var/log/nginx/access.log
> --pid-path=/var/run/nginx/nginx.pid
> --lock-path=/var/run/nginx/nginx.lock
> --http-client-body-temp-path=/var/run/nginx/client_body_temp
> --http-proxy-temp-path=/var/run/nginx/proxy_temp
> --http-fastcgi-temp-path=/var/run/nginx/fastcgi_temp
> --http-uwsgi-temp-path=/var/run/nginx/uwsgi_temp --with-http_dav_module
> --with-http_flv_module --with-mail --with-http_ssl_module
> --with-mail_ssl_module --with-http_stub_status_module --with-ipv6
> --with-http_perl_module --with-perl=/usr/bin/perl
> --with-http_realip_module --with-http_addition_module
> --with-http_sub_module --with-http_gzip_static_module
> 
> $ make
> ...
> gcc -c -pipe -O -W -Wall -Wpointer-arith -Wno-unused-parameter
> -Wunused-function -Wunused-variable -Wno-unused-value -Werror -g
> -I/usr/local/include -O2 -Wno-deprecated-declarations -arch x86_64 -arch
> i386 -arch ppc -g -pipe -fno-common -DPERL_DARWIN -fno-strict-aliasing
> -I/usr/local/include
> -I/System/Library/Perl/5.10.0/darwin-thread-multi-2level/CORE  -I
> src/core -I src/event -I src/event/modules -I src/os/unix -I
> /opt/local/include -I objs -I src/http -I src/http/modules -I
> src/http/modules/perl \
> 		-o objs/src/http/modules/perl/ngx_http_perl_module.o \
> 		src/http/modules/perl/ngx_http_perl_module.c
> In file included from src/http/modules/perl/ngx_http_perl_module.h:17,
>                  from src/http/modules/perl/ngx_http_perl_module.c:10:
> /System/Library/Perl/5.10.0/darwin-thread-multi-2level/CORE/perl.h:3240:
> error: expected specifier-qualifier-list before ?bool?
> ...
> 
> Seems to be a know issue in other contexts.

Looks like problem with perl in your environment.

You may try building basic samples from perlembed manpage to see 
what goes wrong, most likely either ccopts as returned by "perl 
-MExtUtils::Embed -e ccopts" are incorrect, or they correspond to 
other C compiler (check "perl -V" output).

Maxim Dounin


From nginx-forum at nginx.us  Sat Sep 10 11:40:31 2011
From: nginx-forum at nginx.us (carrierdetect)
Date: Sat, 10 Sep 2011 07:40:31 -0400
Subject: "port_in_redirect off" not working.
Message-ID: <d05b25b8d9bed5f6950ec1b7bd449231.NginxMailingListEnglish@forum.nginx.org>

Hello,

I'm running nginx/1.0.5 behing Varnish and serving up static html.
Whenever I visit a URL which is a directory but without a trailing
slash, e.g. /blah, it redirects to the non-80 port which nginx is
running on, e.g. server:5000/blah/. The config is as follows:

http {
    include       mime.types;
    default_type  application/octet-stream;
    sendfile        on;

    server {
        listen       5000;
        server_name  server.com;
        root /path/html;

        location / {
            index  index.html index.htm;
            port_in_redirect off;
       }
    }
}

I've tried moving "port_in_redirect off" to http and server contexts,
and setting "server_name_in_redirect off", to no avail. The Varnish
configuration is as simple as can be and just looks for the request
hostname and forwards to the nginx backend.

Regards,

Andrew

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215059,215059#msg-215059


From igor at sysoev.ru  Sat Sep 10 11:46:26 2011
From: igor at sysoev.ru (Igor Sysoev)
Date: Sat, 10 Sep 2011 15:46:26 +0400
Subject: nginx 1.0.6 perl module build failure
In-Reply-To: <20110910111724.GU1137@mdounin.ru>
References: <a82f61a72ff66a67df6a9885a13e738b.NginxMailingListEnglish@forum.nginx.org>
	<20110910111724.GU1137@mdounin.ru>
Message-ID: <20110910114625.GB18804@nginx.com>

On Sat, Sep 10, 2011 at 03:17:24PM +0400, Maxim Dounin wrote:
> Hello!
> 
> On Fri, Sep 09, 2011 at 09:08:08PM -0400, frostbyte wrote:
> 
> > Snow Leopard 10.6.7, Xcode 4.0.1:
> > 
> > $ cd nginx-1.0.6
> > $ ./configure --prefix=/usr/local --with-cc-opt="-I/usr/local/include
> > -O2 -Wno-deprecated-declarations" --with-ld-opt="-L/usr/local/lib"
> > --conf-path=/etc/nginx/nginx.conf
> > --error-log-path=/var/log/nginx/error.log
> > --http-log-path=/var/log/nginx/access.log
> > --pid-path=/var/run/nginx/nginx.pid
> > --lock-path=/var/run/nginx/nginx.lock
> > --http-client-body-temp-path=/var/run/nginx/client_body_temp
> > --http-proxy-temp-path=/var/run/nginx/proxy_temp
> > --http-fastcgi-temp-path=/var/run/nginx/fastcgi_temp
> > --http-uwsgi-temp-path=/var/run/nginx/uwsgi_temp --with-http_dav_module
> > --with-http_flv_module --with-mail --with-http_ssl_module
> > --with-mail_ssl_module --with-http_stub_status_module --with-ipv6
> > --with-http_perl_module --with-perl=/usr/bin/perl
> > --with-http_realip_module --with-http_addition_module
> > --with-http_sub_module --with-http_gzip_static_module
> > 
> > $ make
> > ...
> > gcc -c -pipe -O -W -Wall -Wpointer-arith -Wno-unused-parameter
> > -Wunused-function -Wunused-variable -Wno-unused-value -Werror -g
> > -I/usr/local/include -O2 -Wno-deprecated-declarations -arch x86_64 -arch
> > i386 -arch ppc -g -pipe -fno-common -DPERL_DARWIN -fno-strict-aliasing
> > -I/usr/local/include
> > -I/System/Library/Perl/5.10.0/darwin-thread-multi-2level/CORE  -I
> > src/core -I src/event -I src/event/modules -I src/os/unix -I
> > /opt/local/include -I objs -I src/http -I src/http/modules -I
> > src/http/modules/perl \
> > 		-o objs/src/http/modules/perl/ngx_http_perl_module.o \
> > 		src/http/modules/perl/ngx_http_perl_module.c
> > In file included from src/http/modules/perl/ngx_http_perl_module.h:17,
> >                  from src/http/modules/perl/ngx_http_perl_module.c:10:
> > /System/Library/Perl/5.10.0/darwin-thread-multi-2level/CORE/perl.h:3240:
> > error: expected specifier-qualifier-list before ?bool?
> > ...
> > 
> > Seems to be a know issue in other contexts.
> 
> Looks like problem with perl in your environment.
> 
> You may try building basic samples from perlembed manpage to see 
> what goes wrong, most likely either ccopts as returned by "perl 
> -MExtUtils::Embed -e ccopts" are incorrect, or they correspond to 
> other C compiler (check "perl -V" output).

No, this is known issue with MacOSX system perl.
This is conflict between "bool" definition included via
libkern/OSAtomic.h and perl's CORE/handy.h.
I do not personaly encounter this since I have perl from macports :)


-- 
Igor Sysoev


From mdounin at mdounin.ru  Sat Sep 10 12:08:26 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Sat, 10 Sep 2011 16:08:26 +0400
Subject: "port_in_redirect off" not working.
In-Reply-To: <d05b25b8d9bed5f6950ec1b7bd449231.NginxMailingListEnglish@forum.nginx.org>
References: <d05b25b8d9bed5f6950ec1b7bd449231.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20110910120826.GV1137@mdounin.ru>

Hello!

On Sat, Sep 10, 2011 at 07:40:31AM -0400, carrierdetect wrote:

> Hello,
> 
> I'm running nginx/1.0.5 behing Varnish and serving up static html.
> Whenever I visit a URL which is a directory but without a trailing
> slash, e.g. /blah, it redirects to the non-80 port which nginx is
> running on, e.g. server:5000/blah/. The config is as follows:
> 
> http {
>     include       mime.types;
>     default_type  application/octet-stream;
>     sendfile        on;
> 
>     server {
>         listen       5000;
>         server_name  server.com;
>         root /path/html;
> 
>         location / {
>             index  index.html index.htm;
>             port_in_redirect off;
>        }
>     }
> }

Works ok here.

> I've tried moving "port_in_redirect off" to http and server contexts,
> and setting "server_name_in_redirect off", to no avail. The Varnish
> configuration is as simple as can be and just looks for the request
> hostname and forwards to the nginx backend.

Try connecting nginx directly or looking though tcpdump to see 
what actually nginx returns.

I suspect it's actually caching issue, either Varnish cached the 
redirect from your previous nginx config without "port_in_redirect 
off;" or your browser did.

Maxim Dounin


From nginx-forum at nginx.us  Sat Sep 10 13:12:13 2011
From: nginx-forum at nginx.us (carrierdetect)
Date: Sat, 10 Sep 2011 09:12:13 -0400
Subject: "port_in_redirect off" not working.
In-Reply-To: <20110910120826.GV1137@mdounin.ru>
References: <20110910120826.GV1137@mdounin.ru>
Message-ID: <ed454f569f9e91e8a8596b402dc05574.NginxMailingListEnglish@forum.nginx.org>

Hi Maxim,

Maxim Dounin Wrote:
-------------------------------------------------------

> I suspect it's actually caching issue, either
> Varnish cached the 
> redirect from your previous nginx config without
> "port_in_redirect 
> off;" or your browser did.

Ah, so it appears that my browser had cached the redirect (and I feel
silly for not having checked this).

Many thanks for your help!

Regards,

Andrew

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215059,215066#msg-215066


From mdounin at mdounin.ru  Sat Sep 10 13:24:04 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Sat, 10 Sep 2011 17:24:04 +0400
Subject: nginx 1.0.6 perl module build failure
In-Reply-To: <20110910114625.GB18804@nginx.com>
References: <a82f61a72ff66a67df6a9885a13e738b.NginxMailingListEnglish@forum.nginx.org>
	<20110910111724.GU1137@mdounin.ru>
	<20110910114625.GB18804@nginx.com>
Message-ID: <20110910132404.GW1137@mdounin.ru>

Hello!

On Sat, Sep 10, 2011 at 03:46:26PM +0400, Igor Sysoev wrote:

> On Sat, Sep 10, 2011 at 03:17:24PM +0400, Maxim Dounin wrote:
> > Hello!
> > 
> > On Fri, Sep 09, 2011 at 09:08:08PM -0400, frostbyte wrote:
> > 
> > > Snow Leopard 10.6.7, Xcode 4.0.1:
> > > 
> > > $ cd nginx-1.0.6
> > > $ ./configure --prefix=/usr/local --with-cc-opt="-I/usr/local/include
> > > -O2 -Wno-deprecated-declarations" --with-ld-opt="-L/usr/local/lib"
> > > --conf-path=/etc/nginx/nginx.conf
> > > --error-log-path=/var/log/nginx/error.log
> > > --http-log-path=/var/log/nginx/access.log
> > > --pid-path=/var/run/nginx/nginx.pid
> > > --lock-path=/var/run/nginx/nginx.lock
> > > --http-client-body-temp-path=/var/run/nginx/client_body_temp
> > > --http-proxy-temp-path=/var/run/nginx/proxy_temp
> > > --http-fastcgi-temp-path=/var/run/nginx/fastcgi_temp
> > > --http-uwsgi-temp-path=/var/run/nginx/uwsgi_temp --with-http_dav_module
> > > --with-http_flv_module --with-mail --with-http_ssl_module
> > > --with-mail_ssl_module --with-http_stub_status_module --with-ipv6
> > > --with-http_perl_module --with-perl=/usr/bin/perl
> > > --with-http_realip_module --with-http_addition_module
> > > --with-http_sub_module --with-http_gzip_static_module
> > > 
> > > $ make
> > > ...
> > > gcc -c -pipe -O -W -Wall -Wpointer-arith -Wno-unused-parameter
> > > -Wunused-function -Wunused-variable -Wno-unused-value -Werror -g
> > > -I/usr/local/include -O2 -Wno-deprecated-declarations -arch x86_64 -arch
> > > i386 -arch ppc -g -pipe -fno-common -DPERL_DARWIN -fno-strict-aliasing
> > > -I/usr/local/include
> > > -I/System/Library/Perl/5.10.0/darwin-thread-multi-2level/CORE  -I
> > > src/core -I src/event -I src/event/modules -I src/os/unix -I
> > > /opt/local/include -I objs -I src/http -I src/http/modules -I
> > > src/http/modules/perl \
> > > 		-o objs/src/http/modules/perl/ngx_http_perl_module.o \
> > > 		src/http/modules/perl/ngx_http_perl_module.c
> > > In file included from src/http/modules/perl/ngx_http_perl_module.h:17,
> > >                  from src/http/modules/perl/ngx_http_perl_module.c:10:
> > > /System/Library/Perl/5.10.0/darwin-thread-multi-2level/CORE/perl.h:3240:
> > > error: expected specifier-qualifier-list before ?bool?
> > > ...
> > > 
> > > Seems to be a know issue in other contexts.
> > 
> > Looks like problem with perl in your environment.
> > 
> > You may try building basic samples from perlembed manpage to see 
> > what goes wrong, most likely either ccopts as returned by "perl 
> > -MExtUtils::Embed -e ccopts" are incorrect, or they correspond to 
> > other C compiler (check "perl -V" output).
> 
> No, this is known issue with MacOSX system perl.
> This is conflict between "bool" definition included via
> libkern/OSAtomic.h and perl's CORE/handy.h.
> I do not personaly encounter this since I have perl from macports :)

Error message suggests that "bool" is not defined.  And looking 
though [1] I don't see bool defined via libkern/OSAtomic.h.  
OSTypes.h defines "Boolean", but not "bool".

While I feel perl's handy.h is utterly wrong in it's aproach to 
"bool" (it's 2011 now, more than 10 years since C99 defined 
stdbool.h!), I still doesn't see what happens here.  Additionally, 
I see Apple hacks handy.h on it's system perl to include 
stdbool.h[2], and this makes me wonder even more.

Could you please confirm that perlembed sample does actually 
compiles fine with Mac OS X system perl without extra includes 
used?

[1] http://www.opensource.apple.com/source/xnu/xnu-1699.22.81/libkern/libkern/OSAtomic.h
[2] http://www.opensource.apple.com/source/perl/perl-73/5.10/fix/handy.h.ed

Maxim Dounin


From igor at sysoev.ru  Sat Sep 10 14:09:46 2011
From: igor at sysoev.ru (Igor Sysoev)
Date: Sat, 10 Sep 2011 18:09:46 +0400
Subject: nginx 1.0.6 perl module build failure
In-Reply-To: <20110910132404.GW1137@mdounin.ru>
References: <a82f61a72ff66a67df6a9885a13e738b.NginxMailingListEnglish@forum.nginx.org>
	<20110910111724.GU1137@mdounin.ru>
	<20110910114625.GB18804@nginx.com>
	<20110910132404.GW1137@mdounin.ru>
Message-ID: <20110910140946.GC18804@nginx.com>

On Sat, Sep 10, 2011 at 05:24:04PM +0400, Maxim Dounin wrote:
> Hello!
> 
> On Sat, Sep 10, 2011 at 03:46:26PM +0400, Igor Sysoev wrote:
> 
> > On Sat, Sep 10, 2011 at 03:17:24PM +0400, Maxim Dounin wrote:
> > > Hello!
> > > 
> > > On Fri, Sep 09, 2011 at 09:08:08PM -0400, frostbyte wrote:
> > > 
> > > > Snow Leopard 10.6.7, Xcode 4.0.1:
> > > > 
> > > > $ cd nginx-1.0.6
> > > > $ ./configure --prefix=/usr/local --with-cc-opt="-I/usr/local/include
> > > > -O2 -Wno-deprecated-declarations" --with-ld-opt="-L/usr/local/lib"
> > > > --conf-path=/etc/nginx/nginx.conf
> > > > --error-log-path=/var/log/nginx/error.log
> > > > --http-log-path=/var/log/nginx/access.log
> > > > --pid-path=/var/run/nginx/nginx.pid
> > > > --lock-path=/var/run/nginx/nginx.lock
> > > > --http-client-body-temp-path=/var/run/nginx/client_body_temp
> > > > --http-proxy-temp-path=/var/run/nginx/proxy_temp
> > > > --http-fastcgi-temp-path=/var/run/nginx/fastcgi_temp
> > > > --http-uwsgi-temp-path=/var/run/nginx/uwsgi_temp --with-http_dav_module
> > > > --with-http_flv_module --with-mail --with-http_ssl_module
> > > > --with-mail_ssl_module --with-http_stub_status_module --with-ipv6
> > > > --with-http_perl_module --with-perl=/usr/bin/perl
> > > > --with-http_realip_module --with-http_addition_module
> > > > --with-http_sub_module --with-http_gzip_static_module
> > > > 
> > > > $ make
> > > > ...
> > > > gcc -c -pipe -O -W -Wall -Wpointer-arith -Wno-unused-parameter
> > > > -Wunused-function -Wunused-variable -Wno-unused-value -Werror -g
> > > > -I/usr/local/include -O2 -Wno-deprecated-declarations -arch x86_64 -arch
> > > > i386 -arch ppc -g -pipe -fno-common -DPERL_DARWIN -fno-strict-aliasing
> > > > -I/usr/local/include
> > > > -I/System/Library/Perl/5.10.0/darwin-thread-multi-2level/CORE  -I
> > > > src/core -I src/event -I src/event/modules -I src/os/unix -I
> > > > /opt/local/include -I objs -I src/http -I src/http/modules -I
> > > > src/http/modules/perl \
> > > > 		-o objs/src/http/modules/perl/ngx_http_perl_module.o \
> > > > 		src/http/modules/perl/ngx_http_perl_module.c
> > > > In file included from src/http/modules/perl/ngx_http_perl_module.h:17,
> > > >                  from src/http/modules/perl/ngx_http_perl_module.c:10:
> > > > /System/Library/Perl/5.10.0/darwin-thread-multi-2level/CORE/perl.h:3240:
> > > > error: expected specifier-qualifier-list before ?bool?
> > > > ...
> > > > 
> > > > Seems to be a know issue in other contexts.
> > > 
> > > Looks like problem with perl in your environment.
> > > 
> > > You may try building basic samples from perlembed manpage to see 
> > > what goes wrong, most likely either ccopts as returned by "perl 
> > > -MExtUtils::Embed -e ccopts" are incorrect, or they correspond to 
> > > other C compiler (check "perl -V" output).
> > 
> > No, this is known issue with MacOSX system perl.
> > This is conflict between "bool" definition included via
> > libkern/OSAtomic.h and perl's CORE/handy.h.
> > I do not personaly encounter this since I have perl from macports :)
> 
> Error message suggests that "bool" is not defined.  And looking 
> though [1] I don't see bool defined via libkern/OSAtomic.h.  
> OSTypes.h defines "Boolean", but not "bool".
> 
> While I feel perl's handy.h is utterly wrong in it's aproach to 
> "bool" (it's 2011 now, more than 10 years since C99 defined 
> stdbool.h!), I still doesn't see what happens here.  Additionally, 
> I see Apple hacks handy.h on it's system perl to include 
> stdbool.h[2], and this makes me wonder even more.
> 
> Could you please confirm that perlembed sample does actually 
> compiles fine with Mac OS X system perl without extra includes 
> used?
> 
> [1] http://www.opensource.apple.com/source/xnu/xnu-1699.22.81/libkern/libkern/OSAtomic.h
> [2] http://www.opensource.apple.com/source/perl/perl-73/5.10/fix/handy.h.ed

It seems last time when I tried to build nginx with MacOSX system perl
it was Leopard 10.5.x. On Snow Leopard 10.6.8 "#undef bool" in
src/os/unix/ngx_atomic is not required, but
   perl -MExtUtils::Embed -e ldopts
issues settings which do not allow to link nginx:
  -arch x86_64 -arch i386 -arch ppc -L/usr/local/lib  -L/System/Library/Perl/5.10.0/darwin-thread-multi-2level/CORE -lperl -ldl -lm -lutil -lc

ld: warning: in objs.gcc/src/core/nginx.o, file was built for unsupported file format which is not the architecture being linked (i386)


-- 
Igor Sysoev


From katmai at keptprivate.com  Sun Sep 11 08:28:06 2011
From: katmai at keptprivate.com (Stefanita Rares Dumitrescu)
Date: Sun, 11 Sep 2011 10:28:06 +0200
Subject: memory leaks on nginx / php-fpm
Message-ID: <4E6C7116.50307@keptprivate.com>

hi guys,

i have a serious problem that i can't make nor heads nor tails of. i am 
using nginx 1.0.6 with php 5.2.17 + fpm.

i run a gallery2 image gallery, pretty busy, and when i redirected some 
traffic from the old server today to the new one, i got a bunch of 
memory leaks messages in the php-fpm log.  they are going like crazy 
filling up the log.

can you please give me some advice?

thanks in advance.

Sep 11 01:24:29.948540 [WARNING] fpm_stdio_child_said(), line 167: child 
22682 (pool default) said into stderr: 
"/usr/ports/lang/php52/work/php-5.2.17/Zend/zend_variables.c(131) : 
Actual location (location was relayed)"
Sep 11 01:24:29.948583 [WARNING] fpm_stdio_child_said(), line 167: child 
22682 (pool default) said into stderr: "Last leak repeated 2 times"
Sep 11 01:24:29.948591 [WARNING] fpm_stdio_child_said(), line 167: child 
22682 (pool default) said into stderr: "[Sun Sep 11 01:24:29 2011] 
Script:  '/home/cityv3/public_html/gallery/main.php'"
Sep 11 01:24:29.948597 [WARNING] fpm_stdio_child_said(), line 167: child 
22682 (pool default) said into stderr: 
"/usr/ports/lang/php52/work/php-5.2.17/Zend/zend_variables.c(132) : 
Freeing 0x80C269680 (512 bytes), 
script=/home/cityv3/public_html/gallery/main.php"
Sep 11 01:24:29.948603 [WARNING] fpm_stdio_child_said(), line 167: child 
22682 (pool default) said into stderr: 
"/usr/ports/lang/php52/work/php-5.2.17/Zend/zend_alloc.c(2391) : Actual 
location (location was relayed)"
Sep 11 01:24:29.948637 [WARNING] fpm_stdio_child_said(), line 167: child 
22682 (pool default) said into stderr: "Last leak repeated 2 times"
Sep 11 01:24:29.948644 [WARNING] fpm_stdio_child_said(), line 167: child 
22682 (pool default) said into stderr: "[Sun Sep 11 01:24:29 2011] 
Script:  '/home/cityv3/public_html/gallery/main.php'"
Sep 11 01:24:29.948650 [WARNING] fpm_stdio_child_said(), line 167: child 
22682 (pool default) said into stderr: 
"/usr/ports/lang/php52/work/php-5.2.17/Zend/zend_vm_execute.h(7535) : 
Freeing 0x80C26CF78 (24 bytes), 
script=/home/cityv3/public_html/gallery/main.php"
Sep 11 01:24:29.948693 [WARNING] fpm_stdio_child_said(), line 167: child 
22682 (pool default) said into stderr: "[Sun Sep 11 01:24:29 2011] 
Script:  '/home/cityv3/public_html/gallery/main.php'"
Sep 11 01:24:29.948699 [WARNING] fpm_stdio_child_said(), line 167: child 
22682 (pool default) said into stderr: 
"/usr/ports/lang/php52/work/php-5.2.17/ext/standard/string.c(756) : 
Freeing 0x80C26D270 (6 bytes), 
script=/home/cityv3/public_html/gallery/main.php"


From nginx-forum at nginx.us  Sun Sep 11 11:19:48 2011
From: nginx-forum at nginx.us (Ensiferous)
Date: Sun, 11 Sep 2011 07:19:48 -0400
Subject: memory leaks on nginx / php-fpm
In-Reply-To: <4E6C7116.50307@keptprivate.com>
References: <4E6C7116.50307@keptprivate.com>
Message-ID: <b194ab88de98b505dd6d6a7b3ea33bb2.NginxMailingListEnglish@forum.nginx.org>

You will want to bring this up to the PHP people as there is no link
between Nginx and PHP-FPM. There is a fastcgi abstraction layer between
them. That said, they have deprecated PHP 5.2.x and PHP-FPM for 5.2.x is
a patch so I honestly think you'll be told to upgrade first and then see
what happens, so you might want to do that first.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215080,215085#msg-215085


From katmai at keptprivate.com  Sun Sep 11 13:42:13 2011
From: katmai at keptprivate.com (Stefanita Rares Dumitrescu)
Date: Sun, 11 Sep 2011 15:42:13 +0200
Subject: memory leaks on nginx / php-fpm
In-Reply-To: <b194ab88de98b505dd6d6a7b3ea33bb2.NginxMailingListEnglish@forum.nginx.org>
References: <4E6C7116.50307@keptprivate.com>
	<b194ab88de98b505dd6d6a7b3ea33bb2.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <4E6CBAB5.3000709@keptprivate.com>

fixed by disabling the apc extension. i have no clue how that got 
installed as i did not build it. it was conflicting with the installed 
eaccelerator. weird stuff.

On 9/11/2011 1:19 PM, Ensiferous wrote:
> You will want to bring this up to the PHP people as there is no link
> between Nginx and PHP-FPM. There is a fastcgi abstraction layer between
> them. That said, they have deprecated PHP 5.2.x and PHP-FPM for 5.2.x is
> a patch so I honestly think you'll be told to upgrade first and then see
> what happens, so you might want to do that first.
>
> Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215080,215085#msg-215085
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx


From nginx-forum at nginx.us  Sun Sep 11 16:47:53 2011
From: nginx-forum at nginx.us (Sylvia)
Date: Sun, 11 Sep 2011 12:47:53 -0400
Subject: memory leaks on nginx / php-fpm
In-Reply-To: <4E6CBAB5.3000709@keptprivate.com>
References: <4E6CBAB5.3000709@keptprivate.com>
Message-ID: <0ededa716762ad41b13602c078dc58d1.NginxMailingListEnglish@forum.nginx.org>

FYI: last version of APC (3.1.9) fixed some memory leaks with FastCGI,
I have no issues with php-fpm 5.3.8 and APC 3.1.9

another recomendation - switch to PHP 5.3 if you can, 5.2 branch is not
supported anymore, and php-fpm SAPI is now bundled into main PHP (5.3),
works fine, just pay attention that config file is not XML anymore.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215080,215091#msg-215091


From gelonida at gmail.com  Sun Sep 11 18:13:32 2011
From: gelonida at gmail.com (Gelonida N)
Date: Sun, 11 Sep 2011 20:13:32 +0200
Subject: problems with mediawiki behind nginx / apache (redirects to wrong url)
Message-ID: <j4itod$3ar$1@dough.gmane.org>

Hi,

I'm having a little trouble to get Mediawiki working in my nginx setup.

The problem is very probably the configuration of Mediawiki, but the
Help pages about Mediawiki don't mention nginx.
So I thought I ask here.


My Problem:
- I setup nginx with https such, that it handles all files, except php
files, which fill be proxied to apache/mod_php at port 8081

- I installed a new Mediawiki

- the installation via the web worked fine and was confirmed
    with a success message

- then I moved config/LocalSedttings.php  to the wiki directory

Now I access https:/servername/wiki2/
and get redirected to http://servername:8081/wiki2/index.php/Main_Page


Below my nginx configuration:


# ============== nginx configuration ========
server {
    listen       443 ssl;
    ssl_session_timeout  10m;
    ssl_session_cache    shared:SSL:10m;

    location ~ \.(php)$ {
        include proxy.conf;
        proxy_pass         http://127.0.0.1:8081;
    }

    # nginx shall serve all files except the php ones
    location / {
        root   /my/doc/root;
        index  index.html index.htm index.php;
    }

# ==================== proxy.conf ==============
proxy_redirect     off;

proxy_set_header   Host             $host;
proxy_set_header   X-Real-IP        $remote_addr;
proxy_set_header   X-Forwarded-For  $proxy_add_x_forwarded_for;

client_max_body_size       10m;
client_body_buffer_size    128k;

proxy_connect_timeout      90;
proxy_send_timeout         90;
proxy_read_timeout         90;

proxy_buffer_size          4k;
proxy_buffers              4 32k;
proxy_busy_buffers_size    64k;
proxy_temp_file_write_size 64k;


Thanks in advance for any pointers.


From gelonida at gmail.com  Sun Sep 11 18:44:58 2011
From: gelonida at gmail.com (Gelonida N)
Date: Sun, 11 Sep 2011 20:44:58 +0200
Subject: problems with mediawiki behind nginx / apache (redirects to wrong
	url)
In-Reply-To: <j4itod$3ar$1@dough.gmane.org>
References: <j4itod$3ar$1@dough.gmane.org>
Message-ID: <j4ivjb$fq4$1@dough.gmane.org>

I have now a working (perhaps not optimized) solution.


1.) Add following variable to Mediawiki's LocalSettings.php:
$wgServer = 'https://servername';

I found this tip finally at
http://www.cyberciti.biz/faq/mediawiki-fix-internal-server-host-names-redirect-using-wgserver/

though they specify the entire server url with the path to the wiki,
whereas I specify just the protocol and the server



2,) As Mediawiki urls look like:
https://servername/wiki/index.php/blablabla
I have to be sure, that all these urls get redirected to apache

this I changed my setup to:

    location ~ \.(php)$ {
        include proxy.conf;
        proxy_pass         http://127.0.0.1:8081;
    }

    location /wiki/index.php {
        include proxy.conf;
        proxy_pass         http://127.0.0.1:8081;
    }

    location / {
        root   /home/noma/web/nomassl;
        index  index.html index.htm index.php;
    }



Perhaps this post can help somebdoy else. I was searching for quite some
time and was already close to giving up.

On 09/11/2011 08:13 PM, Gelonida N wrote:
> Hi,Well I have finally a (non optimized) setting, which is working.


On 09/11/2011 08:13 PM, Gelonida N wrote:
> Hi,
> 
> I'm having a little trouble to get Mediawiki working in my nginx setup.
> 
> The problem is very probably the configuration of Mediawiki, but the
> Help pages about Mediawiki don't mention nginx.
> So I thought I ask here.
> 
> 
> My Problem:
> - I setup nginx with https such, that it handles all files, except php
> files, which fill be proxied to apache/mod_php at port 8081
> 
> - I installed a new Mediawiki
> 
> - the installation via the web worked fine and was confirmed
>     with a success message
> 
> - then I moved config/LocalSedttings.php  to the wiki directory
> 
> Now I access https:/servername/wiki2/
> and get redirected to http://servername:8081/wiki2/index.php/Main_Page
> 
> 
> Below my nginx configuration:
> 
> 
> # ============== nginx configuration ========
> server {
>     listen       443 ssl;
>     ssl_session_timeout  10m;
>     ssl_session_cache    shared:SSL:10m;
> 
>     location ~ \.(php)$ {
>         include proxy.conf;
>         proxy_pass         http://127.0.0.1:8081;
>     }
> 
>     # nginx shall serve all files except the php ones
>     location / {
>         root   /my/doc/root;
>         index  index.html index.htm index.php;
>     }
> 
> # ==================== proxy.conf ==============
> proxy_redirect     off;
> 
> proxy_set_header   Host             $host;
> proxy_set_header   X-Real-IP        $remote_addr;
> proxy_set_header   X-Forwarded-For  $proxy_add_x_forwarded_for;
> 
> client_max_body_size       10m;
> client_body_buffer_size    128k;
> 
> proxy_connect_timeout      90;
> proxy_send_timeout         90;
> proxy_read_timeout         90;
> 
> proxy_buffer_size          4k;
> proxy_buffers              4 32k;
> proxy_busy_buffers_size    64k;
> proxy_temp_file_write_size 64k;
> 
> 
> Thanks in advance for any pointers.


> 
> I'm having a little trouble to get Mediawiki working in my nginx setup.
> 
> The problem is very probably the configuration of Mediawiki, but the
> Help pages about Mediawiki don't mention nginx.
> So I thought I ask here.
> 
> 
> My Problem:
> - I setup nginx with https such, that it handles all files, except php
> files, which fill be proxied to apache/mod_php at port 8081
> 
> - I installed a new Mediawiki
> 
> - the installation via the web worked fine and was confirmed
>     with a success message
> 
> - then I moved config/LocalSedttings.php  to the wiki directory
> 
> Now I access https:/servername/wiki2/
> and get redirected to http://servername:8081/wiki2/index.php/Main_Page
> 
> 
> Below my nginx configuration:
> 
> 
> # ============== nginx configuration ========
> server {
>     listen       443 ssl;
>     ssl_session_timeout  10m;
>     ssl_session_cache    shared:SSL:10m;
> 
>     location ~ \.(php)$ {
>         include proxy.conf;
>         proxy_pass         http://127.0.0.1:8081;
>     }
> 
>     # nginx shall serve all files except the php ones
>     location / {
>         root   /my/doc/root;
>         index  index.html index.htm index.php;
>     }
> 
> # ==================== proxy.conf ==============
> proxy_redirect     off;
> 
> proxy_set_header   Host             $host;
> proxy_set_header   X-Real-IP        $remote_addr;
> proxy_set_header   X-Forwarded-For  $proxy_add_x_forwarded_for;
> 
> client_max_body_size       10m;
> client_body_buffer_size    128k;
> 
> proxy_connect_timeout      90;
> proxy_send_timeout         90;
> proxy_read_timeout         90;
> 
> proxy_buffer_size          4k;
> proxy_buffers              4 32k;
> proxy_busy_buffers_size    64k;
> proxy_temp_file_write_size 64k;
> 
> 
> Thanks in advance for any pointers.



From nginx-forum at nginx.us  Mon Sep 12 00:30:15 2011
From: nginx-forum at nginx.us (frostbyte)
Date: Sun, 11 Sep 2011 20:30:15 -0400
Subject: nginx 1.0.6 perl module build failure
In-Reply-To: <20110910114625.GB18804@nginx.com>
References: <20110910114625.GB18804@nginx.com>
Message-ID: <e50a2f0a8dcaa1de25c4fee42fe490d2.NginxMailingListEnglish@forum.nginx.org>

Same error using perl 5.12 from MacPorts:

$ ./configure --prefix=/usr/local --with-cc-opt="-I/usr/local/include
-O2 -Wno-deprecated-declarations" --with-ld-opt="-L/usr/local/lib"
--conf-path=/etc/nginx/nginx.conf
--error-log-path=/var/log/nginx/error.log
--http-log-path=/var/log/nginx/access.log
--pid-path=/var/run/nginx/nginx.pid
--lock-path=/var/run/nginx/nginx.lock
--http-client-body-temp-path=/var/run/nginx/client_body_temp
--http-proxy-temp-path=/var/run/nginx/proxy_temp
--http-fastcgi-temp-path=/var/run/nginx/fastcgi_temp
--http-uwsgi-temp-path=/var/run/nginx/uwsgi_temp --with-http_dav_module
--with-http_flv_module --with-mail --with-http_ssl_module
--with-mail_ssl_module --with-http_stub_status_module --with-ipv6
--with-http_perl_module --with-perl=/usr/bin/perl
--with-http_realip_module --with-http_addition_module
--with-http_sub_module --with-http_gzip_static_module
--with-perl=/opt/local/bin/perl
. . .
checking for perl
 + perl version: This is perl 5, version 12, subversion 3 (v5.12.3)
built for darwin-multi-2level
 + perl interpreter multiplicity found
. . .

$ make
. . .
gcc -c -pipe -O -W -Wall -Wpointer-arith -Wno-unused-parameter
-Wunused-function -Wunused-variable -Wno-unused-value -Werror -g
-I/usr/local/include -O2 -Wno-deprecated-declarations -pipe -O2 -arch
x86_64 -fno-common -DPERL_DARWIN -I/opt/local/include -no-cpp-precomp
-fno-strict-aliasing -fstack-protector -I/opt/local/include
-I/opt/local/lib/perl5/5.12.3/darwin-multi-2level/CORE  -I src/core -I
src/event -I src/event/modules -I src/os/unix -I /opt/local/include -I
objs -I src/http -I src/http/modules -I src/http/modules/perl \
		-o objs/src/http/modules/perl/ngx_http_perl_module.o \
		src/http/modules/perl/ngx_http_perl_module.c
In file included from src/http/modules/perl/ngx_http_perl_module.h:17,
                 from src/http/modules/perl/ngx_http_perl_module.c:10:
/opt/local/lib/perl5/5.12.3/darwin-multi-2level/CORE/perl.h:3336: error:
expected specifier-qualifier-list before ?bool?
. . .

Actually I started with the MacPorts nginx 1.0.6 install and then tried
the tarball (getting the same result).

Frosty

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215048,215096#msg-215096


From nginx-forum at nginx.us  Mon Sep 12 00:32:04 2011
From: nginx-forum at nginx.us (frostbyte)
Date: Sun, 11 Sep 2011 20:32:04 -0400
Subject: nginx 1.0.6 perl module build failure
In-Reply-To: <e50a2f0a8dcaa1de25c4fee42fe490d2.NginxMailingListEnglish@forum.nginx.org>
References: <20110910114625.GB18804@nginx.com>
	<e50a2f0a8dcaa1de25c4fee42fe490d2.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <2fa3cc7ba9e0e80774ea82240e53fde3.NginxMailingListEnglish@forum.nginx.org>

perl5.12 via MacPorts:

$ perl -MExtUtils::Embed -e ldopts
  -L/opt/local/lib -arch x86_64 -fstack-protector 
-L/opt/local/lib/perl5/5.12.3/darwin-multi-2level/CORE -lperl -ldl -lm
-lutil -lc

$ perl -V
Summary of my perl5 (revision 5 version 12 subversion 3) configuration:
   
  Platform:
    osname=darwin, osvers=10.8.0, archname=darwin-multi-2level
    uname='darwin beta.macosforge.org 10.8.0 darwin kernel version
10.8.0: tue jun 7 16:33:36 pdt 2011; root:xnu-1504.15.3~1release_i386
i386 '
    config_args='-D inc_version_list=5.12.2/darwin-multi-2level 5.12.2
5.12.1/darwin-multi-2level 5.12.1 5.12.0/darwin-multi-2level 5.12.0 -des
-Dprefix=/opt/local -Dscriptdir=/opt/local/bin
-Dcppflags=-I/opt/local/include -Dccflags=-pipe -O2 -arch x86_64
-Dldflags=-L/opt/local/lib -arch x86_64 -Dvendorprefix=/opt/local
-Dusemultiplicity=y -D cc=/usr/bin/gcc-4.2 -D ld=/usr/bin/gcc-4.2 -D
man1ext=1pm -D man3ext=3pm -D man1dir=/opt/local/share/man/man1p -D
man3dir=/opt/local/share/man/man3p -D
siteman1dir=/opt/local/share/man/man1 -D
siteman3dir=/opt/local/share/man/man3 -D
vendorman1dir=/opt/local/share/man/man1 -D
vendorman3dir=/opt/local/share/man/man3 -D pager=/usr/bin/less -sR'
    hint=recommended, useposix=true, d_sigaction=define
    useithreads=undef, usemultiplicity=define
    useperlio=define, d_sfio=undef, uselargefiles=define,
usesocks=undef
    use64bitint=define, use64bitall=define, uselongdouble=undef
    usemymalloc=n, bincompat5005=undef
  Compiler:
    cc='/usr/bin/gcc-4.2', ccflags ='-pipe -O2 -arch x86_64 -fno-common
-DPERL_DARWIN -I/opt/local/include -no-cpp-precomp -fno-strict-aliasing
-fstack-protector -I/opt/local/include',
    optimize='-O3',
    cppflags='-I/opt/local/include -no-cpp-precomp -pipe -O2 -arch
x86_64 -fno-common -DPERL_DARWIN -I/opt/local/include -no-cpp-precomp
-fno-strict-aliasing -fstack-protector -I/opt/local/include'
    ccversion='', gccversion='4.2.1 (Apple Inc. build 5666) (dot 3)',
gccosandvers=''
    intsize=4, longsize=8, ptrsize=8, doublesize=8, byteorder=12345678
    d_longlong=define, longlongsize=8, d_longdbl=define, longdblsize=16
    ivtype='long', ivsize=8, nvtype='double', nvsize=8, Off_t='off_t',
lseeksize=8
    alignbytes=8, prototype=define
  Linker and Libraries:
    ld='env MACOSX_DEPLOYMENT_TARGET=10.6 /usr/bin/gcc-4.2', ldflags
='-L/opt/local/lib -arch x86_64 -fstack-protector'
    libpth=/opt/local/lib /usr/lib
    libs=-lgdbm -ldbm -ldl -lm -lutil -lc
    perllibs=-ldl -lm -lutil -lc
    libc=, so=dylib, useshrplib=false, libperl=libperl.a
    gnulibc_version=''
  Dynamic Linking:
    dlsrc=dl_dlopen.xs, dlext=bundle, d_dlsymun=undef, ccdlflags=' '
    cccdlflags=' ', lddlflags='-L/opt/local/lib -arch x86_64 -bundle
-undefined dynamic_lookup -fstack-protector'


Characteristics of this binary (from libperl): 
  Compile-time options: MULTIPLICITY PERL_DONT_CREATE_GVSV
                        PERL_IMPLICIT_CONTEXT PERL_MALLOC_WRAP
USE_64_BIT_ALL
                        USE_64_BIT_INT USE_LARGE_FILES USE_PERLIO
                        USE_PERL_ATOF
  Built under darwin
  Compiled at Aug 29 2011 07:18:38
  @INC:
    /opt/local/lib/perl5/site_perl/5.12.3/darwin-multi-2level
    /opt/local/lib/perl5/site_perl/5.12.3
    /opt/local/lib/perl5/vendor_perl/5.12.3/darwin-multi-2level
    /opt/local/lib/perl5/vendor_perl/5.12.3
    /opt/local/lib/perl5/5.12.3/darwin-multi-2level
    /opt/local/lib/perl5/5.12.3
    /opt/local/lib/perl5/site_perl
    /opt/local/lib/perl5/vendor_perl
    .

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215048,215097#msg-215097


From mdounin at mdounin.ru  Mon Sep 12 08:39:02 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Mon, 12 Sep 2011 12:39:02 +0400
Subject: nginx 1.0.6 perl module build failure
In-Reply-To: <e50a2f0a8dcaa1de25c4fee42fe490d2.NginxMailingListEnglish@forum.nginx.org>
References: <20110910114625.GB18804@nginx.com>
	<e50a2f0a8dcaa1de25c4fee42fe490d2.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20110912083902.GA1137@mdounin.ru>

Hello!

On Sun, Sep 11, 2011 at 08:30:15PM -0400, frostbyte wrote:

> Same error using perl 5.12 from MacPorts:
> 
> $ ./configure --prefix=/usr/local --with-cc-opt="-I/usr/local/include
> -O2 -Wno-deprecated-declarations" --with-ld-opt="-L/usr/local/lib"
> --conf-path=/etc/nginx/nginx.conf
> --error-log-path=/var/log/nginx/error.log
> --http-log-path=/var/log/nginx/access.log
> --pid-path=/var/run/nginx/nginx.pid
> --lock-path=/var/run/nginx/nginx.lock
> --http-client-body-temp-path=/var/run/nginx/client_body_temp
> --http-proxy-temp-path=/var/run/nginx/proxy_temp
> --http-fastcgi-temp-path=/var/run/nginx/fastcgi_temp
> --http-uwsgi-temp-path=/var/run/nginx/uwsgi_temp --with-http_dav_module
> --with-http_flv_module --with-mail --with-http_ssl_module
> --with-mail_ssl_module --with-http_stub_status_module --with-ipv6
> --with-http_perl_module --with-perl=/usr/bin/perl
> --with-http_realip_module --with-http_addition_module
> --with-http_sub_module --with-http_gzip_static_module
> --with-perl=/opt/local/bin/perl
> . . .
> checking for perl
>  + perl version: This is perl 5, version 12, subversion 3 (v5.12.3)
> built for darwin-multi-2level
>  + perl interpreter multiplicity found
> . . .
> 
> $ make
> . . .
> gcc -c -pipe -O -W -Wall -Wpointer-arith -Wno-unused-parameter
> -Wunused-function -Wunused-variable -Wno-unused-value -Werror -g
> -I/usr/local/include -O2 -Wno-deprecated-declarations -pipe -O2 -arch
> x86_64 -fno-common -DPERL_DARWIN -I/opt/local/include -no-cpp-precomp
> -fno-strict-aliasing -fstack-protector -I/opt/local/include
> -I/opt/local/lib/perl5/5.12.3/darwin-multi-2level/CORE  -I src/core -I
> src/event -I src/event/modules -I src/os/unix -I /opt/local/include -I
> objs -I src/http -I src/http/modules -I src/http/modules/perl \
> 		-o objs/src/http/modules/perl/ngx_http_perl_module.o \
> 		src/http/modules/perl/ngx_http_perl_module.c
> In file included from src/http/modules/perl/ngx_http_perl_module.h:17,
>                  from src/http/modules/perl/ngx_http_perl_module.c:10:
> /opt/local/lib/perl5/5.12.3/darwin-multi-2level/CORE/perl.h:3336: error:
> expected specifier-qualifier-list before ?bool?
> . . .
> 
> Actually I started with the MacPorts nginx 1.0.6 install and then tried
> the tarball (getting the same result).

It looks like macports also patch handy.h in perl 5.12 and perl 
5.14 to use stdbool.h [1][2] since about 3 months.  (The perl 5.8 
from macports still uses perl's native handy.h.)

Please try the following patch:

diff --git a/src/os/unix/ngx_atomic.h b/src/os/unix/ngx_atomic.h
--- a/src/os/unix/ngx_atomic.h
+++ b/src/os/unix/ngx_atomic.h
@@ -46,9 +46,6 @@ typedef volatile ngx_atomic_uint_t  ngx_

 #include <libkern/OSAtomic.h>

-/* "bool" conflicts with perl's CORE/handy.h */
-#undef bool
-

 #define NGX_HAVE_ATOMIC_OPS  1


[1] https://trac.macports.org/browser/trunk/dports/lang/perl5.12/files/patch-handy.h.stdbool.diff
[2] https://trac.macports.org/browser/trunk/dports/lang/perl5.14/files/patch-handy.h.stdbool.diff

Maxim Dounin


From nginx-forum at nginx.us  Mon Sep 12 11:16:19 2011
From: nginx-forum at nginx.us (frostbyte)
Date: Mon, 12 Sep 2011 07:16:19 -0400
Subject: nginx 1.0.6 perl module build failure
In-Reply-To: <2fa3cc7ba9e0e80774ea82240e53fde3.NginxMailingListEnglish@forum.nginx.org>
References: <20110910114625.GB18804@nginx.com>
	<e50a2f0a8dcaa1de25c4fee42fe490d2.NginxMailingListEnglish@forum.nginx.org>
	<2fa3cc7ba9e0e80774ea82240e53fde3.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <301b3dfe2d6e0ed1ad4bc5ca953cda63.NginxMailingListEnglish@forum.nginx.org>

Perfect. Thanks!

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215048,215107#msg-215107


From nginx-forum at nginx.us  Mon Sep 12 14:11:18 2011
From: nginx-forum at nginx.us (advegav)
Date: Mon, 12 Sep 2011 10:11:18 -0400
Subject: too many open files
In-Reply-To: <77432c470909100810s78c3ee1dhc7e0ab1a17d942f0@mail.gmail.com>
References: <77432c470909100810s78c3ee1dhc7e0ab1a17d942f0@mail.gmail.com>
Message-ID: <ab29b7111032851e072f849db2fceb73.NginxMailingListEnglish@forum.nginx.org>

I found the solution for the same problem on my environment, ( Solaris
10 ) here:

http://www.cyberciti.biz/faq/linux-unix-nginx-too-many-open-files/

Quoting:

....

"Then open the nginx configuration file (default: /etc/nginx/nginx.conf)
and the following line right after the ?worker_processes? entry
line

    worker_rlimit_nofile 20480;

Once both changes are made, run the Nginx lint to make sure no errors
have been introduced:

    # nginx -t
    2010/11/01 17:07:46 [info] 9520#0: the configuration file
/etc/nginx/nginx.conf syntax is ok
    2010/11/01 17:07:46 [info] 9520#0: the configuration file
/etc/nginx/nginx.conf was tested successfully

and then restart Nginx"

Thank to them I solved my problem :-)
hope this helps...

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,6300,215117#msg-215117


From nginx-forum at nginx.us  Mon Sep 12 14:22:25 2011
From: nginx-forum at nginx.us (vejpuste)
Date: Mon, 12 Sep 2011 10:22:25 -0400
Subject: Double free or corruption
Message-ID: <379f8d1bec5578f7a40400b736285b84.NginxMailingListEnglish@forum.nginx.org>

Hello,

after start nginx error.log contains :

*** glibc detected *** nginx: master process /usr/local/nginx/sbin/nginx
-c /usr/local/nginx/conf/nginx.conf: double free or corruption (out):
0x09f24128 ***
======= Backtrace: =========
/lib/libc.so.6[0x5f46c5]
/lib/libc.so.6(cfree+0x59)[0x5f4b09]
nginx: master process /usr/local/nginx/sbin/nginx -c
/usr/local/nginx/conf/nginx.conf[0x80cfdc7]
[0x9f17ec0]
======= Memory map: ========
002d9000-002da000 r-xp 002d9000 00:00 0          [vdso]
00352000-00353000 rwxp 00352000 00:00 0
00539000-00554000 r-xp 00000000 fd:00 950349     /lib/ld-2.5.so
00554000-00555000 r-xp 0001a000 fd:00 950349     /lib/ld-2.5.so
00555000-00556000 rwxp 0001b000 fd:00 950349     /lib/ld-2.5.so
00583000-00586000 r-xp 00000000 fd:00 953074     /lib/libdl-2.5.so
00586000-00587000 r-xp 00002000 fd:00 953074     /lib/libdl-2.5.so
00587000-00588000 rwxp 00003000 fd:00 953074     /lib/libdl-2.5.so
0058b000-006de000 r-xp 00000000 fd:00 950482     /lib/libc-2.5.so
006de000-006e0000 r-xp 00153000 fd:00 950482     /lib/libc-2.5.so
006e0000-006e1000 rwxp 00155000 fd:00 950482     /lib/libc-2.5.so
006e1000-006e4000 rwxp 006e1000 00:00 0
006e6000-006fb000 r-xp 00000000 fd:00 953068     /lib/libpthread-2.5.so
006fb000-006fc000 r-xp 00015000 fd:00 953068     /lib/libpthread-2.5.so
006fc000-006fd000 rwxp 00016000 fd:00 953068     /lib/libpthread-2.5.so
006fd000-006ff000 rwxp 006fd000 00:00 0
00701000-00713000 r-xp 00000000 fd:00 498522     /usr/lib/libz.so.1.2.3
00713000-00714000 rwxp 00011000 fd:00 498522     /usr/lib/libz.so.1.2.3
0075c000-0077b000 r-xp 00000000 fd:00 950493     /lib/libpcre.so.0.0.1
0077b000-0077c000 rwxp 0001e000 fd:00 950493     /lib/libpcre.so.0.0.1
00888000-00889000 rwxp 00888000 00:00 0
009a8000-009a9000 rwxp 009a8000 00:00 0
00b24000-00b2e000 r-xp 00000000 fd:00 951090    
/lib/libnss_files-2.5.so
00b2e000-00b2f000 r-xp 00009000 fd:00 951090    
/lib/libnss_files-2.5.so
00b2f000-00b30000 rwxp 0000a000 fd:00 951090    
/lib/libnss_files-2.5.so
00bb9000-00bba000 rwxs 00000000 00:09 17874      /dev/zero (deleted)
03248000-03253000 r-xp 00000000 fd:00 951429    
/lib/libgcc_s-4.1.2-20080825.so.1
03253000-03254000 rwxp 0000a000 fd:00 951429    
/lib/libgcc_s-4.1.2-20080825.so.1
035fd000-03606000 r-xp 00000000 fd:00 953152     /lib/libcrypt-2.5.so
03606000-03607000 r-xp 00008000 fd:00 953152     /lib/libcrypt-2.5.so
03607000-03608000 rwxp 00009000 fd:00 953152     /lib/libcrypt-2.5.so
03608000-0362f000 rwxp 03608000 00:00 0
08048000-081ee000 r-xp 00000000 fd:00 527526    
/usr/local/nginx/sbin/nginx
081ee000-081ff000 rwxp 001a5000 fd:00 527526    
/usr/local/nginx/sbin/nginx
081ff000-0820d000 rwxp 081ff000 00:00 0
09f14000-09f58000 rwxp 09f14000 00:00 0          [heap]
bfebc000-bfed1000 rwxp bffe9000 00:00 0          [stack]


nginx working.

CentOS 5.6 x86_64 kernel 2.6.18-238.19.1.el5PAE,

 /usr/local/nginx/sbin/nginx -V
nginx: nginx version: nginx/1.0.6
nginx: built by gcc 4.1.2 20080704 (Red Hat 4.1.2-50)
nginx: TLS SNI support enabled
nginx: configure arguments: --with-http_stub_status_module
--with-http_ssl_module
--with-openssl=/root/src/Apache/Apache2/openssl-1.0.0d

Thanks,

Libor

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215119,215119#msg-215119


From nginx-forum at nginx.us  Mon Sep 12 17:22:09 2011
From: nginx-forum at nginx.us (ynasser)
Date: Mon, 12 Sep 2011 13:22:09 -0400
Subject: Beginner questions about proxy configuration and white space
In-Reply-To: <17696a632c172de23cb235a2b6fd33c4.NginxMailingListEnglish@forum.nginx.org>
References: <72ca536398bcc048cbca71aa39cb728e.NginxMailingListEnglish@forum.nginx.org>
	<17696a632c172de23cb235a2b6fd33c4.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <05b6d159ad4a279388a76b7ca2d302b2.NginxMailingListEnglish@forum.nginx.org>

It works now. My configuration looks like this:

	server {
		listen 80 default;

		location / {
			proxy_pass http://some_ip:some_port;
			
			proxy_set_header X-h1 h1;
			proxy_set_header X-h2 h2;
			proxy_set_header X-h3 "h3 h3";
			}
		}

I hope this help anyone else with this problem.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,214982,215126#msg-215126


From nginx-forum at nginx.us  Mon Sep 12 19:20:28 2011
From: nginx-forum at nginx.us (meto)
Date: Mon, 12 Sep 2011 15:20:28 -0400
Subject: mp4 module
In-Reply-To: <20110909130812.GI97288@nginx.com>
References: <20110909130812.GI97288@nginx.com>
Message-ID: <b66d2a7e6c245d9d7fd455f0216b7ba1.NginxMailingListEnglish@forum.nginx.org>

Can't wait to see that feature in mainline (dev or stable). It's one
that was bugging me about the license used by current one (commercial
use). Are you planning to include it in your repo packages?

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215011,215129#msg-215129


From nginx-forum at nginx.us  Mon Sep 12 19:23:18 2011
From: nginx-forum at nginx.us (pedroaxl)
Date: Mon, 12 Sep 2011 15:23:18 -0400
Subject: Decode url in rewrite
In-Reply-To: <94b42f69891b4faa8507abd58c54367a.NginxMailingListEnglish@forum.nginx.org>
References: <94b42f69891b4faa8507abd58c54367a.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <58240b4a4fa26ef240d523994d15c043.NginxMailingListEnglish@forum.nginx.org>

Hi guys,

I solved my problem with set-misc module, now I use the set_unescape_uri
to do it.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215047,215130#msg-215130


From andrew at andrewloe.com  Mon Sep 12 23:58:17 2011
From: andrew at andrewloe.com (W. Andrew Loe III)
Date: Mon, 12 Sep 2011 16:58:17 -0700
Subject: Verify Peer of SSL request in Proxy Module
Message-ID: <CA+-9oNdiQ8P9JJJtiWALd4v5=zGChqsaQv0=Bun-O7y4qDK=sg@mail.gmail.com>

When reverse proxying over SSL, I would like to be able to set the
context to VERIFY_PEER and to provide my own CA file. Right now nginx
uses the OpenSSL default of VERIFY_NONE.

There appears to be some code around this in
ngx_ssl_client_certificate in ngx_event_openssl.c, but this is for
validating client certificates, not nginx acting as a client.

I am working on a patch, but if Igor or someone more experienced
already has this working, I would prefer to use that.


From nginx-forum at nginx.us  Tue Sep 13 00:31:04 2011
From: nginx-forum at nginx.us (redonisc)
Date: Mon, 12 Sep 2011 20:31:04 -0400
Subject: Reverse proxy for multiple apps, tomcat
Message-ID: <9515650c3c82727df6bb502886bf3556.NginxMailingListEnglish@forum.nginx.org>

Hi, 

I'm setting up a reverse proxy to be used by different applications
working on a tomcat server. The configuration, is using proxy_pass
directive but it's not working, have searched but I've been able to
figure it out.  

The scenario is like this:


|---------------------------|
| (NginX)                |
-----------------------|-------------------------------------------------|-------------------------------------------------|
|---------------------------|
_____________|____________________________|____________________________|
_____________________(http://www.app1.com) _________
(http://www.app2.com) ________(http://www.app3.com)
______________________________|____________________________|____________________________|
______________________________|____________________________|____________________________|
-----------------------________________\|/____________________________|____________________________|
|_____________|  ------>
http://tomcat1:8080/app1__________________\|/___________________________|
| Tomcat 1_____| 
-------------------------------------------------------->
http://tomcat1:8080/app2__________________\|/
| (192.168.0.1) _| 
------------------------------------------------------------------------------------------------------>
http://tomcat1:8080/app3
-----------------------

If anyone can help with this questions:

1.- Do I need to set different IP for each app on the tomcat server?
currently my apps are on: http://Tomcat:8080/app1,
http://Tomcat:8080/app2, http://Tomcat:8080/app3 all on the same IP.

2.- Currently I've a config but it seems to have problems with the fact
that I need to publish the application as a root let's say
http://www.app1.com but this should point to my tomcat on
http://tomcat1:8080/app1.


Thanks for your help,

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215134,215134#msg-215134


From nginx-forum at nginx.us  Tue Sep 13 02:10:42 2011
From: nginx-forum at nginx.us (doughboy)
Date: Mon, 12 Sep 2011 22:10:42 -0400
Subject: Enabeling SSL causes Redirect Loop errors ???
Message-ID: <be92617b4417a38334e91a54fc1fe059.NginxMailingListEnglish@forum.nginx.org>

I am new to Nginx but learning every day.  I have the server setup
nicely and the regular webpages serve up fine.  I have multiple IP's on
the server (less than 20) and I am trying to enable SSL for 3 different
sites (each on a different IP).  I have read for hours looking at
examples and trying many different things to see if I can get the SSL
pages to serve up and no luck.  I am hoping someone can help me. 
Running debian lenny, nginx 1.0.6, php 5.3.8, php-fpm,   I am only
wanting to run one domain SSL per IP,  (Bogus domain name has been
inserted below)
-------------------------------------------------
nginx.conf

user www-user;
worker_processes 4;
pid /var/run/nginx.pid;



events {
        worker_connections 768;
        # multi_accept on;
}

http {

        ##
        # Basic Settings
        ##

        sendfile on;
        tcp_nopush on;
        tcp_nodelay on;
        keepalive_timeout 60 60;
        types_hash_max_size 2048;
        # server_tokens off;

        # server_names_hash_bucket_size 64;
        # server_name_in_redirect off;

        include /etc/nginx/mime.types;
        default_type application/octet-stream;

        ##
        # Logging Settings
        ##

        access_log /var/log/nginx/access.log;
        error_log /var/log/nginx/error.log;

        ##
        # Gzip Settings
        ##

        gzip on;
        gzip_disable "msie6";

        gzip_vary on;
        # gzip_proxied any;
        gzip_comp_level 6;
        gzip_buffers 16 8k;
        gzip_http_version 1.1;
        gzip_types text/plain text/css application/json
application/x-javascript text/xml application/xml application/xml+rss
text/javascript;

        ##
        # Virtual Host Configs
        ##

        include /etc/nginx/conf.d/*.conf;
        include /etc/nginx/sites-enabled/*;
}

-----------------------------------------------------------
> include /etc/nginx/conf.d/*.conf;

There NO conf files located in the above folder.  I have not set up any
site specific conf files yet.

------------------------------------------------------------
vhost entry (one try)

server {
        listen 80;
        listen 206.201.217.7:443 ssl;
        server_name www.luvinlife.com luvinlife.com;

        ssl_certificate /usr/nginx/certs/luvinlife.pem;
        ssl_certificate_key /usr/nginx/certs/luvinlife.key;

        client_max_body_size 5m;

        root /var/www/luvinlife.com/public_html;
        try_files $uri $uri/ /index.php?$args;
        index index.html index.php;

        location ~ \.php$ {
            #fastcgi_pass       unix:/var/run/php5-fpm.sock;
            fastcgi_pass        127.0.0.1:9000;
            fastcgi_index       index.php;
            include             fastcgi_params;
            fastcgi_param       SCRIPT_FILENAME
$document_root$fastcgi_script_name;
        }
}

-------------------------------------------------
version check :

nginx: nginx version: nginx/1.0.6
nginx: TLS SNI support enabled
nginx: configure arguments: --prefix=/etc/nginx
--conf-path=/etc/nginx/nginx.conf
--error-log-path=/var/log/nginx/error.log
--http-client-body-temp-path=/var/lib/nginx/body
--http-fastcgi-temp-path=/var/lib/nginx/fastcgi
--http-log-path=/var/log/nginx/access.log
--http-proxy-temp-path=/var/lib/nginx/proxy
--http-scgi-temp-path=/var/lib/nginx/scgi
--http-uwsgi-temp-path=/var/lib/nginx/uwsgi
--lock-path=/var/lock/nginx.lock --pid-path=/var/run/nginx.pid
--with-debug --with-http_addition_module --with-http_dav_module
--with-http_geoip_module --with-http_gzip_static_module
--with-http_image_filter_module --with-http_realip_module
--with-http_stub_status_module --with-http_ssl_module
--with-http_sub_module --with-http_xslt_module --with-ipv6
--with-sha1=/usr/include/openssl --with-md5=/usr/include/openssl
--with-mail --with-mail_ssl_module
--add-module=/usr/src/nginx/source/nginx-1.0.6/debian/modules/nginx-echo
--add-module=/usr/src/nginx/source/nginx-1.0.6/debian/modules/nginx-upstream-fair

----------------------------------------------

SSL Page Error message:

The webpage at https://www.luvinlife.com/index.php/customer/account/ has
resulted in too many redirects. Clearing ......

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215136,215136#msg-215136


From Richard.Kearsley at m247.com  Tue Sep 13 10:29:21 2011
From: Richard.Kearsley at m247.com (Richard Kearsley)
Date: Tue, 13 Sep 2011 10:29:21 +0000
Subject: nginx & Range header & proxy cache
Message-ID: <EFC5A19DBF3A2E489904733B467D990B2BFB62@office-server2.m247.local>

Hi
If a Range request is sent to nginx proxy (e.g. header "Range: bytes=50-") and the file isn't in cache yet (cache MISS), the range is ignored and full file sent
When the file has been cached (cache HIT), Range request works as expected

Is this the intended behaviour?

Please consider the following example

# /usr/local/nginx/sbin/nginx -V
nginx: nginx version: nginx/1.0.6
nginx: built by gcc 4.4.5 (Debian 4.4.5-8)
nginx: configure arguments:

error_log  logs/error-small.log;
worker_processes  2;
events
{
    use epoll;
    worker_connections  16384;
}

http
{
        include       mime.types;
        default_type  application/octet-stream;
        keepalive_timeout  65;
        proxy_cache_path /var/cache/nginx  levels=2:2:2  keys_zone=small:512m inactive=24h max_size=128G;
        resolver 8.8.8.8;

        server
        {
                listen       80;
                server_name  _;
                location /
                {
                        proxy_pass         http://$http_X_Backend:$http_X_Backend_Port;
                        proxy_cache        small;
                        proxy_cache_valid  200 302  10y;
                        proxy_cache_valid  404      10m;
                        proxy_cache_key         $uri;
                        proxy_cache_min_uses    10;
                        proxy_cache_use_stale updating;
                        add_header X-Cached $upstream_cache_status;
                }
                error_page   500 502 503 504  /50x.html;
                location = /50x.html
                {
                        root   html;
                }
        }
}


# wget -SO /dev/null --header="X-Backend: download.thinkbroadband.com" --header="X-Backend-Port: 80" --header="Range: bytes=50-" localhost/5MB.zip
--2011-09-13 11:33:45--  http://localhost/5MB.zip
Resolving localhost... ::1, 127.0.0.1
Connecting to localhost|127.0.0.1|:80... connected.
HTTP request sent, awaiting response...
  HTTP/1.1 200 OK
  Server: nginx/1.0.6
  Date: Tue, 13 Sep 2011 10:33:45 GMT
  Content-Type: application/zip
  Connection: keep-alive
  Last-Modified: Mon, 02 Jun 2008 15:30:42 GMT
  ETag: "11f000b-500000-44eb0adaf4c80"
  Accept-Ranges: bytes
  Content-Length: 5242880
  X-Cached: MISS
Length: 5242880 (5.0M) [application/zip]
Saving to: `/dev/null'

(try 9 more times to get it cached)

# wget -SO /dev/null --header="X-Backend: download.thinkbroadband.com" --header="X-Backend-Port: 80" --header="Range: bytes=50-" localhost/5MB.zip
--2011-09-13 11:33:11--  http://localhost/5MB.zip
Resolving localhost... ::1, 127.0.0.1
Connecting to localhost|127.0.0.1|:80... connected.
HTTP request sent, awaiting response...
  HTTP/1.1 206 Partial Content
  Server: nginx/1.0.6
  Date: Tue, 13 Sep 2011 10:33:11 GMT
  Content-Type: application/zip
  Content-Length: 5242830
  Connection: keep-alive
  Last-Modified: Mon, 02 Jun 2008 15:30:42 GMT
  ETag: "11f000b-500000-44eb0adaf4c80"
  X-Cached: HIT
  Content-Range: bytes 50-5242879/5242880
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110913/824d91b9/attachment-0001.html>

From mdounin at mdounin.ru  Tue Sep 13 10:30:23 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Tue, 13 Sep 2011 14:30:23 +0400
Subject: nginx & Range header & proxy cache
In-Reply-To: <EFC5A19DBF3A2E489904733B467D990B2BFB62@office-server2.m247.local>
References: <EFC5A19DBF3A2E489904733B467D990B2BFB62@office-server2.m247.local>
Message-ID: <20110913103023.GH1137@mdounin.ru>

Hello!

On Tue, Sep 13, 2011 at 10:29:21AM +0000, Richard Kearsley wrote:

> Hi
> If a Range request is sent to nginx proxy (e.g. header "Range: bytes=50-") and the file isn't in cache yet (cache MISS), the range is ignored and full file sent
> When the file has been cached (cache HIT), Range request works as expected
> 
> Is this the intended behaviour?

Yes.

Maxim Dounin


From nginx-forum at nginx.us  Tue Sep 13 11:32:06 2011
From: nginx-forum at nginx.us (liushaobo)
Date: Tue, 13 Sep 2011 07:32:06 -0400
Subject: Detects that the server is redirecting the address request cycle
Message-ID: <b29095f38022c7450012f7ab76d1f513.NginxMailingListEnglish@forum.nginx.org>

hi all

add "       if ($http_referer ~* "www.test.com") {
                rewrite    ^/(.*)$       http://www.xxx.com permanent;
        }"

open browser ?? ?	
This redirect page is incorrect
Detects that the server is redirecting the address request cycle?

remove  "       if ($http_referer ~* "www.test.com") {
                rewrite    ^/(.*)$       http://www.xxx.com permanent;
        }"
are normal

nginx server config

server {
        listen 80;
        server_name www.xxx.com;
        index index.htm index.html index.php;
        root /www/xxx;

        log_format  wwwlogs  '$remote_addr - $remote_user [$time_local]
"$request" '
               '$status $body_bytes_sent "$http_referer" '
               '"$http_user_agent" $http_x_forwarded_for';
        access_log      /data0/logs/wwwlogs.log;

        location ^~ /index.php/mo {
                rewrite    ^/(.*)$    http://mo.xxx.com    permanent;
        }

       if ($http_referer ~* "www.test.com") {
                rewrite    ^/(.*)$       http://www.xxx.com permanent;
        }

        location ~* \.php {
                fastcgi_pass 127.0.0.1:9000;
                fastcgi_index index.php;
                include fcgi.conf;
                fastcgi_param REQUEST_URI $fastcgi_script_name;
                fastcgi_param SCRIPT_FILENAME
$document_root$fastcgi_script_name;
                fastcgi_param PATH_INFO $fastcgi_script_name;
        }


        location ~* \.(jpg|png|jpeg|bmp|gif|swf|js|css) {
                expires 7d;
        }

        location ~* \.apk$ {
                limit_rate 80k;
        }

    rewrite ^/web/soft/index/category/(\d+)/?$ /category/$1 permanent;
    rewrite ^/web/allSoft/soft/category/(\d+)/?$ /apk/list/$1
permanent;
    rewrite ^/web/allSoft/soft/category/(\d+)/order/download_count/?$
/apk/list/$1?order=download_count permanent;
    rewrite ^/web/allSoft/soft/category/(\d+)/order/release_time/?$
/apk/list/$1?order=release_time permanent;
    rewrite ^/web/soft/show/id/(\d+)/?$ /apk/detail/$1 permanent;
    rewrite ^/web/default/apk/id/(\d+)/?$ /apk/download/$1 permanent;
    rewrite ^/web/allSoft/rec/?$ /apk/recommended permanent;
    rewrite ^/web/search/all/q/(.*)/?$ /search?q=$1 permanent;
    rewrite ^/web/rom/index/brand/(\d+)/?$ /rom/brand/$1 permanent;
    rewrite ^/web/rom/index/?$ /rom permanent;
    rewrite ^/web/rom/device/id/(\d+)/?/$ /rom/device/$1 permanent;
    rewrite ^/web/rom/list/category/(.*)/device/(\d+)/?$ /rom/list/$1/$2
permanent;
    rewrite ^/web/rom/show/id/(\d+)/?$ /rom/detail/$1 permanent;
    rewrite ^/web/rom/api/level/(.*)/?$ /rom/level/$1 permanent;
    rewrite ^/web/rom/download/id/(\d+)/?$ /rom/download/$1 permanent;

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215139,215139#msg-215139


From mdounin at mdounin.ru  Tue Sep 13 11:35:58 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Tue, 13 Sep 2011 15:35:58 +0400
Subject: Detects that the server is redirecting the address request cycle
In-Reply-To: <b29095f38022c7450012f7ab76d1f513.NginxMailingListEnglish@forum.nginx.org>
References: <b29095f38022c7450012f7ab76d1f513.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20110913113557.GI1137@mdounin.ru>

Hello!

On Tue, Sep 13, 2011 at 07:32:06AM -0400, liushaobo wrote:

> hi all
> 
> add "       if ($http_referer ~* "www.test.com") {
>                 rewrite    ^/(.*)$       http://www.xxx.com permanent;
>         }"
> 
> open browser ?? ?	
> This redirect page is incorrect
> Detects that the server is redirecting the address request cycle?
> 
> remove  "       if ($http_referer ~* "www.test.com") {
>                 rewrite    ^/(.*)$       http://www.xxx.com permanent;
>         }"
> are normal

You've asked nginx to do infinite redirect, and you've got it.  
It's not clear what you are asking about.

Maxim Dounin


From Richard.Kearsley at m247.com  Tue Sep 13 11:57:12 2011
From: Richard.Kearsley at m247.com (Richard Kearsley)
Date: Tue, 13 Sep 2011 11:57:12 +0000
Subject: nginx & Range header & proxy cache
In-Reply-To: <20110913103023.GH1137@mdounin.ru>
References: <EFC5A19DBF3A2E489904733B467D990B2BFB62@office-server2.m247.local>
	<20110913103023.GH1137@mdounin.ru>
Message-ID: <EFC5A19DBF3A2E489904733B467D990B2BFE99@office-server2.m247.local>


> Is this the intended behaviour?

>Yes.

Can your ngx_http_bytes_filter_module be used to get around this problem?


From amoiz.shine at gmail.com  Tue Sep 13 11:58:14 2011
From: amoiz.shine at gmail.com (Sharl.Jimh.Tsin)
Date: Tue, 13 Sep 2011 19:58:14 +0800
Subject: Detects that the server is redirecting the address request cycle
In-Reply-To: <20110913113557.GI1137@mdounin.ru>
References: <b29095f38022c7450012f7ab76d1f513.NginxMailingListEnglish@forum.nginx.org>
	<20110913113557.GI1137@mdounin.ru>
Message-ID: <1315915094.2012.6.camel@sharl-desktop>

? 2011-09-13?? 15:35 +0400?Maxim Dounin???
> Hello!
> 
> On Tue, Sep 13, 2011 at 07:32:06AM -0400, liushaobo wrote:
> 
> > hi all
> > 
> > add "       if ($http_referer ~* "www.test.com") {
> >                 rewrite    ^/(.*)$       http://www.xxx.com permanent;
> >         }"
> > 
> > open browser ?? ?	
> > This redirect page is incorrect
> > Detects that the server is redirecting the address request cycle?
> > 
> > remove  "       if ($http_referer ~* "www.test.com") {
> >                 rewrite    ^/(.*)$       http://www.xxx.com permanent;
> >         }"
> > are normal
> 
> You've asked nginx to do infinite redirect, and you've got it.  
> It's not clear what you are asking about.
> 
> Maxim Dounin
> 
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx

well,the OP make nginx into DEAD-CYCLE while $Http_Refer equals the site
rewrite/redirect to.

i guess that the OP should do something on regex of $Http_refer.

-- 
Best regards,
Sharl.Jimh.Tsin (From China **Obviously Taiwan INCLUDED**)

Using Gmail? Please read this important notice:
http://www.fsf.org/campaigns/jstrap/gmail?10073.


From nginx-forum at nginx.us  Tue Sep 13 12:48:30 2011
From: nginx-forum at nginx.us (MaLo)
Date: Tue, 13 Sep 2011 08:48:30 -0400
Subject: Using Nginx Proxy Pss IPv4 to IPv6 Problem
In-Reply-To: <20101206121231.GH72777@mdounin.ru>
References: <20101206121231.GH72777@mdounin.ru>
Message-ID: <1db6f1a2d0f4995f98baf7b3a218d723.NginxMailingListEnglish@forum.nginx.org>

Hello,

has anything changed at this topic? Is IPv6 now available for the proxy
module?

I need this feature too, and, if not available, i'm wondering why, i
think it would be very useful.

Best regards
Matthias

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,155900,215154#msg-215154


From savages at mozapps.com  Tue Sep 13 16:01:40 2011
From: savages at mozapps.com (Shaun savage)
Date: Wed, 14 Sep 2011 00:01:40 +0800
Subject: search headers_in
Message-ID: <4E6F7E64.5070503@mozapps.com>

I am starting to write a new module that uses a header token.  I have
searched google for the answer about "how to seach in headers for a
name.  I want to search for "X-session-token".  I am surprised that
there is not a library function that just does the search,

As I understand it now every module that wants to search for a header
writes it own search function?
Is there a "library search function"?

once I have the session token I use memcached in the PREACCESS phase to
load the session info, then in the ACCESS phase check the access permission.


From Richard.Kearsley at m247.com  Tue Sep 13 16:04:44 2011
From: Richard.Kearsley at m247.com (Richard Kearsley)
Date: Tue, 13 Sep 2011 16:04:44 +0000
Subject: nginx & Range header & proxy cache
In-Reply-To: <EFC5A19DBF3A2E489904733B467D990B2BFE99@office-server2.m247.local>
References: <EFC5A19DBF3A2E489904733B467D990B2BFB62@office-server2.m247.local>
	<20110913103023.GH1137@mdounin.ru>
	<EFC5A19DBF3A2E489904733B467D990B2BFE99@office-server2.m247.local>
Message-ID: <EFC5A19DBF3A2E489904733B467D990B2C06E6@office-server2.m247.local>

In relation to this,
If the file is incomplete on cache (for some reason nginx kept a failed cache attempt?) . When range is inside the content-length but more than the size of the (incomplete) cached file, it returns 502

I actually posted about it before here but forgot about it http://forum.nginx.org/read.php?2,190276,190276

Is there any possible fix for this?




From lists at ruby-forum.com  Wed Sep 14 06:57:28 2011
From: lists at ruby-forum.com (JAZZ F.)
Date: Wed, 14 Sep 2011 08:57:28 +0200
Subject: Nginx proxy truncates large file downloads to 1.1G
In-Reply-To: <AANLkTimpX+OVM13Osjm2fMuvBU=7uq+uHb0YVMzewe8G@mail.gmail.com>
References: <4CFC3A80.7090709@rattyshack.ca>
	<AANLkTimpX+OVM13Osjm2fMuvBU=7uq+uHb0YVMzewe8G@mail.gmail.com>
Message-ID: <cd419cae77e206b112a805bc3019595d@ruby-forum.com>

Even i am facing the same problem but i m yet to figure out the exact 
cause of it.....The disk space seems to be fine.....In my case when 
downloading a file larger than 1G(1.2 GB) it will truncate to 1 GB...

-- 
Posted via http://www.ruby-forum.com/.


From cliff at develix.com  Wed Sep 14 07:40:38 2011
From: cliff at develix.com (Cliff Wells)
Date: Wed, 14 Sep 2011 00:40:38 -0700
Subject: Nginx proxy truncates large file downloads to 1.1G
In-Reply-To: <4CFC3A80.7090709@rattyshack.ca>
References: <4CFC3A80.7090709@rattyshack.ca>
Message-ID: <1315986038.3331.0.camel@portable-evil>

On Sun, 2010-12-05 at 20:21 -0500, Duane Mulder wrote:
> So nginx so far has been running great. Today however I discovered that
> when downloading a file larger than 1G (ie 1.5G)  it will always fail
> leaving the client with a file of 1.1G and an incomplete download. 

Try disabling gzip and see if that helps.  If so, tuning the gzip
settings might be in order.

Cliff


From mdounin at mdounin.ru  Wed Sep 14 07:53:20 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Wed, 14 Sep 2011 11:53:20 +0400
Subject: Nginx proxy truncates large file downloads to 1.1G
In-Reply-To: <cd419cae77e206b112a805bc3019595d@ruby-forum.com>
References: <4CFC3A80.7090709@rattyshack.ca>
	<AANLkTimpX+OVM13Osjm2fMuvBU=7uq+uHb0YVMzewe8G@mail.gmail.com>
	<cd419cae77e206b112a805bc3019595d@ruby-forum.com>
Message-ID: <20110914075320.GP1137@mdounin.ru>

Hello!

On Wed, Sep 14, 2011 at 08:57:28AM +0200, JAZZ F. wrote:

> Even i am facing the same problem but i m yet to figure out the exact 
> cause of it.....The disk space seems to be fine.....In my case when 
> downloading a file larger than 1G(1.2 GB) it will truncate to 1 GB...

proxy_max_temp_file_size, when reached, causes relative long delay 
in reading from upstream server; this may cause timeout on 
upstream server, seen as "truncation".

There are 3 possible workarounds:

1. Use smaller proxy_max_temp_file_size to reduce delay in 
reading.

2. Use larger proxy_max_temp_file_size to avoid hitting it.

3. Use larger timeout on upstream server.

Maxim Dounin


From agentzh at gmail.com  Wed Sep 14 08:49:43 2011
From: agentzh at gmail.com (agentzh)
Date: Wed, 14 Sep 2011 16:49:43 +0800
Subject: Rewriting the domain part of Set-Cookie in a proxy_pass
In-Reply-To: <e899e00e344f0d3b312389d4a40c2bbe.NginxMailingListEnglish@forum.nginx.org>
References: <e899e00e344f0d3b312389d4a40c2bbe.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <CAB4Tn6OQ5EwEPMvtwWai2judCupvKHtN8BhX9c0D+wMPbtuwfg@mail.gmail.com>

On Wed, Jun 8, 2011 at 9:16 PM, tobia <nginx-forum at nginx.us> wrote:
> Does nginx (or a module) provide functionality equivalent to httpd's
> ProxyPassReverseCookieDomain?
>
> Long explanation: I have a simple nginx reverse proxy
>
> server {
> ?server_name external.domain.com;
> ?location / {
> ? ?proxy_pass http://backend.int/;
> ?}
> }
>
> The problem is that Set-Cookie response headers contain
> ";Domain=backend.int", because the backend does not know it is being
> reverse proxied. Therefore the users' browsers dutifully reject such bad
> cookies.
>
> How can I rewrite the content of the Set-Cookie response headers,
> replacing ";Domain=backend.int" with ";Domain=external.domain.com"?
>

Now with the latest ngx_lua v0.3.1rc3, we can implement this with a
little inlined Lua in nginx.conf:

    server_name external.domain.com;

    location / {
        proxy_pass http://backend.int/;

        header_filter_by_lua '
            local cookies = ngx.header.set_cookie
            if not cookies then return end
            local newcookies = {}
            for i, val in ipairs(cookies) do
                local newval = string.gsub(val, "([dD]omain)=[%w_-\\\\.]+",
                          "%1=external.domain.com")
                table.insert(newcookies, newval)
            end
            ngx.header.set_cookie = newcookies
        ';
    }

We can surely make this more portable by avoiding hard-coding the
"external.domain.com" literal in our Lua code and reference
ngx.var.server_name (i.e., the nginx variable $server_name) instead ;)

See the ngx_lua documentation for more details:

    http://wiki.nginx.org/HttpLuaModule

The most relevant sections are

    http://wiki.nginx.org/HttpLuaModule#header_filter_by_lua

and

    http://wiki.nginx.org/HttpLuaModule#ngx.header.HEADER

Have fun!
-agentzh


From agentzh at gmail.com  Wed Sep 14 09:24:16 2011
From: agentzh at gmail.com (agentzh)
Date: Wed, 14 Sep 2011 17:24:16 +0800
Subject: Rewriting the domain part of Set-Cookie in a proxy_pass
In-Reply-To: <CAB4Tn6OQ5EwEPMvtwWai2judCupvKHtN8BhX9c0D+wMPbtuwfg@mail.gmail.com>
References: <e899e00e344f0d3b312389d4a40c2bbe.NginxMailingListEnglish@forum.nginx.org>
	<CAB4Tn6OQ5EwEPMvtwWai2judCupvKHtN8BhX9c0D+wMPbtuwfg@mail.gmail.com>
Message-ID: <CAB4Tn6OdefnS1fZDSHNqtQyPi6usCP3_Mq9YNzYokv5gMeyZUw@mail.gmail.com>

On Wed, Sep 14, 2011 at 4:49 PM, agentzh <agentzh at gmail.com> wrote:
> ? ?server_name external.domain.com;
>
> ? ?location / {
> ? ? ? ?proxy_pass http://backend.int/;
>
> ? ? ? ?header_filter_by_lua '
> ? ? ? ? ? ?local cookies = ngx.header.set_cookie
> ? ? ? ? ? ?if not cookies then return end

Sorry, one line of Lua should be inserted right here for the case of a
single Set-Cookie response header:

            if type(cookies) ~= "table" then cookies = {cookies} end

that is, ngx.header.set_cookie will just return a single Lua string
instead of a Lua table if there's only one Set-Cookie header.

Regards,
-agentzh


From nginx-forum at nginx.us  Wed Sep 14 10:48:25 2011
From: nginx-forum at nginx.us (coviex)
Date: Wed, 14 Sep 2011 06:48:25 -0400
Subject: 400 Bad request
Message-ID: <9ba035044ed2b01344ed15e1c61b3e46.NginxMailingListEnglish@forum.nginx.org>

Hi,

Having problems with urls containing percent sign. Most of them return
400 and don't even appear neither in error.log nor in access.log.

GET /download-%27.07%%27.html?frame=1 HTTP/1.1
Host: www.my.eu
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:6.0) Gecko/20100101
Firefox/6.0 FirePHP/0.6
Accept: text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8
Accept-Language: en-us,en;q=0.5
Accept-Encoding: gzip, deflate
Accept-Charset: ISO-8859-1,utf-8;q=0.7,*;q=0.7
Connection: keep-alive
x-insight: activate

HTTP/1.1 400 Bad Request
Server: nginx/1.0.5
Date: Wed, 14 Sep 2011 10:36:24 GMT
Content-Type: text/html
Content-Length: 172
Connection: close


What's wrong with them?


Thanks

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215184,215184#msg-215184


From mdounin at mdounin.ru  Wed Sep 14 11:11:58 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Wed, 14 Sep 2011 15:11:58 +0400
Subject: 400 Bad request
In-Reply-To: <9ba035044ed2b01344ed15e1c61b3e46.NginxMailingListEnglish@forum.nginx.org>
References: <9ba035044ed2b01344ed15e1c61b3e46.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20110914111158.GS1137@mdounin.ru>

Hello!

On Wed, Sep 14, 2011 at 06:48:25AM -0400, coviex wrote:

> Hi,
> 
> Having problems with urls containing percent sign. Most of them return
> 400 and don't even appear neither in error.log nor in access.log.
> 
> GET /download-%27.07%%27.html?frame=1 HTTP/1.1
> Host: www.my.eu
> User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:6.0) Gecko/20100101
> Firefox/6.0 FirePHP/0.6
> Accept: text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8
> Accept-Language: en-us,en;q=0.5
> Accept-Encoding: gzip, deflate
> Accept-Charset: ISO-8859-1,utf-8;q=0.7,*;q=0.7
> Connection: keep-alive
> x-insight: activate
> 
> HTTP/1.1 400 Bad Request
> Server: nginx/1.0.5
> Date: Wed, 14 Sep 2011 10:36:24 GMT
> Content-Type: text/html
> Content-Length: 172
> Connection: close
> 
> 
> What's wrong with them?

They aren't valid.

http://tools.ietf.org/html/rfc3986#section-2.4

   Because the percent ("%") character serves as the indicator for
   percent-encoded octets, it must be percent-encoded as "%25" for that
   octet to be used as data within a URI.

Maxim Dounin


From francis at daoine.org  Wed Sep 14 11:13:15 2011
From: francis at daoine.org (Francis Daly)
Date: Wed, 14 Sep 2011 12:13:15 +0100
Subject: 400 Bad request
In-Reply-To: <9ba035044ed2b01344ed15e1c61b3e46.NginxMailingListEnglish@forum.nginx.org>
References: <9ba035044ed2b01344ed15e1c61b3e46.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20110914111315.GB12179@craic.sysops.org>

On Wed, Sep 14, 2011 at 06:48:25AM -0400, coviex wrote:

Hi there,

> Having problems with urls containing percent sign. Most of them return
> 400 and don't even appear neither in error.log nor in access.log.

% is an escape character within a url. It must be followed by exactly
two hex-characters, or the url is invalid and the request is bad.

> GET /download-%27.07%%27.html?frame=1 HTTP/1.1

The first one there is %27 and decodes to '.

The second one is %%2 and is invalid.

> What's wrong with them?

They're broken.

Whatever is generating that url should be fixed so that it is properly
encoded/escaped.

If the filename is download-'.07%'.html, it should appear in the url
as download-%27.07%25%27.html

All the best,

	f
-- 
Francis Daly        francis at daoine.org


From nginx-forum at nginx.us  Wed Sep 14 11:26:12 2011
From: nginx-forum at nginx.us (tm1978)
Date: Wed, 14 Sep 2011 07:26:12 -0400
Subject: Nginx + Tomcat7 for single webapp and multiple alias domain
Message-ID: <3a1f887cd0dc159f1671f69825a3dd72.NginxMailingListEnglish@forum.nginx.org>

I am thinking about using NGINX instead of Apache to deliver a Tomcat7
webapp on Port :80 to the enduser.

I have the following requirements:

1. There is a single  web-app in the root with any urls (its own
internal urlrewriting). It has various URLs pointing to it for several
languages (and partially other content)

2. Another admin-webapp exists


My questions are:

(a) would it be a good idea to use NGINX a without  Apache Http server,
just with Apache Tomcat7 ?
(b) how can I set this up so that the Tomcat App always know the domain
name the user was using when calling the page?
(c) will the tomcat app still be able to know the IPs of every user?


Thanks!

Tobias

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215189,215189#msg-215189


From Richard.Kearsley at m247.com  Wed Sep 14 13:50:39 2011
From: Richard.Kearsley at m247.com (Richard Kearsley)
Date: Wed, 14 Sep 2011 13:50:39 +0000
Subject: $request_uri without args
Message-ID: <EFC5A19DBF3A2E489904733B467D990B2C25BA@office-server2.m247.local>

Hi
Is it possible to access the $request_uri without the args at the end?
I know I can use $uri - but this returns the url-decoded uri.
I need the url-encoded (I.e. unchanged from original request) but with the args stripped


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110914/57db8d91/attachment.html>

From Andrew.Hester at mouser.com  Wed Sep 14 13:55:07 2011
From: Andrew.Hester at mouser.com (Andrew Hester)
Date: Wed, 14 Sep 2011 08:55:07 -0500
Subject: load balance and re-encrypt ssl
Message-ID: <01799974966632498879A6D3F9CB343201CDBAD7E293@EXCL01.mouser.lan>

Hi all,

I have been looking through the docs and have found that nginx will load balance ssl if the cert is setup correctly.  Is there an option to have it re-encrypt before sending to the web servers?  What have others done to encrypt the traffic to their webservers (if not through nginx)?   I will be load balancing in front of IIS.


Decrypting and sending to the webservers in plain text is not an option in my mind.

Thanks in advance.



-Andy



This communication, its contents and any file attachments transmitted with it are intended solely for the addressee(s) and may contain confidential proprietary information.
Access by any other party without the express written permission of the sender is STRICTLY PROHIBITED.
If you have received this communication in error you may not copy, distribute or use the contents, attachments or information in any way.  Please destroy it and contact the sender.


From mdounin at mdounin.ru  Wed Sep 14 14:12:36 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Wed, 14 Sep 2011 18:12:36 +0400
Subject: load balance and re-encrypt ssl
In-Reply-To: <01799974966632498879A6D3F9CB343201CDBAD7E293@EXCL01.mouser.lan>
References: <01799974966632498879A6D3F9CB343201CDBAD7E293@EXCL01.mouser.lan>
Message-ID: <20110914141236.GW1137@mdounin.ru>

Hello!

On Wed, Sep 14, 2011 at 08:55:07AM -0500, Andrew Hester wrote:

> I have been looking through the docs and have found that nginx 
> will load balance ssl if the cert is setup correctly.  Is there 
> an option to have it re-encrypt before sending to the web 
> servers?  What have others done to encrypt the traffic to their 
> webservers (if not through nginx)?   I will be load balancing in 
> front of IIS.
> 
> 
> Decrypting and sending to the webservers in plain text is not an 
> option in my mind.

proxy_pass https://...

Maxim Dounin


From agentzh at gmail.com  Wed Sep 14 14:14:12 2011
From: agentzh at gmail.com (agentzh)
Date: Wed, 14 Sep 2011 22:14:12 +0800
Subject: $request_uri without args
In-Reply-To: <EFC5A19DBF3A2E489904733B467D990B2C25BA@office-server2.m247.local>
References: <EFC5A19DBF3A2E489904733B467D990B2C25BA@office-server2.m247.local>
Message-ID: <CAB4Tn6Pm0TCx=CYPpdKbQSqPbvYmCeMSmPyN-KYVLeaRXVdUZw@mail.gmail.com>

On Wed, Sep 14, 2011 at 9:50 PM, Richard Kearsley
<Richard.Kearsley at m247.com> wrote:
> Hi
>
> Is it possible to access the $request_uri without the args at the end?
>
> I know I can use $uri ? but this returns the url-decoded uri.
>
> I need the url-encoded (I.e. unchanged from original request) but with the
> args stripped
>

Why not strip $request_uri yourself? You can use "if" + pcre regex or
more preferably, set_by_lua/rewrite_by_lua provided by ngx_lua, to
process what in $request_uri :)

Regards,
-agentzh


From Andrew.Hester at mouser.com  Wed Sep 14 14:21:12 2011
From: Andrew.Hester at mouser.com (Andrew Hester)
Date: Wed, 14 Sep 2011 09:21:12 -0500
Subject: load balance and re-encrypt ssl
In-Reply-To: <20110914141236.GW1137@mdounin.ru>
References: <01799974966632498879A6D3F9CB343201CDBAD7E293@EXCL01.mouser.lan>
	<20110914141236.GW1137@mdounin.ru>
Message-ID: <01799974966632498879A6D3F9CB343201CDBAD7E2AE@EXCL01.mouser.lan>

I guess I assumed that this would just pass this through at the tcp level.   Thanks for the help!

-Andy



-----Original Message-----
From: nginx-bounces at nginx.org [mailto:nginx-bounces at nginx.org] On Behalf Of Maxim Dounin
Sent: Wednesday, September 14, 2011 9:13 AM
To: nginx at nginx.org
Subject: Re: load balance and re-encrypt ssl

Hello!

On Wed, Sep 14, 2011 at 08:55:07AM -0500, Andrew Hester wrote:

> I have been looking through the docs and have found that nginx will
> load balance ssl if the cert is setup correctly.  Is there an option
> to have it re-encrypt before sending to the web servers?  What have
> others done to encrypt the traffic to their
> webservers (if not through nginx)?   I will be load balancing in
> front of IIS.
>
>
> Decrypting and sending to the webservers in plain text is not an
> option in my mind.

proxy_pass https://...

Maxim Dounin

_______________________________________________
nginx mailing list
nginx at nginx.org
http://mailman.nginx.org/mailman/listinfo/nginx

This communication, its contents and any file attachments transmitted with it are intended solely for the addressee(s) and may contain confidential proprietary information.
Access by any other party without the express written permission of the sender is STRICTLY PROHIBITED.
If you have received this communication in error you may not copy, distribute or use the contents, attachments or information in any way.  Please destroy it and contact the sender.


From Richard.Kearsley at m247.com  Wed Sep 14 14:30:26 2011
From: Richard.Kearsley at m247.com (Richard Kearsley)
Date: Wed, 14 Sep 2011 14:30:26 +0000
Subject: $request_uri without args
In-Reply-To: <CAB4Tn6Pm0TCx=CYPpdKbQSqPbvYmCeMSmPyN-KYVLeaRXVdUZw@mail.gmail.com>
References: <EFC5A19DBF3A2E489904733B467D990B2C25BA@office-server2.m247.local>
	<CAB4Tn6Pm0TCx=CYPpdKbQSqPbvYmCeMSmPyN-KYVLeaRXVdUZw@mail.gmail.com>
Message-ID: <EFC5A19DBF3A2E489904733B467D990B2C26F5@office-server2.m247.local>

Agentzh, I'm glad you replied :)
I am trying to find out why something is happening - a bug in nginx or lua module (currently finding it hard to reproduce)
I'm using:

location /proxy
{
	proxy_pass http://backend;
}

access_by_lua '
                                local res = ngx.location.capture("/proxy" .. ngx.var.request_uri, { args = { country = ngx.var.geoip_country_code, host = ngx.var.host } } )

if ngx.var.request_url contains args like this:

/something/blah.jpg?34567

It gives 404

But like this:

/something/blah.jps?anything=34567

Works and gets processed ok

I don't need args at backend - but I DO need the url still encoded

Backend when accessed directly with curl/wget with "/something/blah.jpg?34567" returns ok

Any ideas?

-----Original Message-----
From: nginx-bounces at nginx.org [mailto:nginx-bounces at nginx.org] On Behalf Of agentzh
Sent: 14 September 2011 15:14
To: nginx at nginx.org
Subject: Re: $request_uri without args

On Wed, Sep 14, 2011 at 9:50 PM, Richard Kearsley <Richard.Kearsley at m247.com> wrote:
> Hi
>
> Is it possible to access the $request_uri without the args at the end?
>
> I know I can use $uri - but this returns the url-decoded uri.
>
> I need the url-encoded (I.e. unchanged from original request) but with 
> the args stripped
>

Why not strip $request_uri yourself? You can use "if" + pcre regex or more preferably, set_by_lua/rewrite_by_lua provided by ngx_lua, to process what in $request_uri :)

Regards,
-agentzh

_______________________________________________
nginx mailing list
nginx at nginx.org
http://mailman.nginx.org/mailman/listinfo/nginx


From nginx-forum at nginx.us  Wed Sep 14 15:19:05 2011
From: nginx-forum at nginx.us (dullnicker)
Date: Wed, 14 Sep 2011 11:19:05 -0400
Subject: Problem with Hotlink Protection
Message-ID: <0c253481fec8974edd546fbde9b7c383.NginxMailingListEnglish@forum.nginx.org>

Dear all,

I am having a problem getting hotlinking protection to work. I run an
image gallery on a nginx webserver and would like to achieve that
hotlinking to the images on the website is impossible to save
bandwidth.

I have the following code in my location block:

<-- SNIP --->

location ~* \.(png|gif|jpg|jpeg)$ {
    valid_referers server_names;
    if ($invalid_referer) {
    return   403;
  }
}

<-- SNAP -->

That works fine so far. But the problem is that the gallery script
creates image links that look like this:

<-- SNIP --->

http://www.mydomain.com/image.php?twg_album=album-05900-05999&twg_show=image-05995.jpg

<-- SNAP -->

These do not trigger the hotlink protection. I am totally lost and have
no idea how I might be able to protect those image links from
hotlinking.
Any help would be highly appreciated! Thank you all in advance!

Kind regards
-Amitz

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215198,215198#msg-215198


From igor at sysoev.ru  Wed Sep 14 15:19:25 2011
From: igor at sysoev.ru (Igor Sysoev)
Date: Wed, 14 Sep 2011 19:19:25 +0400
Subject: nginx-1.1.3
Message-ID: <20110914151925.GG49199@nginx.com>

Changes with nginx 1.1.3                                         14 Sep 2011

    *) Feature: the module ngx_http_mp4_module.

    *) Bugfix: in Linux AIO combined with open_file_cache.

    *) Bugfix: open_file_cache did not update file info on retest if file
       was not atomically changed.

    *) Bugfix: nginx could not be built on MacOSX 10.7.


-- 
Igor Sysoev


From nginx-forum at nginx.us  Wed Sep 14 15:25:24 2011
From: nginx-forum at nginx.us (coviex)
Date: Wed, 14 Sep 2011 11:25:24 -0400
Subject: 400 Bad request
In-Reply-To: <9ba035044ed2b01344ed15e1c61b3e46.NginxMailingListEnglish@forum.nginx.org>
References: <9ba035044ed2b01344ed15e1c61b3e46.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <5f17e081023105312ec1153d0e3ab89b.NginxMailingListEnglish@forum.nginx.org>

This query string results in 400 on all nginx servers/sites I know of. 
Say, http://nginx.org/download-%27.07%%27.html?frame=1.
Any suggestions?

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215184,215203#msg-215203


From igor at sysoev.ru  Wed Sep 14 15:30:37 2011
From: igor at sysoev.ru (Igor Sysoev)
Date: Wed, 14 Sep 2011 19:30:37 +0400
Subject: 400 Bad request
In-Reply-To: <5f17e081023105312ec1153d0e3ab89b.NginxMailingListEnglish@forum.nginx.org>
References: <9ba035044ed2b01344ed15e1c61b3e46.NginxMailingListEnglish@forum.nginx.org>
	<5f17e081023105312ec1153d0e3ab89b.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20110914153037.GI49199@nginx.com>

On Wed, Sep 14, 2011 at 11:25:24AM -0400, coviex wrote:
> This query string results in 400 on all nginx servers/sites I know of. 
> Say, http://nginx.org/download-%27.07%%27.html?frame=1.

Not only nginx. Apache does (correctly) the same:

$nc apache.org 80
GET /download-%27.07%%27.html HTTP/1.0

HTTP/1.1 400 Bad Request
Date: Wed, 14 Sep 2011 15:29:31 GMT
Server: Apache/2.3.15-dev (Unix) mod_ssl/2.3.15-dev OpenSSL/1.0.0c
Content-Length: 342
Connection: close
Content-Type: text/html; charset=iso-8859-1

[...]

> Any suggestions?

-http://nginx.org/download-%27.07%%27.html?frame=1
+http://nginx.org/download-%27.07%25%27.html?frame=1


-- 
Igor Sysoev


From agentzh at gmail.com  Wed Sep 14 15:39:01 2011
From: agentzh at gmail.com (agentzh)
Date: Wed, 14 Sep 2011 23:39:01 +0800
Subject: $request_uri without args
In-Reply-To: <EFC5A19DBF3A2E489904733B467D990B2C26F5@office-server2.m247.local>
References: <EFC5A19DBF3A2E489904733B467D990B2C25BA@office-server2.m247.local>
	<CAB4Tn6Pm0TCx=CYPpdKbQSqPbvYmCeMSmPyN-KYVLeaRXVdUZw@mail.gmail.com>
	<EFC5A19DBF3A2E489904733B467D990B2C26F5@office-server2.m247.local>
Message-ID: <CAB4Tn6NunDf7YpuhuuZD6eGPzMBf=iQOh0+aUUEas0X6u=dgfw@mail.gmail.com>

On Wed, Sep 14, 2011 at 10:30 PM, Richard Kearsley
<Richard.Kearsley at m247.com> wrote:
>
> I don't need args at backend - but I DO need the url still encoded
>

Use the following line of Lua to strip the query args part (if any)
from ngx.var.request (i.e., the nginx variable $request_uri):

    local uri = string.gsub(ngx.var.request_uri, "?.*", "")

and then you can feed the resulting uri variable into your subrequest
call instead of ngx.var.request_uri:

    local res = ngx.location.capture("/proxy" .. uri,
            { args =
                  { country = ngx.var.geoip_country_code, host = ngx.var.host }
            } )

Hope it works for you :)

Regards,
-agentzh


From francis at daoine.org  Wed Sep 14 15:52:32 2011
From: francis at daoine.org (Francis Daly)
Date: Wed, 14 Sep 2011 16:52:32 +0100
Subject: Problem with Hotlink Protection
In-Reply-To: <0c253481fec8974edd546fbde9b7c383.NginxMailingListEnglish@forum.nginx.org>
References: <0c253481fec8974edd546fbde9b7c383.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20110914155232.GC12179@craic.sysops.org>

On Wed, Sep 14, 2011 at 11:19:05AM -0400, dullnicker wrote:

Hi there,

> I am having a problem getting hotlinking protection to work.

> I have the following code in my location block:

> location ~* \.(png|gif|jpg|jpeg)$ {
>     valid_referers server_names;
>     if ($invalid_referer) {
>     return   403;
>   }
> }

That applies to locations that end in one of those four words.

> http://www.mydomain.com/image.php?twg_album=album-05900-05999&twg_show=image-05995.jpg

That location ends in ".php", since things after the ? don't count for
location matching. And so the above config doesn't apply, as you saw.

> These do not trigger the hotlink protection. I am totally lost and have
> no idea how I might be able to protect those image links from
> hotlinking.

The two simplest (to me) ways of approaching it would be (a) if /image.php
counts as "an image link", then configure it like one -- wherever in
your configuration you match this location, add the "invalid_referer"
check. (If you have something like "location ~* \.php$", and you don't
want all matching urls to have this configuration, then copy the content
to a new "location = /image.php" and add the extra bit there);

or (b) if image.php is your modifiable script, and it has access to
HTTP_REFERER, code the protection in there yourself.

If neither of those applies, then you can probably try mucking around
with $arg_twg_show; but that feels wrong to me.

All the best,

	f
-- 
Francis Daly        francis at daoine.org


From nginx-forum at nginx.us  Wed Sep 14 16:22:30 2011
From: nginx-forum at nginx.us (dullnicker)
Date: Wed, 14 Sep 2011 12:22:30 -0400
Subject: Problem with Hotlink Protection
In-Reply-To: <0c253481fec8974edd546fbde9b7c383.NginxMailingListEnglish@forum.nginx.org>
References: <0c253481fec8974edd546fbde9b7c383.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <694ad8f49315e7d8acba3180d8e80c61.NginxMailingListEnglish@forum.nginx.org>

Dear Francis,

thank you very much indeed for taking the time to help me out. It seems
that - with your help - I made it! :-)
I inserted the "invalid_referrer" check within the "location ~* \.php$"
block like you suggested and it works like a charm now!

It looks like this:
<-- SNIP -->
location ~ \.php(.*)$ {
    valid_referers server_names;
    if ($invalid_referer) {
    return   403;
}

   include fastcgi_params;
    fastcgi_intercept_errors on;
    fastcgi_index index.php;
    fastcgi_param SCRIPT_FILENAME $document_root$fastcgi_script_name;
  try_files $uri =404;
    fastcgi_pass 127.0.0.1:9000;
    error_page 404 /404page.html; #makes nginx return it's default 404
#	page instead of a blank page

}

<--SNAP-->

Again, thank you for your fantastic help!

Kind regards
-Amitz

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215198,215207#msg-215207


From Richard.Kearsley at m247.com  Wed Sep 14 16:33:38 2011
From: Richard.Kearsley at m247.com (Richard Kearsley)
Date: Wed, 14 Sep 2011 16:33:38 +0000
Subject: $request_uri without args
In-Reply-To: <CAB4Tn6NunDf7YpuhuuZD6eGPzMBf=iQOh0+aUUEas0X6u=dgfw@mail.gmail.com>
References: <EFC5A19DBF3A2E489904733B467D990B2C25BA@office-server2.m247.local>
	<CAB4Tn6Pm0TCx=CYPpdKbQSqPbvYmCeMSmPyN-KYVLeaRXVdUZw@mail.gmail.com>
	<EFC5A19DBF3A2E489904733B467D990B2C26F5@office-server2.m247.local>
	<CAB4Tn6NunDf7YpuhuuZD6eGPzMBf=iQOh0+aUUEas0X6u=dgfw@mail.gmail.com>
Message-ID: <EFC5A19DBF3A2E489904733B467D990B2C2BD3@office-server2.m247.local>

Many thanks, this fixed the problem (or got around it at least :) - so not sure where the bug is but I suspect it's the mixture of ?3456 and ?val=3456 in the query string after the lua pass to backend (I wasn't reading "host" at backend properly, hence 404)


>    local uri = string.gsub(ngx.var.request_uri, "?.*", "")
>    local res = ngx.location.capture("/proxy" .. uri,
>            { args =
>                  { country = ngx.var.geoip_country_code, host = ngx.var.host }
>            } )


From nginx-forum at nginx.us  Wed Sep 14 17:45:15 2011
From: nginx-forum at nginx.us (coviex)
Date: Wed, 14 Sep 2011 13:45:15 -0400
Subject: 400 Bad request
In-Reply-To: <20110914153037.GI49199@nginx.com>
References: <20110914153037.GI49199@nginx.com>
Message-ID: <b0048681811a927c153e3804cd386a83.NginxMailingListEnglish@forum.nginx.org>

Funny thing is there are two sibling threads on the forum named
identically and answers to my question wind up here.
http://forum.nginx.org/read.php?2,2757
http://forum.nginx.org/read.php?2,215184
And I didn't know about that so sorry for asking twice.

I get that requests with %% are invalid.
But why they don't appear in logs?
Can I show pretty 404 page for those 400?

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,2757,215210#msg-215210


From igor at sysoev.ru  Wed Sep 14 18:03:40 2011
From: igor at sysoev.ru (Igor Sysoev)
Date: Wed, 14 Sep 2011 22:03:40 +0400
Subject: 400 Bad request
In-Reply-To: <b0048681811a927c153e3804cd386a83.NginxMailingListEnglish@forum.nginx.org>
References: <20110914153037.GI49199@nginx.com>
	<b0048681811a927c153e3804cd386a83.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20110914180340.GA63007@nginx.com>

On Wed, Sep 14, 2011 at 01:45:15PM -0400, coviex wrote:
> Funny thing is there are two sibling threads on the forum named
> identically and answers to my question wind up here.
> http://forum.nginx.org/read.php?2,2757
> http://forum.nginx.org/read.php?2,215184
> And I didn't know about that so sorry for asking twice.
> 
> I get that requests with %% are invalid.
> But why they don't appear in logs?
> Can I show pretty 404 page for those 400?

nginx logs this on "info" level:
"client sent invalid request while reading client request line"

The default error_log level is "error".


-- 
Igor Sysoev


From igor at sysoev.ru  Wed Sep 14 18:41:50 2011
From: igor at sysoev.ru (Igor Sysoev)
Date: Wed, 14 Sep 2011 22:41:50 +0400
Subject: mp4 module
In-Reply-To: <b66d2a7e6c245d9d7fd455f0216b7ba1.NginxMailingListEnglish@forum.nginx.org>
References: <20110909130812.GI97288@nginx.com>
	<b66d2a7e6c245d9d7fd455f0216b7ba1.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20110914184150.GC63007@nginx.com>

On Mon, Sep 12, 2011 at 03:20:28PM -0400, meto wrote:
> Can't wait to see that feature in mainline (dev or stable). It's one
> that was bugging me about the license used by current one (commercial
> use).

Changes with nginx 1.1.3                                         14 Sep 2011

    *) Feature: the module ngx_http_mp4_module.

Or you can use the following patch with nginx-1.0:
http://nginx.org/patches/patch.mp4.3.txt

Documentation is here:
http://nginx.org/en/docs/http/ngx_http_mp4_module.html

> Are you planning to include it in your repo packages?

Yes.


-- 
Igor Sysoev


From francis at daoine.org  Wed Sep 14 18:51:20 2011
From: francis at daoine.org (Francis Daly)
Date: Wed, 14 Sep 2011 19:51:20 +0100
Subject: Problem with Hotlink Protection
In-Reply-To: <694ad8f49315e7d8acba3180d8e80c61.NginxMailingListEnglish@forum.nginx.org>
References: <0c253481fec8974edd546fbde9b7c383.NginxMailingListEnglish@forum.nginx.org>
	<694ad8f49315e7d8acba3180d8e80c61.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20110914185120.GD12179@craic.sysops.org>

On Wed, Sep 14, 2011 at 12:22:30PM -0400, dullnicker wrote:

Hi there,

> thank you very much indeed for taking the time to help me out. It seems
> that - with your help - I made it! :-)

You're welcome; and congratulations :-)

> I inserted the "invalid_referrer" check within the "location ~* \.php$"
> block like you suggested and it works like a charm now!

Good stuff.

> location ~ \.php(.*)$ {

Note that this config applies to all urls that include the four characters
".php"; that certainly includes your /image.php script, but possibly
also includes more.

If you're happy that all of those have this invalid_referer check,
then it's fine. If not, you'll want to adjust the config again.

All the best,

	f
-- 
Francis Daly        francis at daoine.org


From appa at perusio.net  Wed Sep 14 19:56:45 2011
From: appa at perusio.net (=?UTF-8?B?QW50w7NuaW8=?= P. P. Almeida)
Date: Wed, 14 Sep 2011 20:56:45 +0100
Subject: Problem with Hotlink Protection
In-Reply-To: <694ad8f49315e7d8acba3180d8e80c61.NginxMailingListEnglish@forum.nginx.org>
References: <0c253481fec8974edd546fbde9b7c383.NginxMailingListEnglish@forum.nginx.org>
	<694ad8f49315e7d8acba3180d8e80c61.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <87ty8eyk1u.wl%appa@perusio.net>

On 14 Set 2011 17h22 WEST, nginx-forum at nginx.us wrote:

> Dear Francis,
>
> thank you very much indeed for taking the time to help me out. It
> seems that - with your help - I made it! :-) I inserted the
> "invalid_referrer" check within the "location ~* \.php$" block like
> you suggested and it works like a charm now!
>
> It looks like this:
> <-- SNIP -->
> location ~ \.php(.*)$ {
> valid_referers server_names;
> if ($invalid_referer) {
> return   403;
> }
>
> include fastcgi_params;
> fastcgi_intercept_errors on;
> fastcgi_index index.php;
> fastcgi_param SCRIPT_FILENAME $document_root$fastcgi_script_name;
> try_files $uri =404;
> fastcgi_pass 127.0.0.1:9000;
> error_page 404 /404page.html; #makes nginx return it's default 404
> #	page instead of a blank page
>

I think that Francis suggestion is one that you should
consider. Restrict the referer checking to the image.php script only.

location = /image.php {
   valid_referers server_names;
   if ($invalid_referer) {
       return 403;
   }
   include fastcgi_params;
   fastcgi_intercept_errors on;
   fastcgi_index index.php;
   fastcgi_param SCRIPT_FILENAME $document_root$fastcgi_script_name;
   fastcgi_pass 127.0.0.1:9000;
   error_page 404 /404page.html; # makes nginx return it's default 404
                                 # page instead of a blank page
}

HTH,
--- appa


From nginx-forum at nginx.us  Wed Sep 14 22:54:24 2011
From: nginx-forum at nginx.us (magicbear)
Date: Wed, 14 Sep 2011 18:54:24 -0400
Subject: upstream keepalive - call for testing
In-Reply-To: <20110905180210.GL1137@mdounin.ru>
References: <20110905180210.GL1137@mdounin.ru>
Message-ID: <051a01adad30a88a5d6d428e8fbaaafc.NginxMailingListEnglish@forum.nginx.org>

I have run the nginx 1.1.2 via this patch for 7 days, except for one
days have a large DDoS so I restart nginx for several seconds, it was
very stable to work.
Handle about 70million request without problem happen, I think the last
problem may be have a memory corruption, you are right.
I will check that server when have times.
Thanks for your hard work.

MagicBear

Maxim Dounin Wrote:
-------------------------------------------------------
> Hello!
> 
> On Mon, Sep 05, 2011 at 11:42:31PM +0800,
> ????? wrote:
> 
> > (gdb) fr 0
> > #0  ngx_http_upstream_handler
> (ev=0x7fc45735f8a8)
> >     at src/http/ngx_http_upstream.c:915
> > 915    ctx->current_request = r;
> > (gdb) p ngx_cycle->log
> > $1 = (ngx_log_t *) 0x21f19a8
> > (gdb) p *r
> > $2 = {signature = 51686928, connection =
> 0x23b4160, ctx = 0x0,
> 
> [...]
> 
> This looks like memory corruption, but
> unfortunately I don't see 
> any traces of the real cause.  My best quess is
> improper handling 
> of proxy_ignore_client_abort as fixed in 1.1.2. 
> Please try 1.1.2 
> with patches from
> 
> http://nginx.org/patches/patch-nginx-keepalive-ful
> l-5.txt
> 
> It already includes upstream keepalive module, as
> well as all 
> other upstream-keepalive related fixes.  See here
> for details:
> 
> http://mailman.nginx.org/pipermail/nginx-devel/201
> 1-September/001147.html
> 
> Maxim Dounin
> 
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,213207,215217#msg-215217


From agentzh at gmail.com  Thu Sep 15 00:22:26 2011
From: agentzh at gmail.com (agentzh)
Date: Thu, 15 Sep 2011 08:22:26 +0800
Subject: $request_uri without args
In-Reply-To: <EFC5A19DBF3A2E489904733B467D990B2C2BD3@office-server2.m247.local>
References: <EFC5A19DBF3A2E489904733B467D990B2C25BA@office-server2.m247.local>
	<CAB4Tn6Pm0TCx=CYPpdKbQSqPbvYmCeMSmPyN-KYVLeaRXVdUZw@mail.gmail.com>
	<EFC5A19DBF3A2E489904733B467D990B2C26F5@office-server2.m247.local>
	<CAB4Tn6NunDf7YpuhuuZD6eGPzMBf=iQOh0+aUUEas0X6u=dgfw@mail.gmail.com>
	<EFC5A19DBF3A2E489904733B467D990B2C2BD3@office-server2.m247.local>
Message-ID: <CAB4Tn6NEoN4nvBdCWz7PAYTkA4KLgNYZR8A05cr-oDm_e_eOZg@mail.gmail.com>

On Thu, Sep 15, 2011 at 12:33 AM, Richard Kearsley
<Richard.Kearsley at m247.com> wrote:
> Many thanks, this fixed the problem (or got around it at least :) - so not sure where the bug is but I suspect it's the mixture of ?3456 and ?val=3456 in the query string after the lua pass to backend (I wasn't reading "host" at backend properly, hence 404)
>

I believe it's not a bug because it's wrong to pass the query args
part as the URI argument to ngx.location.capture. The behavior is
unspecified. So please avoid that in the first place.

Regards,
-agentzh


From clanherb at gmail.com  Thu Sep 15 03:05:59 2011
From: clanherb at gmail.com (=?UTF-8?B?5p2o6ZWt?=)
Date: Thu, 15 Sep 2011 11:05:59 +0800
Subject: nginx-1.1.3
In-Reply-To: <20110914151925.GG49199@nginx.com>
References: <20110914151925.GG49199@nginx.com>
Message-ID: <CAPKb8pD+3Z4=MEgwX=6sAqmrGQOMy1DjjoPfLZBij4DLC=y_yg@mail.gmail.com>

Cool?people who upgrade to Lion can use nginx again

Bugfix: nginx could not be built on MacOSX 10.7.

On Wed, Sep 14, 2011 at 11:19 PM, Igor Sysoev <igor at sysoev.ru> wrote:
> Changes with nginx 1.1.3 ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? 14 Sep 2011
>
> ? ?*) Feature: the module ngx_http_mp4_module.
>
> ? ?*) Bugfix: in Linux AIO combined with open_file_cache.
>
> ? ?*) Bugfix: open_file_cache did not update file info on retest if file
> ? ? ? was not atomically changed.
>
> ? ?*) Bugfix: nginx could not be built on MacOSX 10.7.
>
>
> --
> Igor Sysoev
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx
>



-- 
lei yang


From webwindy at gmail.com  Thu Sep 15 03:25:36 2011
From: webwindy at gmail.com (C.F)
Date: Thu, 15 Sep 2011 11:25:36 +0800
Subject: mp4 module
In-Reply-To: <20110914184150.GC63007@nginx.com>
References: <20110909130812.GI97288@nginx.com>
	<b66d2a7e6c245d9d7fd455f0216b7ba1.NginxMailingListEnglish@forum.nginx.org>
	<20110914184150.GC63007@nginx.com>
Message-ID: <CAHurYRWg4hodHm3qoBMEap3E1LKn=vauVk_KmSFyECrUpy=pMg@mail.gmail.com>

2011/09/15 11:22:44 [error] 29330#0: *6 no mp4 stco atoms were found in
"/usr/local/opt/nginx_new/html/video/v.mp4", client: 10.10.50.130, server:
localhost, request: "GET /video/v.mp4?start=43.4 HTTP/1.1"




2011/9/15 Igor Sysoev <igor at sysoev.ru>

> On Mon, Sep 12, 2011 at 03:20:28PM -0400, meto wrote:
> > Can't wait to see that feature in mainline (dev or stable). It's one
> > that was bugging me about the license used by current one (commercial
> > use).
>
> Changes with nginx 1.1.3                                         14 Sep
> 2011
>
>    *) Feature: the module ngx_http_mp4_module.
>
> Or you can use the following patch with nginx-1.0:
> http://nginx.org/patches/patch.mp4.3.txt
>
> Documentation is here:
> http://nginx.org/en/docs/http/ngx_http_mp4_module.html
>
> > Are you planning to include it in your repo packages?
>
> Yes.
>
>
> --
> Igor Sysoev
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110915/28a643b6/attachment.html>

From nginx-forum at nginx.us  Thu Sep 15 03:46:53 2011
From: nginx-forum at nginx.us (cfsego)
Date: Wed, 14 Sep 2011 23:46:53 -0400
Subject: Nginx + Tomcat7 for single webapp and multiple alias domain
In-Reply-To: <3a1f887cd0dc159f1671f69825a3dd72.NginxMailingListEnglish@forum.nginx.org>
References: <3a1f887cd0dc159f1671f69825a3dd72.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <8cec3ba96c769fc5cd3cf088ee04b45a.NginxMailingListEnglish@forum.nginx.org>

Yes, It is good for you to NGINX in front of your application server,
for that you will gain extra features, ie. load-control and DoS
tolerance for your stability.

And it is simple for you to configure NGINX to use tomcat as backend.
When your tomcat listens localhost:8080, the nginx configuration may be
like this:

>>> location / {
>>>     proxy_pass 127.0.0.1:8080;
>>>     proxy_set_header        Host $host;
>>>     proxy_set_header        X-Real-IP $remote_addr;
>>>     proxy_set_header        X-Forwarded-For
$proxy_add_x_forwarded_for;
>>> }

so you will get the clients' real IP from the HTTP header "X-Real-IP"

hope this helps

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215189,215225#msg-215225


From igor at sysoev.ru  Thu Sep 15 04:25:11 2011
From: igor at sysoev.ru (Igor Sysoev)
Date: Thu, 15 Sep 2011 08:25:11 +0400
Subject: mp4 module
In-Reply-To: <CAHurYRWg4hodHm3qoBMEap3E1LKn=vauVk_KmSFyECrUpy=pMg@mail.gmail.com>
References: <20110909130812.GI97288@nginx.com>
	<b66d2a7e6c245d9d7fd455f0216b7ba1.NginxMailingListEnglish@forum.nginx.org>
	<20110914184150.GC63007@nginx.com>
	<CAHurYRWg4hodHm3qoBMEap3E1LKn=vauVk_KmSFyECrUpy=pMg@mail.gmail.com>
Message-ID: <20110915042511.GB73767@nginx.com>

On Thu, Sep 15, 2011 at 11:25:36AM +0800, C.F wrote:
> 2011/09/15 11:22:44 [error] 29330#0: *6 no mp4 stco atoms were found in
> "/usr/local/opt/nginx_new/html/video/v.mp4", client: 10.10.50.130, server:
> localhost, request: "GET /video/v.mp4?start=43.4 HTTP/1.1"

Could you create debug log of the request ?
http://nginx.org/en/docs/debugging_log.html

Is this file larger than 4G ?


-- 
Igor Sysoev


From webwindy at gmail.com  Thu Sep 15 04:39:49 2011
From: webwindy at gmail.com (C.F)
Date: Thu, 15 Sep 2011 12:39:49 +0800
Subject: mp4 module
In-Reply-To: <20110915042511.GB73767@nginx.com>
References: <20110909130812.GI97288@nginx.com>
	<b66d2a7e6c245d9d7fd455f0216b7ba1.NginxMailingListEnglish@forum.nginx.org>
	<20110914184150.GC63007@nginx.com>
	<CAHurYRWg4hodHm3qoBMEap3E1LKn=vauVk_KmSFyECrUpy=pMg@mail.gmail.com>
	<20110915042511.GB73767@nginx.com>
Message-ID: <CAHurYRWX7KwkuvooSaw09UkecJT6BSM9w32xs4VYNG3U_SxsFQ@mail.gmail.com>

-rw-r--r-- 1 root root  39493113 09-15 11:20 v.mp4

debug_log

2011/09/15 12:36:58 [debug] 31923#0: *2 http mp4 filename:
"/usr/local/opt/nginx_n1/html/video/v.mp4"
2011/09/15 12:36:58 [debug] 31923#0: *2 add cleanup: 000000001D7A10E0
2011/09/15 12:36:58 [debug] 31923#0: *2 malloc: 000000001D799D00:5752
2011/09/15 12:36:58 [debug] 31923#0: *2 mp4 start:21700
2011/09/15 12:36:58 [debug] 31923#0: *2 malloc: 000000001D7EC230:1048576
2011/09/15 12:36:58 [debug] 31923#0: *2 read: 10, 000000001D7EC230, 1048576,
0
2011/09/15 12:36:58 [debug] 31923#0: *2 mp4 atom: ftyp @0:24
2011/09/15 12:36:58 [debug] 31923#0: *2 mp4 ftyp atom
2011/09/15 12:36:58 [debug] 31923#0: *2 mp4 atom: moov @24:31577
2011/09/15 12:36:58 [debug] 31923#0: *2 mp4 moov atom
2011/09/15 12:36:58 [debug] 31923#0: *2 mp4 atom: mvhd @32:108
2011/09/15 12:36:58 [debug] 31923#0: *2 mp4 mvhd atom
2011/09/15 12:36:58 [debug] 31923#0: *2 mvhd timescale:1000,
duration:104977, time:104.977s
2011/09/15 12:36:58 [debug] 31923#0: *2 mvhd new duration:83277,
time:83.277s
2011/09/15 12:36:58 [debug] 31923#0: *2 mp4 atom: trak @140:19371
2011/09/15 12:36:58 [debug] 31923#0: *2 mp4 trak atom
2011/09/15 12:36:58 [debug] 31923#0: *2 mp4 atom: tkhd @148:92
2011/09/15 12:36:58 [debug] 31923#0: *2 mp4 tkhd atom
2011/09/15 12:36:58 [debug] 31923#0: *2 tkhd duration:104977, time:104.977s
2011/09/15 12:36:58 [debug] 31923#0: *2 tkhd new duration:83277,
time:83.277s
2011/09/15 12:36:58 [debug] 31923#0: *2 mp4 atom: mdia @240:19271
2011/09/15 12:36:58 [debug] 31923#0: *2 process mdia atom
2011/09/15 12:36:58 [debug] 31923#0: *2 mp4 atom: mdhd @248:32
2011/09/15 12:36:58 [debug] 31923#0: *2 mp4 mdhd atom
2011/09/15 12:36:58 [debug] 31923#0: *2 mdhd timescale:44100,
duration:4629504, time:104.977s
2011/09/15 12:36:58 [debug] 31923#0: *2 mdhd new duration:3672534,
time:83.277s
2011/09/15 12:36:58 [debug] 31923#0: *2 mp4 atom: hdlr @280:44
2011/09/15 12:36:58 [debug] 31923#0: *2 mp4 hdlr atom
2011/09/15 12:36:58 [debug] 31923#0: *2 mp4 atom: minf @324:19187
2011/09/15 12:36:58 [debug] 31923#0: *2 process minf atom
2011/09/15 12:36:58 [debug] 31923#0: *2 mp4 atom: smhd @332:16
2011/09/15 12:36:58 [debug] 31923#0: *2 mp4 smhd atom
2011/09/15 12:36:58 [debug] 31923#0: *2 mp4 atom: dinf @348:36
2011/09/15 12:36:58 [debug] 31923#0: *2 mp4 dinf atom
2011/09/15 12:36:58 [debug] 31923#0: *2 mp4 atom: stbl @384:19127
2011/09/15 12:36:58 [debug] 31923#0: *2 process stbl atom
2011/09/15 12:36:58 [debug] 31923#0: *2 mp4 atom: stsd @392:91
2011/09/15 12:36:58 [debug] 31923#0: *2 mp4 stsd atom
2011/09/15 12:36:58 [debug] 31923#0: *2 stsd entries:1, media:mp4a
2011/09/15 12:36:58 [debug] 31923#0: *2 mp4 atom: stts @483:32
2011/09/15 12:36:58 [debug] 31923#0: *2 mp4 stts atom
2011/09/15 12:36:58 [debug] 31923#0: *2 mp4 time-to-sample entries:2
2011/09/15 12:36:58 [debug] 31923#0: *2 mp4 atom: stsz @515:18104
2011/09/15 12:36:58 [debug] 31923#0: *2 mp4 stsz atom
2011/09/15 12:36:58 [debug] 31923#0: *2 sample uniform size:0, entries:4521
2011/09/15 12:36:58 [debug] 31923#0: *2 mp4 atom: stsc @18619:52
2011/09/15 12:36:58 [debug] 31923#0: *2 mp4 stsc atom
2011/09/15 12:36:58 [debug] 31923#0: *2 sample-to-chunk entries:3
2011/09/15 12:36:58 [debug] 31923#0: *2 mp4 atom: co64 @18671:840
2011/09/15 12:36:58 [debug] 31923#0: *2 mp4 trak atom: 0
2011/09/15 12:36:58 [debug] 31923#0: *2 mp4 atom: trak @19511:12090
2011/09/15 12:36:58 [debug] 31923#0: *2 mp4 trak atom
2011/09/15 12:36:58 [debug] 31923#0: *2 mp4 atom: tkhd @19519:92
2011/09/15 12:36:58 [debug] 31923#0: *2 mp4 tkhd atom
2011/09/15 12:36:58 [debug] 31923#0: *2 tkhd duration:104692, time:104.692s
2011/09/15 12:36:58 [debug] 31923#0: *2 tkhd new duration:82992,
time:82.992s
2011/09/15 12:36:58 [debug] 31923#0: *2 mp4 atom: edts @19611:48
2011/09/15 12:36:58 [debug] 31923#0: *2 mp4 atom: mdia @19659:11942
2011/09/15 12:36:58 [debug] 31923#0: *2 process mdia atom
2011/09/15 12:36:58 [debug] 31923#0: *2 mp4 atom: mdhd @19667:32
2011/09/15 12:36:58 [debug] 31923#0: *2 mp4 mdhd atom
2011/09/15 12:36:58 [debug] 31923#0: *2 mdhd timescale:90000,
duration:9422237, time:104.692s
2011/09/15 12:36:58 [debug] 31923#0: *2 mdhd new duration:7469237,
time:82.992s
2011/09/15 12:36:58 [debug] 31923#0: *2 mp4 atom: hdlr @19699:44
2011/09/15 12:36:58 [debug] 31923#0: *2 mp4 hdlr atom
2011/09/15 12:36:58 [debug] 31923#0: *2 mp4 atom: minf @19743:11858
2011/09/15 12:36:58 [debug] 31923#0: *2 process minf atom
2011/09/15 12:36:58 [debug] 31923#0: *2 mp4 atom: vmhd @19751:20
2011/09/15 12:36:58 [debug] 31923#0: *2 mp4 vmhd atom
2011/09/15 12:36:58 [debug] 31923#0: *2 mp4 atom: dinf @19771:36
2011/09/15 12:36:58 [debug] 31923#0: *2 mp4 dinf atom
2011/09/15 12:36:58 [debug] 31923#0: *2 mp4 atom: stbl @19807:11794
2011/09/15 12:36:58 [debug] 31923#0: *2 process stbl atom
2011/09/15 12:36:58 [debug] 31923#0: *2 mp4 atom: stsd @19815:150
2011/09/15 12:36:58 [debug] 31923#0: *2 mp4 stsd atom
2011/09/15 12:36:58 [debug] 31923#0: *2 stsd entries:1, media:avc1
2011/09/15 12:36:58 [debug] 31923#0: *2 mp4 atom: stts @19965:6416
2011/09/15 12:36:58 [debug] 31923#0: *2 mp4 stts atom
2011/09/15 12:36:58 [debug] 31923#0: *2 mp4 time-to-sample entries:800
2011/09/15 12:36:58 [debug] 31923#0: *2 mp4 atom: stss @26381:152
2011/09/15 12:36:58 [debug] 31923#0: *2 mp4 stss atom
2011/09/15 12:36:58 [debug] 31923#0: *2 sync sample entries:34
2011/09/15 12:36:58 [debug] 31923#0: *2 mp4 atom: stsz @26533:4208
2011/09/15 12:36:58 [debug] 31923#0: *2 mp4 stsz atom
2011/09/15 12:36:58 [debug] 31923#0: *2 sample uniform size:0, entries:1047
2011/09/15 12:36:58 [debug] 31923#0: *2 mp4 atom: stsc @30741:76
2011/09/15 12:36:58 [debug] 31923#0: *2 mp4 stsc atom
2011/09/15 12:36:58 [debug] 31923#0: *2 sample-to-chunk entries:5
2011/09/15 12:36:58 [debug] 31923#0: *2 mp4 atom: co64 @30817:784
2011/09/15 12:36:58 [debug] 31923#0: *2 mp4 trak atom: 0
2011/09/15 12:36:58 [debug] 31923#0: *2 mp4 moov atom done
2011/09/15 12:36:58 [debug] 31923#0: *2 mp4 atom: free @31601:778423
2011/09/15 12:36:58 [debug] 31923#0: *2 mp4 atom: mdat @810024:38683089
2011/09/15 12:36:58 [debug] 31923#0: *2 mp4 mdat atom
2011/09/15 12:36:58 [debug] 31923#0: *2 mp4 stts atom update
2011/09/15 12:36:58 [debug] 31923#0: *2 time-to-sample start_time:956970
2011/09/15 12:36:58 [debug] 31923#0: *2 count:1, duration:1024
2011/09/15 12:36:58 [debug] 31923#0: *2 count:4520, duration:1024
2011/09/15 12:36:58 [debug] 31923#0: *2 start_sample:934, new count:3586
2011/09/15 12:36:58 [debug] 31923#0: *2 mp4 stss atom update
2011/09/15 12:36:58 [debug] 31923#0: *2 mp4 ctts atom update
2011/09/15 12:36:58 [debug] 31923#0: *2 mp4 stsc atom update
2011/09/15 12:36:58 [debug] 31923#0: *2 start_sample:934, chunk:1, chunks:1,
samples:46, id:1
2011/09/15 12:36:58 [debug] 31923#0: *2 start_sample:888, chunk:2,
chunks:101, samples:44, id:1
2011/09/15 12:36:58 [debug] 31923#0: *2 start chunk:21, samples:8
2011/09/15 12:36:58 [debug] 31923#0: *2 mp4 stsz atom update
2011/09/15 12:36:58 [debug] 31923#0: *2 chunk samples sizes:1489
2011/09/15 12:36:58 [debug] 31923#0: *2 mp4 stco atom update
2011/09/15 12:36:58 [error] 31923#0: *2 no mp4 stco atoms were found in
"/usr/local/opt/nginx_n1/html/video/v.mp4", client: 10.10.50.130, server:
localhost, request: "GET /video/v.mp4?start=21.7 HTTP/1.1", host: "
192.168.51.31:3080"
2011/09/15 12:36:58 [debug] 31923#0: *2 free: 000000001D7EC230
2011/09/15 12:36:58 [debug] 31923#0: *2 free: 000000001D799D00
2011/09/15 12:36:58 [debug] 31923#0: *2 http finalize request: 500,
"/video/v.mp4?start=21.7" a:1, c:1
2011/09/15 12:36:58 [debug] 31923#0: *2 http special response: 500,
"/video/v.mp4?start=21.7"
2011/09/15 12:36:58 [debug] 31923#0: *2 internal redirect: "/50x.html?"
2011/09/15 12:36:58 [debug] 31923#0: *2 rewrite phase: 0
2011/09/15 12:36:58 [debug] 31923#0: *2 test location: "/"
2011/09/15 12:36:58 [debug] 31923#0: *2 test location: "video/"
2011/09/15 12:36:58 [debug] 31923#0: *2 test location: "50x.html"
2011/09/15 12:36:58 [debug] 31923#0: *2 using configuration "=/50x.html"
2011/09/15 12:36:58 [debug] 31923#0: *2 http cl:-1 max:1048576
2011/09/15 12:36:58 [debug] 31923#0: *2 rewrite phase: 2
2011/09/15 12:36:58 [debug] 31923#0: *2 post rewrite phase: 3
2011/09/15 12:36:58 [debug] 31923#0: *2 generic phase: 4
2011/09/15 12:36:58 [debug] 31923#0: *2 generic phase: 5
2011/09/15 12:36:58 [debug] 31923#0: *2 access phase: 6
2011/09/15 12:36:58 [debug] 31923#0: *2 access phase: 7
2011/09/15 12:36:58 [debug] 31923#0: *2 post access phase: 8
2011/09/15 12:36:58 [debug] 31923#0: *2 content phase: 9
2011/09/15 12:36:58 [debug] 31923#0: *2 content phase: 10
2011/09/15 12:36:58 [debug] 31923#0: *2 content phase: 11
2011/09/15 12:36:58 [debug] 31923#0: *2 http filename:
"/usr/local/opt/nginx_n1/html/50x.html"
2011/09/15 12:36:58 [debug] 31923#0: *2 add cleanup: 000000001D7A1170
2011/09/15 12:36:58 [debug] 31923#0: *2 http static fd: 11
2011/09/15 12:36:58 [debug] 31923#0: *2 http set discard body
2011/09/15 12:36:58 [debug] 31923#0: *2 HTTP/1.1 500 Internal Server Error
Server: nginx/1.1.3
Date: Thu, 15 Sep 2011 04:36:58 GMT
Content-Type: text/html
Content-Length: 383
Connection: close

2011/09/15 12:36:58 [debug] 31923#0: *2 write new buf t:1 f:0
000000001D7A1308, pos 000000001D7A1308, size: 161 file: 0, size: 0
2011/09/15 12:36:58 [debug] 31923#0: *2 http write filter: l:0 f:0 s:161
2011/09/15 12:36:58 [debug] 31923#0: *2 http output filter "/50x.html?"
2011/09/15 12:36:58 [debug] 31923#0: *2 http copy filter: "/50x.html?"
2011/09/15 12:36:58 [debug] 31923#0: *2 http postpone filter "/50x.html?"
00007FFF6568DA30
2011/09/15 12:36:58 [debug] 31923#0: *2 write old buf t:1 f:0
000000001D7A1308, pos 000000001D7A1308, size: 161 file: 0, size: 0
2011/09/15 12:36:58 [debug] 31923#0: *2 write new buf t:0 f:1
0000000000000000, pos 0000000000000000, size: 0 file: 0, size: 383
2011/09/15 12:36:58 [debug] 31923#0: *2 http write filter: l:1 f:0 s:544
2011/09/15 12:36:58 [debug] 31923#0: *2 http write filter limit 0
2011/09/15 12:36:58 [debug] 31923#0: *2 writev: 161
2011/09/15 12:36:58 [debug] 31923#0: *2 sendfile: @0 383
2011/09/15 12:36:58 [debug] 31923#0: *2 sendfile: 383, @0 383:383
2011/09/15 12:36:58 [debug] 31923#0: *2 http write filter 0000000000000000
2011/09/15 12:36:58 [debug] 31923#0: *2 http copy filter: 0 "/50x.html?"
2011/09/15 12:36:58 [debug] 31923#0: *2 http finalize request: 0,
"/50x.html?" a:1, c:2
2011/09/15 12:36:58 [debug] 31923#0: *2 http request count:2 blk:0
2011/09/15 12:36:58 [debug] 31923#0: *2 http finalize request: -4,
"/50x.html?" a:1, c:1
2011/09/15 12:36:58 [debug] 31923#0: *2 event timer add: 3:
5000:1316061423751
2011/09/15 12:36:58 [debug] 31923#0: *2 http lingering close handler
2011/09/15 12:36:58 [debug] 31923#0: *2 recv: fd:3 -1 of 4096
2011/09/15 12:36:58 [debug] 31923#0: *2 recv() not ready (11: Resource
temporarily unavailable)
2011/09/15 12:36:58 [debug] 31923#0: *2 lingering read: -2
2011/09/15 12:36:58 [debug] 31923#0: *2 event timer: 3, old: 1316061423751,
new: 1316061423751
2011/09/15 12:36:58 [debug] 31923#0: *2 http lingering close handler
2011/09/15 12:36:58 [debug] 31923#0: *2 recv: fd:3 0 of 4096
2011/09/15 12:36:58 [debug] 31923#0: *2 lingering read: 0
2011/09/15 12:36:58 [debug] 31923#0: *2 http request count:1 blk:0
2011/09/15 12:36:58 [debug] 31923#0: *2 http close request
2011/09/15 12:36:58 [debug] 31923#0: *2 http log handler
2011/09/15 12:36:58 [debug] 31923#0: *2 run cleanup: 000000001D7A1170
2011/09/15 12:36:58 [debug] 31923#0: *2 file cleanup: fd:11
2011/09/15 12:36:58 [debug] 31923#0: *2 run cleanup: 000000001D7A10E0
2011/09/15 12:36:58 [debug] 31923#0: *2 file cleanup: fd:10
2011/09/15 12:36:58 [debug] 31923#0: *2 free: 0000000000000000
2011/09/15 12:36:58 [debug] 31923#0: *2 free: 0000000000000000
2011/09/15 12:36:58 [debug] 31923#0: *2 free: 000000001D7A06E0, unused: 480
2011/09/15 12:36:58 [debug] 31923#0: *2 close http connection: 3
2011/09/15 12:36:58 [debug] 31923#0: *2 event timer del: 3: 1316061423751
2011/09/15 12:36:58 [debug] 31923#0: *2 reusable connection: 0
2011/09/15 12:36:58 [debug] 31923#0: *2 free: 000000001D7A02D0
2011/09/15 12:36:58 [debug] 31923#0: *2 free: 000000001D79FDE0
2011/09/15 12:36:58 [debug] 31923#0: *2 free: 000000001D7AA640, unused: 8
2011/09/15 12:36:58 [debug] 31923#0: *2 free: 000000001D795390, unused: 128
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110915/775bd87a/attachment-0001.html>

From igor at sysoev.ru  Thu Sep 15 04:53:54 2011
From: igor at sysoev.ru (Igor Sysoev)
Date: Thu, 15 Sep 2011 08:53:54 +0400
Subject: mp4 module
In-Reply-To: <CAHurYRWX7KwkuvooSaw09UkecJT6BSM9w32xs4VYNG3U_SxsFQ@mail.gmail.com>
References: <20110909130812.GI97288@nginx.com>
	<b66d2a7e6c245d9d7fd455f0216b7ba1.NginxMailingListEnglish@forum.nginx.org>
	<20110914184150.GC63007@nginx.com>
	<CAHurYRWg4hodHm3qoBMEap3E1LKn=vauVk_KmSFyECrUpy=pMg@mail.gmail.com>
	<20110915042511.GB73767@nginx.com>
	<CAHurYRWX7KwkuvooSaw09UkecJT6BSM9w32xs4VYNG3U_SxsFQ@mail.gmail.com>
Message-ID: <20110915045354.GD73767@nginx.com>

On Thu, Sep 15, 2011 at 12:39:49PM +0800, C.F wrote:
> -rw-r--r-- 1 root root  39493113 09-15 11:20 v.mp4

> 2011/09/15 12:36:58 [debug] 31923#0: *2 mp4 atom: co64 @30817:784

Thank you. As I supposed it has 64-bit co64 atom, however, it's strange
that this atom was used for so small size file - 39M.
I will make patch today.


-- 
Igor Sysoev


From igor at sysoev.ru  Thu Sep 15 05:26:02 2011
From: igor at sysoev.ru (Igor Sysoev)
Date: Thu, 15 Sep 2011 09:26:02 +0400
Subject: mp4 module
In-Reply-To: <20110915045354.GD73767@nginx.com>
References: <20110909130812.GI97288@nginx.com>
	<b66d2a7e6c245d9d7fd455f0216b7ba1.NginxMailingListEnglish@forum.nginx.org>
	<20110914184150.GC63007@nginx.com>
	<CAHurYRWg4hodHm3qoBMEap3E1LKn=vauVk_KmSFyECrUpy=pMg@mail.gmail.com>
	<20110915042511.GB73767@nginx.com>
	<CAHurYRWX7KwkuvooSaw09UkecJT6BSM9w32xs4VYNG3U_SxsFQ@mail.gmail.com>
	<20110915045354.GD73767@nginx.com>
Message-ID: <20110915052602.GE73767@nginx.com>

On Thu, Sep 15, 2011 at 08:53:54AM +0400, Igor Sysoev wrote:
> On Thu, Sep 15, 2011 at 12:39:49PM +0800, C.F wrote:
> > -rw-r--r-- 1 root root  39493113 09-15 11:20 v.mp4
> 
> > 2011/09/15 12:36:58 [debug] 31923#0: *2 mp4 atom: co64 @30817:784
> 
> Thank you. As I supposed it has 64-bit co64 atom, however, it's strange
> that this atom was used for so small size file - 39M.
> I will make patch today.

Patch for nginx-1.1.3.


-- 
Igor Sysoev
-------------- next part --------------
Index: src/http/modules/ngx_http_mp4_module.c
===================================================================
--- src/http/modules/ngx_http_mp4_module.c	(revision 4105)
+++ src/http/modules/ngx_http_mp4_module.c	(working copy)
@@ -32,8 +32,10 @@
 #define NGX_HTTP_MP4_STSZ_DATA    21
 #define NGX_HTTP_MP4_STCO_ATOM    22
 #define NGX_HTTP_MP4_STCO_DATA    23
+#define NGX_HTTP_MP4_CO64_ATOM    24
+#define NGX_HTTP_MP4_CO64_DATA    25
 
-#define NGX_HTTP_MP4_LAST_ATOM    NGX_HTTP_MP4_STCO_DATA
+#define NGX_HTTP_MP4_LAST_ATOM    NGX_HTTP_MP4_CO64_DATA
 
 
 typedef struct {
@@ -61,7 +63,7 @@
     ngx_uint_t            start_sample;
     ngx_uint_t            start_chunk;
     ngx_uint_t            chunk_samples;
-    ngx_uint_t            chunk_samples_size;
+    uint64_t              chunk_samples_size;
     off_t                 start_offset;
 
     size_t                tkhd_size;
@@ -96,8 +98,10 @@
     ngx_buf_t             stsc_data_buf;
     ngx_buf_t             stsz_atom_buf;
     ngx_buf_t             stsz_data_buf;
-    ngx_buf_t             tsco_atom_buf;
-    ngx_buf_t             tsco_data_buf;
+    ngx_buf_t             stco_atom_buf;
+    ngx_buf_t             stco_data_buf;
+    ngx_buf_t             co64_atom_buf;
+    ngx_buf_t             co64_data_buf;
 
     ngx_mp4_stsc_entry_t  stsc_chunk_entry;
 } ngx_http_mp4_trak_t;
@@ -268,6 +272,12 @@
     ngx_http_mp4_trak_t *trak);
 static void ngx_http_mp4_adjust_stco_atom(ngx_http_mp4_file_t *mp4,
     ngx_http_mp4_trak_t *trak, int32_t adjustment);
+static ngx_int_t ngx_http_mp4_read_co64_atom(ngx_http_mp4_file_t *mp4,
+    uint64_t atom_data_size);
+static ngx_int_t ngx_http_mp4_update_co64_atom(ngx_http_mp4_file_t *mp4,
+    ngx_http_mp4_trak_t *trak);
+static void ngx_http_mp4_adjust_co64_atom(ngx_http_mp4_file_t *mp4,
+    ngx_http_mp4_trak_t *trak, off_t adjustment);
 static char *ngx_http_mp4(ngx_conf_t *cf, ngx_command_t *cmd, void *conf);
 static void *ngx_http_mp4_create_conf(ngx_conf_t *cf);
 static char *ngx_http_mp4_merge_conf(ngx_conf_t *cf, void *parent, void *child);
@@ -373,6 +383,7 @@
     { "stsc", ngx_http_mp4_read_stsc_atom },
     { "stsz", ngx_http_mp4_read_stsz_atom },
     { "stco", ngx_http_mp4_read_stco_atom },
+    { "co64", ngx_http_mp4_read_co64_atom },
     { NULL, NULL }
 };
 
@@ -676,8 +687,15 @@
 
         ngx_http_mp4_update_stsz_atom(mp4, &trak[i]);
 
-        if (ngx_http_mp4_update_stco_atom(mp4, &trak[i]) != NGX_OK) {
-            return NGX_ERROR;
+        if (trak[i].out[NGX_HTTP_MP4_CO64_DATA].buf) {
+            if (ngx_http_mp4_update_co64_atom(mp4, &trak[i]) != NGX_OK) {
+                return NGX_ERROR;
+            }
+
+        } else {
+            if (ngx_http_mp4_update_stco_atom(mp4, &trak[i]) != NGX_OK) {
+                return NGX_ERROR;
+            }
         }
 
         ngx_http_mp4_update_stbl_atom(mp4, &trak[i]);
@@ -721,7 +739,11 @@
                    "mp4 adjustment:%D", adjustment);
 
     for (i = 0; i < mp4->trak.nelts; i++) {
-        ngx_http_mp4_adjust_stco_atom(mp4, &trak[i], (int32_t) adjustment);
+        if (trak[i].out[NGX_HTTP_MP4_CO64_DATA].buf) {
+            ngx_http_mp4_adjust_co64_atom(mp4, &trak[i], adjustment);
+        } else {
+            ngx_http_mp4_adjust_stco_atom(mp4, &trak[i], (int32_t) adjustment);
+        }
     }
 
     return NGX_OK;
@@ -1702,7 +1724,6 @@
 {
     u_char               *atom_header, *atom_table;
     size_t                atom_size;
-    uint32_t              entries;
     ngx_buf_t            *atom;
     ngx_mp4_stsd_atom_t  *stsd_atom;
     ngx_http_mp4_trak_t  *trak;
@@ -1718,10 +1739,9 @@
     ngx_mp4_set_32value(stsd_atom->size, atom_size);
     ngx_mp4_set_atom_name(stsd_atom, 's', 't', 's', 'd');
 
-    entries = ngx_mp4_get_32value(stsd_atom->entries);
-
     if ((uint64_t) (sizeof(ngx_mp4_stsd_atom_t) - sizeof(ngx_mp4_atom_header_t))
-         > atom_data_size) {
+         > atom_data_size)
+    {
         ngx_log_error(NGX_LOG_ERR, mp4->file.log, 0,
                       "\"%s\" mp4 stsd atom too large",
                       mp4->file.name.data);
@@ -2497,7 +2517,7 @@
         }
 
         ngx_log_debug1(NGX_LOG_DEBUG_HTTP, mp4->file.log, 0,
-                       "chunk samples sizes:%uD", trak->chunk_samples_size);
+                       "chunk samples sizes:%uL", trak->chunk_samples_size);
 
         atom_size = sizeof(ngx_mp4_stsz_atom_t) + (data->last - data->pos);
         trak->size += atom_size;
@@ -2554,12 +2574,12 @@
     trak = ngx_mp4_last_trak(mp4);
     trak->chunks = entries;
 
-    atom = &trak->tsco_atom_buf;
+    atom = &trak->stco_atom_buf;
     atom->temporary = 1;
     atom->pos = atom_header;
     atom->last = atom_table;
 
-    data = &trak->tsco_data_buf;
+    data = &trak->stco_data_buf;
     data->temporary = 1;
     data->pos = atom_table;
     data->last = atom_end;
@@ -2648,6 +2668,142 @@
 }
 
 
+typedef struct {
+    u_char    size[4];
+    u_char    name[4];
+    u_char    version[1];
+    u_char    flags[3];
+    u_char    entries[4];
+} ngx_mp4_co64_atom_t;
+
+
+static ngx_int_t
+ngx_http_mp4_read_co64_atom(ngx_http_mp4_file_t *mp4, uint64_t atom_data_size)
+{
+    u_char               *atom_header, *atom_table, *atom_end;
+    uint32_t              entries;
+    ngx_buf_t            *atom, *data;
+    ngx_mp4_co64_atom_t  *co64_atom;
+    ngx_http_mp4_trak_t  *trak;
+
+    /* chunk offsets atom */
+
+    ngx_log_debug0(NGX_LOG_DEBUG_HTTP, mp4->file.log, 0, "mp4 co64 atom");
+
+    atom_header = ngx_mp4_atom_header(mp4);
+    co64_atom = (ngx_mp4_co64_atom_t *) atom_header;
+    ngx_mp4_set_atom_name(co64_atom, 'c', 'o', '6', '4');
+
+    entries = ngx_mp4_get_32value(co64_atom->entries);
+
+    ngx_log_debug1(NGX_LOG_DEBUG_HTTP, mp4->file.log, 0, "chunks:%uD", entries);
+
+    atom_table = atom_header + sizeof(ngx_mp4_co64_atom_t);
+    atom_end = atom_table + entries * sizeof(uint64_t);
+
+    if ((uint64_t) (atom_end - co64_atom->version) > atom_data_size) {
+        ngx_log_error(NGX_LOG_ERR, mp4->file.log, 0,
+                      "\"%s\" mp4 co64 atom too large", mp4->file.name.data);
+        return NGX_ERROR;
+    }
+
+    trak = ngx_mp4_last_trak(mp4);
+    trak->chunks = entries;
+
+    atom = &trak->co64_atom_buf;
+    atom->temporary = 1;
+    atom->pos = atom_header;
+    atom->last = atom_table;
+
+    data = &trak->co64_data_buf;
+    data->temporary = 1;
+    data->pos = atom_table;
+    data->last = atom_end;
+
+    trak->out[NGX_HTTP_MP4_CO64_ATOM].buf = atom;
+    trak->out[NGX_HTTP_MP4_CO64_DATA].buf = data;
+
+    ngx_mp4_atom_next(mp4, atom_data_size);
+
+    return NGX_OK;
+}
+
+
+static ngx_int_t
+ngx_http_mp4_update_co64_atom(ngx_http_mp4_file_t *mp4,
+    ngx_http_mp4_trak_t *trak)
+{
+    size_t                atom_size;
+    ngx_buf_t            *atom, *data;
+    ngx_mp4_co64_atom_t  *co64_atom;
+
+    /*
+     * mdia.minf.stbl.co64 updating requires trak->start_chunk
+     * from mdia.minf.stbl.stsc which depends on value from mdia.mdhd
+     * atom which may reside after mdia.minf
+     */
+
+    ngx_log_debug0(NGX_LOG_DEBUG_HTTP, mp4->file.log, 0,
+                   "mp4 co64 atom update");
+
+    data = trak->out[NGX_HTTP_MP4_CO64_DATA].buf;
+
+    if (data == NULL) {
+        ngx_log_error(NGX_LOG_ERR, mp4->file.log, 0,
+                      "no mp4 co64 atoms were found in \"%s\"",
+                      mp4->file.name.data);
+        return NGX_ERROR;
+    }
+
+    data->pos += trak->start_chunk * sizeof(uint64_t);
+    atom_size = sizeof(ngx_mp4_co64_atom_t) + (data->last - data->pos);
+    trak->size += atom_size;
+
+    trak->start_offset = ngx_mp4_get_64value(data->pos);
+    trak->start_offset += trak->chunk_samples_size;
+    ngx_mp4_set_64value(data->pos, trak->start_offset);
+
+    ngx_log_debug1(NGX_LOG_DEBUG_HTTP, mp4->file.log, 0,
+                   "start chunk offset:%uL", trak->start_offset);
+
+    atom = trak->out[NGX_HTTP_MP4_CO64_ATOM].buf;
+    co64_atom = (ngx_mp4_co64_atom_t *) atom->pos;
+
+    ngx_mp4_set_32value(co64_atom->size, atom_size);
+    ngx_mp4_set_32value(co64_atom->entries, trak->chunks - trak->start_chunk);
+
+    return NGX_OK;
+}
+
+
+static void
+ngx_http_mp4_adjust_co64_atom(ngx_http_mp4_file_t *mp4,
+    ngx_http_mp4_trak_t *trak, off_t adjustment)
+{
+    uint64_t    offset, *entry, *end;
+    ngx_buf_t  *data;
+
+    /*
+     * moov.trak.mdia.minf.stbl.co64 adjustment requires
+     * minimal start offset of all traks and new moov atom size
+     */
+
+    ngx_log_debug0(NGX_LOG_DEBUG_HTTP, mp4->file.log, 0,
+                   "mp4 co64 atom adjustment");
+
+    data = trak->out[NGX_HTTP_MP4_CO64_DATA].buf;
+    entry = (uint64_t *) data->pos;
+    end = (uint64_t *) data->last;
+
+    while (entry < end) {
+        offset = ngx_mp4_get_64value(entry);
+        offset += adjustment;
+        ngx_mp4_set_64value(entry, offset);
+        entry++;
+    }
+}
+
+
 static char *
 ngx_http_mp4(ngx_conf_t *cf, ngx_command_t *cmd, void *conf)
 {

From nginx-forum at nginx.us  Thu Sep 15 05:43:38 2011
From: nginx-forum at nginx.us (qzpmwo)
Date: Thu, 15 Sep 2011 01:43:38 -0400
Subject: How to force nginx to cache HTTP POST response for few seconds
Message-ID: <2df58b853f7760988237ee14c8d662ef.NginxMailingListEnglish@forum.nginx.org>

Hi everybody,

I'm a newbie with nginx and would like to ask your help. On potentially
very high loaded site I would like to cache response to the HTTP POST
for the few seconds. Works perfectly for GETs, but...

nginx version: nginx/0.8.54

Please see below my config:  

proxy_cache_path /tmp/nginx levels=2:2:2 keys_zone=small:1m inactive=1m
max_size=10m;

server {

log_format main '$remote_addr $host $remote_user [$time_local]
      "$request" $status $body_bytes_sent "$http_referer"
      "$http_user_agent" "$gzip_ratio"';

    access_log /var/log/nginx/test_server.access.log main buffer=32k;
    error_log /var/log/nginx/test_server.error.log info;


    listen       80;

    root   /var/www/html/nginx;
    index  index.html index.htm;

    location / {
        try_files $uri @apache_backend;
    }

###    location ~ \.php$ {
###        proxy_pass http://127.0.0.1:88;
###    }

    location @apache_backend {
        proxy_pass http://127.0.0.1:88;
proxy_cache small;
proxy_cache_valid 5s;
proxy_cache_key $uri;
proxy_cache_use_stale updating;
add_header X-Cached $upstream_cache_status;
    }
}


Any help/advices will be appreciated.

Thanks
--->>> qzpmwo <<<---

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215231,215231#msg-215231


From igor at sysoev.ru  Thu Sep 15 05:49:56 2011
From: igor at sysoev.ru (Igor Sysoev)
Date: Thu, 15 Sep 2011 09:49:56 +0400
Subject: How to force nginx to cache HTTP POST response for few seconds
In-Reply-To: <2df58b853f7760988237ee14c8d662ef.NginxMailingListEnglish@forum.nginx.org>
References: <2df58b853f7760988237ee14c8d662ef.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20110915054956.GF73767@nginx.com>

On Thu, Sep 15, 2011 at 01:43:38AM -0400, qzpmwo wrote:
> Hi everybody,
> 
> I'm a newbie with nginx and would like to ask your help. On potentially
> very high loaded site I would like to cache response to the HTTP POST
> for the few seconds. Works perfectly for GETs, but...
> 
> nginx version: nginx/0.8.54
> 
> Please see below my config:  
> 
> proxy_cache_path /tmp/nginx levels=2:2:2 keys_zone=small:1m inactive=1m
> max_size=10m;
> 
> server {
> 
> log_format main '$remote_addr $host $remote_user [$time_local]
>       "$request" $status $body_bytes_sent "$http_referer"
>       "$http_user_agent" "$gzip_ratio"';
> 
>     access_log /var/log/nginx/test_server.access.log main buffer=32k;
>     error_log /var/log/nginx/test_server.error.log info;
> 
> 
>     listen       80;
> 
>     root   /var/www/html/nginx;
>     index  index.html index.htm;
> 
>     location / {
>         try_files $uri @apache_backend;
>     }
> 
> ###    location ~ \.php$ {
> ###        proxy_pass http://127.0.0.1:88;
> ###    }
> 
>     location @apache_backend {
>         proxy_pass http://127.0.0.1:88;
> proxy_cache small;
> proxy_cache_valid 5s;
> proxy_cache_key $uri;
> proxy_cache_use_stale updating;
> add_header X-Cached $upstream_cache_status;
>     }
> }
> 
> 
> Any help/advices will be appreciated.

You should upgrade to at least 0.8.55.
If POST request body is small enough, then you can use:

    location @apache_backend {
        ...
        proxy_cache_methods   POST;
        proxy_cache_key       "$uri|$request_body";
        client_max_body_size  1k;
    }


-- 
Igor Sysoev


From nginx-forum at nginx.us  Thu Sep 15 05:52:05 2011
From: nginx-forum at nginx.us (qzpmwo)
Date: Thu, 15 Sep 2011 01:52:05 -0400
Subject: How to force nginx to cache HTTP POST response for few seconds
In-Reply-To: <2df58b853f7760988237ee14c8d662ef.NginxMailingListEnglish@forum.nginx.org>
References: <2df58b853f7760988237ee14c8d662ef.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <8931a2474c800d8c46b7d511d52d353c.NginxMailingListEnglish@forum.nginx.org>

Hi everybody,

Sorry for not reading documentation properly. proxy_cache_methods is a
saver!

Thanks a lot, case closed :)

--->>> qzpmwo <<< ---

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215231,215235#msg-215235


From nginx-forum at nginx.us  Thu Sep 15 07:09:48 2011
From: nginx-forum at nginx.us (qzpmwo)
Date: Thu, 15 Sep 2011 03:09:48 -0400
Subject: How to force nginx to cache HTTP POST response for few seconds
In-Reply-To: <20110915054956.GF73767@nginx.com>
References: <20110915054956.GF73767@nginx.com>
Message-ID: <3058acfdb341f2753dc5b23c0f12d8e0.NginxMailingListEnglish@forum.nginx.org>

Hi Igor,

Thanks a lot for so fast answer and for so wonderful product! I will try
to use your trick with request_body, though in my case it is an AMF
binary.

I updated to 1.0.5 and have weird problem now (from error log):

http: file cache: /tmp/nginx-cache 0.000M, bsize: 4096
signal 17 (SIGCHLD) recieved
cache loader process 11737 exited with code 0
signal 29 (SIGIO) recieved

:(

Any advices of how to figure out why is it happening?

Thanks again
--->>> qzpmwo <<<---

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215231,215236#msg-215236


From igor at sysoev.ru  Thu Sep 15 09:28:33 2011
From: igor at sysoev.ru (Igor Sysoev)
Date: Thu, 15 Sep 2011 13:28:33 +0400
Subject: How to force nginx to cache HTTP POST response for few seconds
In-Reply-To: <3058acfdb341f2753dc5b23c0f12d8e0.NginxMailingListEnglish@forum.nginx.org>
References: <20110915054956.GF73767@nginx.com>
	<3058acfdb341f2753dc5b23c0f12d8e0.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20110915092833.GA80146@nginx.com>

On Thu, Sep 15, 2011 at 03:09:48AM -0400, qzpmwo wrote:
> Hi Igor,
> 
> Thanks a lot for so fast answer and for so wonderful product! I will try
> to use your trick with request_body, though in my case it is an AMF
> binary.

How large these binaries ?

> I updated to 1.0.5 and have weird problem now (from error log):
> 
> http: file cache: /tmp/nginx-cache 0.000M, bsize: 4096
> signal 17 (SIGCHLD) recieved
> cache loader process 11737 exited with code 0
> signal 29 (SIGIO) recieved
> 
> :(
> 
> Any advices of how to figure out why is it happening?

This is non-erroneous messages logged at "notice" level.


-- 
Igor Sysoev


From ash2kk at gmail.com  Thu Sep 15 09:30:28 2011
From: ash2kk at gmail.com (Mikhail Mazursky)
Date: Thu, 15 Sep 2011 15:30:28 +0600
Subject: Nginx + Tomcat7 for single webapp and multiple alias domain
In-Reply-To: <3a1f887cd0dc159f1671f69825a3dd72.NginxMailingListEnglish@forum.nginx.org>
References: <3a1f887cd0dc159f1671f69825a3dd72.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <CAGg1-iNFQL4+7kb4QxZTgNRCrErKFEuDeZN3Z-3b7pW5mam3Vw@mail.gmail.com>

2011/9/14 tm1978 <nginx-forum at nginx.us>:
> I am thinking about using NGINX instead of Apache to deliver a Tomcat7
> webapp on Port :80 to the enduser.
>
> I have the following requirements:
>
> 1. There is a single ?web-app in the root with any urls (its own
> internal urlrewriting). It has various URLs pointing to it for several
> languages (and partially other content)
>
> 2. Another admin-webapp exists
>
>
> My questions are:
>
> (a) would it be a good idea to use NGINX a without ?Apache Http server,
> just with Apache Tomcat7 ?
> (b) how can I set this up so that the Tomcat App always know the domain
> name the user was using when calling the page?
> (c) will the tomcat app still be able to know the IPs of every user?

For this you will have to setup a RemoveIpValve [1] in your
context.xml/server.xml/context.xml.default like this:

    <Valve
        className="org.apache.catalina.valves.RemoteIpValve"
        remoteIpHeader="X-Forwarded-For"
        protocolHeader="X-Forwarded-Proto" />

In nginx you may use configuration provided by cfsego but don't forget
to add X-Forwarded-Proto if you use https on nginx. Also you may
remove X-Real-IP header because it's not used in this setup:

>>> location / {
>>>     proxy_pass 127.0.0.1:8080;
>>>     proxy_set_header        Host $host;
>>>     proxy_set_header         $remote_addr;

proxy_set_header        X-Forwarded-Proto $scheme;

>>>     proxy_set_header        X-Forwarded-For $proxy_add_x_forwarded_for;
>>> }


This way Tomcat will be aware of https and will use secure cookies for sessions.

[1]: https://tomcat.apache.org/tomcat-7.0-doc/config/valve.html#Remote_IP_Valve

> Thanks!
>
> Tobias


From nginx-forum at nginx.us  Thu Sep 15 09:33:08 2011
From: nginx-forum at nginx.us (qzpmwo)
Date: Thu, 15 Sep 2011 05:33:08 -0400
Subject: How to force nginx to cache HTTP POST response for few seconds
In-Reply-To: <20110915092833.GA80146@nginx.com>
References: <20110915092833.GA80146@nginx.com>
Message-ID: <1ef7fdb8e7d77fa9a9d5f3934aa586bf.NginxMailingListEnglish@forum.nginx.org>

Hi Igor,


Payload size is small around 120 bytes according to my data. Response is
empty. May be that is why? Tomorrow I will try to recompile nginx with
debug enabled and may be get more data.

Thanks a lot for support.
--->>> qzpmwo <<<---

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215231,215244#msg-215244


From nginx-forum at nginx.us  Thu Sep 15 09:44:27 2011
From: nginx-forum at nginx.us (qzpmwo)
Date: Thu, 15 Sep 2011 05:44:27 -0400
Subject: How to force nginx to cache HTTP POST response for few seconds
In-Reply-To: <1ef7fdb8e7d77fa9a9d5f3934aa586bf.NginxMailingListEnglish@forum.nginx.org>
References: <20110915092833.GA80146@nginx.com>
	<1ef7fdb8e7d77fa9a9d5f3934aa586bf.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <4cd892cb0ebd639ee05f69162a2d098b.NginxMailingListEnglish@forum.nginx.org>

Actually I was wrong. Response "Content-Length" header says 66 bytes, so
response is not empty. If I disable caching of that request, application
works. If it is enabled it stops after first cache HIT. As I mentioned
in previous mail I will try to get more data (tcpdump and nginx with
debug) tomorrow.

Thanks
--->>> qzpmwo <<<---

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215231,215246#msg-215246


From kworthington at gmail.com  Thu Sep 15 10:22:10 2011
From: kworthington at gmail.com (Kevin Worthington)
Date: Thu, 15 Sep 2011 06:22:10 -0400
Subject: nginx-1.1.3
In-Reply-To: <CAPKb8pD+3Z4=MEgwX=6sAqmrGQOMy1DjjoPfLZBij4DLC=y_yg@mail.gmail.com>
References: <20110914151925.GG49199@nginx.com>
	<CAPKb8pD+3Z4=MEgwX=6sAqmrGQOMy1DjjoPfLZBij4DLC=y_yg@mail.gmail.com>
Message-ID: <CAGo79UUFEdLCMDX6yZp0Om9t-TJmFNFcouq9HpogF5Q0ntoWCw@mail.gmail.com>

Hello Nginxers,

I just released Nginx 1.1.3 For Windows http://goo.gl/u9O1a (32-bit and
64-bit)

These versions are to support legacy users who are already using Cygwin
based builds of Nginx. Official Windows binaries are at nginx.org

Thanks,
Kevin
--
Kevin Worthington
kworthington (at] gmail {dot) com


On Wed, Sep 14, 2011 at 11:05 PM, ?? <clanherb at gmail.com> wrote:

> Cool?people who upgrade to Lion can use nginx again
>
> Bugfix: nginx could not be built on MacOSX 10.7.
>
> On Wed, Sep 14, 2011 at 11:19 PM, Igor Sysoev <igor at sysoev.ru> wrote:
> > Changes with nginx 1.1.3                                         14 Sep
> 2011
> >
> >    *) Feature: the module ngx_http_mp4_module.
> >
> >    *) Bugfix: in Linux AIO combined with open_file_cache.
> >
> >    *) Bugfix: open_file_cache did not update file info on retest if file
> >       was not atomically changed.
> >
> >    *) Bugfix: nginx could not be built on MacOSX 10.7.
> >
> >
> > --
> > Igor Sysoev
> >
> > _______________________________________________
> > nginx mailing list
> > nginx at nginx.org
> > http://mailman.nginx.org/mailman/listinfo/nginx
> >
>
>
>
> --
> lei yang
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110915/6cafb129/attachment.html>

From hagaia at qwilt.com  Thu Sep 15 14:21:44 2011
From: hagaia at qwilt.com (Hagai Avrahami)
Date: Thu, 15 Sep 2011 16:21:44 +0200
Subject: Help with AIO
In-Reply-To: <20110906073352.GR1137@mdounin.ru>
References: <CAN-ZXpuTdDMmQ7rgo_QP=1W53+j1qc-Bqa=f4=QLNskOubY2dw@mail.gmail.com>
	<20110906073352.GR1137@mdounin.ru>
Message-ID: <CAN-ZXpsTEVi3Dk0fQAjye1-AAxkZ6HQ9m6UOkS051A=0HVUL1A@mail.gmail.com>

Hi

When benchmarking AIO read, I reached more than 1024 concurrent Async IO -
1. I saw a weird error log
2. Nginx did not fallback to regular blocking read

I looked on the source code --  src/os/unix/ngx_linux_aio_read.c  line:105
and I think  there is a problem in errno handling

// Current Code
//----------------------------------------------------------------------------
    *n = io_submit(ngx_aio_ctx, 1, piocb);

    if (n == 1) {
        ev->active = 1;
        ev->ready = 0;
        ev->complete = 0;

        return NGX_AGAIN;
    }

    n = -n;    <---------------------------------------------------------??

    if (n == NGX_EAGAIN) {
        return ngx_read_file(file, buf, size, offset);
    }

    ngx_log_error(NGX_LOG_CRIT, file->log, n,
                  "io_submit(\"%V\") failed", &file->name);

    if (n == NGX_ENOSYS) {
        ngx_file_aio = 0;
        return ngx_read_file(file, buf, size, offset);
    }

    return NGX_ERROR;*



// I think it should be
//----------------------------------------------------------------------------
    *n = io_submit(ngx_aio_ctx, 1, piocb);

    if (n == 1) {
        ev->active = 1;
        ev->ready = 0;
        ev->complete = 0;

        return NGX_AGAIN;
    }

    if (n == 0)
    {
        ngx_log_error(NGX_LOG_CRIT, file->log, 0,
                  "io_submit(\"%V\") failed - 0 - I/O request blocks",
&file->name);

        return NGX_ERROR;
    }
    else
    {
        n = errno;
    }

* *  if (n== NGX_EAGAIN) {
        return ngx_read_file(file, buf, size, offset);
    }

    ngx_log_error(NGX_LOG_CRIT, file->log, n,
                  "io_submit(\"%V\") failed", &file->name);

    if (n== NGX_ENOSYS) {
        ngx_file_aio = 0;
        return ngx_read_file(file, buf, size, offset);
    }

    return NGX_ERROR;*
//----------------------------------------------------------------------------


your help will be appreciated
Thanks
Hagai

On Tue, Sep 6, 2011 at 9:33 AM, Maxim Dounin <mdounin at mdounin.ru> wrote:

> Hello!
>
> On Tue, Sep 06, 2011 at 01:21:00AM +0300, Hagai Avrahami wrote:
>
> > Hi
> >
> > I configured Nginx with *aio *mode *on *and 1 *worker process*
> > In benchmarking I saw limit of 1024 concurrent sessions even configured
> to
> > 4096 *worker connections*
>
> Sure it's not limitation of the benchmarking tool you use?  Most
> tools out there use select() and hence limited to 1024
> concurrent connections.
>
> > When I looked in the source code I could see that  io_setup is called
> with
> > fix value of 1024 max events
> > can anyone please explain me the hard coded limit
> > Is it OK to change it or it is derived from other limit I am not aware of
> > any help will be appreciated
>
> io_setup() maxevents is a limit on number of aio requests allowed
> to be submitted simulteneously; if nginx hits this limit, it will
> fallback to normal blocking read.
>
> If you are going to change it, make sure to keep total for all
> processes lower than /proc/sys/fs/aio-max-nr.
>
> Maxim Dounin
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx
>



-- 

*Hagai Avrahami*
Qwilt | Work: +972-72-2221644| Mobile: +972-54-4895656 |
hagaia at qwilt.com<yoav at qwilt.com>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110915/bb77057f/attachment.html>

From quan.nexthop at gmail.com  Thu Sep 15 15:02:57 2011
From: quan.nexthop at gmail.com (Geoge.Q)
Date: Thu, 15 Sep 2011 08:02:57 -0700
Subject: nginx and tproxy
Message-ID: <CAP_76MYh14zoQ1NVT8KLPiN0FeS0kdjvNZAJy61tcdkmBm6gyg@mail.gmail.com>

Hi all:


Can we let both work together? Anyone can share his experience with us if we
want to integrate them?

thanks in advance

george
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110915/3e0b9d25/attachment.html>

From r at roze.lv  Thu Sep 15 15:19:48 2011
From: r at roze.lv (Reinis Rozitis)
Date: Thu, 15 Sep 2011 18:19:48 +0300
Subject: nginx and tproxy
In-Reply-To: <CAP_76MYh14zoQ1NVT8KLPiN0FeS0kdjvNZAJy61tcdkmBm6gyg@mail.gmail.com>
References: <CAP_76MYh14zoQ1NVT8KLPiN0FeS0kdjvNZAJy61tcdkmBm6gyg@mail.gmail.com>
Message-ID: <241A0BDFBDDA489996B455F7135B0651@DD21>

By tproxy you mean transparent proxying as in doing proxy requests from nginx while using clients IP?

It is far more simple to use the Real IP module http://wiki.nginx.org/HttpRealIpModule ( pretty much all webservers (apache - 
mod_rpaf / lighttpd - mod_extforward etc)  have such modules in one or another form ) rather than trying to achieve it on TCP level.

And it is also straight forward to implement it in any other backend application.

rr


From: Geoge.Q
Sent: Thursday, September 15, 2011 6:02 PM
To: nginx at nginx.org
Subject: nginx and tproxy

Hi all:


Can we let both work together? Anyone can share his experience with us if we want to integrate them?

thanks in advance

george


_______________________________________________
nginx mailing list
nginx at nginx.org
http://mailman.nginx.org/mailman/listinfo/nginx 


From quan.nexthop at gmail.com  Thu Sep 15 16:21:56 2011
From: quan.nexthop at gmail.com (Geoge.Q)
Date: Thu, 15 Sep 2011 09:21:56 -0700
Subject: nginx and tproxy
In-Reply-To: <241A0BDFBDDA489996B455F7135B0651@DD21>
References: <CAP_76MYh14zoQ1NVT8KLPiN0FeS0kdjvNZAJy61tcdkmBm6gyg@mail.gmail.com>
	<241A0BDFBDDA489996B455F7135B0651@DD21>
Message-ID: <CAP_76Mb1hPZvgfCd=uhTDQaxLE95KM2nOMvUYK8_r6mRHJW7ww@mail.gmail.com>

thanks your feedback.

Yes, we want to get the source-ip directly.
However, if we want to get real client ip from L7 layer, we need modify the
backend configuration, in some case, we have no permission to modify the
configuration of Web-server.

Is there other way to get client IP other than x-forward-for/x-real-ip?

thanks
George


On Thu, Sep 15, 2011 at 8:19 AM, Reinis Rozitis <r at roze.lv> wrote:

> By tproxy you mean transparent proxying as in doing proxy requests from
> nginx while using clients IP?
>
> It is far more simple to use the Real IP module http://wiki.nginx.org/**
> HttpRealIpModule <http://wiki.nginx.org/HttpRealIpModule> ( pretty much
> all webservers (apache - mod_rpaf / lighttpd - mod_extforward etc)  have
> such modules in one or another form ) rather than trying to achieve it on
> TCP level.
>
> And it is also straight forward to implement it in any other backend
> application.
>
> rr
>
>
> From: Geoge.Q
> Sent: Thursday, September 15, 2011 6:02 PM
> To: nginx at nginx.org
> Subject: nginx and tproxy
>
>
> Hi all:
>
>
> Can we let both work together? Anyone can share his experience with us if
> we want to integrate them?
>
> thanks in advance
>
> george
>
>
> ______________________________**_________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/**mailman/listinfo/nginx<http://mailman.nginx.org/mailman/listinfo/nginx>
> ______________________________**_________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/**mailman/listinfo/nginx<http://mailman.nginx.org/mailman/listinfo/nginx>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110915/6d97b84d/attachment.html>

From magicbearmo at gmail.com  Thu Sep 15 17:51:25 2011
From: magicbearmo at gmail.com (MagicBear)
Date: Fri, 16 Sep 2011 01:51:25 +0800
Subject: upstream keepalive - call for testing
In-Reply-To: <051a01adad30a88a5d6d428e8fbaaafc.NginxMailingListEnglish@forum.nginx.org>
References: <20110905180210.GL1137@mdounin.ru>
	<051a01adad30a88a5d6d428e8fbaaafc.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <CAENo=s0DrYVS2vhayfs_azwoFC3pOSsUE_uwYzq_RyerP_MZNQ@mail.gmail.com>

Hello,
I have wrote a module to make nginx support 304 to decrease bandwidth usage.
note: I have a newbie for nginx module development, so the above module may
have some problem. Welcome to test it and feedback another problem with me.

You can download full patch file from here:
http://m-b.cc/share/proxy_304.txt

# User MagicBear <magicbearmo at gmail.com>
Upstream:
add $upstream_last_modified variant.
add handler for 304 Unmodified.
Proxy:
change to send If-Modified-Since header.

TODO:
change write TO not block IO.

diff -ruN a/http/modules/ngx_http_proxy_module.c
b/http/modules/ngx_http_proxy_module.c
--- a/http/modules/ngx_http_proxy_module.c      2011-09-15
22:23:03.284431407 +0800
+++ b/http/modules/ngx_http_proxy_module.c      2011-09-16
01:41:44.654428632 +0800
@@ -543,7 +543,7 @@
     { ngx_string("Connection"), ngx_string("close") },
     { ngx_string("Keep-Alive"), ngx_string("") },
     { ngx_string("Expect"), ngx_string("") },
-    { ngx_string("If-Modified-Since"), ngx_string("") },
+    { ngx_string("If-Modified-Since"),
ngx_string("$upstream_last_modified") },
     { ngx_string("If-Unmodified-Since"), ngx_string("") },
     { ngx_string("If-None-Match"), ngx_string("") },
     { ngx_string("If-Match"), ngx_string("") },
diff -ruN a/http/ngx_http_upstream.c b/http/ngx_http_upstream.c
--- a/http/ngx_http_upstream.c  2011-09-15 22:23:03.284431407 +0800
+++ b/http/ngx_http_upstream.c  2011-09-16 01:41:44.654428632 +0800
@@ -16,6 +16,8 @@
     ngx_http_upstream_t *u);
 static ngx_int_t ngx_http_upstream_cache_status(ngx_http_request_t *r,
     ngx_http_variable_value_t *v, uintptr_t data);
+static ngx_int_t ngx_http_upstream_last_modified(ngx_http_request_t *r,
+    ngx_http_variable_value_t *v, uintptr_t data);
 #endif

 static void ngx_http_upstream_init_request(ngx_http_request_t *r);
@@ -342,6 +344,10 @@
       ngx_http_upstream_cache_status, 0,
       NGX_HTTP_VAR_NOCACHEABLE, 0 },

+    { ngx_string("upstream_last_modified"), NULL,
+      ngx_http_upstream_last_modified, 0,
+      NGX_HTTP_VAR_NOCACHEABLE, 0 },
+
 #endif

     { ngx_null_string, NULL, NULL, 0, 0, 0 }
@@ -1618,6 +1624,80 @@
             u->buffer.last = u->buffer.pos;
         }

+#if (NGX_HTTP_CACHE)
+
+        if (u->cache_status == NGX_HTTP_CACHE_EXPIRED &&
+                       u->headers_in.status_n == NGX_HTTP_NOT_MODIFIED &&
+                       ngx_http_file_cache_valid(u->conf->cache_valid,
u->headers_in.status_n))
+        {
+            ngx_int_t  rc;
+
+            rc = u->reinit_request(r);
+
+            if (rc == NGX_OK) {
+                u->cache_status = NGX_HTTP_CACHE_BYPASS;
+                rc = ngx_http_upstream_cache_send(r, u);
+
+                               time_t  now, valid;
+
+                               now = ngx_time();
+
+                               valid = r->cache->valid_sec;
+
+                               if (valid == 0) {
+                                       valid =
ngx_http_file_cache_valid(u->conf->cache_valid,
+
                              u->headers_in.status_n);
+                                       if (valid) {
+                                               r->cache->valid_sec = now +
valid;
+                                       }
+                               }
+
+                               if (valid) {
+                                       r->cache->last_modified =
r->headers_out.last_modified_time;
+                                       r->cache->date = now;
+                                       r->cache->body_start = (u_short)
(u->buffer.pos - u->buffer.start);
+
+                                       // update Header
+                                       ngx_http_file_cache_set_header(r,
u->buffer.start);
+
+                                       ngx_log_debug1(NGX_LOG_DEBUG_HTTP,
r->connection->log, 0,
+
 "update cache \"%s\" header to new expired." , r->cache->file.name.data);
+
+                                       // Reopen file via RW
+                                       ngx_fd_t fd =
ngx_open_file(r->cache->file.name.data, NGX_FILE_RDWR, NGX_FILE_OPEN, 0);
+
+                                       if (fd == NGX_INVALID_FILE) {
+                                               ngx_log_error(NGX_LOG_CRIT,
r->connection->log, ngx_errno,
+
ngx_open_file_n " \"%s\" failed", r->cache->file.name.data);
+                                               return;
+                                       }
+
+                                       // Write cache
+                                       if (write(fd, u->buffer.start,
sizeof(ngx_http_file_cache_header_t)) < 0)
+                                       {
+                                               ngx_log_error(NGX_LOG_CRIT,
r->connection->log, ngx_errno,
+
"write proxy_cache \"%s\" failed", r->cache->file.name.data);
+                                               return;
+                                       }
+
+                                       if (ngx_close_file(fd) ==
NGX_FILE_ERROR) {
+                                               ngx_log_error(NGX_LOG_ALERT,
r->connection->log, ngx_errno,
+
ngx_close_file_n " \"%s\" failed", r->cache->file.name.data);
+                                       }
+                                       ngx_log_debug1(NGX_LOG_DEBUG_HTTP,
r->connection->log, 0,
+
 "update cache \"%s\" header to new expired done." ,
r->cache->file.name.data);
+                               } else {
+                                       u->cacheable = 0;
+                                       r->headers_out.last_modified_time =
-1;
+                               }
+            }
+
+            ngx_http_upstream_finalize_request(r, u, rc);
+            return;
+        }
+
+#endif
+
         if (ngx_http_upstream_test_next(r, u) == NGX_OK) {
             return;
         }
@@ -4006,6 +4086,32 @@

     return NGX_OK;
 }
+
+ngx_int_t
+ngx_http_upstream_last_modified(ngx_http_request_t *r,
+    ngx_http_variable_value_t *v, uintptr_t data)
+{
+    u_char *u;
+
+    if (r->upstream == NULL || r->upstream->cache_status == 0 ||
r->cache==NULL || r->cache->last_modified <= 0) {
+        v->not_found = 1;
+        return NGX_OK;
+    }
+
+    v->valid = 1;
+    v->no_cacheable = 0;
+    v->not_found = 0;
+       u = ngx_pcalloc(r->pool, 30);
+    if (u == NULL) {
+        return NGX_ERROR;
+    }
+
+    v->len = 29;
+       ngx_http_time(u, r->cache->last_modified);
+    v->data = u;
+
+    return NGX_OK;
+}

 #endif


MagicBear

2011/9/15 magicbear <nginx-forum at nginx.us>

> I have run the nginx 1.1.2 via this patch for 7 days, except for one
> days have a large DDoS so I restart nginx for several seconds, it was
> very stable to work.
> Handle about 70million request without problem happen, I think the last
> problem may be have a memory corruption, you are right.
> I will check that server when have times.
> Thanks for your hard work.
>
> MagicBear
>
> Maxim Dounin Wrote:
> -------------------------------------------------------
> > Hello!
> >
> > On Mon, Sep 05, 2011 at 11:42:31PM +0800,
> > ????? wrote:
> >
> > > (gdb) fr 0
> > > #0  ngx_http_upstream_handler
> > (ev=0x7fc45735f8a8)
> > >     at src/http/ngx_http_upstream.c:915
> > > 915    ctx->current_request = r;
> > > (gdb) p ngx_cycle->log
> > > $1 = (ngx_log_t *) 0x21f19a8
> > > (gdb) p *r
> > > $2 = {signature = 51686928, connection =
> > 0x23b4160, ctx = 0x0,
> >
> > [...]
> >
> > This looks like memory corruption, but
> > unfortunately I don't see
> > any traces of the real cause.  My best quess is
> > improper handling
> > of proxy_ignore_client_abort as fixed in 1.1.2.
> > Please try 1.1.2
> > with patches from
> >
> > http://nginx.org/patches/patch-nginx-keepalive-ful
> > l-5.txt
> >
> > It already includes upstream keepalive module, as
> > well as all
> > other upstream-keepalive related fixes.  See here
> > for details:
> >
> > http://mailman.nginx.org/pipermail/nginx-devel/201
> > 1-September/001147.html
> >
> > Maxim Dounin
> >
> > _______________________________________________
> > nginx mailing list
> > nginx at nginx.org
> > http://mailman.nginx.org/mailman/listinfo/nginx
>
> Posted at Nginx Forum:
> http://forum.nginx.org/read.php?2,213207,215217#msg-215217
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110916/095ad33d/attachment-0001.html>
-------------- next part --------------
# User MagicBear <magicbearmo at gmail.com>
Upstream:
add $upstream_last_modified variant.
add handler for 304 Unmodified.
Proxy:
change to send If-Modified-Since header. 

TODO:
change write TO not block IO.

diff -ruN a/http/modules/ngx_http_proxy_module.c b/http/modules/ngx_http_proxy_module.c
--- a/http/modules/ngx_http_proxy_module.c      2011-09-15 22:23:03.284431407 +0800
+++ b/http/modules/ngx_http_proxy_module.c      2011-09-16 01:41:44.654428632 +0800
@@ -543,7 +543,7 @@
     { ngx_string("Connection"), ngx_string("close") },
     { ngx_string("Keep-Alive"), ngx_string("") },
     { ngx_string("Expect"), ngx_string("") },
-    { ngx_string("If-Modified-Since"), ngx_string("") },
+    { ngx_string("If-Modified-Since"), ngx_string("$upstream_last_modified") },
     { ngx_string("If-Unmodified-Since"), ngx_string("") },
     { ngx_string("If-None-Match"), ngx_string("") },
     { ngx_string("If-Match"), ngx_string("") },
diff -ruN a/http/ngx_http_upstream.c b/http/ngx_http_upstream.c
--- a/http/ngx_http_upstream.c  2011-09-15 22:23:03.284431407 +0800
+++ b/http/ngx_http_upstream.c  2011-09-16 01:41:44.654428632 +0800
@@ -16,6 +16,8 @@
     ngx_http_upstream_t *u);
 static ngx_int_t ngx_http_upstream_cache_status(ngx_http_request_t *r,
     ngx_http_variable_value_t *v, uintptr_t data);
+static ngx_int_t ngx_http_upstream_last_modified(ngx_http_request_t *r,
+    ngx_http_variable_value_t *v, uintptr_t data);
 #endif

 static void ngx_http_upstream_init_request(ngx_http_request_t *r);
@@ -342,6 +344,10 @@
       ngx_http_upstream_cache_status, 0,
       NGX_HTTP_VAR_NOCACHEABLE, 0 },

+    { ngx_string("upstream_last_modified"), NULL,
+      ngx_http_upstream_last_modified, 0,
+      NGX_HTTP_VAR_NOCACHEABLE, 0 },
+
 #endif

     { ngx_null_string, NULL, NULL, 0, 0, 0 }
@@ -1618,6 +1624,80 @@
             u->buffer.last = u->buffer.pos;
         }

+#if (NGX_HTTP_CACHE)
+
+        if (u->cache_status == NGX_HTTP_CACHE_EXPIRED &&
+                       u->headers_in.status_n == NGX_HTTP_NOT_MODIFIED &&
+                       ngx_http_file_cache_valid(u->conf->cache_valid, u->headers_in.status_n))
+        {
+            ngx_int_t  rc;
+
+            rc = u->reinit_request(r);
+
+            if (rc == NGX_OK) {
+                u->cache_status = NGX_HTTP_CACHE_BYPASS;
+                rc = ngx_http_upstream_cache_send(r, u);
+
+                               time_t  now, valid;
+
+                               now = ngx_time();
+
+                               valid = r->cache->valid_sec;
+
+                               if (valid == 0) {
+                                       valid = ngx_http_file_cache_valid(u->conf->cache_valid,
+                                                                                                         u->headers_in.status_n);
+                                       if (valid) {
+                                               r->cache->valid_sec = now + valid;
+                                       }
+                               }
+
+                               if (valid) {
+                                       r->cache->last_modified = r->headers_out.last_modified_time;
+                                       r->cache->date = now;
+                                       r->cache->body_start = (u_short) (u->buffer.pos - u->buffer.start);
+
+                                       // update Header
+                                       ngx_http_file_cache_set_header(r, u->buffer.start);
+
+                                       ngx_log_debug1(NGX_LOG_DEBUG_HTTP, r->connection->log, 0,
+                                                                          "update cache \"%s\" header to new expired." , r->cache->file.name.data);
+
+                                       // Reopen file via RW
+                                       ngx_fd_t fd = ngx_open_file(r->cache->file.name.data, NGX_FILE_RDWR, NGX_FILE_OPEN, 0);
+
+                                       if (fd == NGX_INVALID_FILE) {
+                                               ngx_log_error(NGX_LOG_CRIT, r->connection->log, ngx_errno,
+                                                                         ngx_open_file_n " \"%s\" failed", r->cache->file.name.data);
+                                               return;
+                                       }
+
+                                       // Write cache
+                                       if (write(fd, u->buffer.start, sizeof(ngx_http_file_cache_header_t)) < 0)
+                                       {
+                                               ngx_log_error(NGX_LOG_CRIT, r->connection->log, ngx_errno,
+                                                                         "write proxy_cache \"%s\" failed", r->cache->file.name.data);
+                                               return;
+                                       }
+
+                                       if (ngx_close_file(fd) == NGX_FILE_ERROR) {
+                                               ngx_log_error(NGX_LOG_ALERT, r->connection->log, ngx_errno,
+                                                                         ngx_close_file_n " \"%s\" failed", r->cache->file.name.data);
+                                       }
+                                       ngx_log_debug1(NGX_LOG_DEBUG_HTTP, r->connection->log, 0,
+                                                                          "update cache \"%s\" header to new expired done." , r->cache->file.name.data);
+                               } else {
+                                       u->cacheable = 0;
+                                       r->headers_out.last_modified_time = -1;
+                               }
+            }
+
+            ngx_http_upstream_finalize_request(r, u, rc);
+            return;
+        }
+
+#endif
+
         if (ngx_http_upstream_test_next(r, u) == NGX_OK) {
             return;
         }
@@ -4006,6 +4086,32 @@

     return NGX_OK;
 }
+
+ngx_int_t
+ngx_http_upstream_last_modified(ngx_http_request_t *r,
+    ngx_http_variable_value_t *v, uintptr_t data)
+{
+    u_char *u;
+
+    if (r->upstream == NULL || r->upstream->cache_status == 0 || r->cache==NULL || r->cache->last_modified <= 0) {
+        v->not_found = 1;
+        return NGX_OK;
+    }
+
+    v->valid = 1;
+    v->no_cacheable = 0;
+    v->not_found = 0;
+       u = ngx_pcalloc(r->pool, 30);
+    if (u == NULL) {
+        return NGX_ERROR;
+    }
+
+    v->len = 29;
+       ngx_http_time(u, r->cache->last_modified);
+    v->data = u;
+
+    return NGX_OK;
+}

 #endif

From nginx-forum at nginx.us  Thu Sep 15 21:03:41 2011
From: nginx-forum at nginx.us (caoxingk)
Date: Thu, 15 Sep 2011 17:03:41 -0400
Subject: How to reference OS Environment Variales in nginx.conf
Message-ID: <f76c0a1550e252b4ce66fa7e60502791.NginxMailingListEnglish@forum.nginx.org>

Hi there,

After set a variable by `set $name value`, i can reference it like
`$name`,
but when i export a OS Environment Variale by `env name_from_env`, like
http://wiki.nginx.org/CoreModule#env said,
and i am sure the name_from_env is valid which defined form nginx's
parent process.

But, my friends, how to reference it ? 
$name_from_env or ${name_from_env} or %name_from_env% didn't work what
i've tried before.

Thanks a lot.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215269,215269#msg-215269


From nginx-forum at nginx.us  Thu Sep 15 23:22:00 2011
From: nginx-forum at nginx.us (qzpmwo)
Date: Thu, 15 Sep 2011 19:22:00 -0400
Subject: How to force nginx to cache HTTP POST response for few seconds
In-Reply-To: <4cd892cb0ebd639ee05f69162a2d098b.NginxMailingListEnglish@forum.nginx.org>
References: <20110915092833.GA80146@nginx.com>
	<1ef7fdb8e7d77fa9a9d5f3934aa586bf.NginxMailingListEnglish@forum.nginx.org>
	<4cd892cb0ebd639ee05f69162a2d098b.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <4e7738edd4de5780202699e00467e70c.NginxMailingListEnglish@forum.nginx.org>

Hi Igor,

Thanks a lot for your help. Problem was with the request body. Now
everything is working as expected.

Sincerely
--->>> qzpmwo <<<---

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215231,215274#msg-215274


From mat999 at gmail.com  Fri Sep 16 01:42:25 2011
From: mat999 at gmail.com (SplitIce)
Date: Fri, 16 Sep 2011 11:42:25 +1000
Subject: upstream keepalive - call for testing
In-Reply-To: <CAENo=s0DrYVS2vhayfs_azwoFC3pOSsUE_uwYzq_RyerP_MZNQ@mail.gmail.com>
References: <20110905180210.GL1137@mdounin.ru>
	<051a01adad30a88a5d6d428e8fbaaafc.NginxMailingListEnglish@forum.nginx.org>
	<CAENo=s0DrYVS2vhayfs_azwoFC3pOSsUE_uwYzq_RyerP_MZNQ@mail.gmail.com>
Message-ID: <CALFfGYax7FO9zyVvskwpfHmBgjU=2y_sOGgJXBKCbdMZMut-jw@mail.gmail.com>

wow, this + last-modified and etags would really cut into the reason people
use varnish in their nginx + varnish setup.

On Fri, Sep 16, 2011 at 3:51 AM, MagicBear <magicbearmo at gmail.com> wrote:

> Hello,
> I have wrote a module to make nginx support 304 to decrease bandwidth
> usage.
> note: I have a newbie for nginx module development, so the above module may
> have some problem. Welcome to test it and feedback another problem with me.
>
> You can download full patch file from here:
> http://m-b.cc/share/proxy_304.txt
>
> # User MagicBear <magicbearmo at gmail.com>
> Upstream:
> add $upstream_last_modified variant.
> add handler for 304 Unmodified.
> Proxy:
> change to send If-Modified-Since header.
>
> TODO:
> change write TO not block IO.
>
> diff -ruN a/http/modules/ngx_http_proxy_module.c
> b/http/modules/ngx_http_proxy_module.c
> --- a/http/modules/ngx_http_proxy_module.c      2011-09-15
> 22:23:03.284431407 +0800
> +++ b/http/modules/ngx_http_proxy_module.c      2011-09-16
> 01:41:44.654428632 +0800
> @@ -543,7 +543,7 @@
>      { ngx_string("Connection"), ngx_string("close") },
>      { ngx_string("Keep-Alive"), ngx_string("") },
>      { ngx_string("Expect"), ngx_string("") },
> -    { ngx_string("If-Modified-Since"), ngx_string("") },
> +    { ngx_string("If-Modified-Since"),
> ngx_string("$upstream_last_modified") },
>      { ngx_string("If-Unmodified-Since"), ngx_string("") },
>      { ngx_string("If-None-Match"), ngx_string("") },
>      { ngx_string("If-Match"), ngx_string("") },
> diff -ruN a/http/ngx_http_upstream.c b/http/ngx_http_upstream.c
> --- a/http/ngx_http_upstream.c  2011-09-15 22:23:03.284431407 +0800
> +++ b/http/ngx_http_upstream.c  2011-09-16 01:41:44.654428632 +0800
> @@ -16,6 +16,8 @@
>      ngx_http_upstream_t *u);
>  static ngx_int_t ngx_http_upstream_cache_status(ngx_http_request_t *r,
>      ngx_http_variable_value_t *v, uintptr_t data);
> +static ngx_int_t ngx_http_upstream_last_modified(ngx_http_request_t *r,
> +    ngx_http_variable_value_t *v, uintptr_t data);
>  #endif
>
>  static void ngx_http_upstream_init_request(ngx_http_request_t *r);
> @@ -342,6 +344,10 @@
>        ngx_http_upstream_cache_status, 0,
>        NGX_HTTP_VAR_NOCACHEABLE, 0 },
>
> +    { ngx_string("upstream_last_modified"), NULL,
> +      ngx_http_upstream_last_modified, 0,
> +      NGX_HTTP_VAR_NOCACHEABLE, 0 },
> +
>  #endif
>
>      { ngx_null_string, NULL, NULL, 0, 0, 0 }
> @@ -1618,6 +1624,80 @@
>              u->buffer.last = u->buffer.pos;
>          }
>
> +#if (NGX_HTTP_CACHE)
> +
> +        if (u->cache_status == NGX_HTTP_CACHE_EXPIRED &&
> +                       u->headers_in.status_n == NGX_HTTP_NOT_MODIFIED &&
> +                       ngx_http_file_cache_valid(u->conf->cache_valid,
> u->headers_in.status_n))
> +        {
> +            ngx_int_t  rc;
> +
> +            rc = u->reinit_request(r);
> +
> +            if (rc == NGX_OK) {
> +                u->cache_status = NGX_HTTP_CACHE_BYPASS;
> +                rc = ngx_http_upstream_cache_send(r, u);
> +
> +                               time_t  now, valid;
> +
> +                               now = ngx_time();
> +
> +                               valid = r->cache->valid_sec;
> +
> +                               if (valid == 0) {
> +                                       valid =
> ngx_http_file_cache_valid(u->conf->cache_valid,
> +
>                               u->headers_in.status_n);
> +                                       if (valid) {
> +                                               r->cache->valid_sec = now +
> valid;
> +                                       }
> +                               }
> +
> +                               if (valid) {
> +                                       r->cache->last_modified =
> r->headers_out.last_modified_time;
> +                                       r->cache->date = now;
> +                                       r->cache->body_start = (u_short)
> (u->buffer.pos - u->buffer.start);
> +
> +                                       // update Header
> +                                       ngx_http_file_cache_set_header(r,
> u->buffer.start);
> +
> +                                       ngx_log_debug1(NGX_LOG_DEBUG_HTTP,
> r->connection->log, 0,
> +
>  "update cache \"%s\" header to new expired." , r->cache->file.name.data);
> +
> +                                       // Reopen file via RW
> +                                       ngx_fd_t fd =
> ngx_open_file(r->cache->file.name.data, NGX_FILE_RDWR, NGX_FILE_OPEN, 0);
> +
> +                                       if (fd == NGX_INVALID_FILE) {
> +                                               ngx_log_error(NGX_LOG_CRIT,
> r->connection->log, ngx_errno,
> +
> ngx_open_file_n " \"%s\" failed", r->cache->file.name.data);
> +                                               return;
> +                                       }
> +
> +                                       // Write cache
> +                                       if (write(fd, u->buffer.start,
> sizeof(ngx_http_file_cache_header_t)) < 0)
> +                                       {
> +                                               ngx_log_error(NGX_LOG_CRIT,
> r->connection->log, ngx_errno,
> +
> "write proxy_cache \"%s\" failed", r->cache->file.name.data);
> +                                               return;
> +                                       }
> +
> +                                       if (ngx_close_file(fd) ==
> NGX_FILE_ERROR) {
> +
> ngx_log_error(NGX_LOG_ALERT, r->connection->log, ngx_errno,
> +
> ngx_close_file_n " \"%s\" failed", r->cache->file.name.data);
> +                                       }
> +                                       ngx_log_debug1(NGX_LOG_DEBUG_HTTP,
> r->connection->log, 0,
> +
>  "update cache \"%s\" header to new expired done." ,
> r->cache->file.name.data);
> +                               } else {
> +                                       u->cacheable = 0;
> +                                       r->headers_out.last_modified_time =
> -1;
> +                               }
> +            }
> +
> +            ngx_http_upstream_finalize_request(r, u, rc);
> +            return;
> +        }
> +
> +#endif
> +
>          if (ngx_http_upstream_test_next(r, u) == NGX_OK) {
>              return;
>          }
> @@ -4006,6 +4086,32 @@
>
>      return NGX_OK;
>  }
> +
> +ngx_int_t
> +ngx_http_upstream_last_modified(ngx_http_request_t *r,
> +    ngx_http_variable_value_t *v, uintptr_t data)
> +{
> +    u_char *u;
> +
> +    if (r->upstream == NULL || r->upstream->cache_status == 0 ||
> r->cache==NULL || r->cache->last_modified <= 0) {
> +        v->not_found = 1;
> +        return NGX_OK;
> +    }
> +
> +    v->valid = 1;
> +    v->no_cacheable = 0;
> +    v->not_found = 0;
> +       u = ngx_pcalloc(r->pool, 30);
> +    if (u == NULL) {
> +        return NGX_ERROR;
> +    }
> +
> +    v->len = 29;
> +       ngx_http_time(u, r->cache->last_modified);
> +    v->data = u;
> +
> +    return NGX_OK;
> +}
>
>  #endif
>
>
> MagicBear
>
> 2011/9/15 magicbear <nginx-forum at nginx.us>
>
>> I have run the nginx 1.1.2 via this patch for 7 days, except for one
>> days have a large DDoS so I restart nginx for several seconds, it was
>> very stable to work.
>> Handle about 70million request without problem happen, I think the last
>> problem may be have a memory corruption, you are right.
>> I will check that server when have times.
>> Thanks for your hard work.
>>
>> MagicBear
>>
>> Maxim Dounin Wrote:
>> -------------------------------------------------------
>> > Hello!
>> >
>> > On Mon, Sep 05, 2011 at 11:42:31PM +0800,
>> > ????? wrote:
>> >
>> > > (gdb) fr 0
>> > > #0  ngx_http_upstream_handler
>> > (ev=0x7fc45735f8a8)
>> > >     at src/http/ngx_http_upstream.c:915
>> > > 915    ctx->current_request = r;
>> > > (gdb) p ngx_cycle->log
>> > > $1 = (ngx_log_t *) 0x21f19a8
>> > > (gdb) p *r
>> > > $2 = {signature = 51686928, connection =
>> > 0x23b4160, ctx = 0x0,
>> >
>> > [...]
>> >
>> > This looks like memory corruption, but
>> > unfortunately I don't see
>> > any traces of the real cause.  My best quess is
>> > improper handling
>> > of proxy_ignore_client_abort as fixed in 1.1.2.
>> > Please try 1.1.2
>> > with patches from
>> >
>> > http://nginx.org/patches/patch-nginx-keepalive-ful
>> > l-5.txt
>> >
>> > It already includes upstream keepalive module, as
>> > well as all
>> > other upstream-keepalive related fixes.  See here
>> > for details:
>> >
>> > http://mailman.nginx.org/pipermail/nginx-devel/201
>> > 1-September/001147.html
>> >
>> > Maxim Dounin
>> >
>> > _______________________________________________
>> > nginx mailing list
>> > nginx at nginx.org
>> > http://mailman.nginx.org/mailman/listinfo/nginx
>>
>> Posted at Nginx Forum:
>> http://forum.nginx.org/read.php?2,213207,215217#msg-215217
>>
>> _______________________________________________
>> nginx mailing list
>> nginx at nginx.org
>> http://mailman.nginx.org/mailman/listinfo/nginx
>>
>
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110916/4bc2a465/attachment-0001.html>

From nginx-forum at nginx.us  Fri Sep 16 02:41:20 2011
From: nginx-forum at nginx.us (damnlinux)
Date: Thu, 15 Sep 2011 22:41:20 -0400
Subject: Making Wordpress Plugins Work With Nginx
Message-ID: <b9dc7fdee421c04252cde760c9f4e595.NginxMailingListEnglish@forum.nginx.org>

I'm new to Nginx (Moving from Apache). I'm having a problem where the
majority of Wordpress plugin are not working. I'm using Nginx 1.1.3.

Here are my configuration if it can help figuring out the problem

#####nginx.conf 
user www-data;
worker_processes  1;

error_log  /var/log/nginx/error.log;
pid        /var/run/nginx.pid;

events {
    worker_connections  1024;
    # multi_accept on;
}

http {
    include       /etc/nginx/mime.types;

    access_log	/var/log/nginx/access.log;

    sendfile        on;
    #tcp_nopush     on;

    #keepalive_timeout  0;
    keepalive_timeout  65;
    tcp_nodelay        on;

    gzip  on;
    gzip_disable "MSIE [1-6]\.(?!.*SV1)";

    include /etc/nginx/conf.d/*.conf;
    include /etc/nginx/sites-enabled/*;
}

####sites-available/mydomain.com

server {
        listen   80;
        server_name mydomain.com;
 
        access_log  /var/log/nginx/mydomain.com.access.log;
        root   /var/www/mydomain;
        index  index.php;
 
        location / {
                try_files $uri $uri/ @wordpress;
        }
 
        location @wordpress {
                fastcgi_pass 127.0.0.1:9000;
                fastcgi_param SCRIPT_FILENAME
/var/www/mydomain/index.php;
                include /etc/nginx/fastcgi_params;
                fastcgi_param SCRIPT_NAME /index.php;
        }
 
        location ~ \.php$ {
                try_files $uri @wordpress;
                fastcgi_pass    127.0.0.1:9000;
                fastcgi_index   index.php;
                fastcgi_param SCRIPT_FILENAME
/var/www/mydomain$fastcgi_script_name;
                include fastcgi_params;
 
Thanks

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215277,215277#msg-215277


From nginx-forum at nginx.us  Fri Sep 16 02:50:45 2011
From: nginx-forum at nginx.us (caoxingk)
Date: Thu, 15 Sep 2011 22:50:45 -0400
Subject: How to reference OS Environment Variales in nginx.conf
In-Reply-To: <f76c0a1550e252b4ce66fa7e60502791.NginxMailingListEnglish@forum.nginx.org>
References: <f76c0a1550e252b4ce66fa7e60502791.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <a70537c1f951044ec0e7fd6dd9b687e1.NginxMailingListEnglish@forum.nginx.org>

2ed version is here:
http://stackoverflow.com/questions/7439515/how-to-reference-os-environment-variables-in-nginx-conf

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215269,215278#msg-215278


From nginx-forum at nginx.us  Fri Sep 16 03:09:15 2011
From: nginx-forum at nginx.us (qzpmwo)
Date: Thu, 15 Sep 2011 23:09:15 -0400
Subject: How to check that $request_body has some string and use it like a
	caching key?
Message-ID: <3118d31cde1d2d5fe00750552a676846.NginxMailingListEnglish@forum.nginx.org>

Hi everybody,

I have the following situation:

Incoming request with method POST has binary body. I would like to
extract piece(s) from the payload to use it like a caching key. 

Something like:

if ( $request_body ~* (action.*) ) {
    set $key $1;
} 
...
proxy_cache_key $uri|$key

Unfortunately it does not work for me even if I specified (like Igor
advised):

 client_max_body_size 1k;

I also tried client_body_buffer_size 1k; but it did not help either.

In the error.log with debug on I can see that $request_body has value,
but it never matches my condition, so $key is always empty.

Just in case I'm repeating, that request_body is binary. It has 0x00 as
well as everything else. It is AMF data.

Any advices?

Thanks
--->>> qzpmwo <<<---

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215279,215279#msg-215279


From agentzh at gmail.com  Fri Sep 16 04:06:23 2011
From: agentzh at gmail.com (agentzh)
Date: Fri, 16 Sep 2011 12:06:23 +0800
Subject: How to reference OS Environment Variales in nginx.conf
In-Reply-To: <f76c0a1550e252b4ce66fa7e60502791.NginxMailingListEnglish@forum.nginx.org>
References: <f76c0a1550e252b4ce66fa7e60502791.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <CAB4Tn6NOnh1WsYxOmR+EOe=15toD2Et+fTatnVRN39f69qaQqg@mail.gmail.com>

On Fri, Sep 16, 2011 at 5:03 AM, caoxingk <nginx-forum at nginx.us> wrote:
> After set a variable by `set $name value`, i can reference it like
> `$name`,
> but when i export a OS Environment Variale by `env name_from_env`, like
> http://wiki.nginx.org/CoreModule#env said,
> and i am sure the name_from_env is valid which defined form nginx's
> parent process.
>
> But, my friends, how to reference it ?

You can read system environment variables with ngx_lua enabled in your
nginx build: http://wiki.nginx.org/HttpLuaModule

    env PATH;
    http {
        ...
        server {
            location /path {
                set_by_lua $path 'return os.getenv("PATH")';
                ...
            }
        }

BTW, to use the set_by_lua directive, you also need to enable the
ngx_devel_kit module here: https://github.com/simpl/ngx_devel_kit
(it'll be easier if you use the ngx_openresty bundle).

Regards,
-agentzh


From nginx-forum at nginx.us  Fri Sep 16 04:53:28 2011
From: nginx-forum at nginx.us (skyshade)
Date: Fri, 16 Sep 2011 00:53:28 -0400
Subject: nginx-1.1.3
In-Reply-To: <20110914151925.GG49199@nginx.com>
References: <20110914151925.GG49199@nginx.com>
Message-ID: <48a8cbc475ca42c28ec91d441278940d.NginxMailingListEnglish@forum.nginx.org>

Igor

I test 1.1.3 develop version with the new ngx_http_mp4_module, but its
dont work with HttpProxyModule. 
1. This is spectated?
2. If yes. There is any plan to add HttpProxyModule +
ngx_http_mp4_module support?

Will be aweason to have this working on Nginx.


Thanks for you amazing work on nginx.

Sky

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215200,215280#msg-215280


From igor at sysoev.ru  Fri Sep 16 05:47:32 2011
From: igor at sysoev.ru (Igor Sysoev)
Date: Fri, 16 Sep 2011 09:47:32 +0400
Subject: nginx-1.1.3
In-Reply-To: <48a8cbc475ca42c28ec91d441278940d.NginxMailingListEnglish@forum.nginx.org>
References: <20110914151925.GG49199@nginx.com>
	<48a8cbc475ca42c28ec91d441278940d.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20110916054732.GB8567@nginx.com>

On Fri, Sep 16, 2011 at 12:53:28AM -0400, skyshade wrote:
> Igor
> 
> I test 1.1.3 develop version with the new ngx_http_mp4_module, but its
> dont work with HttpProxyModule. 
> 1. This is spectated?
> 2. If yes. There is any plan to add HttpProxyModule +
> ngx_http_mp4_module support?
> 
> Will be aweason to have this working on Nginx.

Currently ngx_http_mp4_module is not a filter module, so it can work
only with static files. There are plans to implement it as filter.


-- 
Igor Sysoev


From patcito at gmail.com  Fri Sep 16 06:32:10 2011
From: patcito at gmail.com (Patrick Aljord)
Date: Fri, 16 Sep 2011 01:32:10 -0500
Subject: can't get cache controle to set an expire date
Message-ID: <CAK9TW4rD2BGnA3s96BH+TyW9es729CVh5zkBci8eaNzOuaLE8Q@mail.gmail.com>

Hi,

This is my nginx config file: http://pastebin.com/TZcfzipk

The problem is that despite setting "expires max" cache sends no
expire date so I must be doing something wrong.
This is what appears as header:

Cache-Control	max-age=0

Any idea how I could fix this?

Thanks in advance,

Pat


From lists at ruby-forum.com  Fri Sep 16 09:49:34 2011
From: lists at ruby-forum.com (Mahalingam Mr)
Date: Fri, 16 Sep 2011 11:49:34 +0200
Subject: can't get cache controle to set an expire date
In-Reply-To: <CAK9TW4rD2BGnA3s96BH+TyW9es729CVh5zkBci8eaNzOuaLE8Q@mail.gmail.com>
References: <CAK9TW4rD2BGnA3s96BH+TyW9es729CVh5zkBci8eaNzOuaLE8Q@mail.gmail.com>
Message-ID: <1c65c6dd406143a6673d311145a09842@ruby-forum.com>

#   location /images/ {
#       # serve from disk and set expires
#       expires max;
#   }
#
#   location /stylesheets/ {
#       expires max;
#   }
#
#   location /javascripts/ {
#       expires max;
#   }


Solution:

Try this instead of giving the folders explicitly.,

location ~* \.(ico|css|js|gif|jpe?g|png)(\?[0-9]+)?$ {
       expires max;
       break;
    }

TML

-- 
Posted via http://www.ruby-forum.com/.


From mdounin at mdounin.ru  Fri Sep 16 10:13:53 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Fri, 16 Sep 2011 14:13:53 +0400
Subject: can't get cache controle to set an expire date
In-Reply-To: <CAK9TW4rD2BGnA3s96BH+TyW9es729CVh5zkBci8eaNzOuaLE8Q@mail.gmail.com>
References: <CAK9TW4rD2BGnA3s96BH+TyW9es729CVh5zkBci8eaNzOuaLE8Q@mail.gmail.com>
Message-ID: <20110916101353.GJ1137@mdounin.ru>

Hello!

On Fri, Sep 16, 2011 at 01:32:10AM -0500, Patrick Aljord wrote:

> Hi,
> 
> This is my nginx config file: http://pastebin.com/TZcfzipk
> 
> The problem is that despite setting "expires max" cache sends no
> expire date so I must be doing something wrong.
> This is what appears as header:
> 
> Cache-Control	max-age=0
> 
> Any idea how I could fix this?

Please show request and response.  (And please make sure you 
actually see request in nginx logs, i.e. don't test your browser's 
cache.)

If request actually hits nginx in question, I see two most likely 
reasons:

1) Your request is processed in location where you don't have 
"expires max;" set.  I.e. request doesn't match /images/ etc.

2) Your request is processed in another server block.

Maxim Dounin


From mdounin at mdounin.ru  Fri Sep 16 10:19:51 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Fri, 16 Sep 2011 14:19:51 +0400
Subject: can't get cache controle to set an expire date
In-Reply-To: <1c65c6dd406143a6673d311145a09842@ruby-forum.com>
References: <CAK9TW4rD2BGnA3s96BH+TyW9es729CVh5zkBci8eaNzOuaLE8Q@mail.gmail.com>
	<1c65c6dd406143a6673d311145a09842@ruby-forum.com>
Message-ID: <20110916101951.GK1137@mdounin.ru>

Hello!

On Fri, Sep 16, 2011 at 11:49:34AM +0200, Mahalingam Mr wrote:

> #   location /images/ {
> #       # serve from disk and set expires
> #       expires max;
> #   }
> #
> #   location /stylesheets/ {
> #       expires max;
> #   }
> #
> #   location /javascripts/ {
> #       expires max;
> #   }
> 
> 
> Solution:
> 
> Try this instead of giving the folders explicitly.,

No, this is bad idea.  Using regexp locations is bad as it makes 
configuration support much harder.  Using separate prefixes for 
static files is much more scalable aproach.

Additionally, this particular snippet is incorrect and misleading 
in serveral ways.

> location ~* \.(ico|css|js|gif|jpe?g|png)(\?[0-9]+)?$ {

Testing "\?..." is meaningless: query string isn't tested by 
"location" directives.

>        expires max;
>        break;

Directive "break" here does nothing but wastes cpu cycles.

>     }

Maxim Dounin


From igor at sysoev.ru  Fri Sep 16 10:39:07 2011
From: igor at sysoev.ru (Igor Sysoev)
Date: Fri, 16 Sep 2011 14:39:07 +0400
Subject: Help with AIO
In-Reply-To: <CAN-ZXpsTEVi3Dk0fQAjye1-AAxkZ6HQ9m6UOkS051A=0HVUL1A@mail.gmail.com>
References: <CAN-ZXpuTdDMmQ7rgo_QP=1W53+j1qc-Bqa=f4=QLNskOubY2dw@mail.gmail.com>
	<20110906073352.GR1137@mdounin.ru>
	<CAN-ZXpsTEVi3Dk0fQAjye1-AAxkZ6HQ9m6UOkS051A=0HVUL1A@mail.gmail.com>
Message-ID: <20110916103907.GC14211@nginx.com>

On Thu, Sep 15, 2011 at 04:21:44PM +0200, Hagai Avrahami wrote:
> Hi
> 
> When benchmarking AIO read, I reached more than 1024 concurrent Async IO -
> 1. I saw a weird error log
> 2. Nginx did not fallback to regular blocking read
> 
> I looked on the source code --  src/os/unix/ngx_linux_aio_read.c  line:105
> and I think  there is a problem in errno handling
> 
> // Current Code
> //----------------------------------------------------------------------------
>     *n = io_submit(ngx_aio_ctx, 1, piocb);
> 
>     if (n == 1) {
>         ev->active = 1;
>         ev->ready = 0;
>         ev->complete = 0;
> 
>         return NGX_AGAIN;
>     }
> 
>     n = -n;    <---------------------------------------------------------??

Thank you, this is a bug. I confused this with _syscall(2).


-- 
Igor Sysoev


From lists at ruby-forum.com  Fri Sep 16 11:13:32 2011
From: lists at ruby-forum.com (Philipp H.)
Date: Fri, 16 Sep 2011 13:13:32 +0200
Subject: proxy_cache_path growing beyond max_size
In-Reply-To: <242477381477ed3a267166929db03b12.NginxMailingListEnglish@forum.nginx.org>
References: <242477381477ed3a267166929db03b12.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <4d58339f158e7bd1ba41229ffff50691@ruby-forum.com>

I have the same problem running nginx 1.0.6 from dotdeb.org

2011/09/16 13:08:50 [alert] 12610#0: epoll_ctl(1, 0) failed (1:
Operation not permitted)
2011/09/16 13:08:50 [alert] 12610#0: failed to register channel handler
while initializing push module worker (1: Operation not permitted)
2011/09/16 13:08:50 [alert] 12611#0: epoll_ctl(1, 0) failed (1:
Operation not permitted)
2011/09/16 13:08:50 [alert] 12611#0: failed to register channel handler
while initializing push module worker (1: Operation not permitted)
2011/09/16 13:08:50 [alert] 12608#0: cache manager process 12610 exited
with fatal code 2 and can not be respawn

There are several posts in the Internet concerning the same problem but
there is no solution :/

https://github.com/slact/nginx_http_push_module/issues/33
http://forum.nginx.org/read.php?21,185003,185003#msg-185003

-- 
Posted via http://www.ruby-forum.com/.


From nginx-forum at nginx.us  Fri Sep 16 11:35:07 2011
From: nginx-forum at nginx.us (Ensiferous)
Date: Fri, 16 Sep 2011 07:35:07 -0400
Subject: Making Wordpress Plugins Work With Nginx
In-Reply-To: <b9dc7fdee421c04252cde760c9f4e595.NginxMailingListEnglish@forum.nginx.org>
References: <b9dc7fdee421c04252cde760c9f4e595.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <a105fe3cb0de536a8c6250f483550771.NginxMailingListEnglish@forum.nginx.org>

Hi,

Nginx does not embed PHP into its own process like Apache does, this
mean that Nginx has no effect on the execution of PHP and if word press
plugins do not work then it's either due to the plugins relying on
Apache via mod_rewrite or because you have misconfigured PHP.

If it's a mod_rewrite problem then we can potentially help, but we'll
need to know what it expects from .htaccess, if it's the latter then
this is probably the wrong mailing list.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215277,215295#msg-215295


From nginx-forum at nginx.us  Fri Sep 16 11:37:18 2011
From: nginx-forum at nginx.us (philipp)
Date: Fri, 16 Sep 2011 07:37:18 -0400
Subject: proxy_cache_path growing beyond max_size
In-Reply-To: <20110825154418.GA1137@mdounin.ru>
References: <20110825154418.GA1137@mdounin.ru>
Message-ID: <734d026f7afc29036645027b8700618d.NginxMailingListEnglish@forum.nginx.org>

I have the same problem here. How can we fix it?

Is this problem known by the developers?

Caching in general is working but without having a working cache manager
I am afraid of running out of disk space...

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,214067,215296#msg-215296


From mdounin at mdounin.ru  Fri Sep 16 12:03:37 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Fri, 16 Sep 2011 16:03:37 +0400
Subject: proxy_cache_path growing beyond max_size
In-Reply-To: <734d026f7afc29036645027b8700618d.NginxMailingListEnglish@forum.nginx.org>
References: <20110825154418.GA1137@mdounin.ru>
	<734d026f7afc29036645027b8700618d.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20110916120336.GN1137@mdounin.ru>

Hello!

On Fri, Sep 16, 2011 at 07:37:18AM -0400, philipp wrote:

> I have the same problem here. How can we fix it?
> 
> Is this problem known by the developers?
> 
> Caching in general is working but without having a working cache manager
> I am afraid of running out of disk space...

Recompiling nginx without third party push module will solve it.  
If you want this to be fixed, you may want to CC push module 
author.

Maxim Dounin


From akunz at wishmedia.de  Fri Sep 16 12:12:22 2011
From: akunz at wishmedia.de (Alexander Kunz)
Date: Fri, 16 Sep 2011 14:12:22 +0200
Subject: proxy_cache_path growing beyond max_size
In-Reply-To: <4d58339f158e7bd1ba41229ffff50691@ruby-forum.com>
References: <242477381477ed3a267166929db03b12.NginxMailingListEnglish@forum.nginx.org>
	<4d58339f158e7bd1ba41229ffff50691@ruby-forum.com>
Message-ID: <4E733D26.60700@wishmedia.de>

Hello,

Am 16.09.2011 13:13, schrieb Philipp H.:
> I have the same problem running nginx 1.0.6 from dotdeb.org
>
> 2011/09/16 13:08:50 [alert] 12610#0: epoll_ctl(1, 0) failed (1:
> Operation not permitted)
> 2011/09/16 13:08:50 [alert] 12610#0: failed to register channel handler
> while initializing push module worker (1: Operation not permitted)
> 2011/09/16 13:08:50 [alert] 12611#0: epoll_ctl(1, 0) failed (1:
> Operation not permitted)
> 2011/09/16 13:08:50 [alert] 12611#0: failed to register channel handler
> while initializing push module worker (1: Operation not permitted)
> 2011/09/16 13:08:50 [alert] 12608#0: cache manager process 12610 exited
> with fatal code 2 and can not be respawn
>
> There are several posts in the Internet concerning the same problem but
> there is no solution :/
>
> https://github.com/slact/nginx_http_push_module/issues/33
> http://forum.nginx.org/read.php?21,185003,185003#msg-185003
>

if you cant fix this issue, take a look at the new http-push-stream-module

https://github.com/wandenberg/nginx-push-stream-module

since version 0.3.0  it also supports long-polling and polling.

Alexander Kunz




From igor at sysoev.ru  Fri Sep 16 14:10:02 2011
From: igor at sysoev.ru (Igor Sysoev)
Date: Fri, 16 Sep 2011 18:10:02 +0400
Subject: Help with AIO
In-Reply-To: <20110916103907.GC14211@nginx.com>
References: <CAN-ZXpuTdDMmQ7rgo_QP=1W53+j1qc-Bqa=f4=QLNskOubY2dw@mail.gmail.com>
	<20110906073352.GR1137@mdounin.ru>
	<CAN-ZXpsTEVi3Dk0fQAjye1-AAxkZ6HQ9m6UOkS051A=0HVUL1A@mail.gmail.com>
	<20110916103907.GC14211@nginx.com>
Message-ID: <20110916141002.GH14211@nginx.com>

On Fri, Sep 16, 2011 at 02:39:07PM +0400, Igor Sysoev wrote:
> On Thu, Sep 15, 2011 at 04:21:44PM +0200, Hagai Avrahami wrote:
> > Hi
> > 
> > When benchmarking AIO read, I reached more than 1024 concurrent Async IO -
> > 1. I saw a weird error log
> > 2. Nginx did not fallback to regular blocking read
> > 
> > I looked on the source code --  src/os/unix/ngx_linux_aio_read.c  line:105
> > and I think  there is a problem in errno handling
> > 
> > // Current Code
> > //----------------------------------------------------------------------------
> >     *n = io_submit(ngx_aio_ctx, 1, piocb);
> > 
> >     if (n == 1) {
> >         ev->active = 1;
> >         ev->ready = 0;
> >         ev->complete = 0;
> > 
> >         return NGX_AGAIN;
> >     }
> > 
> >     n = -n;    <---------------------------------------------------------??
> 
> Thank you, this is a bug. I confused this with _syscall(2).

Linux AIO fixed in these two commits:

http://trac.nginx.org/nginx/changeset/4130/nginx
http://trac.nginx.org/nginx/changeset/4131/nginx

Also I'm going to introduce the directive:

events {
    worker_aio_events 100;
}

to set how much aio events worker may request to the kernel.
I'm going to set default value to 32 events per worker.
Currenly this value is hardcoded to 1024 and it seems it is too large,
since default /proc/sys/fs/aio-max-nr value is 65536 and this means
that you can run no more than 64 workers.


-- 
Igor Sysoev


From wendal1985 at gmail.com  Fri Sep 16 14:34:59 2011
From: wendal1985 at gmail.com (Wendal Chen)
Date: Fri, 16 Sep 2011 22:34:59 +0800
Subject: Segmentation fault in drizzle-nginx-module 0.1.0
Message-ID: <CAM79QPVPtt6oEZyB23sk=dSCRHgNLEXeerP=OjnPpOiOgT_U+Q@mail.gmail.com>

Env:
CentOS 5.6
nginx 0.8.54
rds-json-nginx-module-0.11
mysql 5.0
drizzle7-2011.08.25

nginx.conf

        location =/mysql_status {
            drizzle_status;
        }

        #Common SQL Query point
        location =/mysql_query {
          #internal;
            set_by_lua $sql "return ngx.unescape_uri(ngx.var.arg_sql)";
            drizzle_query $sql;
            drizzle_pass mysqls;
            drizzle_module_header off;
            rds_json on;
        }

===========================================================================================================

request path : localhost/mysql_query?sql=SELECT%201

print out by gdb:

Program terminated with signal 11, Segmentation fault.
#0  0x080c2551 in ngx_http_upstream_dbd_connect (r=0x8badcd0, u=0x8bae2d4)
at ../drizzle-nginx-module/src/ngx_http_drizzle_util.c:604
604         r->connection->log->action = "connecting to upstream";
(gdb) bt
#0  0x080c2551 in ngx_http_upstream_dbd_connect (r=0x8badcd0, u=0x8bae2d4)
at ../drizzle-nginx-module/src/ngx_http_drizzle_util.c:604
#1  0x080c2539 in ngx_http_upstream_dbd_init_request (r=0x8badcd0) at
../drizzle-nginx-module/src/ngx_http_drizzle_util.c:593
#2  0x080c2066 in ngx_http_upstream_dbd_init (r=0x8badcd0) at
../drizzle-nginx-module/src/ngx_http_drizzle_util.c:390
#3  0x080be58e in ngx_http_drizzle_handler (r=0x8badcd0) at
../drizzle-nginx-module/src/ngx_http_drizzle_handler.c:164
#4  0x0807d610 in ngx_http_core_content_phase (r=0x8badcd0, ph=0x8bbb6d8) at
src/http/ngx_http_core_module.c:1350
#5  0x0807c8a9 in ngx_http_core_run_phases (r=0x8badcd0) at
src/http/ngx_http_core_module.c:852
#6  0x0807c844 in ngx_http_handler (r=0x8badcd0) at
src/http/ngx_http_core_module.c:835
#7  0x08086a03 in ngx_http_run_posted_requests (c=0xb6a900b8) at
src/http/ngx_http_request.c:1828
#8  0x080866b1 in ngx_http_process_request (r=0x8ba6368) at
src/http/ngx_http_request.c:1643
#9  0x080855b9 in ngx_http_process_request_headers (rev=0xb657b070) at
src/http/ngx_http_request.c:1084
#10 0x08084f4e in ngx_http_process_request_line (rev=0xb657b070) at
src/http/ngx_http_request.c:889
#11 0x0808493b in ngx_http_init_request (rev=0xb657b070) at
src/http/ngx_http_request.c:514
#12 0x0806f59f in ngx_event_process_posted (cycle=0x8ba6e98,
posted=0x80df6cc) at src/event/ngx_event_posted.c:39
#13 0x0806ddf3 in ngx_process_events_and_timers (cycle=0x8ba6e98) at
src/event/ngx_event.c:272
#14 0x08076c50 in ngx_worker_process_cycle (cycle=0x8ba6e98, data=0x0) at
src/os/unix/ngx_process_cycle.c:795
#15 0x080746a2 in ngx_spawn_process (cycle=0x8ba6e98, proc=0x8076b3a
<ngx_worker_process_cycle>, data=0x0, name=0x80cdc44 "worker process",
respawn=-3)
    at src/os/unix/ngx_process.c:196
#16 0x080761b9 in ngx_start_worker_processes (cycle=0x8ba6e98, n=2, type=-3)
at src/os/unix/ngx_process_cycle.c:355
#17 0x08075ac2 in ngx_master_process_cycle (cycle=0x8ba6e98) at
src/os/unix/ngx_process_cycle.c:136
#18 0x08053faf in main (argc=1, argv=0xbfe971b4) at src/core/nginx.c:401

============================================================================================================

request path : /mysql_status

===========================================================================================================
#0  0x080bef9b in ngx_http_drizzle_status_handler (r=0x8ba6368) at
../drizzle-nginx-module/src/ngx_http_drizzle_handler.c:604
604             ngx_log_error(NGX_LOG_ALERT, r->connection->log, 0,
(gdb) bt
#0  0x080bef9b in ngx_http_drizzle_status_handler (r=0x8ba6368) at
../drizzle-nginx-module/src/ngx_http_drizzle_handler.c:604
#1  0x0807d610 in ngx_http_core_content_phase (r=0x8ba6368, ph=0x8bbb6d8) at
src/http/ngx_http_core_module.c:1350
#2  0x0807c8a9 in ngx_http_core_run_phases (r=0x8ba6368) at
src/http/ngx_http_core_module.c:852
#3  0x0807c844 in ngx_http_handler (r=0x8ba6368) at
src/http/ngx_http_core_module.c:835
#4  0x080866a6 in ngx_http_process_request (r=0x8ba6368) at
src/http/ngx_http_request.c:1641
#5  0x080855b9 in ngx_http_process_request_headers (rev=0xb657b070) at
src/http/ngx_http_request.c:1084
#6  0x08084f4e in ngx_http_process_request_line (rev=0xb657b070) at
src/http/ngx_http_request.c:889
#7  0x0808493b in ngx_http_init_request (rev=0xb657b070) at
src/http/ngx_http_request.c:514
#8  0x0806f59f in ngx_event_process_posted (cycle=0x8ba6e98,
posted=0x80df6cc) at src/event/ngx_event_posted.c:39
#9  0x0806ddf3 in ngx_process_events_and_timers (cycle=0x8ba6e98) at
src/event/ngx_event.c:272
#10 0x08076c50 in ngx_worker_process_cycle (cycle=0x8ba6e98, data=0x0) at
src/os/unix/ngx_process_cycle.c:795
#11 0x080746a2 in ngx_spawn_process (cycle=0x8ba6e98, proc=0x8076b3a
<ngx_worker_process_cycle>, data=0x0, name=0x80cdc44 "worker process",
respawn=-3)
    at src/os/unix/ngx_process.c:196
#12 0x080761b9 in ngx_start_worker_processes (cycle=0x8ba6e98, n=2, type=-3)
at src/os/unix/ngx_process_cycle.c:355
#13 0x08075ac2 in ngx_master_process_cycle (cycle=0x8ba6e98) at
src/os/unix/ngx_process_cycle.c:136
#14 0x08053faf in main (argc=1, argv=0xbfe971b4) at src/core/nginx.c:401

===========================================================================================================
Thank any help !!!

-- 
Wendal Chen
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110916/17012b9b/attachment.html>

From wendal1985 at gmail.com  Fri Sep 16 16:00:27 2011
From: wendal1985 at gmail.com (Wendal Chen)
Date: Sat, 17 Sep 2011 00:00:27 +0800
Subject: Segmentation fault in drizzle-nginx-module 0.1.0
In-Reply-To: <CAM79QPVPtt6oEZyB23sk=dSCRHgNLEXeerP=OjnPpOiOgT_U+Q@mail.gmail.com>
References: <CAM79QPVPtt6oEZyB23sk=dSCRHgNLEXeerP=OjnPpOiOgT_U+Q@mail.gmail.com>
Message-ID: <CAM79QPU7A_Kp7-zwu89bbK4Qo7isuJgBAZ9GZaP-RWrSW0-6Aw@mail.gmail.com>

switch to:

ngx_openresty-1.0.5.1 + drizzle7-2011.08.25

work find.



2011/9/16 Wendal Chen <wendal1985 at gmail.com>

> Env:
> CentOS 5.6
> nginx 0.8.54
> rds-json-nginx-module-0.11
> mysql 5.0
> drizzle7-2011.08.25
>
> nginx.conf
>
>         location =/mysql_status {
>             drizzle_status;
>         }
>
>         #Common SQL Query point
>         location =/mysql_query {
>           #internal;
>             set_by_lua $sql "return ngx.unescape_uri(ngx.var.arg_sql)";
>             drizzle_query $sql;
>             drizzle_pass mysqls;
>             drizzle_module_header off;
>             rds_json on;
>         }
>
>
> ===========================================================================================================
>
> request path : localhost/mysql_query?sql=SELECT%201
>
> print out by gdb:
>
> Program terminated with signal 11, Segmentation fault.
> #0  0x080c2551 in ngx_http_upstream_dbd_connect (r=0x8badcd0, u=0x8bae2d4)
> at ../drizzle-nginx-module/src/ngx_http_drizzle_util.c:604
> 604         r->connection->log->action = "connecting to upstream";
> (gdb) bt
> #0  0x080c2551 in ngx_http_upstream_dbd_connect (r=0x8badcd0, u=0x8bae2d4)
> at ../drizzle-nginx-module/src/ngx_http_drizzle_util.c:604
> #1  0x080c2539 in ngx_http_upstream_dbd_init_request (r=0x8badcd0) at
> ../drizzle-nginx-module/src/ngx_http_drizzle_util.c:593
> #2  0x080c2066 in ngx_http_upstream_dbd_init (r=0x8badcd0) at
> ../drizzle-nginx-module/src/ngx_http_drizzle_util.c:390
> #3  0x080be58e in ngx_http_drizzle_handler (r=0x8badcd0) at
> ../drizzle-nginx-module/src/ngx_http_drizzle_handler.c:164
> #4  0x0807d610 in ngx_http_core_content_phase (r=0x8badcd0, ph=0x8bbb6d8)
> at src/http/ngx_http_core_module.c:1350
> #5  0x0807c8a9 in ngx_http_core_run_phases (r=0x8badcd0) at
> src/http/ngx_http_core_module.c:852
> #6  0x0807c844 in ngx_http_handler (r=0x8badcd0) at
> src/http/ngx_http_core_module.c:835
> #7  0x08086a03 in ngx_http_run_posted_requests (c=0xb6a900b8) at
> src/http/ngx_http_request.c:1828
> #8  0x080866b1 in ngx_http_process_request (r=0x8ba6368) at
> src/http/ngx_http_request.c:1643
> #9  0x080855b9 in ngx_http_process_request_headers (rev=0xb657b070) at
> src/http/ngx_http_request.c:1084
> #10 0x08084f4e in ngx_http_process_request_line (rev=0xb657b070) at
> src/http/ngx_http_request.c:889
> #11 0x0808493b in ngx_http_init_request (rev=0xb657b070) at
> src/http/ngx_http_request.c:514
> #12 0x0806f59f in ngx_event_process_posted (cycle=0x8ba6e98,
> posted=0x80df6cc) at src/event/ngx_event_posted.c:39
> #13 0x0806ddf3 in ngx_process_events_and_timers (cycle=0x8ba6e98) at
> src/event/ngx_event.c:272
> #14 0x08076c50 in ngx_worker_process_cycle (cycle=0x8ba6e98, data=0x0) at
> src/os/unix/ngx_process_cycle.c:795
> #15 0x080746a2 in ngx_spawn_process (cycle=0x8ba6e98, proc=0x8076b3a
> <ngx_worker_process_cycle>, data=0x0, name=0x80cdc44 "worker process",
> respawn=-3)
>     at src/os/unix/ngx_process.c:196
> #16 0x080761b9 in ngx_start_worker_processes (cycle=0x8ba6e98, n=2,
> type=-3) at src/os/unix/ngx_process_cycle.c:355
> #17 0x08075ac2 in ngx_master_process_cycle (cycle=0x8ba6e98) at
> src/os/unix/ngx_process_cycle.c:136
> #18 0x08053faf in main (argc=1, argv=0xbfe971b4) at src/core/nginx.c:401
>
>
> ============================================================================================================
>
> request path : /mysql_status
>
>
> ===========================================================================================================
> #0  0x080bef9b in ngx_http_drizzle_status_handler (r=0x8ba6368) at
> ../drizzle-nginx-module/src/ngx_http_drizzle_handler.c:604
> 604             ngx_log_error(NGX_LOG_ALERT, r->connection->log, 0,
> (gdb) bt
> #0  0x080bef9b in ngx_http_drizzle_status_handler (r=0x8ba6368) at
> ../drizzle-nginx-module/src/ngx_http_drizzle_handler.c:604
> #1  0x0807d610 in ngx_http_core_content_phase (r=0x8ba6368, ph=0x8bbb6d8)
> at src/http/ngx_http_core_module.c:1350
> #2  0x0807c8a9 in ngx_http_core_run_phases (r=0x8ba6368) at
> src/http/ngx_http_core_module.c:852
> #3  0x0807c844 in ngx_http_handler (r=0x8ba6368) at
> src/http/ngx_http_core_module.c:835
> #4  0x080866a6 in ngx_http_process_request (r=0x8ba6368) at
> src/http/ngx_http_request.c:1641
> #5  0x080855b9 in ngx_http_process_request_headers (rev=0xb657b070) at
> src/http/ngx_http_request.c:1084
> #6  0x08084f4e in ngx_http_process_request_line (rev=0xb657b070) at
> src/http/ngx_http_request.c:889
> #7  0x0808493b in ngx_http_init_request (rev=0xb657b070) at
> src/http/ngx_http_request.c:514
> #8  0x0806f59f in ngx_event_process_posted (cycle=0x8ba6e98,
> posted=0x80df6cc) at src/event/ngx_event_posted.c:39
> #9  0x0806ddf3 in ngx_process_events_and_timers (cycle=0x8ba6e98) at
> src/event/ngx_event.c:272
> #10 0x08076c50 in ngx_worker_process_cycle (cycle=0x8ba6e98, data=0x0) at
> src/os/unix/ngx_process_cycle.c:795
> #11 0x080746a2 in ngx_spawn_process (cycle=0x8ba6e98, proc=0x8076b3a
> <ngx_worker_process_cycle>, data=0x0, name=0x80cdc44 "worker process",
> respawn=-3)
>     at src/os/unix/ngx_process.c:196
> #12 0x080761b9 in ngx_start_worker_processes (cycle=0x8ba6e98, n=2,
> type=-3) at src/os/unix/ngx_process_cycle.c:355
> #13 0x08075ac2 in ngx_master_process_cycle (cycle=0x8ba6e98) at
> src/os/unix/ngx_process_cycle.c:136
> #14 0x08053faf in main (argc=1, argv=0xbfe971b4) at src/core/nginx.c:401
>
>
> ===========================================================================================================
> Thank any help !!!
>
> --
> Wendal Chen
>
>
>


-- 
Wendal Chen
<http://t.sina.com.cn/1068040557?s=6uyXnP>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110917/81d0092c/attachment-0001.html>

From nginx-forum at nginx.us  Fri Sep 16 16:16:56 2011
From: nginx-forum at nginx.us (martijnsenden)
Date: Fri, 16 Sep 2011 12:16:56 -0400
Subject: change nginx user in config >> 403 error index.html forbidden
Message-ID: <17e343cd5e06fc4d31ab27ccf4006c74.NginxMailingListEnglish@forum.nginx.org>

Hi All,

I have just installed NGINX 1.1.2. I made my own configuration and
makefile and installed the nginx server.
When I set the config to:
user root root;
I get the hello world index.html file provided by the nginx installer
(./html/index.html).

When I set the config to:
user nginx nginx;
(which is what I want), I get a 403 message. The error.log says the
index.html is forbidden.

These are the file permissions I have set on the files in the ./html
directory:
total 16
-rw-r--r-- 1 someowner nginx  383 Sep 16 15:27 50x.html
-rw-r--r-- 1 someowner nginx  151 Sep 16 15:27 index.html
-rw-r--r-- 1 someowner nginx 6971 Sep 16 15:27 ngx_core_module.html

The nginx user is a member of the nginx group (which has the required
permission to view the index.html):
$ groups nginx
returns:
nginx : nginx

So, what should I do to give the worker processes of nginx access to
these files? I'm at a loss here.

I would greatly appreciate any help!

Best regards,
Martijn.

ps this is the content of my nginx.conf:

user  root root;
worker_processes  4;

#error_log  logs/error.log;
#error_log  logs/error.log  notice;
#error_log  logs/error.log  info;

#pid        logs/nginx.pid;


events {
    worker_connections  128;
}


http {
    include       mime.types;
    default_type  application/octet-stream;


    #log_format  main  '$remote_addr - $remote_user [$time_local]
"$request" '
    #                  '$status $body_bytes_sent "$http_referer" '
    #                  '"$http_user_agent" "$http_x_forwarded_for"';

    #access_log  logs/access.log  main;

    sendfile        on;
    #tcp_nopush     on;

    #keepalive_timeout  0;
    keepalive_timeout  65;

    #gzip  on;

    server {
        listen       80;
        server_name  localhost;

        #charset koi8-r;

        #access_log  logs/host.access.log  main;

        location / {
            root   html;
            index  index.html index.htm;
        }

        #error_page  404              /404.html;

        # redirect server error pages to the static page /50x.html
        #
        error_page   500 502 503 504  /50x.html;
        location = /50x.html {
            root   html;
        }

        # proxy the PHP scripts to Apache listening on 127.0.0.1:80
        #
        #location ~ \.php$ {
        #    proxy_pass   http://127.0.0.1;
        #}

        # pass the PHP scripts to FastCGI server listening on
127.0.0.1:9000
        #
        #location ~ \.php$ {
        #    root           html;
        #    fastcgi_pass   127.0.0.1:9000;
        #    fastcgi_index  index.php;
        #    fastcgi_param  SCRIPT_FILENAME 
/scripts$fastcgi_script_name;
        #    include        fastcgi_params;
        #}

        # deny access to .htaccess files, if Apache's document root
        # concurs with nginx's one
        #
        #location ~ /\.ht {
        #
        error_page   500 502 503 504  /50x.html;
        location = /50x.html {
            root   html;
        }

        # proxy the PHP scripts to Apache listening on 127.0.0.1:80
        #
        #location ~ \.php$ {
        #    proxy_pass   http://127.0.0.1;
        #}

        # pass the PHP scripts to FastCGI server listening on
127.0.0.1:9000
        #
        #location ~ \.php$ {
        #    root           html;
        #    fastcgi_pass   127.0.0.1:9000;
        #    fastcgi_index  index.php;
        #    fastcgi_param  SCRIPT_FILENAME 
/scripts$fastcgi_script_name;
        #    include        fastcgi_params;
        #}

        # deny access to .htaccess files, if Apache's document root
        # concurs with nginx's one
        #
        #location ~ /\.ht {
        #    deny  all;
        #}
    }


    # another virtual host using mix of IP-, name-, and port-based
configuration
    #
    #server {
    #    listen       8000;
    #    listen       somename:8080;
    #    server_name  somename  alias  another.alias;

    #    location / {
    #        root   html;
    #        index  index.html index.htm;
    #    }
    #}


    # HTTPS server
    #
    #server {
    #    listen       443;
    #    server_name  localhost;

    #    ssl                  on;
    #    ssl_certificate      cert.pem;
    #    ssl_certificate_key  cert.key;

    #    ssl_session_timeout  5m;

    #    ssl_protocols  SSLv2 SSLv3 TLSv1;
    #    ssl_ciphers  HIGH:!aNULL:!MD5;
    #    ssl_prefer_server_ciphers   on;

    #    location / {
    #        root   html;
    #        index  index.html index.htm;
    #    }
    #}

}

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215310,215310#msg-215310


From mdounin at mdounin.ru  Fri Sep 16 16:24:35 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Fri, 16 Sep 2011 20:24:35 +0400
Subject: change nginx user in config >> 403 error index.html forbidden
In-Reply-To: <17e343cd5e06fc4d31ab27ccf4006c74.NginxMailingListEnglish@forum.nginx.org>
References: <17e343cd5e06fc4d31ab27ccf4006c74.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20110916162435.GT1137@mdounin.ru>

Hello!

On Fri, Sep 16, 2011 at 12:16:56PM -0400, martijnsenden wrote:

> Hi All,
> 
> I have just installed NGINX 1.1.2. I made my own configuration and
> makefile and installed the nginx server.
> When I set the config to:
> user root root;
> I get the hello world index.html file provided by the nginx installer
> (./html/index.html).
> 
> When I set the config to:
> user nginx nginx;
> (which is what I want), I get a 403 message. The error.log says the
> index.html is forbidden.
> 
> These are the file permissions I have set on the files in the ./html
> directory:
> total 16
> -rw-r--r-- 1 someowner nginx  383 Sep 16 15:27 50x.html
> -rw-r--r-- 1 someowner nginx  151 Sep 16 15:27 index.html
> -rw-r--r-- 1 someowner nginx 6971 Sep 16 15:27 ngx_core_module.html
> 
> The nginx user is a member of the nginx group (which has the required
> permission to view the index.html):
> $ groups nginx
> returns:
> nginx : nginx
> 
> So, what should I do to give the worker processes of nginx access to
> these files? I'm at a loss here.

Make sure nginx use have at least search permissions for all 
directories above document root.

Maxim Dounin


From lists at ruby-forum.com  Fri Sep 16 16:25:58 2011
From: lists at ruby-forum.com (dhaval deshpande)
Date: Fri, 16 Sep 2011 18:25:58 +0200
Subject: Passing variables from nginx to sinatra
Message-ID: <39ffba646ba6d273ee6425b528fb6d0f@ruby-forum.com>

Hello,
       I am trying to pass a uID($uid_got value) that has been generated
by Nginx httpuseridmodule to my sinatra app.

Basically I could have taken from the cookie but the cookie does not get
set on the browser unless the response headers have been sent back.

Can anyone give any suggestions regarding that?

Thanks,
Dhaval.

-- 
Posted via http://www.ruby-forum.com/.


From nginx-forum at nginx.us  Fri Sep 16 18:47:49 2011
From: nginx-forum at nginx.us (dullnicker)
Date: Fri, 16 Sep 2011 14:47:49 -0400
Subject: Problem with Hotlink Protection
In-Reply-To: <0c253481fec8974edd546fbde9b7c383.NginxMailingListEnglish@forum.nginx.org>
References: <0c253481fec8974edd546fbde9b7c383.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <beea1075bf2bd47f45365417af71e225.NginxMailingListEnglish@forum.nginx.org>

Thank you both! I totally agree with you, but as soon as I restrict
referer checking to the image.php script only, there is a flaw
appearing: image.php then gets downloaded by the browser instead of
displaying it. Do you have any idea why this happens? It would be great
if I could restrict the check only to this one .php file, but - as said
- it does not work here...

Thank you again for your kind help and support,
Amitz

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215198,215316#msg-215316


From fud.theturtle at gmail.com  Fri Sep 16 19:54:08 2011
From: fud.theturtle at gmail.com (Chris Zimmerman)
Date: Fri, 16 Sep 2011 15:54:08 -0400
Subject: Problem with Hotlink Protection
In-Reply-To: <beea1075bf2bd47f45365417af71e225.NginxMailingListEnglish@forum.nginx.org>
References: <0c253481fec8974edd546fbde9b7c383.NginxMailingListEnglish@forum.nginx.org>
	<beea1075bf2bd47f45365417af71e225.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <CAOo_=6mpmkYCE3HF1YbUx4bv885vsZkyEXWVRLD_3DrF0Qyv+Q@mail.gmail.com>

Please post the config you are using when it does this.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110916/e135efff/attachment.html>

From lists at ruby-forum.com  Fri Sep 16 20:29:03 2011
From: lists at ruby-forum.com (dhaval deshpande)
Date: Fri, 16 Sep 2011 22:29:03 +0200
Subject: Passing variables from nginx to sinatra
In-Reply-To: <39ffba646ba6d273ee6425b528fb6d0f@ruby-forum.com>
References: <39ffba646ba6d273ee6425b528fb6d0f@ruby-forum.com>
Message-ID: <210b425c0fcd6df385284e7ba61b67ff@ruby-forum.com>

ok never mind I figured out my self. Just in case if anyone wants to
know here is how I did it.

passenger_set_cgi_param Variable Value

-- 
Posted via http://www.ruby-forum.com/.


From patcito at gmail.com  Fri Sep 16 20:29:23 2011
From: patcito at gmail.com (Patrick Aljord)
Date: Fri, 16 Sep 2011 15:29:23 -0500
Subject: can't get cache controle to set an expire date
In-Reply-To: <20110916101951.GK1137@mdounin.ru>
References: <CAK9TW4rD2BGnA3s96BH+TyW9es729CVh5zkBci8eaNzOuaLE8Q@mail.gmail.com>
	<1c65c6dd406143a6673d311145a09842@ruby-forum.com>
	<20110916101951.GK1137@mdounin.ru>
Message-ID: <CAK9TW4oTtoUBs0EaToS8Mv3pnKgKeEhdoBB9uVcenhvKtU+w-A@mail.gmail.com>

Hey Maxim,

Ok so after investigation I realized that this one worked:

/images/q-icon-medium.png

But these two did not have expire cache:

/packages/base.js
/_files/groups/medium/4af798a119ce955bd1000001/0.png

So I added this to the conf:


? ?location /packages/ {? ? ? ?expires max;? ?}
   location /_files/ {
       expires max;
   }

This fixed the '/packages/base.js' cache but I started getting a 404
on the '/_files/groups/medium/4af798a119ce955bd1000001/0.png'. So I
removed the "location /_files/" from the conf. The thing is that this
file does not come from the hard drive. It comes from the web app (I'm
using unicorn for rails + nginx). Any idea how I should set the expire
headers for those assets that come from the app? Should I set them in
the rails app?

Thanks,

Pat


From mdounin at mdounin.ru  Fri Sep 16 22:51:42 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Sat, 17 Sep 2011 02:51:42 +0400
Subject: can't get cache controle to set an expire date
In-Reply-To: <CAK9TW4oTtoUBs0EaToS8Mv3pnKgKeEhdoBB9uVcenhvKtU+w-A@mail.gmail.com>
References: <CAK9TW4rD2BGnA3s96BH+TyW9es729CVh5zkBci8eaNzOuaLE8Q@mail.gmail.com>
	<1c65c6dd406143a6673d311145a09842@ruby-forum.com>
	<20110916101951.GK1137@mdounin.ru>
	<CAK9TW4oTtoUBs0EaToS8Mv3pnKgKeEhdoBB9uVcenhvKtU+w-A@mail.gmail.com>
Message-ID: <20110916225142.GU1137@mdounin.ru>

Hello!

On Fri, Sep 16, 2011 at 03:29:23PM -0500, Patrick Aljord wrote:

> Hey Maxim,
> 
> Ok so after investigation I realized that this one worked:
> 
> /images/q-icon-medium.png
> 
> But these two did not have expire cache:
> 
> /packages/base.js
> /_files/groups/medium/4af798a119ce955bd1000001/0.png
> 
> So I added this to the conf:
> 
> 
> ? ?location /packages/ {? ? ? ?expires max;? ?}
>    location /_files/ {
>        expires max;
>    }
> 
> This fixed the '/packages/base.js' cache but I started getting a 404
> on the '/_files/groups/medium/4af798a119ce955bd1000001/0.png'. So I
> removed the "location /_files/" from the conf. The thing is that this
> file does not come from the hard drive. It comes from the web app (I'm
> using unicorn for rails + nginx). Any idea how I should set the expire
> headers for those assets that come from the app? Should I set them in
> the rails app?

Either set it from app, or pass to app while still setting 
expires.  Something like this should work:

    location /_files/ {
        expires max;
        proxy_pass http://foo_app_server;
        proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
        proxy_set_header Host $http_host;
    }

If some data in /_files/ may be actually static files on disk, you 
may want to add try_files with named location, much like in 
"location /" + "location @app" in your config, but with expires 
set.  I.e. something like this:

    location /_files/ {
        expires max;
        try_files ... @appstatic;
    }

    location @appstatic {
        expires max;
        proxy_pass http://foo_app_server;
        proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
        proxy_set_header Host $http_host;
    }

Maxim Dounin


From patcito at gmail.com  Sat Sep 17 00:09:32 2011
From: patcito at gmail.com (Patrick Aljord)
Date: Fri, 16 Sep 2011 19:09:32 -0500
Subject: can't get cache controle to set an expire date
In-Reply-To: <20110916225142.GU1137@mdounin.ru>
References: <CAK9TW4rD2BGnA3s96BH+TyW9es729CVh5zkBci8eaNzOuaLE8Q@mail.gmail.com>
	<1c65c6dd406143a6673d311145a09842@ruby-forum.com>
	<20110916101951.GK1137@mdounin.ru>
	<CAK9TW4oTtoUBs0EaToS8Mv3pnKgKeEhdoBB9uVcenhvKtU+w-A@mail.gmail.com>
	<20110916225142.GU1137@mdounin.ru>
Message-ID: <CAK9TW4pW9LvMa2VoyzOvGi1DuXxF1VDd0uueA3CFzcbXvG2HEQ@mail.gmail.com>

2011/9/16 Maxim Dounin <mdounin at mdounin.ru>:
> ? ?location /_files/ {
> ? ? ? ?expires max;
> ? ? ? ?proxy_pass http://foo_app_server;
> ? ? ? ?proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
> ? ? ? ?proxy_set_header Host $http_host;
> ? ?}
>


Thanks! That worked great.


From nginx-forum at nginx.us  Sat Sep 17 00:25:01 2011
From: nginx-forum at nginx.us (damnlinux)
Date: Fri, 16 Sep 2011 20:25:01 -0400
Subject: Making Wordpress Plugins Work With Nginx
In-Reply-To: <a105fe3cb0de536a8c6250f483550771.NginxMailingListEnglish@forum.nginx.org>
References: <b9dc7fdee421c04252cde760c9f4e595.NginxMailingListEnglish@forum.nginx.org>
	<a105fe3cb0de536a8c6250f483550771.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <f3dc419f96fde2015b2cfa5ca7eb265d.NginxMailingListEnglish@forum.nginx.org>

Thank you Ensiferous.

I don't think it's a PHP issue because Wordpress itself works fine it is
quite possible though that it's a mod_rewrite issue. I'm not familiar
with Nginx access rewrite my .htaccess is really basic


# BEGIN WordPress
<IfModule mod_rewrite.c>
RewriteEngine On
RewriteBase /
RewriteRule ^index\.php$ - [L]
RewriteCond %{REQUEST_FILENAME} !-f
RewriteCond %{REQUEST_FILENAME} !-d
RewriteRule . /index.php [L]
</IfModule>

# END WordPress
RewriteEngine On
RewriteCond %{HTTP_REFERER} !^http://(www\.)?mydomain\.com/ [NC]
RewriteCond %{HTTP_REFERER} ^http:// [NC]
RewriteCond %{HTTP_REFERER} !^$
RewriteRule \.(jpe?g|gif|png)$ - [F]

Thanks again for the help

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215277,215319#msg-215319


From nginx-forum at nginx.us  Sat Sep 17 04:00:30 2011
From: nginx-forum at nginx.us (Sylvia)
Date: Sat, 17 Sep 2011 00:00:30 -0400
Subject: Making Wordpress Plugins Work With Nginx
In-Reply-To: <f3dc419f96fde2015b2cfa5ca7eb265d.NginxMailingListEnglish@forum.nginx.org>
References: <b9dc7fdee421c04252cde760c9f4e595.NginxMailingListEnglish@forum.nginx.org>
	<a105fe3cb0de536a8c6250f483550771.NginxMailingListEnglish@forum.nginx.org>
	<f3dc419f96fde2015b2cfa5ca7eb265d.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <18010d6d655555324c6c51a637892311.NginxMailingListEnglish@forum.nginx.org>

http://wordpress.org/extend/plugins/nginx-compatibility/
try installing this plugin when WP detects nginx and "not compilant"
SAPI it assumes that rewrite is impossible and turns it off,
this plugins works around that problem. (but do not use it with Apache
if you move your WP somewhere else later)

~ cheers

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215277,215320#msg-215320


From yaniv at aknin.name  Sat Sep 17 04:40:19 2011
From: yaniv at aknin.name (Yaniv Aknin)
Date: Sat, 17 Sep 2011 07:40:19 +0300
Subject: daemon off in production
Message-ID: <CAPzVZKk8HJOoo6adQsuq2KVkpA_t3qad-rBZ0EGdpSS6D1Wikg@mail.gmail.com>

Hi,

The documentations says I should never run nginx with the daemon off setting
in production. I would like to do so anyway so my Python/twisted based
process manager will have an easier time knowing if nginx terminated or not
(I know there are other ways to do it, but this seems like the easiest; if
I'll be forced to, I'll use another way).

I'm aware of SIGWINCH and nginx in general and especially after version
1.1.1, and I don't mind taking care of guaranteeing nginx will run without a
controlling terminal (I will fork()/setsid() before exec()ing it), run in a
sensible current working directory, etc. I just don't want it to do the
daemonizing double fork.

>From skimming the code, it appears to me that nothing bad should happen.
What am I missing? Why is it so bad to run nginx in the foreground in
production?

Thanks,
 - Yaniv
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110917/7bbfb04f/attachment.html>

From agentzh at gmail.com  Sat Sep 17 04:42:56 2011
From: agentzh at gmail.com (agentzh)
Date: Sat, 17 Sep 2011 12:42:56 +0800
Subject: Segmentation fault in drizzle-nginx-module 0.1.0
In-Reply-To: <CAM79QPVPtt6oEZyB23sk=dSCRHgNLEXeerP=OjnPpOiOgT_U+Q@mail.gmail.com>
References: <CAM79QPVPtt6oEZyB23sk=dSCRHgNLEXeerP=OjnPpOiOgT_U+Q@mail.gmail.com>
Message-ID: <CAB4Tn6Nuvx63wjPh1M9BDb=kvgcj3pTxMXteQBqGWn169sEhww@mail.gmail.com>

On Fri, Sep 16, 2011 at 10:34 PM, Wendal Chen <wendal1985 at gmail.com> wrote:
>
> Program terminated with signal 11, Segmentation fault.
> #0? 0x080c2551 in ngx_http_upstream_dbd_connect (r=0x8badcd0, u=0x8bae2d4)
> at ../drizzle-nginx-module/src/ngx_http_drizzle_util.c:604
> 604???????? r->connection->log->action = "connecting to upstream";
> (gdb) bt

This was a known issue in ngx_drizzle v0.1.0 and earlier on 32-bit
systems and has already been fixed in ngx_drizzle v0.1.1rc3. I'll
release ngx_drizzle v0.1.1 soon :)

Regards,
-agentzh


From agentzh at gmail.com  Sat Sep 17 04:44:48 2011
From: agentzh at gmail.com (agentzh)
Date: Sat, 17 Sep 2011 12:44:48 +0800
Subject: Segmentation fault in drizzle-nginx-module 0.1.0
In-Reply-To: <CAM79QPU7A_Kp7-zwu89bbK4Qo7isuJgBAZ9GZaP-RWrSW0-6Aw@mail.gmail.com>
References: <CAM79QPVPtt6oEZyB23sk=dSCRHgNLEXeerP=OjnPpOiOgT_U+Q@mail.gmail.com>
	<CAM79QPU7A_Kp7-zwu89bbK4Qo7isuJgBAZ9GZaP-RWrSW0-6Aw@mail.gmail.com>
Message-ID: <CAB4Tn6O=eEzBktonq_y4GCcv1D0riedsR2Yj28cWBTDX9dF+zQ@mail.gmail.com>

On Sat, Sep 17, 2011 at 12:00 AM, Wendal Chen <wendal1985 at gmail.com> wrote:
>
> switch to:
>
> ngx_openresty-1.0.5.1 + drizzle7-2011.08.25
>
> work find.
>

Cool :)

The ngx_openresty 1.0.5.1 release bundles ngx_drizzle 0.1.1rc4, which
already contains the fix for 32-bit systems. See
http://openresty.org/#ChangeLog1000005

Regards,
-agentzh


From agentzh at gmail.com  Sat Sep 17 05:14:14 2011
From: agentzh at gmail.com (agentzh)
Date: Sat, 17 Sep 2011 13:14:14 +0800
Subject: [ANN] ngx_drizzle v0.1.1: fixed segfaults on 32-bit systems
In-Reply-To: <CAB4Tn6PTe6p9LtcWxpMFrshDt64ova+2ziugVN1sTw-SJR8AQA@mail.gmail.com>
References: <CAB4Tn6PTe6p9LtcWxpMFrshDt64ova+2ziugVN1sTw-SJR8AQA@mail.gmail.com>
Message-ID: <CAB4Tn6M9uAdbWGy8hc8C2evOLB5ySSb+UX749m3w9oDo0WS9Ww@mail.gmail.com>

Hi, folks!

I'm happy to announce that ngx_drizzle v0.1.1 has been released. You
can download the latest release tarball from

??? https://github.com/chaoslawful/drizzle-nginx-module/downloads

This release fixed a long-standing segmentation fault on 32-bit
systems, as well as some other issues. Here goes the complete change
log:

* fixed segmentation faults on 32-bit systems. thanks @??? and
@stefanli for reporting this issue.
* documented the $drizzle_thread_id variable.
* added lots of debug outputs (enabled by the --with-debug option
while building Nginx or OpenResty), inspired by github issue #10.
* fixed issues regarding defining global variables in C header files:
we should have defined rds_rough_col_type_t as a type rather than a
global variable. thanks @???.
* documented the python -> python3 pitfall while building libdrizzle
1.0 on at least ArchLinux.
* fixed the automatic libdrizzle searching algorithm in the config
file: now we should look under libdrizzle-1.0/ instead. thanks ???
(Calio) for reporting this issue.

ngx_drizzle is an nginx upstream module integrating libdrizzle into
Nginx in a non-blocking and streamming way.

Essentially it provides a very efficient and flexible way for nginx
internals to access MySQL, Drizzle, as well as other RDBMS's that
support the Drizzle or MySQL wired protocol. Also it can serve as a
direct REST interface to those RDBMS backends.

This module does not generate human-readable outputs, rather, in a
binary format called Resty-DBD-Stream (RDS) designed by ourselves. You
usually need other components, like ngx_rds_json [1], ngx_rds_csv [2],
or lua-rds-parser [3], to work with this module. See the RDS output
format spec for more details:
http://wiki.nginx.org/HttpDrizzleModule#Output_Format

You can find the complete documentation for ngx_drizzle on Nginx Wiki:

??? http://wiki.nginx.org/HttpDrizzleModule

and the latest source is on GitHub:

??? https://github.com/chaoslawful/drizzle-nginx-module

This module is also bundled (though not enabled by default) by the
ngx_openresty bundle:

??? http://openresty.org/

You can check out this wiki page to enable this component while
building ngx_openresty:

??? http://openresty.org/#DrizzleNginxModule

Have fun!
-agentzh

[1] https://github.com/agentzh/rds-json-nginx-module
[2] https://github.com/agentzh/rds-csv-nginx-module
[3] https://github.com/agentzh/lua-rds-parser


From nginx at nginxuser.net  Sat Sep 17 08:53:31 2011
From: nginx at nginxuser.net (Nginx User)
Date: Sat, 17 Sep 2011 11:53:31 +0300
Subject: Openresty downstream bundle maintenance requests
Message-ID: <CAN=WzV2phRfvp5dZ8Qm0EQSYOaMxPgyLDPfnwuHx+8b_pdxAug@mail.gmail.com>

Hi.

A few things to consider with respect to downstream maintenance support.

The configure script under the "for my $opt (@ARGV) {" loop covers a few
typical items and any others cause the "die "Invalid option $opt\n";" line
to be triggered. However, when in a Fedora/Enterprise Linux build
environment, the "%configure" macro in the spec file adds several other
switches (which work fine when building nginx directly).

Can the upstream please consider modifying  the configure script to allow
for these?

A possiblity is:

for my $opt (@ARGV) {
...

    } elsif ($opt =~ /^--build=(.*)/) {
        push @ngx_opts, $opt;

    } elsif ($opt =~ /^--host=(.*)/) {
        push @ngx_opts, $opt;

    } elsif ($opt =~ /^--target=(.*)/) {
        push @ngx_opts, $opt;

    } elsif ($opt =~ /^--program-prefix=(.*)/) {
        push @ngx_opts, $opt;

    } elsif ($opt =~ /^--exec-prefix=(.*)/) {
        push @ngx_opts, $opt;

    } elsif ($opt =~ /^--bindir=(.*)/) {
        push @ngx_opts, $opt;

    } elsif ($opt =~ /^--sbindir=(.*)/) {
        push @ngx_opts, $opt;

    } elsif ($opt =~ /^--sysconfdir=(.*)/) {
        push @ngx_opts, $opt;

    } elsif ($opt =~ /^--datadir=(.*)/) {
        push @ngx_opts, $opt;

    } elsif ($opt =~ /^--includedir=(.*)/) {
        push @ngx_opts, $opt;

    } elsif ($opt =~ /^--libexecdir=(.*)/) {
        push @ngx_opts, $opt;

    } elsif ($opt =~ /^--localstatedir=(.*)/) {
        push @ngx_opts, $opt;

    } elsif ($opt =~ /^--sharedstatedir=(.*)/) {
        push @ngx_opts, $opt;

    } elsif ($opt =~ /^--mandir=(.*)/) {
        push @ngx_opts, $opt;

    } elsif ($opt =~ /^--infodir=(.*)/) {
        push @ngx_opts, $opt;

    } else {
        die "Invalid option $opt\n";
    }
}

May be even merge all the "dir" ones into "} elsif ($opt =~
/^--(.*)dir=(.*)/) {"

Also, can the upstream consider dropping version numbers from the folders in
the "bundle" directory and record the version numbers in a README? Would
save having to edit the spec file for every release.

Finally, can the upstream consider allowing the with luajit option to be
selected but not to try to build luajit? I.E. set openresty up to use luajit
but have the luajit build done separately and not try to replace this by the
standard lua. I suppose the same applies to the standard lua. Basically,
third party applications should ideally be built separately and we would
look into creating seperate rpms for those. Can foresee some potential
knotty issues to overcome with this but if distributing an rpm, then there
are issues anyway.

Thanks
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110917/3c3f7bb5/attachment.html>

From lists at ruby-forum.com  Sat Sep 17 10:10:36 2011
From: lists at ruby-forum.com (Martijn Senden)
Date: Sat, 17 Sep 2011 12:10:36 +0200
Subject: change nginx user in config >> 403 error index.html forbidden
In-Reply-To: <17e343cd5e06fc4d31ab27ccf4006c74.NginxMailingListEnglish@forum.nginx.org>
References: <17e343cd5e06fc4d31ab27ccf4006c74.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <82d209499b597d34e352b99a365f737e@ruby-forum.com>

Thanks! I guess by search permissions you mean read permission, not 
write or execute?

Regards,
Martijn Senden.

-- 
Posted via http://www.ruby-forum.com/.


From lists at ruby-forum.com  Sat Sep 17 10:16:13 2011
From: lists at ruby-forum.com (Martijn Senden)
Date: Sat, 17 Sep 2011 12:16:13 +0200
Subject: change nginx user in config >> 403 error index.html forbidden
In-Reply-To: <17e343cd5e06fc4d31ab27ccf4006c74.NginxMailingListEnglish@forum.nginx.org>
References: <17e343cd5e06fc4d31ab27ccf4006c74.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <cc1f9dcac4b590ddee243eac74acccfc@ruby-forum.com>

Some googling told me search permission is often used as an alternate 
name for execute permission. Thanks, I'll try this!

-- 
Posted via http://www.ruby-forum.com/.


From lists at ruby-forum.com  Sat Sep 17 10:27:40 2011
From: lists at ruby-forum.com (Martijn Senden)
Date: Sat, 17 Sep 2011 12:27:40 +0200
Subject: change nginx user in config >> 403 error index.html forbidden
In-Reply-To: <17e343cd5e06fc4d31ab27ccf4006c74.NginxMailingListEnglish@forum.nginx.org>
References: <17e343cd5e06fc4d31ab27ccf4006c74.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <8c5570cad3538b32e0c137d2dff3a112@ruby-forum.com>

Thanks! With this I got it to work! I changed the group of the server 
folder and the nginx folder to nginx with the chown command and set the 
permissions to (at least) execute, and now I get the welcome page! Thank 
you very much! This was starting to give me headaches yesterday. ;-)

-- 
Posted via http://www.ruby-forum.com/.


From mdounin at mdounin.ru  Sat Sep 17 10:30:29 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Sat, 17 Sep 2011 14:30:29 +0400
Subject: daemon off in production
In-Reply-To: <CAPzVZKk8HJOoo6adQsuq2KVkpA_t3qad-rBZ0EGdpSS6D1Wikg@mail.gmail.com>
References: <CAPzVZKk8HJOoo6adQsuq2KVkpA_t3qad-rBZ0EGdpSS6D1Wikg@mail.gmail.com>
Message-ID: <20110917103028.GV1137@mdounin.ru>

Hello!

On Sat, Sep 17, 2011 at 07:40:19AM +0300, Yaniv Aknin wrote:

> Hi,
> 
> The documentations says I should never run nginx with the daemon off setting
> in production. I would like to do so anyway so my Python/twisted based
> process manager will have an easier time knowing if nginx terminated or not
> (I know there are other ways to do it, but this seems like the easiest; if
> I'll be forced to, I'll use another way).
> 
> I'm aware of SIGWINCH and nginx in general and especially after version
> 1.1.1, and I don't mind taking care of guaranteeing nginx will run without a
> controlling terminal (I will fork()/setsid() before exec()ing it), run in a
> sensible current working directory, etc. I just don't want it to do the
> daemonizing double fork.
> 
> From skimming the code, it appears to me that nothing bad should happen.
> What am I missing? Why is it so bad to run nginx in the foreground in
> production?

Documentation[1] says "daemon" is generally used for development, 
it doesn't say it shouldn't be used in production.  It's safe to 
run with "daemon off;" if you use some controlling process 
manager.  Note that you don't really need to setsid(), nginx will 
take care of this even with "daemon off;".

Please don't confuse with "master_process off;".  It's *unsafe* to 
run with "master_process off;" and it's *only* used for 
development.

[1] http://sysoev.ru/nginx/docs/ngx_core_module.html#daemon

Maxim Dounin


From igor at sysoev.ru  Sat Sep 17 10:49:03 2011
From: igor at sysoev.ru (Igor Sysoev)
Date: Sat, 17 Sep 2011 14:49:03 +0400
Subject: daemon off in production
In-Reply-To: <CAPzVZKk8HJOoo6adQsuq2KVkpA_t3qad-rBZ0EGdpSS6D1Wikg@mail.gmail.com>
References: <CAPzVZKk8HJOoo6adQsuq2KVkpA_t3qad-rBZ0EGdpSS6D1Wikg@mail.gmail.com>
Message-ID: <20110917104903.GA50310@nginx.com>

On Sat, Sep 17, 2011 at 07:40:19AM +0300, Yaniv Aknin wrote:
> Hi,
> 
> The documentations says I should never run nginx with the daemon off setting
> in production. I would like to do so anyway so my Python/twisted based
> process manager will have an easier time knowing if nginx terminated or not
> (I know there are other ways to do it, but this seems like the easiest; if
> I'll be forced to, I'll use another way).
> 
> I'm aware of SIGWINCH and nginx in general and especially after version
> 1.1.1, and I don't mind taking care of guaranteeing nginx will run without a
> controlling terminal (I will fork()/setsid() before exec()ing it), run in a
> sensible current working directory, etc. I just don't want it to do the
> daemonizing double fork.
> 
> From skimming the code, it appears to me that nothing bad should happen.
> What am I missing? Why is it so bad to run nginx in the foreground in
> production?

There are two directives: "master_process on|off" and "daemon on|off".
Both are used to simplify nginx development: while testing a bug/feature
I usually run nginx in foreground without master process and simply
press ^C to stop nginx. This is an analog of Apache's -X command line switch.
I did not want to type it in commanf line, so these two directives
have been introduced.

There was thought to allow nginx to run in single process mode in production
as many lightweight server do. So there is some incomplete code to support
this fucntionality. But this code is incomplete, so you should never
use "master_process off" in production.

As to "daemon off", I did not intend to use it in production (and never did),
but many people do it. I believe it is safe to use "daemon off" in
prodcution. You can apply SIGWINCH fix using:

http://trac.nginx.org/nginx/changeset/4020/nginx


-- 
Igor Sysoev


From igor at sysoev.ru  Sat Sep 17 10:57:25 2011
From: igor at sysoev.ru (Igor Sysoev)
Date: Sat, 17 Sep 2011 14:57:25 +0400
Subject: daemon off in production
In-Reply-To: <20110917104903.GA50310@nginx.com>
References: <CAPzVZKk8HJOoo6adQsuq2KVkpA_t3qad-rBZ0EGdpSS6D1Wikg@mail.gmail.com>
	<20110917104903.GA50310@nginx.com>
Message-ID: <20110917105725.GB50310@nginx.com>

On Sat, Sep 17, 2011 at 02:49:03PM +0400, Igor Sysoev wrote:
> On Sat, Sep 17, 2011 at 07:40:19AM +0300, Yaniv Aknin wrote:
> > Hi,
> > 
> > The documentations says I should never run nginx with the daemon off setting
> > in production. I would like to do so anyway so my Python/twisted based
> > process manager will have an easier time knowing if nginx terminated or not
> > (I know there are other ways to do it, but this seems like the easiest; if
> > I'll be forced to, I'll use another way).
> > 
> > I'm aware of SIGWINCH and nginx in general and especially after version
> > 1.1.1, and I don't mind taking care of guaranteeing nginx will run without a
> > controlling terminal (I will fork()/setsid() before exec()ing it), run in a
> > sensible current working directory, etc. I just don't want it to do the
> > daemonizing double fork.
> > 
> > From skimming the code, it appears to me that nothing bad should happen.
> > What am I missing? Why is it so bad to run nginx in the foreground in
> > production?
> 
> There are two directives: "master_process on|off" and "daemon on|off".
> Both are used to simplify nginx development: while testing a bug/feature
> I usually run nginx in foreground without master process and simply
> press ^C to stop nginx. This is an analog of Apache's -X command line switch.
> I did not want to type it in commanf line, so these two directives
> have been introduced.
> 
> There was thought to allow nginx to run in single process mode in production
> as many lightweight server do. So there is some incomplete code to support
> this fucntionality. But this code is incomplete, so you should never
> use "master_process off" in production.
> 
> As to "daemon off", I did not intend to use it in production (and never did),
> but many people do it. I believe it is safe to use "daemon off" in
> prodcution. You can apply SIGWINCH fix using:
> 
> http://trac.nginx.org/nginx/changeset/4020/nginx

The single option that is not available in "daemon off" mode is online
upgrade, because after the upgrade you supervior can not control new
nginx instance.


-- 
Igor Sysoev


From igor at sysoev.ru  Sat Sep 17 10:59:01 2011
From: igor at sysoev.ru (Igor Sysoev)
Date: Sat, 17 Sep 2011 14:59:01 +0400
Subject: Help with AIO
In-Reply-To: <20110916141002.GH14211@nginx.com>
References: <CAN-ZXpuTdDMmQ7rgo_QP=1W53+j1qc-Bqa=f4=QLNskOubY2dw@mail.gmail.com>
	<20110906073352.GR1137@mdounin.ru>
	<CAN-ZXpsTEVi3Dk0fQAjye1-AAxkZ6HQ9m6UOkS051A=0HVUL1A@mail.gmail.com>
	<20110916103907.GC14211@nginx.com>
	<20110916141002.GH14211@nginx.com>
Message-ID: <20110917105900.GC50310@nginx.com>

On Fri, Sep 16, 2011 at 06:10:02PM +0400, Igor Sysoev wrote:
> On Fri, Sep 16, 2011 at 02:39:07PM +0400, Igor Sysoev wrote:
> > On Thu, Sep 15, 2011 at 04:21:44PM +0200, Hagai Avrahami wrote:
> > > Hi
> > > 
> > > When benchmarking AIO read, I reached more than 1024 concurrent Async IO -
> > > 1. I saw a weird error log
> > > 2. Nginx did not fallback to regular blocking read
> > > 
> > > I looked on the source code --  src/os/unix/ngx_linux_aio_read.c  line:105
> > > and I think  there is a problem in errno handling
> > > 
> > > // Current Code
> > > //----------------------------------------------------------------------------
> > >     *n = io_submit(ngx_aio_ctx, 1, piocb);
> > > 
> > >     if (n == 1) {
> > >         ev->active = 1;
> > >         ev->ready = 0;
> > >         ev->complete = 0;
> > > 
> > >         return NGX_AGAIN;
> > >     }
> > > 
> > >     n = -n;    <---------------------------------------------------------??
> > 
> > Thank you, this is a bug. I confused this with _syscall(2).
> 
> Linux AIO fixed in these two commits:
> 
> http://trac.nginx.org/nginx/changeset/4130/nginx
> http://trac.nginx.org/nginx/changeset/4131/nginx
> 
> Also I'm going to introduce the directive:
> 
> events {
>     worker_aio_events 100;
> }
> 
> to set how much aio events worker may request to the kernel.
> I'm going to set default value to 32 events per worker.
> Currenly this value is hardcoded to 1024 and it seems it is too large,
> since default /proc/sys/fs/aio-max-nr value is 65536 and this means
> that you can run no more than 64 workers.

I think it should be renamed to

events {
    worker_aio_requests 100;
}


-- 
Igor Sysoev


From agentzh at gmail.com  Sat Sep 17 13:58:14 2011
From: agentzh at gmail.com (agentzh)
Date: Sat, 17 Sep 2011 21:58:14 +0800
Subject: Openresty downstream bundle maintenance requests
In-Reply-To: <CAN=WzV2phRfvp5dZ8Qm0EQSYOaMxPgyLDPfnwuHx+8b_pdxAug@mail.gmail.com>
References: <CAN=WzV2phRfvp5dZ8Qm0EQSYOaMxPgyLDPfnwuHx+8b_pdxAug@mail.gmail.com>
Message-ID: <CAB4Tn6OqrcNftf7KM2VoHWSeNsQauhre=Ytv82TDJABSHbvXVw@mail.gmail.com>

On Sat, Sep 17, 2011 at 4:53 PM, Nginx User <nginx at nginxuser.net> wrote:
> The configure script under the "for my $opt (@ARGV) {" loop covers a few
> typical items and any others cause the "die "Invalid option $opt\n";" line
> to be triggered. However, when in a Fedora/Enterprise Linux build
> environment, the "%configure" macro in the spec file adds several other
> switches (which work fine when building nginx directly).

Which options do not work for you? The last part in that loop is actually:

    } elsif ($opt =~ /^--\w.*/) {
        push @ngx_opts, $opt;

    } else {
        die "Invalid option $opt\n";
    }

That is, any options starting with a /\w/ will be matched in the last
2nd condition and passed to nginx's configure script.

> Can the upstream please consider?modifying? the configure script to allow
> for these?
> A possiblity is:
> for my $opt (@ARGV) {
> ...
> ? ? } elsif ($opt =~ /^--build=(.*)/) {
> ? ? ? ? push @ngx_opts, $opt;
> ? ? } elsif ($opt =~ /^--host=(.*)/) {
> ? ? ? ? push @ngx_opts, $opt;
> ? ? } elsif ($opt =~ /^--target=(.*)/) {
> ? ? ? ? push @ngx_opts, $opt;
> ? ? } elsif ($opt =~ /^--program-prefix=(.*)/) {
> ? ? ? ? push @ngx_opts, $opt;
> ? ? } elsif ($opt =~ /^--exec-prefix=(.*)/) {
> ? ? ? ? push @ngx_opts, $opt;
> ? ? } elsif ($opt =~ /^--bindir=(.*)/) {
> ? ? ? ? push @ngx_opts, $opt;
> ? ? } elsif ($opt =~ /^--sbindir=(.*)/) {
> ? ? ? ? push @ngx_opts, $opt;
> ? ? } elsif ($opt =~ /^--sysconfdir=(.*)/) {
> ? ? ? ? push @ngx_opts, $opt;
> ? ? } elsif ($opt =~ /^--datadir=(.*)/) {
> ? ? ? ? push @ngx_opts, $opt;
> ? ? } elsif ($opt =~ /^--includedir=(.*)/) {
> ? ? ? ? push @ngx_opts, $opt;
> ? ? } elsif ($opt =~ /^--libexecdir=(.*)/) {
> ? ? ? ? push @ngx_opts, $opt;
> ? ? } elsif ($opt =~ /^--localstatedir=(.*)/) {
> ? ? ? ? push @ngx_opts, $opt;
> ? ? } elsif ($opt =~ /^--sharedstatedir=(.*)/) {
> ? ? ? ? push @ngx_opts, $opt;
> ? ? } elsif ($opt =~ /^--mandir=(.*)/) {
> ? ? ? ? push @ngx_opts, $opt;
> ? ? } elsif ($opt =~ /^--infodir=(.*)/) {
> ? ? ? ? push @ngx_opts, $opt;
> ? ? } else {
> ? ? ? ? die "Invalid option $opt\n";
> ? ? }
> }

These options should be happily accepted by the last 2nd option. If
not, please let me know :)

> Also, can the upstream consider dropping version numbers from the folders in
> the "bundle" directory and record the version numbers in a README? Would
> save having to edit the spec file for every release.

Okay, I'll remove version numbers in my build script in the next
release of ngx_openresty :)

> Finally, can the upstream consider allowing the with luajit option to be
> selected but not to try to build luajit? I.E. set openresty up to use luajit
> but have the luajit build done separately and not try to replace this by the
> standard lua. I suppose the same applies to the standard lua. Basically,
> third party applications should ideally be built separately and we would
> look into creating seperate rpms for those.

Will adding new options like --with-luajit=PATH and --with-lua51=PATH
work for you? That is, telling ngx_openresty to use an external luajit
or lua 5.1 build located by PATH?

By default, ngx_openresty will continue try to build its bundled
version of LuaJIT or Lua 5.1 and use it whenever possible and that's
the whole point of a bundle ;) But I'm still willing to add special
options to help RPM/Debian/etc packaging for sure :)

> Can foresee some potential
> knotty issues to overcome with this but if?distributing?an rpm, then there
> are issues anyway.

I'm very willing to help and solve any issues along the way ;)

Thanks!
-agentzh


From agentzh at gmail.com  Sat Sep 17 14:08:18 2011
From: agentzh at gmail.com (agentzh)
Date: Sat, 17 Sep 2011 22:08:18 +0800
Subject: Openresty downstream bundle maintenance requests
In-Reply-To: <CAN=WzV2phRfvp5dZ8Qm0EQSYOaMxPgyLDPfnwuHx+8b_pdxAug@mail.gmail.com>
References: <CAN=WzV2phRfvp5dZ8Qm0EQSYOaMxPgyLDPfnwuHx+8b_pdxAug@mail.gmail.com>
Message-ID: <CAB4Tn6MYvnLy_tncUbJEV+x3dRCScMEM1FpqF0TBTqN2_kW9Yw@mail.gmail.com>

On Sat, Sep 17, 2011 at 4:53 PM, Nginx User <nginx at nginxuser.net> wrote:
> Also, can the upstream consider dropping version numbers from the folders in
> the "bundle" directory and record the version numbers in a README? Would
> save having to edit the spec file for every release.

Just out of curiosity, why will you hard-code the directory names
under bundle/ in your rpm spec file?

Regards,
-agentzh


From nginx at nginxuser.net  Sat Sep 17 14:43:17 2011
From: nginx at nginxuser.net (Nginx User)
Date: Sat, 17 Sep 2011 17:43:17 +0300
Subject: Openresty downstream bundle maintenance requests
In-Reply-To: <CAB4Tn6OqrcNftf7KM2VoHWSeNsQauhre=Ytv82TDJABSHbvXVw@mail.gmail.com>
References: <CAN=WzV2phRfvp5dZ8Qm0EQSYOaMxPgyLDPfnwuHx+8b_pdxAug@mail.gmail.com>
	<CAB4Tn6OqrcNftf7KM2VoHWSeNsQauhre=Ytv82TDJABSHbvXVw@mail.gmail.com>
Message-ID: <CAN=WzV1GakpH5O7vCktUFcC05x3QyxGXp_70EUjP3M3aHx7Zjw@mail.gmail.com>

On Sat, Sep 17, 2011 at 4:58 PM, agentzh <agentzh at gmail.com> wrote:

> On Sat, Sep 17, 2011 at 4:53 PM, Nginx User <nginx at nginxuser.net> wrote:
> > The configure script under the "for my $opt (@ARGV) {" loop covers a few
> > typical items and any others cause the "die "Invalid option $opt\n";"
> line
> > to be triggered. However, when in a Fedora/Enterprise Linux build
> > environment, the "%configure" macro in the spec file adds several other
> > switches (which work fine when building nginx directly).
>
> Which options do not work for you? The last part in that loop is actually:
>
>    } elsif ($opt =~ /^--\w.*/) {
>         push @ngx_opts, $opt;
>
>    } else {
>        die "Invalid option $opt\n";
>    }
>
> That is, any options starting with a /\w/ will be matched in the last
> 2nd condition and passed to nginx's configure script.
>

Correct and it matches all the "--with XYZ" and "--without XYZ" flags for
nginx.


> > Can the upstream please consider modifying  the configure script to allow
> > for these?
> > A possiblity is:
> > for my $opt (@ARGV) {
> > ...
> >     } elsif ($opt =~ /^--build=(.*)/) {
> >         push @ngx_opts, $opt;
> >     } elsif ($opt =~ /^--host=(.*)/) {
> >         push @ngx_opts, $opt;
> >     } elsif ($opt =~ /^--target=(.*)/) {
> >         push @ngx_opts, $opt;
> >     } elsif ($opt =~ /^--program-prefix=(.*)/) {
> >         push @ngx_opts, $opt;
> >     } elsif ($opt =~ /^--exec-prefix=(.*)/) {
> >         push @ngx_opts, $opt;
> >     } elsif ($opt =~ /^--bindir=(.*)/) {
> >         push @ngx_opts, $opt;
> >     } elsif ($opt =~ /^--sbindir=(.*)/) {
> >         push @ngx_opts, $opt;
> >     } elsif ($opt =~ /^--sysconfdir=(.*)/) {
> >         push @ngx_opts, $opt;
> >     } elsif ($opt =~ /^--datadir=(.*)/) {
> >         push @ngx_opts, $opt;
> >     } elsif ($opt =~ /^--includedir=(.*)/) {
> >         push @ngx_opts, $opt;
> >     } elsif ($opt =~ /^--libexecdir=(.*)/) {
> >         push @ngx_opts, $opt;
> >     } elsif ($opt =~ /^--localstatedir=(.*)/) {
> >         push @ngx_opts, $opt;
> >     } elsif ($opt =~ /^--sharedstatedir=(.*)/) {
> >         push @ngx_opts, $opt;
> >     } elsif ($opt =~ /^--mandir=(.*)/) {
> >         push @ngx_opts, $opt;
> >     } elsif ($opt =~ /^--infodir=(.*)/) {
> >         push @ngx_opts, $opt;
> >     } else {
> >         die "Invalid option $opt\n";
> >     }
> > }
>
> These options should be happily accepted by the last 2nd option. If
> not, please let me know :)
>

They are matched by the last option of 'die "Invalid option $opt\n";' .


>
> > Also, can the upstream consider dropping version numbers from the folders
> in
> > the "bundle" directory and record the version numbers in a README? Would
> > save having to edit the spec file for every release.
>
> Okay, I'll remove version numbers in my build script in the next
> release of ngx_openresty :)
>

No longer required thanks.


>
> > Finally, can the upstream consider allowing the with luajit option to be
> > selected but not to try to build luajit? I.E. set openresty up to use
> luajit
> > but have the luajit build done separately and not try to replace this by
> the
> > standard lua. I suppose the same applies to the standard lua. Basically,
> > third party applications should ideally be built separately and we would
> > look into creating seperate rpms for those.
>
> Will adding new options like --with-luajit=PATH and --with-lua51=PATH
> work for you? That is, telling ngx_openresty to use an external luajit
> or lua 5.1 build located by PATH?
>

This should work. We can then switch piar this with --without luajit. We
obviously want to keep the lua-rds-parser and lua-redis options.


>
> By default, ngx_openresty will continue try to build its bundled
> version of LuaJIT or Lua 5.1 and use it whenever possible and that's
> the whole point of a bundle ;) But I'm still willing to add special
> options to help RPM/Debian/etc packaging for sure :)
>
> > Can foresee some potential
> > knotty issues to overcome with this but if distributing an rpm, then
> there
> > are issues anyway.
>
> I'm very willing to help and solve any issues along the way ;)
>
> Thanks!
> -agentzh
>

Thanks!
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110917/8e0bcac9/attachment.html>

From agentzh at gmail.com  Sat Sep 17 15:22:03 2011
From: agentzh at gmail.com (agentzh)
Date: Sat, 17 Sep 2011 23:22:03 +0800
Subject: Openresty downstream bundle maintenance requests
In-Reply-To: <CAN=WzV1GakpH5O7vCktUFcC05x3QyxGXp_70EUjP3M3aHx7Zjw@mail.gmail.com>
References: <CAN=WzV2phRfvp5dZ8Qm0EQSYOaMxPgyLDPfnwuHx+8b_pdxAug@mail.gmail.com>
	<CAB4Tn6OqrcNftf7KM2VoHWSeNsQauhre=Ytv82TDJABSHbvXVw@mail.gmail.com>
	<CAN=WzV1GakpH5O7vCktUFcC05x3QyxGXp_70EUjP3M3aHx7Zjw@mail.gmail.com>
Message-ID: <CAB4Tn6O0JPbOjGN1kKUrwCTmTtFMpoM8sA0U9K30a4CE0xAH8A@mail.gmail.com>

On Sat, Sep 17, 2011 at 10:43 PM, Nginx User <nginx at nginxuser.net> wrote:
>>
>> These options should be happily accepted by the last 2nd option. If
>> not, please let me know :)
>
>
> They are matched by the last option of 'die "Invalid option $opt\n";' .
>

I've just confirmed that at least "--includedir=PATH" and
"--datadir=PATH" are accepted by ngx_openresty's configure script but
rejected by nginx's configure script:

    ~/work/nginx-1.0.6 $ ./configure --includedir=/tmp/blah
    ./configure: error: invalid option "--includedir=/tmp/blah"

    ~/work/nginx-1.0.6 $ ./configure --datadir=/tmp/blah
    ./configure: error: invalid option "--datadir=/tmp/blah"

Please note that the 'error: invalid option ".*"' messages
*different* from the one output by 'die "Invalid option $opt\n";'.

Regards,
-agentzh


From nginx at nginxuser.net  Sat Sep 17 16:04:58 2011
From: nginx at nginxuser.net (Nginx User)
Date: Sat, 17 Sep 2011 19:04:58 +0300
Subject: Openresty downstream bundle maintenance requests
In-Reply-To: <CAN=WzV0u96EWixVbVoBaZx8v5fqhFh5peaYkjAqqMp9KUcOUOw@mail.gmail.com>
References: <CAN=WzV2phRfvp5dZ8Qm0EQSYOaMxPgyLDPfnwuHx+8b_pdxAug@mail.gmail.com>
	<CAB4Tn6OqrcNftf7KM2VoHWSeNsQauhre=Ytv82TDJABSHbvXVw@mail.gmail.com>
	<CAN=WzV1GakpH5O7vCktUFcC05x3QyxGXp_70EUjP3M3aHx7Zjw@mail.gmail.com>
	<CAB4Tn6O0JPbOjGN1kKUrwCTmTtFMpoM8sA0U9K30a4CE0xAH8A@mail.gmail.com>
	<CAN=WzV0u96EWixVbVoBaZx8v5fqhFh5peaYkjAqqMp9KUcOUOw@mail.gmail.com>
Message-ID: <CAN=WzV3GU7zYjanuWO8A_mTkp6UDgkzARtgKR+bODnkMDT6abw@mail.gmail.com>

On Sat, Sep 17, 2011 at 6:42 PM, Nginx User <nginx at nginxuser.net> wrote:

>
>
> On Sat, Sep 17, 2011 at 6:22 PM, agentzh <agentzh at gmail.com> wrote:
>
>> On Sat, Sep 17, 2011 at 10:43 PM, Nginx User <nginx at nginxuser.net> wrote:
>> >>
>> >> These options should be happily accepted by the last 2nd option. If
>> >> not, please let me know :)
>> >
>> >
>> > They are matched by the last option of 'die "Invalid option $opt\n";' .
>> >
>>
>> I've just confirmed that at least "--includedir=PATH" and
>> "--datadir=PATH" are accepted by ngx_openresty's configure script but
>> rejected by nginx's configure script:
>>
>>    ~/work/nginx-1.0.6 $ ./configure --includedir=/tmp/blah
>>    ./configure: error: invalid option "--includedir=/tmp/blah"
>>
>>    ~/work/nginx-1.0.6 $ ./configure --datadir=/tmp/blah
>>    ./configure: error: invalid option "--datadir=/tmp/blah"
>>
>> Please note that the 'error: invalid option ".*"' messages
>> *different* from the one output by 'die "Invalid option $opt\n";'.
>>
>> Regards,
>> -agentzh
>>
>
> Nginx configure script has to be patched for rpm build environment.
> Below is the standard patch used.
>
> --- auto/cc/gcc.orig 2007-03-22 08:34:53.000000000 -0600
> +++ auto/cc/gcc 2007-03-22 08:58:47.000000000 -0600
> @@ -172,7 +172,9 @@
>
>
>  # stop on warning
> -CFLAGS="$CFLAGS -Werror"
> +# This combined with Fedora's FORTIFY_SOURCE=2 option causes it nginx
> +# to not compile.
> +#CFLAGS="$CFLAGS -Werror"
>
>  # debug
>  CFLAGS="$CFLAGS -g"
>
>
> The standard outputs from the rpm %configure macro including all those
> elements build fine after this when building nginx by itself. An example of
> the typical spec file with this patch is
> http://www4.atomicorp.com/channels/source/nginx/nginx.spec (they are not
> up to date but it does work as is for 1.0.6 and should for the 1.1.x branch
> as well.
>
> However things crash with the openresty configure as said. Specifically,
> the message returned is for on "--build" which is the first one encountered.
> Recreating the build machine at the moment so can't give specifics right
> now. I'll try changing the text in 'die "Invalid option $opt\n";' when
> having another go to verify whether it is actually this or not. Looking at
> the code though, it seems clear that these extra options will fail at the
> specified point.
>
> Thanks.
>


Hi.

Just looking at the spec file example I linked to, I can see it is not using
the rpm %configure macro.  So you were on to something there.

I suppose the only remaining valid request from my original post is the lua
one.

Thanks
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110917/bb625ccc/attachment-0001.html>

From appa at perusio.net  Sat Sep 17 22:49:01 2011
From: appa at perusio.net (=?UTF-8?B?QW50w7NuaW8=?= P. P. Almeida)
Date: Sat, 17 Sep 2011 23:49:01 +0100
Subject: Problem with Hotlink Protection
In-Reply-To: <beea1075bf2bd47f45365417af71e225.NginxMailingListEnglish@forum.nginx.org>
References: <0c253481fec8974edd546fbde9b7c383.NginxMailingListEnglish@forum.nginx.org>
	<beea1075bf2bd47f45365417af71e225.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <87d3eyyeci.wl%appa@perusio.net>

On 16 Set 2011 19h47 WEST, nginx-forum at nginx.us wrote:

> Thank you both! I totally agree with you, but as soon as I restrict
> referer checking to the image.php script only, there is a flaw
> appearing: image.php then gets downloaded by the browser instead of
> displaying it. Do you have any idea why this happens? It would be
> great if I could restrict the check only to this one .php file, but
> - as said
> - it does not work here...

No I don't. If you replicated the FastCGI stuff it should work. The
request should be sent upstream to the FastCGI listener that you have
on port 9000.

Your config had a strange thing. You used a try_files directive. That
doesn't make sense unless you're serving the file directly or using
another location (try_files does an internal redirect).

My suggestion omitted that. The $fastcgi_script_name variable takes as
value the $uri. In this case image.php.

If you request
http://www.mydomain.com/image.php?twg_album=album-05900-05999&twg_show=image-05995.jpg

this location is used and the request is handled upstream bu the
FastCGI process that will use image.php and whatever arguments you
give.

--- appa


From agentzh at gmail.com  Sun Sep 18 04:28:17 2011
From: agentzh at gmail.com (agentzh)
Date: Sun, 18 Sep 2011 12:28:17 +0800
Subject: Openresty downstream bundle maintenance requests
In-Reply-To: <CAN=WzV0u96EWixVbVoBaZx8v5fqhFh5peaYkjAqqMp9KUcOUOw@mail.gmail.com>
References: <CAN=WzV2phRfvp5dZ8Qm0EQSYOaMxPgyLDPfnwuHx+8b_pdxAug@mail.gmail.com>
	<CAB4Tn6OqrcNftf7KM2VoHWSeNsQauhre=Ytv82TDJABSHbvXVw@mail.gmail.com>
	<CAN=WzV1GakpH5O7vCktUFcC05x3QyxGXp_70EUjP3M3aHx7Zjw@mail.gmail.com>
	<CAB4Tn6O0JPbOjGN1kKUrwCTmTtFMpoM8sA0U9K30a4CE0xAH8A@mail.gmail.com>
	<CAN=WzV0u96EWixVbVoBaZx8v5fqhFh5peaYkjAqqMp9KUcOUOw@mail.gmail.com>
Message-ID: <CAB4Tn6NnSHNXK+s7ZfAQ=DpCOA_KsVAOfWinXCDZAEAvZw9S_g@mail.gmail.com>

On Sat, Sep 17, 2011 at 11:42 PM, Nginx User <nginx at nginxuser.net> wrote:
> Nginx configure script has to be patched for rpm build environment.
> Below is the standard patch used.
>
> --- auto/cc/gcc.orig 2007-03-22 08:34:53.000000000 -0600
> +++ auto/cc/gcc 2007-03-22 08:58:47.000000000 -0600
> @@ -172,7 +172,9 @@
>
>
> ?# stop on warning
> -CFLAGS="$CFLAGS -Werror"
> +# This combined with Fedora's FORTIFY_SOURCE=2 option causes it nginx
> +# to not compile.
> +#CFLAGS="$CFLAGS -Werror"
>
> ?# debug
> ?CFLAGS="$CFLAGS -g"
>

The nginx core bundled by ngx_openresty has this patch as well.

> The standard outputs from the rpm %configure macro including all those
> elements build fine after this when building nginx by itself. An example of
> the typical spec file with this patch
> is?http://www4.atomicorp.com/channels/source/nginx/nginx.spec (they are not
> up to date but it does work as is for 1.0.6 and should for the 1.1.x branch
> as well.
> However things crash with the openresty configure as said. Specifically, the
> message returned is for on "--build" which is the first one encountered.

The configure command in the RPM spec file you link to does not crash
on my slackware linux x86_64 for ngx_openresty 1.0.6.5. I wonder what
error messages are you getting?

> Recreating the build machine at the moment so can't give specifics right
> now. I'll try changing the text in?'die "Invalid option $opt\n";' when
> having another go to verify whether it is actually this or not. Looking at
> the code though, it seems clear that these extra options will fail at the
> specified point.

Why won't "--build" be matched by /^--\w.*/ in ngx_openresty's
configure script? I've said extra options would be captured by the
last 2nd condition and be passed verbatim to nginx's configure script
and would not reach the last "die" statement there.

Regards,
-agentzh


From agentzh at gmail.com  Sun Sep 18 05:34:28 2011
From: agentzh at gmail.com (agentzh)
Date: Sun, 18 Sep 2011 13:34:28 +0800
Subject: Openresty downstream bundle maintenance requests
In-Reply-To: <CAN=WzV1GakpH5O7vCktUFcC05x3QyxGXp_70EUjP3M3aHx7Zjw@mail.gmail.com>
References: <CAN=WzV2phRfvp5dZ8Qm0EQSYOaMxPgyLDPfnwuHx+8b_pdxAug@mail.gmail.com>
	<CAB4Tn6OqrcNftf7KM2VoHWSeNsQauhre=Ytv82TDJABSHbvXVw@mail.gmail.com>
	<CAN=WzV1GakpH5O7vCktUFcC05x3QyxGXp_70EUjP3M3aHx7Zjw@mail.gmail.com>
Message-ID: <CAB4Tn6O99Oe8TQeki8KT4me0Gh8zAj2HbzHN4d9nO3yxJoAz7g@mail.gmail.com>

On Sat, Sep 17, 2011 at 10:43 PM, Nginx User <nginx at nginxuser.net> wrote:
> On Sat, Sep 17, 2011 at 4:58 PM, agentzh <agentzh at gmail.com> wrote:
>>
>> Will adding new options like --with-luajit=PATH and --with-lua51=PATH
>> work for you? That is, telling ngx_openresty to use an external luajit
>> or lua 5.1 build located by PATH?
>
> This should work. We can then switch piar this with --without luajit. We
> obviously want to keep the lua-rds-parser and lua-redis options.
>

I've just added the --with-luajit=PATH and --with-lua51=PATH options
to ngx_openresty's configure script and this update has already been
included in the latest devel release, 1.0.6.7:

    http://openresty.org/#Download

Please test it out to see if it fits your requirements :) All of those
Lua modules bundled by ngx_openresty will still be built by default
when either --with-luajit=PATH or --with-lua51=PATH is specified
unless being explicitly disabled.

Enjoy!
-agentzh


From nginx-forum at nginx.us  Sun Sep 18 13:50:33 2011
From: nginx-forum at nginx.us (SergeyQw)
Date: Sun, 18 Sep 2011 09:50:33 -0400
Subject: nginx 1.0.6 perl module build failure
In-Reply-To: <301b3dfe2d6e0ed1ad4bc5ca953cda63.NginxMailingListEnglish@forum.nginx.org>
References: <20110910114625.GB18804@nginx.com>
	<e50a2f0a8dcaa1de25c4fee42fe490d2.NginxMailingListEnglish@forum.nginx.org>
	<2fa3cc7ba9e0e80774ea82240e53fde3.NginxMailingListEnglish@forum.nginx.org>
	<301b3dfe2d6e0ed1ad4bc5ca953cda63.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <75a98594b894ac906e6e9591c20459f9.NginxMailingListEnglish@forum.nginx.org>

?????? ????. ?? ?? ????????.

Mac OS X Lion.
?????? ??????? ????? ????????? perl 5.12 ??
MacPorts.

Summary of my perl5 (revision 5 version 12 subversion 3) configuration:
   
  Platform:
    osname=darwin, osvers=11.1.0, archname=darwin-multi-2level
    uname='darwin macbook-pro-sergey.local 11.1.0 darwin kernel version
11.1.0: tue jul 26 16:07:11 pdt 2011;
root:xnu-1699.22.81~1release_x86_64 x86_64 '
    config_args='-D inc_version_list=5.12.2/darwin-multi-2level 5.12.2
5.12.1/darwin-multi-2level 5.12.1 5.12.0/darwin-multi-2level 5.12.0 -des
-Dprefix=/opt/local -Dscriptdir=/opt/local/bin
-Dcppflags=-I/opt/local/include -Dccflags=-pipe -O2 -arch x86_64
-Dldflags=-L/opt/local/lib -arch x86_64 -Dvendorprefix=/opt/local
-Dusemultiplicity=y -D cc=/Developer/usr/bin/llvm-gcc-4.2 -D
ld=/Developer/usr/bin/llvm-gcc-4.2 -D man1ext=1pm -D man3ext=3pm -D
man1dir=/opt/local/share/man/man1p -D man3dir=/opt/local/share/man/man3p
-D siteman1dir=/opt/local/share/man/man1 -D
siteman3dir=/opt/local/share/man/man3 -D
vendorman1dir=/opt/local/share/man/man1 -D
vendorman3dir=/opt/local/share/man/man3 -D pager=/usr/bin/less -sR'
    hint=recommended, useposix=true, d_sigaction=define
    useithreads=undef, usemultiplicity=define
    useperlio=define, d_sfio=undef, uselargefiles=define,
usesocks=undef
    use64bitint=define, use64bitall=define, uselongdouble=undef
    usemymalloc=n, bincompat5005=undef
  Compiler:
    cc='/Developer/usr/bin/llvm-gcc-4.2', ccflags ='-pipe -O2 -arch
x86_64 -fno-common -DPERL_DARWIN -I/opt/local/include -no-cpp-precomp
-fno-strict-aliasing -fstack-protector -I/opt/local/include',
    optimize='-O3',
    cppflags='-I/opt/local/include -no-cpp-precomp -pipe -O2 -arch
x86_64 -fno-common -DPERL_DARWIN -I/opt/local/include -no-cpp-precomp
-fno-strict-aliasing -fstack-protector -I/opt/local/include'
    ccversion='', gccversion='4.2.1 (Based on Apple Inc. build 5658)
(LLVM build 2335.15.00)', gccosandvers=''
    intsize=4, longsize=8, ptrsize=8, doublesize=8, byteorder=12345678
    d_longlong=define, longlongsize=8, d_longdbl=define, longdblsize=16
    ivtype='long', ivsize=8, nvtype='double', nvsize=8, Off_t='off_t',
lseeksize=8
    alignbytes=8, prototype=define
  Linker and Libraries:
    ld='env MACOSX_DEPLOYMENT_TARGET=10.3
/Developer/usr/bin/llvm-gcc-4.2', ldflags ='-L/opt/local/lib -arch
x86_64 -fstack-protector'
    libpth=/opt/local/lib /usr/lib
    libs=-lgdbm -ldbm -ldl -lm -lutil -lc
    perllibs=-ldl -lm -lutil -lc
    libc=, so=dylib, useshrplib=false, libperl=libperl.a
    gnulibc_version=''
  Dynamic Linking:
    dlsrc=dl_dlopen.xs, dlext=bundle, d_dlsymun=undef, ccdlflags=' '
    cccdlflags=' ', lddlflags='-L/opt/local/lib -arch x86_64 -bundle
-undefined dynamic_lookup -fstack-protector'


Characteristics of this binary (from libperl): 
  Compile-time options: MULTIPLICITY PERL_DONT_CREATE_GVSV
                        PERL_IMPLICIT_CONTEXT PERL_MALLOC_WRAP
USE_64_BIT_ALL
                        USE_64_BIT_INT USE_LARGE_FILES USE_PERLIO
                        USE_PERL_ATOF
  Built under darwin
  Compiled at Sep 18 2011 13:11:44
  @INC:
    /opt/local/lib/perl5/site_perl/5.12.3/darwin-multi-2level
    /opt/local/lib/perl5/site_perl/5.12.3
    /opt/local/lib/perl5/vendor_perl/5.12.3/darwin-multi-2level
    /opt/local/lib/perl5/vendor_perl/5.12.3
    /opt/local/lib/perl5/5.12.3/darwin-multi-2level
    /opt/local/lib/perl5/5.12.3
    /opt/local/lib/perl5/site_perl
    /opt/local/lib/perl5/vendor_perl



------------
???????? nginx 1.0.6  ? nginx 1.1.3 ????????? ??
?????????? ? perl ???????. 
??? ?????? ??? ???????.

./configure \
    --sbin-path=/usr/local/nginx/nginx \
    --conf-path=/usr/local/nginx/nginx.conf \
    --error-log-path=/var/log/nginx/error.log \
    --http-log-path=/var/log/nginx/access.log \
    --pid-path=/var/run/nginx.pid \
    --lock-path=/var/lock/nginx.lock \
    --with-http_perl_module --with-perl=/usr/bin/perl \
    --with-http_ssl_module \
    --with-pcre=../pcre-8.12 \
    --with-zlib=../zlib-1.2.5


make

In file included from src/http/modules/perl/ngx_http_perl_module.h:17,
                 from src/http/modules/perl/ngx_http_perl_module.c:10:
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/perl.h:3336:
error: expected specifier-qualifier-list before ?bool?
In file included from
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/perl.h:3419,
                 from src/http/modules/perl/ngx_http_perl_module.h:17,
                 from src/http/modules/perl/ngx_http_perl_module.c:10:
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/regexp.h:612:
error: expected specifier-qualifier-list before ?bool?
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/regexp.h:640:
error: expected specifier-qualifier-list before ?bool?
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/regexp.h:700:
error: expected specifier-qualifier-list before ?bool?
In file included from
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/perl.h:3428,
                 from src/http/modules/perl/ngx_http_perl_module.h:17,
                 from src/http/modules/perl/ngx_http_perl_module.c:10:
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/cop.h:35:
error: expected specifier-qualifier-list before ?bool?
In file included from
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/perl.h:3449,
                 from src/http/modules/perl/ngx_http_perl_module.h:17,
                 from src/http/modules/perl/ngx_http_perl_module.c:10:
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/parser.h:47:
error: expected specifier-qualifier-list before ?bool?
In file included from src/http/modules/perl/ngx_http_perl_module.h:17,
                 from src/http/modules/perl/ngx_http_perl_module.c:10:
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/perl.h:4196:
error: expected declaration specifiers or ?...? before ?*?
token
cc1: warnings being treated as errors
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/perl.h:4196:
warning: type defaults to ?int? in declaration of ?bool?
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/perl.h:4196:
error: ?bool? declared as function returning a function
In file included from
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/perl.h:4913,
                 from src/http/modules/perl/ngx_http_perl_module.h:17,
                 from src/http/modules/perl/ngx_http_perl_module.c:10:
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/intrpvar.h:168:
error: field ?Iutf8locale? declared as a function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/intrpvar.h:169:
error: field ?Irehash_seed_set? declared as a function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/intrpvar.h:205:
error: field ?Icolorset? declared as a function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/intrpvar.h:206:
error: field ?Idirty? declared as a function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/intrpvar.h:209:
error: field ?Itainted? declared as a function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/intrpvar.h:233:
error: field ?Iminus_c? declared as a function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/intrpvar.h:234:
error: field ?Iminus_n? declared as a function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/intrpvar.h:235:
error: field ?Iminus_p? declared as a function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/intrpvar.h:236:
error: field ?Iminus_l? declared as a function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/intrpvar.h:237:
error: field ?Iminus_a? declared as a function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/intrpvar.h:238:
error: field ?Iminus_F? declared as a function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/intrpvar.h:239:
error: field ?Idoswitches? declared as a function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/intrpvar.h:240:
error: field ?Iminus_E? declared as a function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/intrpvar.h:252:
error: field ?Idoextract? declared as a function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/intrpvar.h:253:
error: field ?Isawampersand? declared as a function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/intrpvar.h:254:
error: field ?Iunsafe? declared as a function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/intrpvar.h:256:
error: field ?Isrand_called? declared as a function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/intrpvar.h:258:
error: field ?Itainting? declared as a function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/intrpvar.h:377:
error: field ?Icv_has_eval? declared as a function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/intrpvar.h:378:
error: field ?Itaint_warn? declared as a function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/intrpvar.h:421:
error: field ?Iin_clean_objs? declared as a function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/intrpvar.h:422:
error: field ?Iin_clean_all? declared as a function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/intrpvar.h:423:
error: field ?Inomemok? declared as a function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/intrpvar.h:424:
error: field ?Isavebegin? declared as a function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/intrpvar.h:494:
error: field ?Icollation_standard? declared as a function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/intrpvar.h:510:
error: field ?Inumeric_standard? declared as a function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/intrpvar.h:512:
error: field ?Inumeric_local? declared as a function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/intrpvar.h:558:
error: field ?Ipad_reset_pending? declared as a function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/intrpvar.h:710:
error: expected specifier-qualifier-list before
?destroyable_proc_t?
In file included from
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/perl.h:4946,
                 from src/http/modules/perl/ngx_http_perl_module.h:17,
                 from src/http/modules/perl/ngx_http_perl_module.c:10:
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:63:
error: ?Perl_doing_taint? declared as function returning a function
In file included from
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/perl.h:4946,
                 from src/http/modules/perl/ngx_http_perl_module.h:17,
                 from src/http/modules/perl/ngx_http_perl_module.c:10:
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:176:
error: ?Perl_av_exists? declared as function returning a function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:299:
error: ?Perl_cando? declared as function returning a function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:544:
error: ?Perl_do_aexec5? declared as function returning a function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:559:
error: ?Perl_do_close? declared as function returning a function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:561:
error: ?Perl_do_eof? declared as function returning a function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:600:
error: ?Perl_do_exec3? declared as function returning a function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:669:
error: ?Perl_do_open9? declared as function returning a function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:675:
error: ?Perl_do_openn? declared as function returning a function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:680:
error: ?Perl_do_print? declared as function returning a function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:692:
error: ?Perl_do_seek? declared as function returning a function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1142:
error: ?Perl_io_close? declared as function returning a function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1166:
error: ?Perl_is_uni_alnum? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1170:
error: ?Perl_is_uni_idfirst? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1174:
error: ?Perl_is_uni_alpha? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1178:
error: ?Perl_is_uni_ascii? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1182:
error: ?Perl_is_uni_space? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1186:
error: ?Perl_is_uni_cntrl? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1190:
error: ?Perl_is_uni_graph? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1194:
error: ?Perl_is_uni_digit? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1198:
error: ?Perl_is_uni_upper? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1202:
error: ?Perl_is_uni_lower? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1206:
error: ?Perl_is_uni_print? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1210:
error: ?Perl_is_uni_punct? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1214:
error: ?Perl_is_uni_xdigit? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1242:
error: ?Perl_is_uni_alnum_lc? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1246:
error: ?Perl_is_uni_idfirst_lc? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1250:
error: ?Perl_is_uni_alpha_lc? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1254:
error: ?Perl_is_uni_ascii_lc? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1258:
error: ?Perl_is_uni_space_lc? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1262:
error: ?Perl_is_uni_cntrl_lc? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1266:
error: ?Perl_is_uni_graph_lc? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1270:
error: ?Perl_is_uni_digit_lc? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1274:
error: ?Perl_is_uni_upper_lc? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1278:
error: ?Perl_is_uni_lower_lc? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1282:
error: ?Perl_is_uni_print_lc? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1286:
error: ?Perl_is_uni_punct_lc? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1290:
error: ?Perl_is_uni_xdigit_lc? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1293:
error: ?Perl_is_ascii_string? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1303:
error: ?Perl_is_utf8_string? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1313:
error: ?Perl_is_utf8_string_loclen? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1319:
error: ?Perl_is_utf8_alnum? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1325:
error: ?Perl_is_utf8_idfirst? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1331:
error: ?Perl_is_utf8_idcont? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1337:
error: ?Perl_is_utf8_alpha? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1343:
error: ?Perl_is_utf8_ascii? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1349:
error: ?Perl_is_utf8_space? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1355:
error: ?Perl_is_utf8_perl_space? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1361:
error: ?Perl_is_utf8_perl_word? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1367:
error: ?Perl_is_utf8_cntrl? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1373:
error: ?Perl_is_utf8_digit? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1379:
error: ?Perl_is_utf8_posix_digit? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1385:
error: ?Perl_is_utf8_graph? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1391:
error: ?Perl_is_utf8_upper? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1397:
error: ?Perl_is_utf8_lower? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1403:
error: ?Perl_is_utf8_print? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1409:
error: ?Perl_is_utf8_punct? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1415:
error: ?Perl_is_utf8_xdigit? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1421:
error: ?Perl_is_utf8_mark? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1427:
error: ?Perl_is_utf8_X_begin? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1433:
error: ?Perl_is_utf8_X_extend? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1439:
error: ?Perl_is_utf8_X_prepend? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1445:
error: ?Perl_is_utf8_X_non_hangul? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1451:
error: ?Perl_is_utf8_X_L? declared as function returning a function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1457:
error: ?Perl_is_utf8_X_LV? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1463:
error: ?Perl_is_utf8_X_LVT? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1469:
error: ?Perl_is_utf8_X_LV_LVT_V? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1475:
error: ?Perl_is_utf8_X_T? declared as function returning a function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1481:
error: ?Perl_is_utf8_X_V? declared as function returning a function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1511:
error: ?Perl_lex_bufutf8? declared as function returning a function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1538:
error: ?Perl_lex_next_chunk? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1606:
error: ?Perl_grok_numeric_radix? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:2410:
error: ?Perl_vverify? declared as function returning a function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:2750:
error: ?Perl_reg_named_buff_exists? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:3135:
error: ?Perl_sv_2bool? declared as function returning a function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:3310:
error: ?Perl_sv_derived_from? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:3317:
error: ?Perl_sv_does? declared as function returning a function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:3421:
error: ?Perl_sv_cat_decode? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:3524:
error: ?Perl_sv_tainted? declared as function returning a function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:4132:
error: ?Perl_sv_utf8_downgrade? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:4142:
error: ?Perl_sv_utf8_decode? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:4331:
error: ?Perl_sv_destroyable? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:6496:
error: ?Perl_is_gv_magical_sv? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:6504:
error: ?Perl_stashpv_hvname_match? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:6523:
error: ?Perl_ckwarn? declared as function returning a function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:6524:
error: ?Perl_ckwarn_d? declared as function returning a function
In file included from src/http/modules/perl/ngx_http_perl_module.h:17,
                 from src/http/modules/perl/ngx_http_perl_module.c:10:
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/perl.h:3336:
error: expected specifier-qualifier-list before ?bool?
In file included from
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/perl.h:3419,
                 from src/http/modules/perl/ngx_http_perl_module.h:17,
                 from src/http/modules/perl/ngx_http_perl_module.c:10:
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/regexp.h:612:
error: expected specifier-qualifier-list before ?bool?
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/regexp.h:640:
error: expected specifier-qualifier-list before ?bool?
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/regexp.h:700:
error: expected specifier-qualifier-list before ?bool?
In file included from
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/perl.h:3428,
                 from src/http/modules/perl/ngx_http_perl_module.h:17,
                 from src/http/modules/perl/ngx_http_perl_module.c:10:
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/cop.h:35:
error: expected specifier-qualifier-list before ?bool?
In file included from
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/perl.h:3449,
                 from src/http/modules/perl/ngx_http_perl_module.h:17,
                 from src/http/modules/perl/ngx_http_perl_module.c:10:
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/parser.h:47:
error: expected specifier-qualifier-list before ?bool?
In file included from src/http/modules/perl/ngx_http_perl_module.h:17,
                 from src/http/modules/perl/ngx_http_perl_module.c:10:
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/perl.h:4196:
error: expected declaration specifiers or ?...? before ?*?
token
cc1: warnings being treated as errors
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/perl.h:4196:
warning: type defaults to ?int? in declaration of ?bool?
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/perl.h:4196:
error: ?bool? declared as function returning a function
In file included from
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/perl.h:4913,
                 from src/http/modules/perl/ngx_http_perl_module.h:17,
                 from src/http/modules/perl/ngx_http_perl_module.c:10:
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/intrpvar.h:168:
error: field ?Iutf8locale? declared as a function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/intrpvar.h:169:
error: field ?Irehash_seed_set? declared as a function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/intrpvar.h:205:
error: field ?Icolorset? declared as a function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/intrpvar.h:206:
error: field ?Idirty? declared as a function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/intrpvar.h:209:
error: field ?Itainted? declared as a function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/intrpvar.h:233:
error: field ?Iminus_c? declared as a function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/intrpvar.h:234:
error: field ?Iminus_n? declared as a function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/intrpvar.h:235:
error: field ?Iminus_p? declared as a function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/intrpvar.h:236:
error: field ?Iminus_l? declared as a function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/intrpvar.h:237:
error: field ?Iminus_a? declared as a function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/intrpvar.h:238:
error: field ?Iminus_F? declared as a function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/intrpvar.h:239:
error: field ?Idoswitches? declared as a function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/intrpvar.h:240:
error: field ?Iminus_E? declared as a function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/intrpvar.h:252:
error: field ?Idoextract? declared as a function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/intrpvar.h:253:
error: field ?Isawampersand? declared as a function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/intrpvar.h:254:
error: field ?Iunsafe? declared as a function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/intrpvar.h:256:
error: field ?Isrand_called? declared as a function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/intrpvar.h:258:
error: field ?Itainting? declared as a function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/intrpvar.h:377:
error: field ?Icv_has_eval? declared as a function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/intrpvar.h:378:
error: field ?Itaint_warn? declared as a function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/intrpvar.h:421:
error: field ?Iin_clean_objs? declared as a function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/intrpvar.h:422:
error: field ?Iin_clean_all? declared as a function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/intrpvar.h:423:
error: field ?Inomemok? declared as a function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/intrpvar.h:424:
error: field ?Isavebegin? declared as a function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/intrpvar.h:494:
error: field ?Icollation_standard? declared as a function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/intrpvar.h:510:
error: field ?Inumeric_standard? declared as a function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/intrpvar.h:512:
error: field ?Inumeric_local? declared as a function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/intrpvar.h:558:
error: field ?Ipad_reset_pending? declared as a function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/intrpvar.h:710:
error: expected specifier-qualifier-list before
?destroyable_proc_t?
In file included from
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/perl.h:4946,
                 from src/http/modules/perl/ngx_http_perl_module.h:17,
                 from src/http/modules/perl/ngx_http_perl_module.c:10:
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:63:
error: ?Perl_doing_taint? declared as function returning a function
In file included from
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/perl.h:4946,
                 from src/http/modules/perl/ngx_http_perl_module.h:17,
                 from src/http/modules/perl/ngx_http_perl_module.c:10:
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:176:
error: ?Perl_av_exists? declared as function returning a function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:299:
error: ?Perl_cando? declared as function returning a function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:544:
error: ?Perl_do_aexec5? declared as function returning a function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:559:
error: ?Perl_do_close? declared as function returning a function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:561:
error: ?Perl_do_eof? declared as function returning a function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:600:
error: ?Perl_do_exec3? declared as function returning a function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:669:
error: ?Perl_do_open9? declared as function returning a function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:675:
error: ?Perl_do_openn? declared as function returning a function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:680:
error: ?Perl_do_print? declared as function returning a function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:692:
error: ?Perl_do_seek? declared as function returning a function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1142:
error: ?Perl_io_close? declared as function returning a function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1166:
error: ?Perl_is_uni_alnum? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1170:
error: ?Perl_is_uni_idfirst? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1174:
error: ?Perl_is_uni_alpha? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1178:
error: ?Perl_is_uni_ascii? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1182:
error: ?Perl_is_uni_space? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1186:
error: ?Perl_is_uni_cntrl? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1190:
error: ?Perl_is_uni_graph? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1194:
error: ?Perl_is_uni_digit? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1198:
error: ?Perl_is_uni_upper? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1202:
error: ?Perl_is_uni_lower? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1206:
error: ?Perl_is_uni_print? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1210:
error: ?Perl_is_uni_punct? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1214:
error: ?Perl_is_uni_xdigit? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1242:
error: ?Perl_is_uni_alnum_lc? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1246:
error: ?Perl_is_uni_idfirst_lc? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1250:
error: ?Perl_is_uni_alpha_lc? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1254:
error: ?Perl_is_uni_ascii_lc? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1258:
error: ?Perl_is_uni_space_lc? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1262:
error: ?Perl_is_uni_cntrl_lc? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1266:
error: ?Perl_is_uni_graph_lc? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1270:
error: ?Perl_is_uni_digit_lc? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1274:
error: ?Perl_is_uni_upper_lc? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1278:
error: ?Perl_is_uni_lower_lc? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1282:
error: ?Perl_is_uni_print_lc? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1286:
error: ?Perl_is_uni_punct_lc? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1290:
error: ?Perl_is_uni_xdigit_lc? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1293:
error: ?Perl_is_ascii_string? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1303:
error: ?Perl_is_utf8_string? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1313:
error: ?Perl_is_utf8_string_loclen? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1319:
error: ?Perl_is_utf8_alnum? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1325:
error: ?Perl_is_utf8_idfirst? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1331:
error: ?Perl_is_utf8_idcont? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1337:
error: ?Perl_is_utf8_alpha? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1343:
error: ?Perl_is_utf8_ascii? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1349:
error: ?Perl_is_utf8_space? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1355:
error: ?Perl_is_utf8_perl_space? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1361:
error: ?Perl_is_utf8_perl_word? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1367:
error: ?Perl_is_utf8_cntrl? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1373:
error: ?Perl_is_utf8_digit? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1379:
error: ?Perl_is_utf8_posix_digit? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1385:
error: ?Perl_is_utf8_graph? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1391:
error: ?Perl_is_utf8_upper? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1397:
error: ?Perl_is_utf8_lower? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1403:
error: ?Perl_is_utf8_print? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1409:
error: ?Perl_is_utf8_punct? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1415:
error: ?Perl_is_utf8_xdigit? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1421:
error: ?Perl_is_utf8_mark? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1427:
error: ?Perl_is_utf8_X_begin? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1433:
error: ?Perl_is_utf8_X_extend? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1439:
error: ?Perl_is_utf8_X_prepend? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1445:
error: ?Perl_is_utf8_X_non_hangul? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1451:
error: ?Perl_is_utf8_X_L? declared as function returning a function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1457:
error: ?Perl_is_utf8_X_LV? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1463:
error: ?Perl_is_utf8_X_LVT? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1469:
error: ?Perl_is_utf8_X_LV_LVT_V? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1475:
error: ?Perl_is_utf8_X_T? declared as function returning a function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1481:
error: ?Perl_is_utf8_X_V? declared as function returning a function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1511:
error: ?Perl_lex_bufutf8? declared as function returning a function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1538:
error: ?Perl_lex_next_chunk? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:1606:
error: ?Perl_grok_numeric_radix? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:2410:
error: ?Perl_vverify? declared as function returning a function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:2750:
error: ?Perl_reg_named_buff_exists? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:3135:
error: ?Perl_sv_2bool? declared as function returning a function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:3310:
error: ?Perl_sv_derived_from? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:3317:
error: ?Perl_sv_does? declared as function returning a function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:3421:
error: ?Perl_sv_cat_decode? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:3524:
error: ?Perl_sv_tainted? declared as function returning a function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:4132:
error: ?Perl_sv_utf8_downgrade? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:4142:
error: ?Perl_sv_utf8_decode? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:4331:
error: ?Perl_sv_destroyable? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:6496:
error: ?Perl_is_gv_magical_sv? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:6504:
error: ?Perl_stashpv_hvname_match? declared as function returning a
function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:6523:
error: ?Perl_ckwarn? declared as function returning a function
/System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/proto.h:6524:
error: ?Perl_ckwarn_d? declared as function returning a function


? ?????? ?? ??????? ????? ???  ???????
????????? ??? ???????? ????.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215048,215355#msg-215355


From mdounin at mdounin.ru  Sun Sep 18 14:10:54 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Sun, 18 Sep 2011 18:10:54 +0400
Subject: nginx 1.0.6 perl module build failure
In-Reply-To: <75a98594b894ac906e6e9591c20459f9.NginxMailingListEnglish@forum.nginx.org>
References: <20110910114625.GB18804@nginx.com>
	<e50a2f0a8dcaa1de25c4fee42fe490d2.NginxMailingListEnglish@forum.nginx.org>
	<2fa3cc7ba9e0e80774ea82240e53fde3.NginxMailingListEnglish@forum.nginx.org>
	<301b3dfe2d6e0ed1ad4bc5ca953cda63.NginxMailingListEnglish@forum.nginx.org>
	<75a98594b894ac906e6e9591c20459f9.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20110918141054.GW1137@mdounin.ru>

Hello!

On Sun, Sep 18, 2011 at 09:50:33AM -0400, SergeyQw wrote:

> ?????? ????. ?? ?? ????????.

This is English mailing list, please don't write in Russian here.

> Mac OS X Lion.
> ?????? ??????? ????? ????????? perl 5.12 ??
> MacPorts.

[...]

> ./configure \
>     --sbin-path=/usr/local/nginx/nginx \
>     --conf-path=/usr/local/nginx/nginx.conf \
>     --error-log-path=/var/log/nginx/error.log \
>     --http-log-path=/var/log/nginx/access.log \
>     --pid-path=/var/run/nginx.pid \
>     --lock-path=/var/lock/nginx.lock \
>     --with-http_perl_module --with-perl=/usr/bin/perl \

You are building with system perl, not MacPorts one.  Make sure to 
use path to correct perl binary in --with-perl configure argument.

>     --with-http_ssl_module \
>     --with-pcre=../pcre-8.12 \
>     --with-zlib=../zlib-1.2.5
> 
> 
> make
> 
> In file included from src/http/modules/perl/ngx_http_perl_module.h:17,
>                  from src/http/modules/perl/ngx_http_perl_module.c:10:
> /System/Library/Perl/5.12/darwin-thread-multi-2level/CORE/perl.h:3336:
> error: expected specifier-qualifier-list before ?bool?

[...]

> ? ?????? ?? ??????? ????? ???  ???????
> ????????? ??? ???????? ????.

1. You need to build with perl from MacPorts.

2. With Perl 5.8 and 5.10 from MacPorts it builds fine out of the 
box.  For Perl 5.12 and 5.14 (from MacPorts) you need patch posted 
earlier in this thread.

Maxim Dounin


From nginx at nginxuser.net  Sun Sep 18 16:18:20 2011
From: nginx at nginxuser.net (Nginx User)
Date: Sun, 18 Sep 2011 19:18:20 +0300
Subject: Openresty downstream bundle maintenance requests
In-Reply-To: <CAB4Tn6O99Oe8TQeki8KT4me0Gh8zAj2HbzHN4d9nO3yxJoAz7g@mail.gmail.com>
References: <CAN=WzV2phRfvp5dZ8Qm0EQSYOaMxPgyLDPfnwuHx+8b_pdxAug@mail.gmail.com>
	<CAB4Tn6OqrcNftf7KM2VoHWSeNsQauhre=Ytv82TDJABSHbvXVw@mail.gmail.com>
	<CAN=WzV1GakpH5O7vCktUFcC05x3QyxGXp_70EUjP3M3aHx7Zjw@mail.gmail.com>
	<CAB4Tn6O99Oe8TQeki8KT4me0Gh8zAj2HbzHN4d9nO3yxJoAz7g@mail.gmail.com>
Message-ID: <CAN=WzV0D9eJSXygPeL2jmgOQZ-M=pqyHODTomep52zxdMksuCw@mail.gmail.com>

On Sun, Sep 18, 2011 at 8:34 AM, agentzh <agentzh at gmail.com> wrote:

> On Sat, Sep 17, 2011 at 10:43 PM, Nginx User <nginx at nginxuser.net> wrote:
> > On Sat, Sep 17, 2011 at 4:58 PM, agentzh <agentzh at gmail.com> wrote:
> >>
> >> Will adding new options like --with-luajit=PATH and --with-lua51=PATH
> >> work for you? That is, telling ngx_openresty to use an external luajit
> >> or lua 5.1 build located by PATH?
> >
> > This should work. We can then switch piar this with --without luajit. We
> > obviously want to keep the lua-rds-parser and lua-redis options.
> >
>
> I've just added the --with-luajit=PATH and --with-lua51=PATH options
> to ngx_openresty's configure script and this update has already been
> included in the latest devel release, 1.0.6.7:
>
>    http://openresty.org/#Download
>
> Please test it out to see if it fits your requirements :) All of those
> Lua modules bundled by ngx_openresty will still be built by default
> when either --with-luajit=PATH or --with-lua51=PATH is specified
> unless being explicitly disabled.
>
> Enjoy!
> -agentzh
>


Hi.

I'll have a go later.

Just as an update, please ignore the other requests. The only one that
needed attention was this lua related one.

Thanks
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110918/9fdec260/attachment.html>

From nginx-forum at nginx.us  Sun Sep 18 18:05:36 2011
From: nginx-forum at nginx.us (jthaddeusc)
Date: Sun, 18 Sep 2011 14:05:36 -0400
Subject: GET, POST, COOKIE methods same as Apache?
Message-ID: <b711befc1ccf95c510481ba6bbb28fd0.NginxMailingListEnglish@forum.nginx.org>

Hi,

New to Nginx and Rails.

Are GET, POST, and COOKIE the same as Apache? Will using POST in a form
convert data into a hash/array?

Thanks!

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215359,215359#msg-215359


From magicbearmo at gmail.com  Sun Sep 18 19:54:15 2011
From: magicbearmo at gmail.com (MagicBear)
Date: Mon, 19 Sep 2011 03:54:15 +0800
Subject: GET, POST, COOKIE methods same as Apache?
In-Reply-To: <b711befc1ccf95c510481ba6bbb28fd0.NginxMailingListEnglish@forum.nginx.org>
References: <b711befc1ccf95c510481ba6bbb28fd0.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <CAENo=s02gDF1GqkDn=KKPuzHLbadkc5MLH5X3jY-xL9e0Jc=XA@mail.gmail.com>

yes, that is standard

MagicBear

2011/9/19 jthaddeusc <nginx-forum at nginx.us>:
> Hi,
>
> New to Nginx and Rails.
>
> Are GET, POST, and COOKIE the same as Apache? Will using POST in a form
> convert data into a hash/array?
>
> Thanks!
>
> Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215359,215359#msg-215359
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx
>


From nginx-forum at nginx.us  Sun Sep 18 23:28:29 2011
From: nginx-forum at nginx.us (raffo)
Date: Sun, 18 Sep 2011 19:28:29 -0400
Subject: $request_filename rename in download
Message-ID: <6b99d39b79d9df564cfc861c894d6efd.NginxMailingListEnglish@forum.nginx.org>

Hi, 

I have created a public proxy with nginx but when i download files like
"proxy.*****.com/?url=google.com/images/logo_sm.gif" the filename
"logo_sm.gif" it's not save in browse as the real name but with the
name: "download". 

I need a rewrite rules that rename the filename with $args of url. 
Eg: 
site.tld/?url=proxed.tld/file.zip 

the file.zip will be downloaded as "download" name, not with file.zip
name. So i need a rule that rename the file with another with: 
site.tld/?url=proxed.tld/file.zip&=filename.zip 

This is very easy for who know rewrite of nginx, i searn a lot of
combination but i loss only time..

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215361,215361#msg-215361


From wendal1985 at gmail.com  Mon Sep 19 05:56:28 2011
From: wendal1985 at gmail.com (Wendal Chen)
Date: Mon, 19 Sep 2011 13:56:28 +0800
Subject: [ANN] ngx_drizzle v0.1.1: fixed segfaults on 32-bit systems
In-Reply-To: <CAB4Tn6M9uAdbWGy8hc8C2evOLB5ySSb+UX749m3w9oDo0WS9Ww@mail.gmail.com>
References: <CAB4Tn6PTe6p9LtcWxpMFrshDt64ova+2ziugVN1sTw-SJR8AQA@mail.gmail.com>
	<CAB4Tn6M9uAdbWGy8hc8C2evOLB5ySSb+UX749m3w9oDo0WS9Ww@mail.gmail.com>
Message-ID: <CAM79QPX7VXnBJ416ayeZS2EqHBxd8E_4hXNnvTpg1PaRPF6pBg@mail.gmail.com>

Great

2011/9/17 agentzh <agentzh at gmail.com>

> Hi, folks!
>
> I'm happy to announce that ngx_drizzle v0.1.1 has been released. You
> can download the latest release tarball from
>
>     https://github.com/chaoslawful/drizzle-nginx-module/downloads
>
> This release fixed a long-standing segmentation fault on 32-bit
> systems, as well as some other issues. Here goes the complete change
> log:
>
> * fixed segmentation faults on 32-bit systems. thanks @??? and
> @stefanli for reporting this issue.
> * documented the $drizzle_thread_id variable.
> * added lots of debug outputs (enabled by the --with-debug option
> while building Nginx or OpenResty), inspired by github issue #10.
> * fixed issues regarding defining global variables in C header files:
> we should have defined rds_rough_col_type_t as a type rather than a
> global variable. thanks @???.
> * documented the python -> python3 pitfall while building libdrizzle
> 1.0 on at least ArchLinux.
> * fixed the automatic libdrizzle searching algorithm in the config
> file: now we should look under libdrizzle-1.0/ instead. thanks ???
> (Calio) for reporting this issue.
>
> ngx_drizzle is an nginx upstream module integrating libdrizzle into
> Nginx in a non-blocking and streamming way.
>
> Essentially it provides a very efficient and flexible way for nginx
> internals to access MySQL, Drizzle, as well as other RDBMS's that
> support the Drizzle or MySQL wired protocol. Also it can serve as a
> direct REST interface to those RDBMS backends.
>
> This module does not generate human-readable outputs, rather, in a
> binary format called Resty-DBD-Stream (RDS) designed by ourselves. You
> usually need other components, like ngx_rds_json [1], ngx_rds_csv [2],
> or lua-rds-parser [3], to work with this module. See the RDS output
> format spec for more details:
> http://wiki.nginx.org/HttpDrizzleModule#Output_Format
>
> You can find the complete documentation for ngx_drizzle on Nginx Wiki:
>
>     http://wiki.nginx.org/HttpDrizzleModule
>
> and the latest source is on GitHub:
>
>     https://github.com/chaoslawful/drizzle-nginx-module
>
> This module is also bundled (though not enabled by default) by the
> ngx_openresty bundle:
>
>     http://openresty.org/
>
> You can check out this wiki page to enable this component while
> building ngx_openresty:
>
>     http://openresty.org/#DrizzleNginxModule
>
> Have fun!
> -agentzh
>
> [1] https://github.com/agentzh/rds-json-nginx-module
> [2] https://github.com/agentzh/rds-csv-nginx-module
> [3] https://github.com/agentzh/lua-rds-parser
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx




-- 
Wendal Chen
<http://t.sina.com.cn/1068040557?s=6uyXnP>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110919/27a09f16/attachment.html>

From amoiz.shine at gmail.com  Mon Sep 19 05:59:42 2011
From: amoiz.shine at gmail.com (Sharl.Jimh.Tsin)
Date: Mon, 19 Sep 2011 13:59:42 +0800
Subject: [ANN] ngx_drizzle v0.1.1: fixed segfaults on 32-bit systems
In-Reply-To: <CAM79QPX7VXnBJ416ayeZS2EqHBxd8E_4hXNnvTpg1PaRPF6pBg@mail.gmail.com>
References: <CAB4Tn6PTe6p9LtcWxpMFrshDt64ova+2ziugVN1sTw-SJR8AQA@mail.gmail.com>
	<CAB4Tn6M9uAdbWGy8hc8C2evOLB5ySSb+UX749m3w9oDo0WS9Ww@mail.gmail.com>
	<CAM79QPX7VXnBJ416ayeZS2EqHBxd8E_4hXNnvTpg1PaRPF6pBg@mail.gmail.com>
Message-ID: <1316411982.5414.1.camel@sharl-desktop>

? 2011-09-19?? 13:56 +0800?Wendal Chen???
> Great
> 
> 2011/9/17 agentzh <agentzh at gmail.com>
>         Hi, folks!
>         
>         I'm happy to announce that ngx_drizzle v0.1.1 has been
>         released. You
>         can download the latest release tarball from
>         
>         
>         https://github.com/chaoslawful/drizzle-nginx-module/downloads
>         
>         This release fixed a long-standing segmentation fault on
>         32-bit
>         systems, as well as some other issues. Here goes the complete
>         change
>         log:
>         
>         * fixed segmentation faults on 32-bit systems. thanks @???
>         and
>         @stefanli for reporting this issue.
>         * documented the $drizzle_thread_id variable.
>         * added lots of debug outputs (enabled by the --with-debug
>         option
>         while building Nginx or OpenResty), inspired by github issue
>         #10.
>         * fixed issues regarding defining global variables in C header
>         files:
>         we should have defined rds_rough_col_type_t as a type rather
>         than a
>         global variable. thanks @???.
>         * documented the python -> python3 pitfall while building
>         libdrizzle
>         1.0 on at least ArchLinux.
>         * fixed the automatic libdrizzle searching algorithm in the
>         config
>         file: now we should look under libdrizzle-1.0/ instead. thanks
>         ???
>         (Calio) for reporting this issue.
>         
>         ngx_drizzle is an nginx upstream module integrating libdrizzle
>         into
>         Nginx in a non-blocking and streamming way.
>         
>         Essentially it provides a very efficient and flexible way for
>         nginx
>         internals to access MySQL, Drizzle, as well as other RDBMS's
>         that
>         support the Drizzle or MySQL wired protocol. Also it can serve
>         as a
>         direct REST interface to those RDBMS backends.
>         
>         This module does not generate human-readable outputs, rather,
>         in a
>         binary format called Resty-DBD-Stream (RDS) designed by
>         ourselves. You
>         usually need other components, like ngx_rds_json [1],
>         ngx_rds_csv [2],
>         or lua-rds-parser [3], to work with this module. See the RDS
>         output
>         format spec for more details:
>         http://wiki.nginx.org/HttpDrizzleModule#Output_Format
>         
>         You can find the complete documentation for ngx_drizzle on
>         Nginx Wiki:
>         
>             http://wiki.nginx.org/HttpDrizzleModule
>         
>         and the latest source is on GitHub:
>         
>             https://github.com/chaoslawful/drizzle-nginx-module
>         
>         This module is also bundled (though not enabled by default) by
>         the
>         ngx_openresty bundle:
>         
>             http://openresty.org/
>         
>         You can check out this wiki page to enable this component
>         while
>         building ngx_openresty:
>         
>             http://openresty.org/#DrizzleNginxModule
>         
>         Have fun!
>         -agentzh
>         
>         [1] https://github.com/agentzh/rds-json-nginx-module
>         [2] https://github.com/agentzh/rds-csv-nginx-module
>         [3] https://github.com/agentzh/lua-rds-parser
>         
>         _______________________________________________
>         nginx mailing list
>         nginx at nginx.org
>         http://mailman.nginx.org/mailman/listinfo/nginx
> 
> 
> 
> -- 
> Wendal Chen
> 
> 
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx

Thanks for your great work of mod_ngx_drizzle!

???????????????????? ?????????

-- 
Best regards,
Sharl.Jimh.Tsin (From China **Obviously Taiwan INCLUDED**)

Using Gmail? Please read this important notice:
http://www.fsf.org/campaigns/jstrap/gmail?10073.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 490 bytes
Desc: This is a digitally signed message part
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110919/e80ac05f/attachment-0001.bin>

From wendal1985 at gmail.com  Mon Sep 19 06:02:40 2011
From: wendal1985 at gmail.com (Wendal Chen)
Date: Mon, 19 Sep 2011 14:02:40 +0800
Subject: [ANN] ngx_drizzle v0.1.1: fixed segfaults on 32-bit systems
In-Reply-To: <1316411982.5414.1.camel@sharl-desktop>
References: <CAB4Tn6PTe6p9LtcWxpMFrshDt64ova+2ziugVN1sTw-SJR8AQA@mail.gmail.com>
	<CAB4Tn6M9uAdbWGy8hc8C2evOLB5ySSb+UX749m3w9oDo0WS9Ww@mail.gmail.com>
	<CAM79QPX7VXnBJ416ayeZS2EqHBxd8E_4hXNnvTpg1PaRPF6pBg@mail.gmail.com>
	<1316411982.5414.1.camel@sharl-desktop>
Message-ID: <CAM79QPXUe=j=ivj4KbTZRWGO02jOSambWdXJeRCRz7Xv8dbncw@mail.gmail.com>

??????????, ?????????,????, cool~~

2011/9/19 Sharl.Jimh.Tsin <amoiz.shine at gmail.com>

> ? 2011-09-19?? 13:56 +0800?Wendal Chen???
> > Great
> >
> > 2011/9/17 agentzh <agentzh at gmail.com>
> >         Hi, folks!
> >
> >         I'm happy to announce that ngx_drizzle v0.1.1 has been
> >         released. You
> >         can download the latest release tarball from
> >
> >
> >         https://github.com/chaoslawful/drizzle-nginx-module/downloads
> >
> >         This release fixed a long-standing segmentation fault on
> >         32-bit
> >         systems, as well as some other issues. Here goes the complete
> >         change
> >         log:
> >
> >         * fixed segmentation faults on 32-bit systems. thanks @???
> >         and
> >         @stefanli for reporting this issue.
> >         * documented the $drizzle_thread_id variable.
> >         * added lots of debug outputs (enabled by the --with-debug
> >         option
> >         while building Nginx or OpenResty), inspired by github issue
> >         #10.
> >         * fixed issues regarding defining global variables in C header
> >         files:
> >         we should have defined rds_rough_col_type_t as a type rather
> >         than a
> >         global variable. thanks @???.
> >         * documented the python -> python3 pitfall while building
> >         libdrizzle
> >         1.0 on at least ArchLinux.
> >         * fixed the automatic libdrizzle searching algorithm in the
> >         config
> >         file: now we should look under libdrizzle-1.0/ instead. thanks
> >         ???
> >         (Calio) for reporting this issue.
> >
> >         ngx_drizzle is an nginx upstream module integrating libdrizzle
> >         into
> >         Nginx in a non-blocking and streamming way.
> >
> >         Essentially it provides a very efficient and flexible way for
> >         nginx
> >         internals to access MySQL, Drizzle, as well as other RDBMS's
> >         that
> >         support the Drizzle or MySQL wired protocol. Also it can serve
> >         as a
> >         direct REST interface to those RDBMS backends.
> >
> >         This module does not generate human-readable outputs, rather,
> >         in a
> >         binary format called Resty-DBD-Stream (RDS) designed by
> >         ourselves. You
> >         usually need other components, like ngx_rds_json [1],
> >         ngx_rds_csv [2],
> >         or lua-rds-parser [3], to work with this module. See the RDS
> >         output
> >         format spec for more details:
> >         http://wiki.nginx.org/HttpDrizzleModule#Output_Format
> >
> >         You can find the complete documentation for ngx_drizzle on
> >         Nginx Wiki:
> >
> >             http://wiki.nginx.org/HttpDrizzleModule
> >
> >         and the latest source is on GitHub:
> >
> >             https://github.com/chaoslawful/drizzle-nginx-module
> >
> >         This module is also bundled (though not enabled by default) by
> >         the
> >         ngx_openresty bundle:
> >
> >             http://openresty.org/
> >
> >         You can check out this wiki page to enable this component
> >         while
> >         building ngx_openresty:
> >
> >             http://openresty.org/#DrizzleNginxModule
> >
> >         Have fun!
> >         -agentzh
> >
> >         [1] https://github.com/agentzh/rds-json-nginx-module
> >         [2] https://github.com/agentzh/rds-csv-nginx-module
> >         [3] https://github.com/agentzh/lua-rds-parser
> >
> >         _______________________________________________
> >         nginx mailing list
> >         nginx at nginx.org
> >         http://mailman.nginx.org/mailman/listinfo/nginx
> >
> >
> >
> > --
> > Wendal Chen
> >
> >
> > _______________________________________________
> > nginx mailing list
> > nginx at nginx.org
> > http://mailman.nginx.org/mailman/listinfo/nginx
>
> Thanks for your great work of mod_ngx_drizzle!
>
> ???????????????????? ?????????
>
> --
> Best regards,
> Sharl.Jimh.Tsin (From China **Obviously Taiwan INCLUDED**)
>
> Using Gmail? Please read this important notice:
> http://www.fsf.org/campaigns/jstrap/gmail?10073.
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx
>



-- 
Wendal Chen
<http://t.sina.com.cn/1068040557?s=6uyXnP>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110919/3c4d562f/attachment.html>

From nginx-forum at nginx.us  Mon Sep 19 09:26:34 2011
From: nginx-forum at nginx.us (liushaobo)
Date: Mon, 19 Sep 2011 05:26:34 -0400
Subject: nginx : segfault at
Message-ID: <c1552055c5bbc89aca7fc0a32133c685.NginxMailingListEnglish@forum.nginx.org>

hi,all

message log ,  I ask why ? use nginx-1.0.4-1 ?thanks,  

Sep 18 15:43:29 web kernel: nginx[794]: segfault at 00007fff4ca24fd8 rip
0000003ea681398b rsp 00007fff4ca24fb0 error 6
Sep 18 16:39:03 web kernel: nginx[800]: segfault at 00007fff4ca24ff8 rip
0000003ea6812a65 rsp 00007fff4ca25000 error 6
Sep 18 21:59:17 web kernel: nginx[799]: segfault at 00007fff4ca24ff8 rip
0000003ea6812a65 rsp 00007fff4ca25000 error 6
Sep 18 23:18:47 web kernel: nginx[10573]: segfault at 00007fff4ca24ff8
rip 0000003ea68139c2 rsp 00007fff4ca25000 error 6

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215370,215370#msg-215370


From nginx-forum at nginx.us  Mon Sep 19 09:31:59 2011
From: nginx-forum at nginx.us (liushaobo)
Date: Mon, 19 Sep 2011 05:31:59 -0400
Subject: nginx : segfault at
In-Reply-To: <c1552055c5bbc89aca7fc0a32133c685.NginxMailingListEnglish@forum.nginx.org>
References: <c1552055c5bbc89aca7fc0a32133c685.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <2b6d0227c6c5ad5f93364f18d255a7de.NginxMailingListEnglish@forum.nginx.org>

# nginx -V
nginx: nginx version: nginx/1.0.4
nginx: built by gcc 4.1.2 20080704 (Red Hat 4.1.2-50)
nginx: TLS SNI support disabled
nginx: configure arguments: --user=www --group=www
--prefix=/usr/share/nginx --sbin-path=/usr/sbin/nginx
--conf-path=/etc/nginx/nginx.conf
--error-log-path=/var/log/nginx/error.log
--http-log-path=/var/log/nginx/access.log
--http-client-body-temp-path=/var/lib/nginx/tmp/client_body
--http-proxy-temp-path=/var/lib/nginx/tmp/proxy
--http-fastcgi-temp-path=/var/lib/nginx/tmp/fastcgi
--pid-path=/var/run/nginx.pid --lock-path=/var/lock/subsys/nginx
--with-http_ssl_module --with-http_realip_module
--with-http_addition_module --with-http_sub_module
--with-http_dav_module --with-http_flv_module
--with-http_gzip_static_module --with-http_stub_status_module
--with-http_perl_module --with-mail --with-mail_ssl_module
--with-cc-opt='-O2 -g -pipe -Wall -Wp,-D_FORTIFY_SOURCE=2 -fexceptions
-fstack-protector --param=ssp-buffer-size=4 -m64 -mtune=generic'
--add-module=/usr/src/redhat/BUILD/nginx-1.0.4/nginx-upstream-fair

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215370,215371#msg-215371


From webwindy at gmail.com  Mon Sep 19 10:02:00 2011
From: webwindy at gmail.com (C.F)
Date: Mon, 19 Sep 2011 18:02:00 +0800
Subject: $request_filename rename in download
In-Reply-To: <6b99d39b79d9df564cfc861c894d6efd.NginxMailingListEnglish@forum.nginx.org>
References: <6b99d39b79d9df564cfc861c894d6efd.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <CAHurYRX6TEtaTUt9HzrucqxZ_GtOHsTVOMoYoCGBwnQXFEZiJA@mail.gmail.com>

try it
key=filename

          add_header    Content-Disposition "attachment; filename=$arg_key";




2011/9/19 raffo <nginx-forum at nginx.us>

> Hi,
>
> I have created a public proxy with nginx but when i download files like
> "proxy.*****.com/?url=google.com/images/logo_sm.gif" the filename
> "logo_sm.gif" it's not save in browse as the real name but with the
> name: "download".
>
> I need a rewrite rules that rename the filename with $args of url.
> Eg:
> site.tld/?url=proxed.tld/file.zip
>
> the file.zip will be downloaded as "download" name, not with file.zip
> name. So i need a rule that rename the file with another with:
> site.tld/?url=proxed.tld/file.zip&=filename.zip
>
> This is very easy for who know rewrite of nginx, i searn a lot of
> combination but i loss only time..
>
> Posted at Nginx Forum:
> http://forum.nginx.org/read.php?2,215361,215361#msg-215361
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110919/32e2a21a/attachment.html>

From mdounin at mdounin.ru  Mon Sep 19 11:03:13 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Mon, 19 Sep 2011 15:03:13 +0400
Subject: nginx : segfault at
In-Reply-To: <c1552055c5bbc89aca7fc0a32133c685.NginxMailingListEnglish@forum.nginx.org>
References: <c1552055c5bbc89aca7fc0a32133c685.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20110919110312.GY1137@mdounin.ru>

Hello!

On Mon, Sep 19, 2011 at 05:26:34AM -0400, liushaobo wrote:

> hi,all
> 
> message log ,  I ask why ? use nginx-1.0.4-1 ?thanks,  
> 
> Sep 18 15:43:29 web kernel: nginx[794]: segfault at 00007fff4ca24fd8 rip
> 0000003ea681398b rsp 00007fff4ca24fb0 error 6
> Sep 18 16:39:03 web kernel: nginx[800]: segfault at 00007fff4ca24ff8 rip
> 0000003ea6812a65 rsp 00007fff4ca25000 error 6
> Sep 18 21:59:17 web kernel: nginx[799]: segfault at 00007fff4ca24ff8 rip
> 0000003ea6812a65 rsp 00007fff4ca25000 error 6
> Sep 18 23:18:47 web kernel: nginx[10573]: segfault at 00007fff4ca24ff8
> rip 0000003ea68139c2 rsp 00007fff4ca25000 error 6

Please upgrade to at least 1.0.6 (latest stable), there are 
several segmentation faults fixed in it since 1.0.4.

If it doesn't help, please upgrade to 1.1.3 (latest devel), there 
are fixes for several segmentation faults which weren't yet merged 
into stable branch.

If it doesn't help either, please follow procedure described here:

http://wiki.nginx.org/Debugging#Asking_for_help

Notably, please provide nginx -V output, full config and stack 
trace for segfaults you see.

Maxim Dounin


From nginx-forum at nginx.us  Mon Sep 19 14:29:29 2011
From: nginx-forum at nginx.us (dullnicker)
Date: Mon, 19 Sep 2011 10:29:29 -0400
Subject: How to tune nginx for 800-1200 concurrent connections?
Message-ID: <ef98ae9bdd90878e490c36035da42860.NginxMailingListEnglish@forum.nginx.org>

Dear all,

I run nginx as stand-alone webserver (not as Apache proxy) on an image
gallery website. The server has to handle up to 1,200 concurrent
connections on Port 80, the average number throughout the day is around
500-600. During peak times, the server suffers a bit under its load and
I wonder whether there is anything that I can do to decrease the load by
tuning the nginx config. I should add that Apache went nuts handling the
site and therefore nginx is already a great relief to have. The site has
just around 3,000 unique visitors but up to 250,000 pageviews per day.

Here are the relevant server specs:
--------------------------------------------
4 cores at 2.1 GHz
1 GB RAM (average free RAM is around 800 MB even during peak times)

Here is my current nginx.conf:
-------------------------------------

user              nginx nginx;
worker_processes  8; # default: 2


error_log	logs/error.log;

pid               logs/nginx.pid;

events {
    worker_connections  2048; # default: 1024
}

http {
    include       mime.types;
    default_type  application/octet-stream;
    client_max_body_size 64M;
    sendfile        on;
    tcp_nopush      on;

    keepalive_timeout  20; #default: 3

    gzip  on;
    gzip_comp_level 2;
    gzip_proxied any;
    gzip_types      text/plain text/css application/x-javascript
text/xml application/xml application/xml+rss text/javascript;

    server_tokens off;

    include /etc/nginx/conf.d/*;
} 

Here are the fastcgi settings (The gallery is heavily PHP-driven):
------------------------------------------------------------------------------

   include fastcgi_params;
    fastcgi_intercept_errors on;
    fastcgi_index index.php;
    fastcgi_param SCRIPT_FILENAME $document_root$fastcgi_script_name;
    try_files $uri =404;
    fastcgi_pass 127.0.0.1:9000;

I would appreciate your comments and suggestions a lot - Thank you very
much in advance!

Kind regards
-A

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215383,215383#msg-215383


From mat999 at gmail.com  Mon Sep 19 14:52:08 2011
From: mat999 at gmail.com (SplitIce)
Date: Tue, 20 Sep 2011 00:52:08 +1000
Subject: How to tune nginx for 800-1200 concurrent connections?
In-Reply-To: <ef98ae9bdd90878e490c36035da42860.NginxMailingListEnglish@forum.nginx.org>
References: <ef98ae9bdd90878e490c36035da42860.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <CALFfGYaVQwDHhaVLxgCmA31zq8pzv+BkViJMWGcFxDeP+VbEPw@mail.gmail.com>

are the images served through PHP? if so make them static and it will
increase thoughrougput by tons.

Additionally you could possibly reduce worker_processes to 4-6 to reduce
context switching.

On Tue, Sep 20, 2011 at 12:29 AM, dullnicker <nginx-forum at nginx.us> wrote:

> Dear all,
>
> I run nginx as stand-alone webserver (not as Apache proxy) on an image
> gallery website. The server has to handle up to 1,200 concurrent
> connections on Port 80, the average number throughout the day is around
> 500-600. During peak times, the server suffers a bit under its load and
> I wonder whether there is anything that I can do to decrease the load by
> tuning the nginx config. I should add that Apache went nuts handling the
> site and therefore nginx is already a great relief to have. The site has
> just around 3,000 unique visitors but up to 250,000 pageviews per day.
>
> Here are the relevant server specs:
> --------------------------------------------
> 4 cores at 2.1 GHz
> 1 GB RAM (average free RAM is around 800 MB even during peak times)
>
> Here is my current nginx.conf:
> -------------------------------------
>
> user              nginx nginx;
> worker_processes  8; # default: 2
>
>
> error_log       logs/error.log;
>
> pid               logs/nginx.pid;
>
> events {
>    worker_connections  2048; # default: 1024
> }
>
> http {
>    include       mime.types;
>    default_type  application/octet-stream;
>    client_max_body_size 64M;
>    sendfile        on;
>    tcp_nopush      on;
>
>    keepalive_timeout  20; #default: 3
>
>    gzip  on;
>    gzip_comp_level 2;
>    gzip_proxied any;
>    gzip_types      text/plain text/css application/x-javascript
> text/xml application/xml application/xml+rss text/javascript;
>
>    server_tokens off;
>
>    include /etc/nginx/conf.d/*;
> }
>
> Here are the fastcgi settings (The gallery is heavily PHP-driven):
>
> ------------------------------------------------------------------------------
>
>   include fastcgi_params;
>    fastcgi_intercept_errors on;
>    fastcgi_index index.php;
>    fastcgi_param SCRIPT_FILENAME $document_root$fastcgi_script_name;
>    try_files $uri =404;
>    fastcgi_pass 127.0.0.1:9000;
>
> I would appreciate your comments and suggestions a lot - Thank you very
> much in advance!
>
> Kind regards
> -A
>
> Posted at Nginx Forum:
> http://forum.nginx.org/read.php?2,215383,215383#msg-215383
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110920/50938fa0/attachment.html>

From al-nginx at none.at  Mon Sep 19 19:56:19 2011
From: al-nginx at none.at (Aleksandar Lazic)
Date: Mon, 19 Sep 2011 21:56:19 +0200
Subject: ssl benchmarks with and without nginx
Message-ID: <20110919195619.GB6932@none.at>

Dear Reader,

today on the haproxy list was this link posted

http://blog.exceliance.fr/2011/09/16/benchmarking_ssl_performance/

I have asked why they not have tried to use nginx the answer was another
link

http://vincent.bernat.im/en/blog/2011-ssl-benchmark.html

Maybe someone can verify if there is really a problem in nginx or this
difference is due to the fact that nginx is a REAL Webserver and the
other both 'only' SSL-Terminator.

BR
Aleks


From j at jjb.cc  Mon Sep 19 20:40:05 2011
From: j at jjb.cc (John Joseph Bachir)
Date: Mon, 19 Sep 2011 16:40:05 -0400
Subject: How to tune nginx for 800-1200 concurrent connections?
In-Reply-To: <CALFfGYaVQwDHhaVLxgCmA31zq8pzv+BkViJMWGcFxDeP+VbEPw@mail.gmail.com>
References: <ef98ae9bdd90878e490c36035da42860.NginxMailingListEnglish@forum.nginx.org>
	<CALFfGYaVQwDHhaVLxgCmA31zq8pzv+BkViJMWGcFxDeP+VbEPw@mail.gmail.com>
Message-ID: <CAOf2Z5vaqkUNrH7YVNo89-D9UYDJ2ZxD6x7Aid2EqXhEyO1f+A@mail.gmail.com>

* might consider turning up keepalive to 65
* depending on the nature of your content you might want to turn up when it
expires -- example: expires 10m;
* looks like you have 2 workers per core -- typically a worker process per
core is recommended (or maybe your have intel cores with "hyerthreding", in
which case maybe 8 is the more relevant #
 * what's your cpu load like? is your server cpu-bound or is IO the problem?
if you can find some way to do fewer disk reads and more ram reads that will
help
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110919/6f124417/attachment.html>

From mdounin at mdounin.ru  Mon Sep 19 22:20:14 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Tue, 20 Sep 2011 02:20:14 +0400
Subject: ssl benchmarks with and without nginx
In-Reply-To: <20110919195619.GB6932@none.at>
References: <20110919195619.GB6932@none.at>
Message-ID: <20110919222014.GH1137@mdounin.ru>

Hello!

On Mon, Sep 19, 2011 at 09:56:19PM +0200, Aleksandar Lazic wrote:

> Dear Reader,
> 
> today on the haproxy list was this link posted
> 
> http://blog.exceliance.fr/2011/09/16/benchmarking_ssl_performance/
> 
> I have asked why they not have tried to use nginx the answer was another
> link
> 
> http://vincent.bernat.im/en/blog/2011-ssl-benchmark.html
> 
> Maybe someone can verify if there is really a problem in nginx or this
> difference is due to the fact that nginx is a REAL Webserver and the
> other both 'only' SSL-Terminator.

The main problem I see is nginx configuration.  By default nginx 
is configured to handle real-word load, not for ssl session 
microbenchmarks.  And the config used in particular test makes 
things even worse due to "multi_accept on" used.  For 
microbenchmarking ssl one want to use

    events {
        accept_mutext off;
        multi_accept off;
        ...
    }

to ensure better distribution of connections between worker 
processes, this should resolve observed latency issues.

Maxim Dounin

p.s. Stunnel in the same test was obviously benchmarked with 
non-threaded OpenSSL, and that's why it doesn't scale to multiple 
CPUs.  With properly compiled OpenSSL it will scale fine.


From zjay1987 at gmail.com  Tue Sep 20 06:22:18 2011
From: zjay1987 at gmail.com (li zJay)
Date: Tue, 20 Sep 2011 14:22:18 +0800
Subject: How to tune nginx for 800-1200 concurrent connections?
In-Reply-To: <CAOf2Z5vaqkUNrH7YVNo89-D9UYDJ2ZxD6x7Aid2EqXhEyO1f+A@mail.gmail.com>
References: <ef98ae9bdd90878e490c36035da42860.NginxMailingListEnglish@forum.nginx.org>
	<CALFfGYaVQwDHhaVLxgCmA31zq8pzv+BkViJMWGcFxDeP+VbEPw@mail.gmail.com>
	<CAOf2Z5vaqkUNrH7YVNo89-D9UYDJ2ZxD6x7Aid2EqXhEyO1f+A@mail.gmail.com>
Message-ID: <CABPr_TbfDJObxb-_=iEkQ08Hi2uhZFhY+wO=HNrhKHMsCzXu=A@mail.gmail.com>

I think the bottleneck may be the PHP, it is a piece of cake for ngx to
handle 1k concurrent connections  without any special tuning.

On Tue, Sep 20, 2011 at 4:40 AM, John Joseph Bachir <j at jjb.cc> wrote:

> * might consider turning up keepalive to 65
> * depending on the nature of your content you might want to turn up when it
> expires -- example: expires 10m;
> * looks like you have 2 workers per core -- typically a worker process per
> core is recommended (or maybe your have intel cores with "hyerthreding", in
> which case maybe 8 is the more relevant #
>  * what's your cpu load like? is your server cpu-bound or is IO the
> problem? if you can find some way to do fewer disk reads and more ram reads
> that will help
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx
>



--
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110920/21899ac3/attachment.html>

From tomlove at gmail.com  Tue Sep 20 08:14:16 2011
From: tomlove at gmail.com (Thomas Love)
Date: Tue, 20 Sep 2011 10:14:16 +0200
Subject: How to tune nginx for 800-1200 concurrent connections?
In-Reply-To: <ef98ae9bdd90878e490c36035da42860.NginxMailingListEnglish@forum.nginx.org>
References: <ef98ae9bdd90878e490c36035da42860.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <CAJCGDi9puVZLcp14HNhDtrSvOPaChnQhUL4xgb9XogOYMVYeTA@mail.gmail.com>

On 19 September 2011 16:29, dullnicker <nginx-forum at nginx.us> wrote:
> Dear all,
>
> I run nginx as stand-alone webserver (not as Apache proxy) on an image
> gallery website. The server has to handle up to 1,200 concurrent
> connections on Port 80, the average number throughout the day is around
> 500-600.

> The site has
> just around 3,000 unique visitors but up to 250,000 pageviews per day.
>

So 80+ hits per person per day and 3 per second, but 550 concurrent
connections? I haven't seen a pattern like this before and I'd be
interested to know how it works. There may be ways to optimize the
application for it. I wouldn't expect nginx to be the problem though.

If any significant portion of that concurrency is being held by PHP,
that'll be tying up a lot of resources.


Thomas


From igor at sysoev.ru  Tue Sep 20 11:22:25 2011
From: igor at sysoev.ru (Igor Sysoev)
Date: Tue, 20 Sep 2011 15:22:25 +0400
Subject: nginx-1.1.4
Message-ID: <20110920112225.GC54733@nginx.com>

Changes with nginx 1.1.4                                         20 Sep 2011

    *) Feature: the ngx_http_upstream_keepalive module.

    *) Feature: the "proxy_http_version" directive.

    *) Feature: the "fastcgi_keep_conn" directive.

    *) Feature: the "worker_aio_requests" directive.

    *) Bugfix: if nginx was built --with-file-aio it could not be run on
       Linux kernel which did not support AIO.

    *) Bugfix: in Linux AIO error processing.
       Thanks to Hagai Avrahami.

    *) Bugfix: reduced memory consumption for long-lived requests.

    *) Bugfix: the module ngx_http_mp4_module did not support 64-bit MP4
       "co64" atom.


-- 
Igor Sysoev


From mdounin at mdounin.ru  Tue Sep 20 11:40:04 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Tue, 20 Sep 2011 15:40:04 +0400
Subject: nginx-1.1.4
In-Reply-To: <20110920112225.GC54733@nginx.com>
References: <20110920112225.GC54733@nginx.com>
Message-ID: <20110920114004.GO1137@mdounin.ru>

Hello!

On Tue, Sep 20, 2011 at 03:22:25PM +0400, Igor Sysoev wrote:

> Changes with nginx 1.1.4                                         20 Sep 2011
> 
>     *) Feature: the ngx_http_upstream_keepalive module.

[...]

>     *) Bugfix: reduced memory consumption for long-lived requests.

[...]

Note to module developers: this release introduces several API 
changes which may affect 3rd party modules.  Please review svn 
commits for details.  Most intresting commits are:

http://trac.nginx.org/nginx/changeset/4115/nginx
(ngx_chain_update_chains() api change)

http://trac.nginx.org/nginx/changeset/4117/nginx
http://trac.nginx.org/nginx/changeset/4118/nginx
http://trac.nginx.org/nginx/changeset/4119/nginx
http://trac.nginx.org/nginx/changeset/4120/nginx
(various upstream module related changes)

Maxim Dounin


From b.yordanov at exsisto.com  Tue Sep 20 12:12:00 2011
From: b.yordanov at exsisto.com (Boyko Yordanov)
Date: Tue, 20 Sep 2011 15:12:00 +0300
Subject: nginx-1.1.4
In-Reply-To: <20110920114004.GO1137@mdounin.ru>
References: <20110920112225.GC54733@nginx.com>
	<20110920114004.GO1137@mdounin.ru>
Message-ID: <32453ED0-15C3-406F-9EA5-92B05E7B0F60@exsisto.com>

Thanks to both of you,

Boyko


From usirsiwal at verivue.com  Tue Sep 20 12:29:06 2011
From: usirsiwal at verivue.com (Sirsiwal, Umesh)
Date: Tue, 20 Sep 2011 08:29:06 -0400
Subject: nginx-1.1.4
In-Reply-To: <32453ED0-15C3-406F-9EA5-92B05E7B0F60@exsisto.com>
References: <20110920112225.GC54733@nginx.com>
	<20110920114004.GO1137@mdounin.ru>,
	<32453ED0-15C3-406F-9EA5-92B05E7B0F60@exsisto.com>
Message-ID: <B293D35E9E20694A8D9DFFE1BC655C556AB265E773@vvexch.verivue.com>

Thanks to both of you. Great work.

If I understand correctly, the only way to use upstream_keepalive module is with upstream module. This implies I cannot use upstream_keepalive where the upstream is identified by a variable. For example:

proxy_pass http://$host$request_uri;

Is my understand correct? Is there a way around this restriction?

-Umesh


________________________________________
From: nginx-bounces at nginx.org [nginx-bounces at nginx.org] On Behalf Of Boyko Yordanov [b.yordanov at exsisto.com]
Sent: Tuesday, September 20, 2011 8:12 AM
To: nginx at nginx.org
Cc: nginx-devel at nginx.org
Subject: Re: nginx-1.1.4

Thanks to both of you,

Boyko

_______________________________________________
nginx mailing list
nginx at nginx.org
http://mailman.nginx.org/mailman/listinfo/nginx


From mdounin at mdounin.ru  Tue Sep 20 12:52:19 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Tue, 20 Sep 2011 16:52:19 +0400
Subject: nginx-1.1.4
In-Reply-To: <B293D35E9E20694A8D9DFFE1BC655C556AB265E773@vvexch.verivue.com>
References: <20110920112225.GC54733@nginx.com>
	<20110920114004.GO1137@mdounin.ru>
	<32453ED0-15C3-406F-9EA5-92B05E7B0F60@exsisto.com>
	<B293D35E9E20694A8D9DFFE1BC655C556AB265E773@vvexch.verivue.com>
Message-ID: <20110920125219.GP1137@mdounin.ru>

Hello!

On Tue, Sep 20, 2011 at 08:29:06AM -0400, Sirsiwal, Umesh wrote:

> Thanks to both of you. Great work.
> 
> If I understand correctly, the only way to use 
> upstream_keepalive module is with upstream module. This implies 
> I cannot use upstream_keepalive where the upstream is identified 
> by a variable. For example:
> 
> proxy_pass http://$host$request_uri;
> 
> Is my understand correct?

Yes.

> Is there a way around this restriction?

Currently no.

Maxim Dounin


From wendal1985 at gmail.com  Tue Sep 20 14:06:11 2011
From: wendal1985 at gmail.com (Wendal Chen)
Date: Tue, 20 Sep 2011 22:06:11 +0800
Subject: nginx-1.1.4
In-Reply-To: <20110920125219.GP1137@mdounin.ru>
References: <20110920112225.GC54733@nginx.com>
	<20110920114004.GO1137@mdounin.ru>
	<32453ED0-15C3-406F-9EA5-92B05E7B0F60@exsisto.com>
	<B293D35E9E20694A8D9DFFE1BC655C556AB265E773@vvexch.verivue.com>
	<20110920125219.GP1137@mdounin.ru>
Message-ID: <CAM79QPX7cBCUKkG43c+RM7onm8f0LXvThf8ygTc4g2Tqa2JJaQ@mail.gmail.com>

Great !!

finally, upstream_keepalive is supported!!

2011/9/20 Maxim Dounin <mdounin at mdounin.ru>

> Hello!
>
> On Tue, Sep 20, 2011 at 08:29:06AM -0400, Sirsiwal, Umesh wrote:
>
> > Thanks to both of you. Great work.
> >
> > If I understand correctly, the only way to use
> > upstream_keepalive module is with upstream module. This implies
> > I cannot use upstream_keepalive where the upstream is identified
> > by a variable. For example:
> >
> > proxy_pass http://$host$request_uri;
> >
> > Is my understand correct?
>
> Yes.
>
> > Is there a way around this restriction?
>
> Currently no.
>
> Maxim Dounin
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx
>



-- 
Wendal Chen
<http://t.sina.com.cn/1068040557?s=6uyXnP>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110920/f77f777c/attachment.html>

From nginx-forum at nginx.us  Tue Sep 20 15:13:01 2011
From: nginx-forum at nginx.us (etrader)
Date: Tue, 20 Sep 2011 11:13:01 -0400
Subject: 504 Gateway Time-out
Message-ID: <954983e3e6e1d481fc1b6feaf871a5f1.NginxMailingListEnglish@forum.nginx.org>

I have a php script which takes too long to be executed (a form for
uploading large files). Upong running this I get "504 Gateway Time-out"
error. Since I do not want to update php.ini for the whole server, I put
set_time_limit(0); at the top of my php script; but still I get the same
time-out error. What is the problem? I'm on Nginx 1.0.4.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215445,215445#msg-215445


From nginx-forum at nginx.us  Tue Sep 20 15:26:34 2011
From: nginx-forum at nginx.us (locojohn)
Date: Tue, 20 Sep 2011 11:26:34 -0400
Subject: 504 Gateway Time-out
In-Reply-To: <954983e3e6e1d481fc1b6feaf871a5f1.NginxMailingListEnglish@forum.nginx.org>
References: <954983e3e6e1d481fc1b6feaf871a5f1.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <0c334ee1406b911ee68ef3c8d1e66b26.NginxMailingListEnglish@forum.nginx.org>

Hi, 

Search the manual for the following directives:

http {

  [..]
  client_header_timeout 10m;
  client_body_timeout 10m;
  send_timeout 10m;

  fastcgi_read_timeout 600;
  fastcgi_send_timeout 600;

}

Then experiment and specify values that are appropriate for your server
or virtual host.  

Andrejs

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215445,215446#msg-215446


From nginx-forum at nginx.us  Tue Sep 20 15:32:36 2011
From: nginx-forum at nginx.us (devdemi)
Date: Tue, 20 Sep 2011 11:32:36 -0400
Subject: upstream timed out (110: Connection timed out)
Message-ID: <9cf6ec328a38e9b6197206d805b2c3a6.NginxMailingListEnglish@forum.nginx.org>

I installed on Debian Lenny nginx 0.6.32 and php-fpm and php file opens
as well but went trying to do post request i get error in my
/var/log/nginx/error.log
"upstream timed out (110: Connection timed out) while reading response
header from upstream, ....."
My config of host
----------------------
server {
  listen 80;
  server_name domain.ru www.domain.ru;

  set $www_folder '/var/www';
  set $root_path '$www_folder';
  root $root_path;
  index index.htm index.html index.php;

  location ~ \.php$ {
        include /etc/nginx/fastcgi_params;
        fastcgi_param SCRIPT_FILENAME $root_path/$fastcgi_script_name;
        fastcgi_param DOCUMENT_ROOT $root_path;
        fastcgi_pass php-fpm;
  }
  location @phpscripts {
        include /etc/nginx/fastcgi_params;
        fastcgi_param SCRIPT_FILENAME $root_path/$fastcgi_script_name;
        fastcgi_param DOCUMENT_ROOT $root_path;
        fastcgi_pass php-fpm;
  }

  location / {
     default_type text/html;
     root  $root_path;

     if (!-e $request_filename) {
        return 404;
     }

     error_page      404 502 504 403 405 = @php;
  }

  location @php {
        include /etc/nginx/fastcgi_params;
        fastcgi_param SCRIPT_FILENAME $root_path/index.php;
        fastcgi_pass php-fpm;
  }

  error_page    502 = /502.htm;
  location = /502.htm {
         root  $www_folder;
  }  

  location ~* \.(css|js|swf|ico|png|jpg|gif|jpeg|mp3|xml|html)$ {
        root  $root_path;
        access_log   off;
        expires      30d;
  }

  location ~ /\.ht {
        deny  all;
  }

  location ~ /.svn/ {
        deny all;
  } 

}
----------------------
and my /etc/nginx/nginx.conf
----------------------
worker_processes  6;
# worker_priority -1;

error_log  /var/log/nginx/error.log;
pid        /var/run/nginx.pid;

events {
    worker_connections  5024;
    # multi_accept on;
}

http {
    include       /etc/nginx/mime.types;

    access_log  /var/log/nginx/access.log;

    sendfile        on;
    #tcp_nopush     on;

    #keepalive_timeout  0;
    keepalive_timeout  65;
    tcp_nodelay        on;

    gzip  on;
    gzip_disable "MSIE [1-6]\.(?!.*SV1)";

        # PHP-FPM (backend)
        upstream php-fpm {
           server 127.0.0.1:9000;
        }


    include /etc/nginx/conf.d/*.conf;
    include /etc/nginx/sites-enabled/*;
}
----------------------

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215447,215447#msg-215447


From nginx-forum at nginx.us  Tue Sep 20 15:36:59 2011
From: nginx-forum at nginx.us (devdemi)
Date: Tue, 20 Sep 2011 11:36:59 -0400
Subject: upstream timed out (110: Connection timed out)
In-Reply-To: <9cf6ec328a38e9b6197206d805b2c3a6.NginxMailingListEnglish@forum.nginx.org>
References: <9cf6ec328a38e9b6197206d805b2c3a6.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <e70cc28df70d6c7d613b5f8b4a67dc28.NginxMailingListEnglish@forum.nginx.org>

size of post requst is not large, it's couple kb.
php script is login form.
so post request handles very slow.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215447,215448#msg-215448


From r at roze.lv  Tue Sep 20 16:01:48 2011
From: r at roze.lv (Reinis Rozitis)
Date: Tue, 20 Sep 2011 19:01:48 +0300
Subject: upstream timed out (110: Connection timed out)
In-Reply-To: <e70cc28df70d6c7d613b5f8b4a67dc28.NginxMailingListEnglish@forum.nginx.org>
References: <9cf6ec328a38e9b6197206d805b2c3a6.NginxMailingListEnglish@forum.nginx.org>
	<e70cc28df70d6c7d613b5f8b4a67dc28.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <52F8668188974A1CB300C0A2453716C6@DD21>

> php script is login form.
> so post request handles very slow.


What does the php-fpm log show?

If its not enabled see where is your 'php-fpm.conf' is and under:

[global]
error_log = /path/to/your/php-fpm.log
slowlog =  /path/to/your/slow.log


Optionaly (if not present) add something like:

request_slowlog_timeout = 10s

(it will log/backtrace all php requests which take longer than 10 seconds).


rr


From r at roze.lv  Tue Sep 20 16:02:58 2011
From: r at roze.lv (Reinis Rozitis)
Date: Tue, 20 Sep 2011 19:02:58 +0300
Subject: upstream timed out (110: Connection timed out)
In-Reply-To: <52F8668188974A1CB300C0A2453716C6@DD21>
References: <9cf6ec328a38e9b6197206d805b2c3a6.NginxMailingListEnglish@forum.nginx.org><e70cc28df70d6c7d613b5f8b4a67dc28.NginxMailingListEnglish@forum.nginx.org>
	<52F8668188974A1CB300C0A2453716C6@DD21>
Message-ID: <C097A9F7109848CD903BCC5CC4DCBFD3@DD21>

> Optionaly (if not present) add something like:
> request_slowlog_timeout = 10s

Of course after the changes restart php-fpm.

rr


From smartin at openclinica.com  Tue Sep 20 17:32:33 2011
From: smartin at openclinica.com (Shaun Martin)
Date: Tue, 20 Sep 2011 13:32:33 -0400
Subject: Timeout issues
Message-ID: <CA9E4671.279AB%smartin@openclinica.com>

Hi All,

Ok I am having issues with NGINX and it timing out. My scenario is that I am
using NGINX as a web server front end for my application. I have two folder
being served as alias?s for static content and one location doing a
proxy_pass to a java server.

My issue is that on one of the request that goes through the proxy_pass to
the local java web server will time out through nginx within 10 minutes. The
timeout is a browser timeout. It takes 11-12 minutes to load the page
without nginx, so I need nginx to not kill the session until that time has
passed. I have tried every timeout setting imaginable that I could find and
I still get a browser timeout at 10 minutes. Note this is an SSL connection,
the test without nginx in the way was also an SSL connection.

Below are my configs, note all the timeout options have been slowly added
over time and lots of testing.....they are not my wanted scenario but I am
trying to figure out how this is done.

NGINX.conf

user www-data;
worker_processes  4;

error_log  /var/log/nginx/error.log;
pid        /var/run/nginx.pid;

events {
    worker_connections  1024;
    # multi_accept on;
}

http {
    include       /etc/nginx/mime.types;

    access_log  /var/log/nginx/access.log;

    sendfile        on;
    #tcp_nopush     on;

    #keepalive_timeout  0;
    keepalive_timeout  900 900;
 client_body_timeout      900;
  client_header_timeout    900;
  send_timeout             900;


    tcp_nodelay        on;

    gzip  on;
    gzip_disable "MSIE [1-6]\.(?!.*SV1)";
    gzip_types text/plain text/html text/css text/javascript image/png
image/x-icon application/x-javascript application/xml image/gif;
    
    ssl_session_cache    shared:SSL:100k;
    ssl_session_timeout  20m;


    include /etc/nginx/conf.d/*.conf;
    include /etc/nginx/sites-enabled/*;
}

Site Conf

server {
        listen       80;
        server_name  localhost;

        #charset koi8-r;

        access_log  /var/log/nginx/80-app.access.log;
        error_log  /var/log/nginx/80-app.error.log;

        location / {
            rewrite ^/(.*)$ https://$host/$1 redirect;
        }

        error_page  404              /404.html;

        # redirect server error pages to the static page /50x.html
        #
        error_page   500 502 503 504  /50x.html;
        location = /50x.html {
            root   html;
        }
}
server {        
        listen       443;
        server_name  localhost;


        access_log  /var/log/nginx/app.access.log;
        error_log  /var/log/nginx/app.error.log;

        ssl                  on;
        ssl_certificate      /usr/local/oc/install/server.com.crt;
        ssl_certificate_key  /usr/local/oc/install/server.key;

#        fail_timeout 3600s;
    keepalive_timeout  900 900;
 client_body_timeout      900;
  client_header_timeout    900;
  send_timeout             900;
        # directly serve the static files in the `images` directory
        location ~ ^/APP/includes/(.*)$ {
            # add future expiry date to force caching of the file on the
client
            expires max;
            add_header Cache-Control "public";
            alias /usr/local/tomcat/webapps/APP/includes/$1;
        }

        location ~ ^/APP/images/(.*)$ {
            # add future expiry date to force caching of the file on the
client 
            expires max;
            add_header Cache-Control "public";
            alias /usr/local/tomcat/webapps/APP/images/$1;
        }
        # pass all other requests to Tomcat
        location /APP {
            proxy_read_timeout 3600s;
            proxy_connect_timeout 3600s;
            proxy_send_timeout 3600s;

            proxy_pass http://127.0.0.1:8080/APP;
            proxy_set_header Host $host;
            proxy_set_header X-Real-IP $remote_addr;
            proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
        }
}


Any help is greatly appreciated. I wouldn?t even mind if it is possible to
disable timeout all together just as a test.

Thanks,
Shaun
-- 


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110920/cbbd8576/attachment-0001.html>

From nginx-forum at nginx.us  Tue Sep 20 18:44:34 2011
From: nginx-forum at nginx.us (etrader)
Date: Tue, 20 Sep 2011 14:44:34 -0400
Subject: 504 Gateway Time-out
In-Reply-To: <954983e3e6e1d481fc1b6feaf871a5f1.NginxMailingListEnglish@forum.nginx.org>
References: <954983e3e6e1d481fc1b6feaf871a5f1.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <3d3bd3ec463b876488540989383c46a3.NginxMailingListEnglish@forum.nginx.org>

Thanks locojohn! It perfectly worked :)

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215445,215453#msg-215453


From nginx-forum at nginx.us  Tue Sep 20 18:47:26 2011
From: nginx-forum at nginx.us (ynasser)
Date: Tue, 20 Sep 2011 14:47:26 -0400
Subject: WebDriver and Nginx
Message-ID: <4bd63487c17a749c842d5e6c46c46985.NginxMailingListEnglish@forum.nginx.org>

Has anyone tried using nginx and webdriver together? Before I used nginx
as a reverse proxy, I never had any problems, but now I can't get the
browser being controlled by webdriver to get webpages.

An example from the python shell: 

>>> import selenium
>>> from selenium import webdriver
>>> profile = webdriver.FirefoxProfile()
>>> profile.set_preference("network.proxy.type", 1)
>>> profile.set_preference("network.proxy.http", "127.0.0.1")
>>> profile.set_preference("network.proxy.http_port", "80")
>>> driver = webdriver.Firefox(firefox_profile=profile)
>>> driver.get("http://www.google.com") 

And then nothing happens.

Since it works when nginx isn't running, I'm guessing the problem is
from the nginx side. Has anyone experienced something similar?

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215454,215454#msg-215454


From smartin at openclinica.com  Tue Sep 20 18:53:40 2011
From: smartin at openclinica.com (Shaun Martin)
Date: Tue, 20 Sep 2011 14:53:40 -0400
Subject: Timeout issues
In-Reply-To: <CA9E4671.279AB%smartin@openclinica.com>
Message-ID: <CA9E5974.279D3%smartin@openclinica.com>

Hi All,

Oh yah I forgot to mention there are no entries in the error log saying
anything about connection timed out.

Thanks,
Shaun

On 9/20/11 01:32 PM, "Shaun Martin" <smartin at openclinica.com> wrote:

> Hi All,
> 
> Ok I am having issues with NGINX and it timing out. My scenario is that I am
> using NGINX as a web server front end for my application. I have two folder
> being served as alias?s for static content and one location doing a proxy_pass
> to a java server.
> 
> My issue is that on one of the request that goes through the proxy_pass to the
> local java web server will time out through nginx within 10 minutes. The
> timeout is a browser timeout. It takes 11-12 minutes to load the page without
> nginx, so I need nginx to not kill the session until that time has passed. I
> have tried every timeout setting imaginable that I could find and I still get
> a browser timeout at 10 minutes. Note this is an SSL connection, the test
> without nginx in the way was also an SSL connection.
> 
> Below are my configs, note all the timeout options have been slowly added over
> time and lots of testing.....they are not my wanted scenario but I am trying
> to figure out how this is done.
> 
> NGINX.conf
> 
> user www-data;
> worker_processes  4;
> 
> error_log  /var/log/nginx/error.log;
> pid        /var/run/nginx.pid;
> 
> events {
>     worker_connections  1024;
>     # multi_accept on;
> }
> 
> http {
>     include       /etc/nginx/mime.types;
> 
>     access_log  /var/log/nginx/access.log;
> 
>     sendfile        on;
>     #tcp_nopush     on;
> 
>     #keepalive_timeout  0;
>     keepalive_timeout  900 900;
>  client_body_timeout      900;
>   client_header_timeout    900;
>   send_timeout             900;
> 
> 
>     tcp_nodelay        on;
> 
>     gzip  on;
>     gzip_disable "MSIE [1-6]\.(?!.*SV1)";
>     gzip_types text/plain text/html text/css text/javascript image/png
> image/x-icon application/x-javascript application/xml image/gif;
>     
>     ssl_session_cache    shared:SSL:100k;
>     ssl_session_timeout  20m;
> 
> 
>     include /etc/nginx/conf.d/*.conf;
>     include /etc/nginx/sites-enabled/*;
> }
> 
> Site Conf
> 
> server {
>         listen       80;
>         server_name  localhost;
> 
>         #charset koi8-r;
> 
>         access_log  /var/log/nginx/80-app.access.log;
>         error_log  /var/log/nginx/80-app.error.log;
> 
>         location / {
>             rewrite ^/(.*)$ https://$host/$1 redirect;
>         }
> 
>         error_page  404              /404.html;
> 
>         # redirect server error pages to the static page /50x.html
>         #
>         error_page   500 502 503 504  /50x.html;
>         location = /50x.html {
>             root   html;
>         }
> }
> server {        
>         listen       443;
>         server_name  localhost;
> 
> 
>         access_log  /var/log/nginx/app.access.log;
>         error_log  /var/log/nginx/app.error.log;
> 
>         ssl                  on;
>         ssl_certificate      /usr/local/oc/install/server.com.crt;
>         ssl_certificate_key  /usr/local/oc/install/server.key;
> 
> #        fail_timeout 3600s;
>     keepalive_timeout  900 900;
>  client_body_timeout      900;
>   client_header_timeout    900;
>   send_timeout             900;
>         # directly serve the static files in the `images` directory
>         location ~ ^/APP/includes/(.*)$ {
>             # add future expiry date to force caching of the file on the
> client
>             expires max;
>             add_header Cache-Control "public";
>             alias /usr/local/tomcat/webapps/APP/includes/$1;
>         }
> 
>         location ~ ^/APP/images/(.*)$ {
>             # add future expiry date to force caching of the file on the
> client 
>             expires max;
>             add_header Cache-Control "public";
>             alias /usr/local/tomcat/webapps/APP/images/$1;
>         }
>         # pass all other requests to Tomcat
>         location /APP {
>             proxy_read_timeout 3600s;
>             proxy_connect_timeout 3600s;
>             proxy_send_timeout 3600s;
> 
>             proxy_pass http://127.0.0.1:8080/APP;
>             proxy_set_header Host $host;
>             proxy_set_header X-Real-IP $remote_addr;
>             proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
>         }
> }
> 
> 
> Any help is greatly appreciated. I wouldn?t even mind if it is possible to
> disable timeout all together just as a test.
> 
> Thanks,
> Shaun
> 
> -- 
> 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110920/c86eb568/attachment.html>

From mdounin at mdounin.ru  Tue Sep 20 18:54:46 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Tue, 20 Sep 2011 22:54:46 +0400
Subject: WebDriver and Nginx
In-Reply-To: <4bd63487c17a749c842d5e6c46c46985.NginxMailingListEnglish@forum.nginx.org>
References: <4bd63487c17a749c842d5e6c46c46985.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20110920185446.GX1137@mdounin.ru>

Hello!

On Tue, Sep 20, 2011 at 02:47:26PM -0400, ynasser wrote:

> Has anyone tried using nginx and webdriver together? Before I used nginx
> as a reverse proxy, I never had any problems, but now I can't get the
> browser being controlled by webdriver to get webpages.
> 
> An example from the python shell: 
> 
> >>> import selenium
> >>> from selenium import webdriver
> >>> profile = webdriver.FirefoxProfile()
> >>> profile.set_preference("network.proxy.type", 1)
> >>> profile.set_preference("network.proxy.http", "127.0.0.1")
> >>> profile.set_preference("network.proxy.http_port", "80")
> >>> driver = webdriver.Firefox(firefox_profile=profile)
> >>> driver.get("http://www.google.com") 
> 
> And then nothing happens.
> 
> Since it works when nginx isn't running, I'm guessing the problem is
> from the nginx side. Has anyone experienced something similar?

Since I really doubt www.google.com running nginx, I suppose you 
are using nginx on 127.0.0.1:80, right?

The problem is that nginx isn't a forward proxy and never was.

Maxim Dounin


From nginx-forum at nginx.us  Tue Sep 20 21:38:32 2011
From: nginx-forum at nginx.us (akamaka)
Date: Tue, 20 Sep 2011 17:38:32 -0400
Subject: need rewrite help
Message-ID: <19a5c00618e68b56baa128ca45971fcd.NginxMailingListEnglish@forum.nginx.org>

Hi,

please help me to rewrite this mod_rewrite rule:


RewriteRule    (img/(.+))$ webroot/$1     [L]


Thanks

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215460,215460#msg-215460


From appa at perusio.net  Tue Sep 20 21:50:44 2011
From: appa at perusio.net (=?UTF-8?B?QW50w7NuaW8=?= P. P. Almeida)
Date: Tue, 20 Sep 2011 22:50:44 +0100
Subject: need rewrite help
In-Reply-To: <19a5c00618e68b56baa128ca45971fcd.NginxMailingListEnglish@forum.nginx.org>
References: <19a5c00618e68b56baa128ca45971fcd.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <87mxdyzxvv.wl%appa@perusio.net>

On 20 Set 2011 22h38 WEST, nginx-forum at nginx.us wrote:

> Hi,
>
> please help me to rewrite this mod_rewrite rule:
>

> RewriteRule    (img/(.+))$ webroot/$1     [L]

Try:

location ^~ /img/ {
    location ~* /img/(.+)$ {
       return 301 /webroot/$1;
   }
}

Or 302 if you want it to be temporary.

--- appa


From latypoff at yandex.ru  Tue Sep 20 21:57:10 2011
From: latypoff at yandex.ru (Denis F. Latypoff)
Date: Wed, 21 Sep 2011 04:57:10 +0700
Subject: need rewrite help
In-Reply-To: <87mxdyzxvv.wl%appa@perusio.net>
References: <19a5c00618e68b56baa128ca45971fcd.NginxMailingListEnglish@forum.nginx.org>
	<87mxdyzxvv.wl%appa@perusio.net>
Message-ID: <638501316555830@web47.yandex.ru>

21.09.2011, 04:50, "Ant?nio P. P. Almeida" <appa at perusio.net>:
> On 20 Set 2011 22h38 WEST, nginx-forum at nginx.us wrote:
>
>> ?Hi,
>>
>> ?please help me to rewrite this mod_rewrite rule:
>> ?RewriteRule ???(img/(.+))$ webroot/$1 ????[L]
>
> Try:
>
> location ^~ /img/ {
> ????location ~* /img/(.+)$ {
> ???????return 301 /webroot/$1;
> ???}
> }
>
> Or 302 if you want it to be temporary.

There is no [R] flag in the rewrite rule, so it should be internal redirect

location /img/ {
    alias /path/to/webroot/;
}

-- 
br, Denis F. Latypoff.


From yomnanasser at gmail.com  Tue Sep 20 21:59:59 2011
From: yomnanasser at gmail.com (Yomna Nasser)
Date: Tue, 20 Sep 2011 17:59:59 -0400
Subject: WebDriver and Nginx
In-Reply-To: <20110920185446.GX1137@mdounin.ru>
References: <4bd63487c17a749c842d5e6c46c46985.NginxMailingListEnglish@forum.nginx.org>
	<20110920185446.GX1137@mdounin.ru>
Message-ID: <CAAba4RZ0V822jM=oSbCZ3nQnOvhLPK6P1epYO5CqTVTvVBxCyQ@mail.gmail.com>

The problem isn't that it can't access google.com (or whatever url),
but that it won't even try. If my code was successful, an error
message should be displayed in the browser saying that it can't access
google because nginx is running, or something along those lines (so,
exactly what you said).

The problem is that I can't control the instance of the browser once
it's opened. It won't respond to any navigation requests and I think
that nginx is somehow interfering, but I have no idea how. It doesn't
make much sense that a web server is interfering with the ability to
control a web browser via software :/

On Tue, Sep 20, 2011 at 2:54 PM, Maxim Dounin <mdounin at mdounin.ru> wrote:
> Hello!
>
> On Tue, Sep 20, 2011 at 02:47:26PM -0400, ynasser wrote:
>
>> Has anyone tried using nginx and webdriver together? Before I used nginx
>> as a reverse proxy, I never had any problems, but now I can't get the
>> browser being controlled by webdriver to get webpages.
>>
>> An example from the python shell:
>>
>> >>> import selenium
>> >>> from selenium import webdriver
>> >>> profile = webdriver.FirefoxProfile()
>> >>> profile.set_preference("network.proxy.type", 1)
>> >>> profile.set_preference("network.proxy.http", "127.0.0.1")
>> >>> profile.set_preference("network.proxy.http_port", "80")
>> >>> driver = webdriver.Firefox(firefox_profile=profile)
>> >>> driver.get("http://www.google.com")
>>
>> And then nothing happens.
>>
>> Since it works when nginx isn't running, I'm guessing the problem is
>> from the nginx side. Has anyone experienced something similar?
>
> Since I really doubt www.google.com running nginx, I suppose you
> are using nginx on 127.0.0.1:80, right?
>
> The problem is that nginx isn't a forward proxy and never was.
>
> Maxim Dounin
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx
>


From nginx-forum at nginx.us  Tue Sep 20 22:39:08 2011
From: nginx-forum at nginx.us (sailorfred)
Date: Tue, 20 Sep 2011 18:39:08 -0400
Subject: Need client cert common name as user
Message-ID: <8c35b9a8bdee80d739cf8f6e9d68e3cb.NginxMailingListEnglish@forum.nginx.org>

I'm in the process of converting from lighttpd to nginx, but I'm not
finding an equivalent of the lighttpd directive:

ssl.verifyclient.username = "SSL_CLIENT_S_DN_CN"

This sets the REMOTE_USER environment variable for the backend FastCGI
process, and logs the common name into the access log.

Is it possible to do this in nginx?

Thanks,

Fred

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215461,215461#msg-215461


From mdounin at mdounin.ru  Tue Sep 20 23:05:26 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Wed, 21 Sep 2011 03:05:26 +0400
Subject: Need client cert common name as user
In-Reply-To: <8c35b9a8bdee80d739cf8f6e9d68e3cb.NginxMailingListEnglish@forum.nginx.org>
References: <8c35b9a8bdee80d739cf8f6e9d68e3cb.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20110920230526.GD1137@mdounin.ru>

Hello!

On Tue, Sep 20, 2011 at 06:39:08PM -0400, sailorfred wrote:

> I'm in the process of converting from lighttpd to nginx, but I'm not
> finding an equivalent of the lighttpd directive:
> 
> ssl.verifyclient.username = "SSL_CLIENT_S_DN_CN"
> 
> This sets the REMOTE_USER environment variable for the backend FastCGI
> process, and logs the common name into the access log.
> 
> Is it possible to do this in nginx?

There is no direct replacement.  If you are ok with full cert subject 
DN as a user name, use something like this:

    fastcgi_pass ...
    fastcgi_param REMOTE_USER $ssl_client_s_dn;
    ...

Extracting only CN should be possible with "if" directive, though 
will require writing some regexp to parse CN from DN.  (And please 
make sure to read http://wiki.nginx.org/IfIsEvil as "if" may cause 
problems by itself if used in location context.)

Maxim Dounin


From igor at sysoev.ru  Wed Sep 21 03:38:02 2011
From: igor at sysoev.ru (Igor Sysoev)
Date: Wed, 21 Sep 2011 07:38:02 +0400
Subject: Need client cert common name as user
In-Reply-To: <20110920230526.GD1137@mdounin.ru>
References: <8c35b9a8bdee80d739cf8f6e9d68e3cb.NginxMailingListEnglish@forum.nginx.org>
	<20110920230526.GD1137@mdounin.ru>
Message-ID: <20110921033802.GB77063@nginx.com>

On Wed, Sep 21, 2011 at 03:05:26AM +0400, Maxim Dounin wrote:
> Hello!
> 
> On Tue, Sep 20, 2011 at 06:39:08PM -0400, sailorfred wrote:
> 
> > I'm in the process of converting from lighttpd to nginx, but I'm not
> > finding an equivalent of the lighttpd directive:
> > 
> > ssl.verifyclient.username = "SSL_CLIENT_S_DN_CN"
> > 
> > This sets the REMOTE_USER environment variable for the backend FastCGI
> > process, and logs the common name into the access log.
> > 
> > Is it possible to do this in nginx?
> 
> There is no direct replacement.  If you are ok with full cert subject 
> DN as a user name, use something like this:
> 
>     fastcgi_pass ...
>     fastcgi_param REMOTE_USER $ssl_client_s_dn;
>     ...
> 
> Extracting only CN should be possible with "if" directive, though 
> will require writing some regexp to parse CN from DN.  (And please 
> make sure to read http://wiki.nginx.org/IfIsEvil as "if" may cause 
> problems by itself if used in location context.)

The "map" directive with regex can be used instead of "if",
something like this:

      map  $ssl_client_s_dn  $ssl_client_s_dn_cn {
           default           "";
           ~/CN=(?<CN>[^/]+) $CN;
      };


-- 
Igor Sysoev


From igor at sysoev.ru  Wed Sep 21 05:01:44 2011
From: igor at sysoev.ru (Igor Sysoev)
Date: Wed, 21 Sep 2011 09:01:44 +0400
Subject: need rewrite help
In-Reply-To: <638501316555830@web47.yandex.ru>
References: <19a5c00618e68b56baa128ca45971fcd.NginxMailingListEnglish@forum.nginx.org>
	<87mxdyzxvv.wl%appa@perusio.net> <638501316555830@web47.yandex.ru>
Message-ID: <20110921050144.GE77063@nginx.com>

On Wed, Sep 21, 2011 at 04:57:10AM +0700, Denis F. Latypoff wrote:
> 21.09.2011, 04:50, "Ant?nio P. P. Almeida" <appa at perusio.net>:
> > On 20 Set 2011 22h38 WEST, nginx-forum at nginx.us wrote:
> >
> >> ?Hi,
> >>
> >> ?please help me to rewrite this mod_rewrite rule:
> >> ?RewriteRule ???(img/(.+))$ webroot/$1 ????[L]
> >
> > Try:
> >
> > location ^~ /img/ {
> > ????location ~* /img/(.+)$ {
> > ???????return 301 /webroot/$1;
> > ???}
> > }
> >
> > Or 302 if you want it to be temporary.
> 
> There is no [R] flag in the rewrite rule, so it should be internal redirect
> 
> location /img/ {
>     alias /path/to/webroot/;
> }

I think $1 in RewriteRule is a whole "img/...":

$ pcretest 
PCRE version 7.9 2009-04-11

  re> #(img/(.+))$#
data> img/qqq
 0: img/qqq
 1: img/qqq
 2: qqq
data> ^D

So it should be "root":

location /img/ {
    root  /path/to/webroot/;
}


-- 
Igor Sysoev


From quan.nexthop at gmail.com  Wed Sep 21 09:58:59 2011
From: quan.nexthop at gmail.com (Geoge.Q)
Date: Wed, 21 Sep 2011 02:58:59 -0700
Subject: nginx-1.1.4
In-Reply-To: <20110920125219.GP1137@mdounin.ru>
References: <20110920112225.GC54733@nginx.com>
	<20110920114004.GO1137@mdounin.ru>
	<32453ED0-15C3-406F-9EA5-92B05E7B0F60@exsisto.com>
	<B293D35E9E20694A8D9DFFE1BC655C556AB265E773@vvexch.verivue.com>
	<20110920125219.GP1137@mdounin.ru>
Message-ID: <CAP_76MaQnmum_2a-vf6ojFUL_AwZyD7qXj3G9fJj8c8jCKgENA@mail.gmail.com>

When is it ported to stable version? I means the upstream keep-alive feature?

Thanks
George

2011/9/20, Maxim Dounin <mdounin at mdounin.ru>:
> Hello!
>
> On Tue, Sep 20, 2011 at 08:29:06AM -0400, Sirsiwal, Umesh wrote:
>
>> Thanks to both of you. Great work.
>>
>> If I understand correctly, the only way to use
>> upstream_keepalive module is with upstream module. This implies
>> I cannot use upstream_keepalive where the upstream is identified
>> by a variable. For example:
>>
>> proxy_pass http://$host$request_uri;
>>
>> Is my understand correct?
>
> Yes.
>
>> Is there a way around this restriction?
>
> Currently no.
>
> Maxim Dounin
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx
>

-- 
?????????


From mdounin at mdounin.ru  Wed Sep 21 10:20:46 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Wed, 21 Sep 2011 14:20:46 +0400
Subject: nginx-1.1.4
In-Reply-To: <CAP_76MaQnmum_2a-vf6ojFUL_AwZyD7qXj3G9fJj8c8jCKgENA@mail.gmail.com>
References: <20110920112225.GC54733@nginx.com>
	<20110920114004.GO1137@mdounin.ru>
	<32453ED0-15C3-406F-9EA5-92B05E7B0F60@exsisto.com>
	<B293D35E9E20694A8D9DFFE1BC655C556AB265E773@vvexch.verivue.com>
	<20110920125219.GP1137@mdounin.ru>
	<CAP_76MaQnmum_2a-vf6ojFUL_AwZyD7qXj3G9fJj8c8jCKgENA@mail.gmail.com>
Message-ID: <20110921102045.GH1137@mdounin.ru>

Hello!

On Wed, Sep 21, 2011 at 02:58:59AM -0700, Geoge.Q wrote:

> When is it ported to stable version? I means the upstream keep-alive feature?

Most likely never: there are several API changes which are 
required for this.  Breaking API in stable is a bad thing.

Maxim Dounin

> 
> Thanks
> George
> 
> 2011/9/20, Maxim Dounin <mdounin at mdounin.ru>:
> > Hello!
> >
> > On Tue, Sep 20, 2011 at 08:29:06AM -0400, Sirsiwal, Umesh wrote:
> >
> >> Thanks to both of you. Great work.
> >>
> >> If I understand correctly, the only way to use
> >> upstream_keepalive module is with upstream module. This implies
> >> I cannot use upstream_keepalive where the upstream is identified
> >> by a variable. For example:
> >>
> >> proxy_pass http://$host$request_uri;
> >>
> >> Is my understand correct?
> >
> > Yes.
> >
> >> Is there a way around this restriction?
> >
> > Currently no.
> >
> > Maxim Dounin
> >
> > _______________________________________________
> > nginx mailing list
> > nginx at nginx.org
> > http://mailman.nginx.org/mailman/listinfo/nginx
> >
> 
> -- 
> ?????????
> 
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx


From igor at sysoev.ru  Wed Sep 21 10:26:29 2011
From: igor at sysoev.ru (Igor Sysoev)
Date: Wed, 21 Sep 2011 14:26:29 +0400
Subject: nginx-1.1.4
In-Reply-To: <20110921102045.GH1137@mdounin.ru>
References: <20110920112225.GC54733@nginx.com>
	<20110920114004.GO1137@mdounin.ru>
	<32453ED0-15C3-406F-9EA5-92B05E7B0F60@exsisto.com>
	<B293D35E9E20694A8D9DFFE1BC655C556AB265E773@vvexch.verivue.com>
	<20110920125219.GP1137@mdounin.ru>
	<CAP_76MaQnmum_2a-vf6ojFUL_AwZyD7qXj3G9fJj8c8jCKgENA@mail.gmail.com>
	<20110921102045.GH1137@mdounin.ru>
Message-ID: <6C5C3A6B-D538-46BA-8825-F869297856FA@sysoev.ru>

On Sep 21, 2011, at 14:20 , Maxim Dounin wrote:

> Hello!
> 
> On Wed, Sep 21, 2011 at 02:58:59AM -0700, Geoge.Q wrote:
> 
>> When is it ported to stable version? I means the upstream keep-alive feature?
> 
> Most likely never: there are several API changes which are 
> required for this.  Breaking API in stable is a bad thing.

At some point 1.1.x will just become stable :)


--
Igor Sysoev
http://sysoev.ru/en/


From gmm at csdoc.com  Wed Sep 21 10:51:34 2011
From: gmm at csdoc.com (Gena Makhomed)
Date: Wed, 21 Sep 2011 13:51:34 +0300
Subject: nginx-1.1.4
In-Reply-To: <6C5C3A6B-D538-46BA-8825-F869297856FA@sysoev.ru>
References: <20110920112225.GC54733@nginx.com>
	<20110920114004.GO1137@mdounin.ru>
	<32453ED0-15C3-406F-9EA5-92B05E7B0F60@exsisto.com>
	<B293D35E9E20694A8D9DFFE1BC655C556AB265E773@vvexch.verivue.com>
	<20110920125219.GP1137@mdounin.ru>
	<CAP_76MaQnmum_2a-vf6ojFUL_AwZyD7qXj3G9fJj8c8jCKgENA@mail.gmail.com>
	<20110921102045.GH1137@mdounin.ru>
	<6C5C3A6B-D538-46BA-8825-F869297856FA@sysoev.ru>
Message-ID: <4E79C1B6.50808@csdoc.com>

On 21.09.2011 13:26, Igor Sysoev wrote:

>>> When is it ported to stable version? I means the upstream keep-alive feature?
>>
>> Most likely never: there are several API changes which are
>> required for this.  Breaking API in stable is a bad thing.
>
> At some point 1.1.x will just become stable :)

major.minor.revision

nginx will not use odd minor version numbers to denote development 
releases and even minor version numbers to denote stable releases?

1.0.x - stable branch
1.1.x - devel branch

1.2.x - stable branch
1.3.x - devel branch

and so on?

if not - it will be many confusion, which version 1.1.x is stable,
and which nginx 1.2.x version is devel or stable.

use odd minor version numbers to denote development releases
and even minor version numbers to denote stable releases -
imho is very useful and clear version numbering scheme.

(but looks like apache/squid/haproxy not use this scheme)

-- 
Best regards,
  Gena


From mdounin at mdounin.ru  Wed Sep 21 11:11:08 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Wed, 21 Sep 2011 15:11:08 +0400
Subject: nginx-1.1.4
In-Reply-To: <4E79C1B6.50808@csdoc.com>
References: <20110920112225.GC54733@nginx.com>
	<20110920114004.GO1137@mdounin.ru>
	<32453ED0-15C3-406F-9EA5-92B05E7B0F60@exsisto.com>
	<B293D35E9E20694A8D9DFFE1BC655C556AB265E773@vvexch.verivue.com>
	<20110920125219.GP1137@mdounin.ru>
	<CAP_76MaQnmum_2a-vf6ojFUL_AwZyD7qXj3G9fJj8c8jCKgENA@mail.gmail.com>
	<20110921102045.GH1137@mdounin.ru>
	<6C5C3A6B-D538-46BA-8825-F869297856FA@sysoev.ru>
	<4E79C1B6.50808@csdoc.com>
Message-ID: <20110921111108.GI1137@mdounin.ru>

Hello!

On Wed, Sep 21, 2011 at 01:51:34PM +0300, Gena Makhomed wrote:

> On 21.09.2011 13:26, Igor Sysoev wrote:
> 
> >>>When is it ported to stable version? I means the upstream keep-alive feature?
> >>
> >>Most likely never: there are several API changes which are
> >>required for this.  Breaking API in stable is a bad thing.
> >
> >At some point 1.1.x will just become stable :)
> 
> major.minor.revision
> 
> nginx will not use odd minor version numbers to denote development
> releases and even minor version numbers to denote stable releases?
> 
> 1.0.x - stable branch
> 1.1.x - devel branch
> 
> 1.2.x - stable branch
> 1.3.x - devel branch
> 
> and so on?
> 
> if not - it will be many confusion, which version 1.1.x is stable,
> and which nginx 1.2.x version is devel or stable.
> 
> use odd minor version numbers to denote development releases
> and even minor version numbers to denote stable releases -
> imho is very useful and clear version numbering scheme.

Yes, I've already proposed this to Igor to simplify stable vs. 
devel distinction.  He isn't convinced yet, but we have some time 
to think about this.

Maxim Dounin


From nginx-forum at nginx.us  Wed Sep 21 12:39:42 2011
From: nginx-forum at nginx.us (Hety)
Date: Wed, 21 Sep 2011 08:39:42 -0400
Subject: $server_name in error_log logfile name
Message-ID: <d920539cb3e93201537aebedad3f2e96.NginxMailingListEnglish@forum.nginx.org>

I'm currently trying to make a template for future virtual hosts. To
minimize the amount of editing required i am using $server_name in some
parts of the template.

For example:
server_name example.com
access_log  /var/log/www/$server_name-access  combined;
error_log   /var/log/www/$server_name-error   error;

While it works great for access_log directive(logfile is named
example.com-access), error log file name is, actually
$server_name-error. 
I'm running nginx 1.0.6

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215503,215503#msg-215503


From nginx-forum at nginx.us  Wed Sep 21 13:04:44 2011
From: nginx-forum at nginx.us (Hety)
Date: Wed, 21 Sep 2011 09:04:44 -0400
Subject: Enabeling SSL causes Redirect Loop errors ???
In-Reply-To: <be92617b4417a38334e91a54fc1fe059.NginxMailingListEnglish@forum.nginx.org>
References: <be92617b4417a38334e91a54fc1fe059.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <d58f8884f3b903d887a6cb75aed7d98e.NginxMailingListEnglish@forum.nginx.org>

There are two things to be aware of here:
1. fastcgi_param HTTPS on; - add this to your location ~ \.php$ {}.
2. If you use php-fpm chroot set:
fastcgi_param SCRIPT_FILENAME $fastcgi_script_name;
fastcgi_param PATH_TRANSLATED $fastcgi_script_name;
Thats because your document root will be just /, but only if your chroot
dir = your actual site dir.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215136,215505#msg-215505


From mdounin at mdounin.ru  Wed Sep 21 13:31:50 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Wed, 21 Sep 2011 17:31:50 +0400
Subject: $server_name in error_log logfile name
In-Reply-To: <d920539cb3e93201537aebedad3f2e96.NginxMailingListEnglish@forum.nginx.org>
References: <d920539cb3e93201537aebedad3f2e96.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20110921133150.GQ1137@mdounin.ru>

Hello!

On Wed, Sep 21, 2011 at 08:39:42AM -0400, Hety wrote:

> I'm currently trying to make a template for future virtual hosts. To
> minimize the amount of editing required i am using $server_name in some
> parts of the template.
> 
> For example:
> server_name example.com
> access_log  /var/log/www/$server_name-access  combined;
> error_log   /var/log/www/$server_name-error   error;
> 
> While it works great for access_log directive(logfile is named
> example.com-access), error log file name is, actually
> $server_name-error. 
> I'm running nginx 1.0.6

The "error_log" directive doesn't support variables.

Addtionally, this is not intended use of variables.  They are 
run-time and evaluated for each request.  You want to use some 
config generator instead.

Maxim Dounin


From nginx-forum at nginx.us  Wed Sep 21 14:18:31 2011
From: nginx-forum at nginx.us (ahu)
Date: Wed, 21 Sep 2011 10:18:31 -0400
Subject: mp4 module
In-Reply-To: <20110915052602.GE73767@nginx.com>
References: <20110915052602.GE73767@nginx.com>
Message-ID: <072588df6d07290a30ba33099451c5f9.NginxMailingListEnglish@forum.nginx.org>

I am very curious about what the difference between this modle and the
3rd
modle(http://h264.code-shop.com/trac/wiki/Mod-H264-Streaming-Nginx-Version2).
Which one should I use?

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215011,215506#msg-215506


From agentzh at gmail.com  Wed Sep 21 15:35:52 2011
From: agentzh at gmail.com (agentzh)
Date: Wed, 21 Sep 2011 23:35:52 +0800
Subject: [ANN] ngx_openresty 1.0.6.12 released
In-Reply-To: <CAB4Tn6Ok9zpm1a8B51fJ-wsVa2zyt5SBmy6LDww7-avyAbPUsQ@mail.gmail.com>
References: <CAB4Tn6Ok9zpm1a8B51fJ-wsVa2zyt5SBmy6LDww7-avyAbPUsQ@mail.gmail.com>
Message-ID: <CAB4Tn6N7R3+fZ1E=UvM3779P3G8HC1kNayXF_8bCZAUGD4Hb2w@mail.gmail.com>

Hi, folks!

I've just kicked the ngx_openresty 1.0.6.12 release out of the door, which
is the first stable release of ngx_openresty based on the nginx 1.0.6 core:

    http://openresty.org/#Download

Here goes the complete change log, as compared to the last stable release
1.0.5.1:

   - upgraded RdsJsonNginxModule to v0.12rc4.
      - made rds_json_ret honor rds_json_success_property and
      rds_json_user_property. thanks Liseen Wan (???)
      - only register our output filters when the rds_json directive is
      actually used in nginx.conf.
   - upgraded RdsCsvNginxModule to v0.03.
      - only register our output filters when the rds_csv directive is
      actually used in nginx.conf.


   - added new options --with-luajit=PATH and --with-lua51=PATH to the
   ./configure script. thanks NginxUser.
   - upgraded DrizzleNginxModule to v0.1.1.


   - upgraded RdsJsonNginxModule to 0.12rc3.
      - implemented new directive rds_json_root.
      - implemented new directive rds_json_success_property.
      - implemented new directive rds_json_user_property.


   - upgraded LuaNginxModule to 0.3.1rc3.
      - implemented and documented the API for reading response headers from
      within Lua: value = ngx.header.HEADER.
      - fixed a bug when setting a multi-value response header to a single
      value (via writing to ngx.header.HEADER): the single value will be
      repeated on each old value.
   - upgraded EchoNginxModule to 0.37rc4.
      - fixed a bug in echo_after_body: when network is not perfect, data
      truncation might occur. we should have taken into account
NGX_AGAINreturned by the downstream output filters. thanks Sparsh
Gupta.
   - upgraded HeadersMoreNginxModule to v0.16rc2.
      - fixed a bug when setting a multi-value response header to a single
      value: the single value will be repeated on each old value.
   - applied the patch from Maxim Dounin to fix a bug in the standard
   ngx_gzip module when dealing with empty flush buffers:
   http://mailman.nginx.org/pipermail/nginx-devel/2011-February/000730.html
   - updated the no-pool-patch to eliminate the -Wset-but-not-used warnings
   issued by gcc 4.6.0.


   - upgraded LuaNginxModule to 0.3.1rc1.
      - fixed a bug when the both the main request and the subrequest are
      POST requests with a body: we should not forward the main request's
      Content-Length headers to the user subrequests. thanks ??.


   - upgraded HeadersMoreNginxModule to 0.16rc1.
      - fixed on-demand hander/filter registration trick for HUP signal
      restarts.
      - added some debugging outputs that can be enabled by the
--with-debugoption when building Nginx or
      OpenResty.
   - Upgraded the nginx core to 1.0.6.

Enjoy!
-agentzh
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110921/99eab4ba/attachment.html>

From avishai.ish-shalom at mail.huji.ac.il  Wed Sep 21 20:03:31 2011
From: avishai.ish-shalom at mail.huji.ac.il (Avishai Ish-Shalom)
Date: Wed, 21 Sep 2011 23:03:31 +0300
Subject: process corruption on CentOS 5.6
Message-ID: <4E7A4313.9090703@mail.huji.ac.il>

On CentOS 5.6, nginx 1.1.4 configured to work with php-fpm using fastcgi
port. nginx was compiled with: configure --prefix=/usr/local/nginx
--with-http_ssl_module --with-http_realip_module
--with-http_geoip_module --with-http_flv_module
--with-http_gzip_static_module --with-http_stub_status_module
--with-http_addition_module --with-http_image_filter_module
--with-http_sub_module --with-http_degradation_module
--add-module=addons/ngx_upstream_fair --with-cc-opt="-D
NGX_UPSTREAM_KEEPALIVE_PATCHED"

every few seconds, i get:

*** glibc detected *** nginx: worker process: double free or corruption
(!prev): 0x000000001d069800 ***

======= Backtrace: =========
/lib64/libc.so.6[0x3cd4c7245f]
/lib64/libc.so.6(cfree+0x4b)[0x3cd4c728bb]
nginx: worker process[0x407628]
nginx: worker process[0x43c6e1]
nginx: worker process[0x43ec4a]
nginx: worker process[0x43e691]
nginx: worker process[0x43eca3]
nginx: worker process[0x43ed2e]
nginx: worker process[0x43d26d]
nginx: worker process[0x421a96]
nginx: worker process[0x41ab68]
nginx: worker process[0x420720]
nginx: worker process[0x41f057]
nginx: worker process[0x420f09]
nginx: worker process[0x406a7a]
/lib64/libc.so.6(__libc_start_main+0xf4)[0x3cd4c1d994]
nginx: worker process[0x405249]
======= Memory map: ========
00400000-00482000 r-xp 00000000 68:05 6882046                           
/usr/local/nginx/sbin/nginx
00681000-00693000 rw-p 00081000 68:05 6882046                           
/usr/local/nginx/sbin/nginx
00693000-006a1000 rw-p 00693000 00:00 0
1d041000-1d22b000 rw-p 1d041000 00:00 0                                 
[heap]
3cd4800000-3cd481c000 r-xp 00000000 68:05 6389762                       
/lib64/ld-2.5.so
3cd4a1c000-3cd4a1d000 r--p 0001c000 68:05 6389762                       
/lib64/ld-2.5.so
3cd4a1d000-3cd4a1e000 rw-p 0001d000 68:05 6389762                       
/lib64/ld-2.5.so
3cd4c00000-3cd4d4e000 r-xp 00000000 68:05 6389764                       
/lib64/libc-2.5.so
3cd4d4e000-3cd4f4e000 ---p 0014e000 68:05 6389764                       
/lib64/libc-2.5.so
3cd4f4e000-3cd4f52000 r--p 0014e000 68:05 6389764                       
/lib64/libc-2.5.so
3cd4f52000-3cd4f53000 rw-p 00152000 68:05 6389764                       
/lib64/libc-2.5.so
3cd4f53000-3cd4f58000 rw-p 3cd4f53000 00:00 0
3cd5000000-3cd501e000 r-xp 00000000 68:05 6391224                       
/lib64/libpcre.so.0.0.1
3cd501e000-3cd521e000 ---p 0001e000 68:05 6391224                       
/lib64/libpcre.so.0.0.1
3cd521e000-3cd521f000 rw-p 0001e000 68:05 6391224                       
/lib64/libpcre.so.0.0.1
3cd5400000-3cd5416000 r-xp 00000000 68:05 6389787                       
/lib64/libpthread-2.5.so
3cd5416000-3cd5615000 ---p 00016000 68:05 6389787                       
/lib64/libpthread-2.5.so
3cd5615000-3cd5616000 r--p 00015000 68:05 6389787                       
/lib64/libpthread-2.5.so
3cd5616000-3cd5617000 rw-p 00016000 68:05 6389787                       
/lib64/libpthread-2.5.so
3cd5617000-3cd561b000 rw-p 3cd5617000 00:00 0
3cd5800000-3cd5882000 r-xp 00000000 68:05 6389811                       
/lib64/libm-2.5.so
3cd5882000-3cd5a81000 ---p 00082000 68:05 6389811                       
/lib64/libm-2.5.so
3cd5a81000-3cd5a82000 r--p 00081000 68:05 6389811                       
/lib64/libm-2.5.so
3cd5a82000-3cd5a83000 rw-p 00082000 68:05 6389811                       
/lib64/libm-2.5.so
3cd5c00000-3cd5c14000 r-xp 00000000 68:05 6260001                       
/usr/lib64/libz.so.1.2.3
3cd5c14000-3cd5e13000 ---p 00014000 68:05 6260001                       
/usr/lib64/libz.so.1.2.3
3cd5e13000-3cd5e14000 rw-p 00013000 68:05 6260001                       
/usr/lib64/libz.so.1.2.3
3cd6000000-3cd6020000 r-xp 00000000 68:05 6272360                       
/usr/lib64/libgd.so.2.0.0
3cd6020000-3cd621f000 ---p 00020000 68:05 6272360                       
/usr/lib64/libgd.so.2.0.0
3cd621f000-3cd6240000 rw-p 0001f000 68:05 6272360                       
/usr/lib64/libgd.so.2.0.0
3cd6240000-3cd6254000 rw-p 3cd6240000 00:00 0
3cd6400000-3cd6415000 r-xp 00000000 68:05 6391132                       
/lib64/libselinux.so.1
3cd6415000-3cd6615000 ---p 00015000 68:05 6391132                       
/lib64/libselinux.so.1
3cd6615000-3cd6617000 rw-p 00015000 68:05 6391132                       
/lib64/libselinux.so.1
3cd6617000-3cd6618000 rw-p 3cd6617000 00:00 0
3cd6c00000-3cd6c0d000 r-xp 00000000 68:05 6391157                       
/lib64/libgcc_s-4.1.2-20080825.so.1
3cd6c0d000-3cd6e0d000 ---p 0000d000 68:05 6391157                       
/lib64/libgcc_s-4.1.2-20080825.so.1
3cd6e0d000-3cd6e0e000 rw-p 0000d000 68:05 6391157                       
/lib64/libgcc_s-4.1.2-20080825.so.1
3cd7400000-3cd7409000 r-xp 00000000 68:05 6391136                       
/lib64/libcrypt-2.5.so
3cd7409000-3cd7608000 ---p 00009000 68:05 6391136                       
/lib64/libcrypt-2.5.so
3cd7608000-3cd7609000 r--p 00008000 68:05 6391136                       
/lib64/libcrypt-2.5.so
3cd7609000-3cd760a000 rw-p 00009000 68:05 6391136                       
/lib64/libcrypt-2.5.so
3cd760a000-3cd7638000 rw-p 3cd760a000 00:00 0
3cd7800000-3cd787f000 r-xp 00000000 68:05 6260010                       
/usr/lib64/libfreetype.so.6.3.10
3cd787f000-3cd7a7f000 ---p 0007f000 68:05 6260010                       
/usr/lib64/libfreetype.so.6.3.10
3cd7a7f000-3cd7a84000 rw-p 0007f000 68:05 6260010                       
/usr/lib64/libfreetype.so.6.3.10
3cd8000000-3cd8021000 r-xp 00000000 68:05 6260000                       
/usr/lib64/libjpeg.so.62.0.0
3cd8021000-3cd8220000 ---p 00021000 68:05 6260000                       
/usr/lib64/libjpeg.so.62.0.0
3cd8220000-3cd8221000 rw-p 00020000 68:05 6260000                       
/usr/lib64/libjpeg.so.62.0.0
3cd8400000-3cd8429000 r-xp 00000000 68:05 6260011                       
/usr/lib64/libfontconfig.so.1.1.0
3cd8429000-3cd8629000 ---p 00029000 68:05 6260011                       
/usr/lib64/libfontconfig.so.1.1.0
3cd8629000-3cd8633000 rw-p 00029000 68:05 6260011                       
/usr/lib64/libfontconfig.so.1.1.0
3cd8633000-3cd8634000 rw-p 3cd8633000 00:00 0
3cd8800000-3cd8905000 r-xp 00000000 68:05 6260015                       
/usr/lib64/libX11.so.6.2.0
3cd8905000-3cd8b05000 ---p 00105000 68:05 6260015                       
/usr/lib64/libX11.so.6.2.0
3cd8b05000-3cd8b0c000 rw-p 00105000 68:05 6260015                       
/usr/lib64/libX11.so.6.2.0
3cd8c00000-3cd8c02000 r-xp 00000000 68:05 6260013                       
/usr/lib64/libXau.so.6.0.0
3cd8c02000-3cd8e01000 ---p 00002000 68:05 6260013                       
/usr/lib64/libXau.so.6.0.0
3cd8e01000-3cd8e02000 rw-p 00001000 68:05 6260013                       
/usr/lib64/libXau.so.6.0.0
3cd9400000-3cd9420000 r-xp 00000000 68:05 6391150                       
/lib64/libexpat.so.0.5.0
3cd9420000-3cd961f000 ---p 00020000 68:05 6391150                       
/lib64/libexpat.so.0.5.0
3cd961f000-3cd9622000 rw-p 0001f000 68:05 6391150                       
/lib64/libexpat.so.0.5.0
3cd9c00000-3cd9c05000 r-xp 00000000 68:05 6260014                       
/usr/lib64/libXdmcp.so.6.0.0
3cd9c05000-3cd9e04000 ---p 00005000 68:05 6260014                       
/usr/lib64/libXdmcp.so.6.0.0
3cd9e04000-3cd9e05000 rw-p 00004000 68:05 6260014                       
/usr/lib64/libXdmcp.so.6.0.0
3cda400000-3cda402000 r-xp 00000000 68:05 6391216                       
/lib64/libkeyutils-1.2.so
3cda402000-3cda601000 ---p 00002000 68:05 6391216                       
/lib64/libkeyutils-1.2.so
3cda601000-3cda602000 rw-p 00001000 68:05 6391216                       
/lib64/libkeyutils-1.2.so
3cda800000-3cda811000 r-xp 00000000 68:05 6391217                       
/lib64/libresolv-2.5.so
3cda811000-3cdaa11000 ---p 00011000 68:05 6391217                       
/lib64/libresolv-2.5.so
3cdaa11000-3cdaa12000 r--p 00011000 68:05 6391217                       
/lib64/libresolv-2.5.so
3cdaa12000-3cdaa13000 rw-p 00012000 68:05 6391217                       
/lib64/libresolv-2.5.so
3cdaa13000-3cdaa15000 rw-p 3cdaa13000 00:00 0
3cdb000000-3cdb002000 r-xp 00000000 68:05 6391218                       
/lib64/libcom_err.so.2.1
3cdb002000-3cdb201000 ---p 00002000 68:05 6391218                       
/lib64/libcom_err.so.2.1
3cdb201000-3cdb202000 rw-p 00001000 68:05 6391218                       
/lib64/libcom_err.so.2.1
3cdc000000-3cdc023000 r-xp 00000000 68:05 6260012                       
/usr/lib64/libpng12.so.0.10.0
3cdc023000-3cdc223000 ---p 00023000 68:05 6260012                       
/usr/lib64/libpng12.so.0.10.0
3cdc223000-3cdc224000 rw-p 00023000 68:05 6260012                       
/usr/lib64/libpng12.so.0.10.0
3cdc400000-3cdc52d000 r-xp 00000000 68:05 6391146                       
/lib64/libcrypto.so.0.9.8e
3cdc52d000-3cdc72c000 ---p 0012d000 68:05 6391146                       
/lib64/libcrypto.so.0.9.8e
3cdc72c000-3cdc74d000 rw-p 0012c000 68:05 6391146                       
/lib64/libcrypto.so.0.9.8e
3cdc74d000-3cdc751000 rw-p 3cdc74d000 00:00 0
3cdc800000-3cdc892000 r-xp 00000000 68:05 6270008                       
/usr/lib64/libkrb5.so.3.3
3cdc892000-3cdca91000 ---p 00092000 68:05 6270008                       
/usr/lib64/libkrb5.so.3.3
3cdca91000-3cdca95000 rw-p 00091000 68:05 6270008                       
/usr/lib64/libkrb5.so.3.3
3cdcc00000-3cdcc24000 r-xp 00000000 68:05 6270001                       
/usr/lib64/libk5crypto.so.3.1
3cdcc24000-3cdce23000 ---p 00024000 68:05 6270001                       
/usr/lib64/libk5crypto.so.3.1
3cdce23000-3cdce25000 rw-p 00023000 68:05 6270001                       
/usr/lib64/libk5crypto.so.3.1
3cdd000000-3cdd008000 r-xp 00000000 68:05 6269995                       
/usr/lib64/libkrb5support.so.0.1
3cdd008000-3cdd207000 ---p 00008000 68:05 6269995                       
/usr/lib64/libkrb5support.so.0.1
3cdd207000-3cdd208000 rw-p 00007000 68:05 6269995                       
/usr/lib64/libkrb5support.so.0.1
3cdd400000-3cdd42c000 r-xp 00000000 68:05 6270239                       
/usr/lib64/libgssapi_krb5.so.2.2
3cdd42c000-3cdd62c000 ---p 0002c000 68:05 6270239                       
/usr/lib64/libgssapi_krb5.so.2.2
3cdd62c000-3cdd62e000 rw-p 0002c000 68:05 6270239                       
/usr/lib64/libgssapi_krb5.so.2.2
3cddc00000-3cddc46000 r-xp 00000000 68:05 6391219                       
/lib64/libssl.so.0.9.8e
3cddc46000-3cdde46000 ---p 00046000 68:05 6391219                       
/lib64/libssl.so.0.9.8e
3cdde46000-3cdde4c000 rw-p 00046000 68:05 6391219                       
/lib64/libssl.so.0.9.8e
3fee800000-3fee832000 r-xp 00000000 68:05 6276884                       
/usr/lib64/libGeoIP.so.1.4.6
3fee832000-3feea32000 ---p 00032000 68:05 6276884                       
/usr/lib64/libGeoIP.so.1.4.6
3feea32000-3feea33000 rw-p 00032000 68:05 6276884                       
/usr/lib64/libGeoIP.so.1.4.6
2b71700eb000-2b71700ed000 rw-p 2b71700eb000 00:00 0
2b71700ed000-2b71700f5000 rw-s 00000000 00:09 3585823051                
/dev/zero (deleted)
2b71700f5000-2b71700f6000 rw-s 00000000 00:09 3585823054                
/dev/zero (deleted)
2b71700fc000-2b71700fd000 rw-p 2b71700fc000 00:00 0
2b71700fd000-2b71700ff000 r-xp 00000000 68:05 6389771                   
/lib64/libdl-2.5.so
2b71700ff000-2b71702ff000 ---p 00002000 68:05 6389771                   
/lib64/libdl-2.5.so
2b71702ff000-2b7170300000 r--p 00002000 68:05 6389771                   
/lib64/libdl-2.5.so
2b7170300000-2b7170301000 rw-p 00003000 68:05 6389771                   
/lib64/libdl-2.5.so
2b7170301000-2b7170304000 rw-p 2b7170301000 00:00 0
2b7170304000-2b7170314000 r-xp 00000000 68:05 6269359                   
/usr/lib64/libXpm.so.4.11.0
2b7170314000-2b7170514000 ---p 00010000 68:05 6269359                   
/usr/lib64/libXpm.so.4.11.0
2b7170514000-2b7170515000 rw-p 00010000 68:05 6269359                   
/usr/lib64/libXpm.so.4.11.0
2b7170515000-2b717051a000 rw-p 2b7170515000 00:00 0
2b717051a000-2b7170555000 r-xp 00000000 68:05 6389821                   
/lib64/libsepol.so.1
2b7170555000-2b7170755000 ---p 0003b000 68:05 6389821                   
/lib64/libsepol.so.1
2b7170755000-2b7170756000 rw-p 0003b000 68:05 6389821                   
/lib64/libsepol.so.1
2b7170756000-2b7170763000 rw-p 2b7170756000 00:00 0
2b7170763000-2b717076d000 r-xp 00000000 68:05 6389786                   
/lib64/libnss_files-2.5.so
2b717076d000-2b717096c000 ---p 0000a000 68:05 6389786                   
/lib64/libnss_files-2.5.so
2b717096c000-2b717096d000 r--p 00009000 68:05 6389786                   
/lib64/libnss_files-2.5.so
2b717096d000-2b717096e000 rw-p 0000a000 68:05 6389786                   
/lib64/libnss_files-2.5.so
2b717096e000-2b7170c91000 rw-p 2b717096e000 00:00 0
7fffa9293000-7fffa92a8000 rw-p 7ffffffe9000 00:00 0                     
[stack]
7fffa92a9000-7fffa92ad000 r-xp 7fffa92a9000 00:00 0                     
[vdso]
ffffffffff600000-ffffffffffe00000 ---p 00000000 00:00 0                 
[vsyscall]
2011/09/21 22:00:01 [alert] 29216#0: worker process 29786 exited on signal 6


-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/pkcs7-signature
Size: 4376 bytes
Desc: S/MIME Cryptographic Signature
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110921/f8538e8b/attachment-0001.bin>

From mdounin at mdounin.ru  Wed Sep 21 20:18:10 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Thu, 22 Sep 2011 00:18:10 +0400
Subject: process corruption on CentOS 5.6
In-Reply-To: <4E7A4313.9090703@mail.huji.ac.il>
References: <4E7A4313.9090703@mail.huji.ac.il>
Message-ID: <20110921201810.GA1137@mdounin.ru>

Hello!

On Wed, Sep 21, 2011 at 11:03:31PM +0300, Avishai Ish-Shalom wrote:

> On CentOS 5.6, nginx 1.1.4 configured to work with php-fpm using fastcgi
> port. nginx was compiled with: configure --prefix=/usr/local/nginx
> --with-http_ssl_module --with-http_realip_module
> --with-http_geoip_module --with-http_flv_module
> --with-http_gzip_static_module --with-http_stub_status_module
> --with-http_addition_module --with-http_image_filter_module
> --with-http_sub_module --with-http_degradation_module
> --add-module=addons/ngx_upstream_fair --with-cc-opt="-D
> NGX_UPSTREAM_KEEPALIVE_PATCHED"

Just a side note: you don't need --with-cc-opt="-D 
NGX_UPSTREAM_KEEPALIVE_PATCHED"'.

> 
> every few seconds, i get:
> 
> *** glibc detected *** nginx: worker process: double free or corruption
> (!prev): 0x000000001d069800 ***

Could you please provide backtrace?  See here for details:
http://wiki.nginx.org/Debugging

Alternatively, please try the patch from here:
http://mailman.nginx.org/pipermail/nginx-devel/2011-September/001233.html

(And provide backtrace if it doesn't help.)

Maxim Dounin


From info at bizhat.com  Wed Sep 21 20:44:04 2011
From: info at bizhat.com (BizHat.com Support)
Date: Thu, 22 Sep 2011 02:14:04 +0530
Subject: Problem with MediaWiki File page
Message-ID: <CAGWcB0N1RvaiFLgJJGc7en0FJxbbd0b_k5nSF-JQ=kPN2s+u6A@mail.gmail.com>

HI,

I have following nginx configuration for a MediaWiki based site


server {
        listen 174.36.130.243:80;
        server_name  webhostingneeds.com www.webhostingneeds.com;

        location ~* \.(js|css|rdf|xml|ico|txt|gif|jpg|png|jpeg|html)$ {
        root   /home/webhost/public_html/;
        access_log  /var/log/nginx/webhostingneeds.com_static.log;
        }

        location / {
            proxy_pass         http://174.36.130.243:81/;
            proxy_redirect     off;
            proxy_set_header   Host             $host;
            proxy_set_header   X-Real-IP        $remote_addr;
            proxy_set_header   X-Forwarded-For  $proxy_add_x_forwarded_for;
            client_max_body_size       1m;
            client_body_buffer_size    128k;
            proxy_connect_timeout      90;
            proxy_send_timeout         90;
            proxy_read_timeout         90;
            proxy_buffer_size          4k;
            proxy_buffers              4 32k;
            proxy_busy_buffers_size    64k;
            proxy_temp_file_write_size 64k;
        }
}

The problem is when a file is uploaded to MediaWIki, i shows a page like

http://webhostingneeds.com/File:FILENAME.exten

For example

http://webhostingneeds.com/File:Averttv_device_property.jpg

Since the file extension is .jpg, nginx try to service this as static content.

How i can set nginx to serve static files, pass anything starting with
/File: to apache ?

Thanks,

Santhosh


From abioy.sun at gmail.com  Wed Sep 21 23:36:31 2011
From: abioy.sun at gmail.com (Abioy Sun)
Date: Thu, 22 Sep 2011 07:36:31 +0800
Subject: returning NGX_AGAIN kills performance?
Message-ID: <CAM6b-v=zgtE=nSkgKDL2_ZUb1sCe4OzopBJUuzq=SZagOhxr=Q@mail.gmail.com>

hi,

i currently dev a module which makes multi subrequest sequently in the
access phase handler, via returning NGX_AGAIN.

my question is, will this kill the performance? and how far could it go in
killing performance.

thx!

Abioy
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110922/e3fa2fe2/attachment.html>

From kworthington at gmail.com  Thu Sep 22 01:10:49 2011
From: kworthington at gmail.com (Kevin Worthington)
Date: Wed, 21 Sep 2011 21:10:49 -0400
Subject: nginx-1.1.4
In-Reply-To: <20110920112225.GC54733@nginx.com>
References: <20110920112225.GC54733@nginx.com>
Message-ID: <CAGo79UVQVWVYGcGKgkzjHY-iLa3dM_+RzgwMy5nd3LKvEN5tEg@mail.gmail.com>

Hello Nginx Users,

I just released Nginx 1.1.4 For Windows http://goo.gl/WzEeA (32-bit and
64-bit)

These versions are to support legacy users who are already using Cygwin
based builds of Nginx. Official Windows binaries are at nginx.org

Thanks,
Kevin
--
Kevin Worthington
kworthington (at] gmail {dot) com


On Tue, Sep 20, 2011 at 7:22 AM, Igor Sysoev <igor at sysoev.ru> wrote:

> Changes with nginx 1.1.4                                         20 Sep
> 2011
>
>    *) Feature: the ngx_http_upstream_keepalive module.
>
>    *) Feature: the "proxy_http_version" directive.
>
>    *) Feature: the "fastcgi_keep_conn" directive.
>
>    *) Feature: the "worker_aio_requests" directive.
>
>    *) Bugfix: if nginx was built --with-file-aio it could not be run on
>       Linux kernel which did not support AIO.
>
>    *) Bugfix: in Linux AIO error processing.
>       Thanks to Hagai Avrahami.
>
>    *) Bugfix: reduced memory consumption for long-lived requests.
>
>    *) Bugfix: the module ngx_http_mp4_module did not support 64-bit MP4
>       "co64" atom.
>
>
> --
> Igor Sysoev
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110921/783937fa/attachment.html>

From pierig at linuxdataflow.org  Wed Sep 21 14:52:29 2011
From: pierig at linuxdataflow.org (Pierig Le Saux)
Date: Wed, 21 Sep 2011 16:52:29 +0200 (CEST)
Subject: mp4 module
In-Reply-To: <072588df6d07290a30ba33099451c5f9.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <26b417a4-12dc-4c21-bf9d-7cbb89fbc51f@mail1.linuxdataflow.org>

To my understanding both mod-h264 (code-shop) and mod-mp4 (nginx) are very similar, as they allow seeking in mp4 files.

The mod_smooth_streaming for Code-shop is the one that is different.

Regards,

Pierig Le Saux


----- Mail original -----
De: "ahu" <nginx-forum at nginx.us>
?: nginx at nginx.org
Envoy?: Mercredi 21 Septembre 2011 16:18:31
Objet: Re: mp4 module

I am very curious about what the difference between this modle and the
3rd
modle(http://h264.code-shop.com/trac/wiki/Mod-H264-Streaming-Nginx-Version2).
Which one should I use?

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215011,215506#msg-215506

_______________________________________________
nginx mailing list
nginx at nginx.org
http://mailman.nginx.org/mailman/listinfo/nginx


From kingsley at internode.com.au  Thu Sep 22 08:07:24 2011
From: kingsley at internode.com.au (Kingsley Foreman)
Date: Thu, 22 Sep 2011 08:07:24 +0000
Subject: keepalive_requests on backend 
Message-ID: <6FA1B8C1117F444AB55F3E36E9314EC841591BF6@EXCHMBX-ADL6-01.staff.internode.com.au>

Hi guys

I use

        upstream backend{
                server xxx.xxx.xxx.xxx:80;
                keepalive 40;
        }

I was wondering if this listened to "keepalive_requests"

I would like to be able to make sure the backend will only maintain a connection for say 1000 queries before it reconnects.

Kingsley
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110922/4fc53e84/attachment.html>

From mdounin at mdounin.ru  Thu Sep 22 08:37:05 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Thu, 22 Sep 2011 12:37:05 +0400
Subject: keepalive_requests on backend
In-Reply-To: <6FA1B8C1117F444AB55F3E36E9314EC841591BF6@EXCHMBX-ADL6-01.staff.internode.com.au>
References: <6FA1B8C1117F444AB55F3E36E9314EC841591BF6@EXCHMBX-ADL6-01.staff.internode.com.au>
Message-ID: <20110922083705.GD1137@mdounin.ru>

Hello!

On Thu, Sep 22, 2011 at 08:07:24AM +0000, Kingsley Foreman wrote:

> Hi guys
> 
> I use
> 
>         upstream backend{
>                 server xxx.xxx.xxx.xxx:80;
>                 keepalive 40;
>         }
> 
> I was wondering if this listened to "keepalive_requests"
> 
> I would like to be able to make sure the backend will only 
> maintain a connection for say 1000 queries before it reconnects.

Right now nginx doesn't set any limits on number of requests to 
backend via single keepalive connection.  Backend may do so.

Maxim Dounin


From atistler at gmail.com  Thu Sep 22 09:17:18 2011
From: atistler at gmail.com (Adam Tistler)
Date: Thu, 22 Sep 2011 05:17:18 -0400
Subject: ssl redirect issue
Message-ID: <20A49295-A19F-49A2-8AFB-C4C6145DB2C7@gmail.com>

Hi,

I have an ssl nginx setup which was working fine with one server { ssl on }.  However I have legacy ssl server_names that I would like to have a 'catch all' for that should be redirected back to the 'main' ssl server name.  Config is pasted here.

http://pastebin.com/dXU7BHaB      ( I removed some sections for brevity )


When the following lines are enabled:

    server {
        listen 443 default;
        server_name _;
        rewrite ^(.*) https://my.example.net$1 permanent;
    }

I will get:

SSL received a record that exceeded the maximum permissible length.
(Error code: ssl_error_rx_record_too_long)

If i comment those lines out, the request will go thru just fine.  However since:

    server {
        listen 443;
        server_name  my.example.net;


is treated as the default vhost,  even ssl requests for myold.example.net are serviced by this server {} and are not redirected to https://my.example.net (obviously not what I want).

Thanks for the help in advanced.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110922/60b929a4/attachment-0001.html>

From nginx-forum at nginx.us  Thu Sep 22 09:25:05 2011
From: nginx-forum at nginx.us (lys0212)
Date: Thu, 22 Sep 2011 05:25:05 -0400
Subject: nginx use memc
Message-ID: <2e626c380c15420f5b6177d971a07eae.NginxMailingListEnglish@forum.nginx.org>

Hello!
     Encountered in the use of memc such a problem, I do not know how to
solve. In the nginx configuration file and php file using memc, can be
normal cache. However, if the cache of php file include jump statement,
such as @header, then jump failure. How to solve this problem?
nginx.conf 
server {
...
 location  ~ /async/(.*) {
                set_md5                 $md5               
$host$request_uri;
                set $memc_key $md5;
                memc_pass memcbackend;
                memc_flags_to_last_modified on;
                memc_connect_timeout  30s;
                memc_send_timeout 30s;
                memc_read_timeout 30s;
                access_log   off;
                expires      max;
                error_page     404 = /async/index.php?$query_string ;
        }
...
}
php file /async/index.php
<?php 
  
  $url = "http://www.baidu.com";
  @header ( "HTTP/1.1 301 Moved Permanently" );
  @header ( "Location: " . $url );
  exit();
         
  $m = new Memcached();
  $m->addServer('192.168.3.101', 3000);
  $m->setOption(Memcached::OPT_COMPRESSION, false);
  $time =  time();
  $urlInfo = parse_url($_SERVER ['REQUEST_URI']);
  $key = $_SERVER ['SERVER_NAME'].$_SERVER ['REQUEST_URI'];
  echo $key;
  $m->set(md5($key), $time ,60);
?>
If the access address http://www.test.com/async/test.php
/async/ directory is not include test.phpd  file, it will be 404
redirected to /async/ index.php, but the @header  in /async/index.php 
will not be effective.
  How to solve this problem? Thank you very much!

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215565,215565#msg-215565


From agentzh at gmail.com  Thu Sep 22 09:44:43 2011
From: agentzh at gmail.com (agentzh)
Date: Thu, 22 Sep 2011 17:44:43 +0800
Subject: nginx use memc
In-Reply-To: <2e626c380c15420f5b6177d971a07eae.NginxMailingListEnglish@forum.nginx.org>
References: <2e626c380c15420f5b6177d971a07eae.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <CAB4Tn6PufTGNhiM-0Lx2qN4=9QHSiwLs3OAaxmA8oyzTx4JSnQ@mail.gmail.com>

On Thu, Sep 22, 2011 at 5:25 PM, lys0212 <nginx-forum at nginx.us> wrote:
> Hello!
> ? ? Encountered in the use of memc such a problem, I do not know how to
> solve. In the nginx configuration file and php file using memc, can be
> normal cache. However, if the cache of php file include jump statement,
> such as @header, then jump failure. How to solve this problem?

Please do not abuse the error_page directive with ngx_memc to do
response caching; it's problematic. I suggest you take a look at the
ngx_srcache module which can serve as a generic cache layer and can be
used safely with backend storage modules like ngx_memc:

    http://wiki.nginx.org/HttpSRCacheModule

Regards,
-agentzh


From nginx-forum at nginx.us  Thu Sep 22 10:44:19 2011
From: nginx-forum at nginx.us (Lintu)
Date: Thu, 22 Sep 2011 06:44:19 -0400
Subject: php-fastcgi always exits.. resulting in 502 bad gateway
Message-ID: <97d9cbc8779183130f5ec6d9cf5ec82e.NginxMailingListEnglish@forum.nginx.org>

Hello everyone,

I'm using nginx since some days and I got a little trouble. It seems
that after some while of working fine nginx exits the fastcgi-daemon and
whenever I try to call a php site then, it shows me 502 Bad Gateway.

Then I have to "/usr/bin/php-fastcgi start" and everything runs fine
again.

So basically my question: How do I make the php-fastcgi not exit itself
or how do I make it automatically restart on exit? It's annoying that my
website everytime turns to "502 Bad Gateway" to everyone and I don't
even know about that!

Can anyone help me please? :(

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215576,215576#msg-215576


From nginx-forum at nginx.us  Thu Sep 22 11:03:57 2011
From: nginx-forum at nginx.us (Ensiferous)
Date: Thu, 22 Sep 2011 07:03:57 -0400
Subject: mp4 module
In-Reply-To: <20110909115107.GH97288@nginx.com>
References: <20110909115107.GH97288@nginx.com>
Message-ID: <76c3136923f6db464986dd2fd3c4fa01.NginxMailingListEnglish@forum.nginx.org>

To quote Igors original post: "If you use 3rd-party mp4 module, it
should be removed."

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215011,215577#msg-215577


From atistler at gmail.com  Thu Sep 22 11:23:46 2011
From: atistler at gmail.com (Adam Tistler)
Date: Thu, 22 Sep 2011 07:23:46 -0400
Subject: ssl redirect issue
In-Reply-To: <20A49295-A19F-49A2-8AFB-C4C6145DB2C7@gmail.com>
References: <20A49295-A19F-49A2-8AFB-C4C6145DB2C7@gmail.com>
Message-ID: <AB9FCA5E-82FA-43A6-9D32-C8AF5EE20B7B@gmail.com>

Actually, I can work around this issue by adding:

        if ($host !~* ^(my.example.net)$ ) {
           rewrite ^(.*) https://my.example.net$1 permanent;
        }

I am curious however this does not work when I put the redirect in the default ( '_' ) server config

On Sep 22, 2011, at 5:17 AM, Adam Tistler wrote:

> Hi,
> 
> I have an ssl nginx setup which was working fine with one server { ssl on }.  However I have legacy ssl server_names that I would like to have a 'catch all' for that should be redirected back to the 'main' ssl server name.  Config is pasted here.
> 
> http://pastebin.com/dXU7BHaB      ( I removed some sections for brevity )
> 
> 
> When the following lines are enabled:
> 
>     server {
>         listen 443 default;
>         server_name _;
>         rewrite ^(.*) https://my.example.net$1 permanent;
>     }
> 
> I will get:
> 
> SSL received a record that exceeded the maximum permissible length.
> (Error code: ssl_error_rx_record_too_long)
> 
> If i comment those lines out, the request will go thru just fine.  However since:
> 
>     server {
>         listen 443;
>         server_name  my.example.net;
> 
> 
> is treated as the default vhost,  even ssl requests for myold.example.net are serviced by this server {} and are not redirected to https://my.example.net (obviously not what I want).
> 
> Thanks for the help in advanced.

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110922/95a79cdd/attachment.html>

From kingsley at internode.com.au  Thu Sep 22 11:26:42 2011
From: kingsley at internode.com.au (Kingsley Foreman)
Date: Thu, 22 Sep 2011 11:26:42 +0000
Subject: keepalive_requests on backend
In-Reply-To: <20110922083705.GD1137@mdounin.ru>
References: <6FA1B8C1117F444AB55F3E36E9314EC841591BF6@EXCHMBX-ADL6-01.staff.internode.com.au>,
	<20110922083705.GD1137@mdounin.ru>
Message-ID: <6FA1B8C1117F444AB55F3E36E9314EC841591DE7@EXCHMBX-ADL6-01.staff.internode.com.au>

Is it worth making it a feature request?

Kingsley

________________________________________
From: nginx-bounces at nginx.org [nginx-bounces at nginx.org] on behalf of Maxim Dounin [mdounin at mdounin.ru]
Sent: 22 September 2011 18:07
To: nginx at nginx.org
Subject: Re: keepalive_requests on backend

Hello!

On Thu, Sep 22, 2011 at 08:07:24AM +0000, Kingsley Foreman wrote:

> Hi guys
>
> I use
>
>         upstream backend{
>                 server xxx.xxx.xxx.xxx:80;
>                 keepalive 40;
>         }
>
> I was wondering if this listened to "keepalive_requests"
>
> I would like to be able to make sure the backend will only
> maintain a connection for say 1000 queries before it reconnects.

Right now nginx doesn't set any limits on number of requests to
backend via single keepalive connection.  Backend may do so.

Maxim Dounin

_______________________________________________
nginx mailing list
nginx at nginx.org
http://mailman.nginx.org/mailman/listinfo/nginx


From r at roze.lv  Thu Sep 22 11:28:36 2011
From: r at roze.lv (Reinis Rozitis)
Date: Thu, 22 Sep 2011 14:28:36 +0300
Subject: php-fastcgi always exits.. resulting in 502 bad gateway
In-Reply-To: <97d9cbc8779183130f5ec6d9cf5ec82e.NginxMailingListEnglish@forum.nginx.org>
References: <97d9cbc8779183130f5ec6d9cf5ec82e.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <3F87DBC5FDD04B4CAA08159D02DF9147@DD21>

> So basically my question: How do I make the php-fastcgi not exit itself or how do I make it automatically restart on exit?

Best way to do it is to use inbuilt process manager PHP-FPM http://www.php.net/manual/en/install.fpm.php (since 5.3.x its in php 
core).

rr 


From francis at daoine.org  Thu Sep 22 11:29:51 2011
From: francis at daoine.org (Francis Daly)
Date: Thu, 22 Sep 2011 12:29:51 +0100
Subject: Problem with MediaWiki File page
In-Reply-To: <CAGWcB0N1RvaiFLgJJGc7en0FJxbbd0b_k5nSF-JQ=kPN2s+u6A@mail.gmail.com>
References: <CAGWcB0N1RvaiFLgJJGc7en0FJxbbd0b_k5nSF-JQ=kPN2s+u6A@mail.gmail.com>
Message-ID: <20110922112951.GF12179@craic.sysops.org>

On Thu, Sep 22, 2011 at 02:14:04AM +0530, BizHat.com Support wrote:

Hi there,

>         location ~* \.(js|css|rdf|xml|ico|txt|gif|jpg|png|jpeg|html)$ {
>         root   /home/webhost/public_html/;
>         access_log  /var/log/nginx/webhostingneeds.com_static.log;
>         }

>         location / {
>             proxy_pass         http://174.36.130.243:81/;

(plus extra proxy settings)

>         }

> The problem is when a file is uploaded to MediaWIki, i shows a page like
> 
> http://webhostingneeds.com/File:FILENAME.exten
> 
> For example
> 
> http://webhostingneeds.com/File:Averttv_device_property.jpg
> 
> Since the file extension is .jpg, nginx try to service this as static content.
> 
> How i can set nginx to serve static files, pass anything starting with
> /File: to apache ?

nginx configuration is based on location blocks. One request is handled
by one location.

Details at http://wiki.nginx.org/HttpCoreModule#location

Your current config has one location for urls that match a regex, such
that they end in one of 11 specific strings; plus one location for
everything else.

You want a new location for urls that start with one specific string,
and you want regex locations not to be tested if this location matches.

So do exactly that:

  location ^~ /File: {
  }

and inside that location block, put your full proxy_pass and other
relevant configuration.

All the best,

	f
-- 
Francis Daly        francis at daoine.org


From nginx-forum at nginx.us  Thu Sep 22 11:34:02 2011
From: nginx-forum at nginx.us (tomp)
Date: Thu, 22 Sep 2011 07:34:02 -0400
Subject: max_fails includes read_timeout even if timeout is excluded from
	proxy_next_upstream
Message-ID: <bfdd4cafdd96aaf7a8fc3c5fc6469dc8.NginxMailingListEnglish@forum.nginx.org>

Hi,

I am having some trouble with the latest stable of Nginx's proxy
upstream support.

I would like Nginx to NOT count read timeouts towards the max_fails in
the upstream definition.

Here is what I have:

proxy_cache_path        /var/cache/nginx/tptest levels=1:2  
keys_zone=tptest:10m;

upstream tptestBackend
{
        server tp.dev.infinity.local max_fails=1 fail_timeout=30s;
        server dn.dev.infinity.local max_fails=1 fail_timeout=30s;
}

server
{
        listen                  172.16.0.64:80;
        server_name             tp.nginx.dev.infinity.local;

        proxy_cache             tptest;
        proxy_set_header        Host 'tp.dev.infinity.local';
        proxy_set_header        X-Real-IP $remote_addr;
        proxy_cache_key         http://$host$request_uri;
        proxy_connect_timeout   1;
        proxy_cache_use_stale   error timeout updating invalid_header
http_500 http_502 http_503 http_504;
        proxy_next_upstream     error http_500 http_502 http_503
http_504;
        proxy_read_timeout      5;

        location /
        {
                proxy_pass      http://tptestBackend;
        }
}

I have removed "timeout" from the proxy_next_upstream directive, however
when I try with a script that sleeps forever, the timeout kicks in, but
it does count towards the max_fails, and so next time only the second
server is used.

Is this a bug?

Thanks

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215585,215585#msg-215585


From mdounin at mdounin.ru  Thu Sep 22 11:34:36 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Thu, 22 Sep 2011 15:34:36 +0400
Subject: keepalive_requests on backend
In-Reply-To: <6FA1B8C1117F444AB55F3E36E9314EC841591DE7@EXCHMBX-ADL6-01.staff.internode.com.au>
References: <6FA1B8C1117F444AB55F3E36E9314EC841591BF6@EXCHMBX-ADL6-01.staff.internode.com.au>
	<20110922083705.GD1137@mdounin.ru>
	<6FA1B8C1117F444AB55F3E36E9314EC841591DE7@EXCHMBX-ADL6-01.staff.internode.com.au>
Message-ID: <20110922113436.GH1137@mdounin.ru>

Hello!

On Thu, Sep 22, 2011 at 11:26:42AM +0000, Kingsley Foreman wrote:

> Is it worth making it a feature request?

As long as you'll explain why do you think it's usefull.

Maxim Dounin

> 
> Kingsley
> 
> ________________________________________
> From: nginx-bounces at nginx.org [nginx-bounces at nginx.org] on behalf of Maxim Dounin [mdounin at mdounin.ru]
> Sent: 22 September 2011 18:07
> To: nginx at nginx.org
> Subject: Re: keepalive_requests on backend
> 
> Hello!
> 
> On Thu, Sep 22, 2011 at 08:07:24AM +0000, Kingsley Foreman wrote:
> 
> > Hi guys
> >
> > I use
> >
> >         upstream backend{
> >                 server xxx.xxx.xxx.xxx:80;
> >                 keepalive 40;
> >         }
> >
> > I was wondering if this listened to "keepalive_requests"
> >
> > I would like to be able to make sure the backend will only
> > maintain a connection for say 1000 queries before it reconnects.
> 
> Right now nginx doesn't set any limits on number of requests to
> backend via single keepalive connection.  Backend may do so.
> 
> Maxim Dounin
> 
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx
> 
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx


From nginx-forum at nginx.us  Thu Sep 22 11:44:37 2011
From: nginx-forum at nginx.us (Lintu)
Date: Thu, 22 Sep 2011 07:44:37 -0400
Subject: php-fastcgi always exits.. resulting in 502 bad gateway
In-Reply-To: <3F87DBC5FDD04B4CAA08159D02DF9147@DD21>
References: <3F87DBC5FDD04B4CAA08159D02DF9147@DD21>
Message-ID: <2d5f13bd1416db57ed25a2212d34bf94.NginxMailingListEnglish@forum.nginx.org>

I'm running Debian Lenny which has only php 5.2 :(

I tried apt-get install php5-fpm and php-fpm, but both don't work. How
to get it on Debian Lenny? I cant compile php on my own... :(

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215576,215588#msg-215588


From edho at myconan.net  Thu Sep 22 11:55:44 2011
From: edho at myconan.net (Edho P Arief)
Date: Thu, 22 Sep 2011 18:55:44 +0700
Subject: php-fastcgi always exits.. resulting in 502 bad gateway
In-Reply-To: <2d5f13bd1416db57ed25a2212d34bf94.NginxMailingListEnglish@forum.nginx.org>
References: <3F87DBC5FDD04B4CAA08159D02DF9147@DD21>
	<2d5f13bd1416db57ed25a2212d34bf94.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <CABTjkK=TKgiKm2EjOKqdmGpxrntAM42_w=hr7DvjJ5FLgxYFZQ@mail.gmail.com>

On Sep 22, 2011 6:44 PM, "Lintu" <nginx-forum at nginx.us> wrote:
>
> I'm running Debian Lenny which has only php 5.2 :(
>
> I tried apt-get install php5-fpm and php-fpm, but both don't work. How
> to get it on Debian Lenny? I cant compile php on my own...

Install supervisord. You can then create as many php-cgi processes using
supervisord which will be automatically restarted upon killed.

I usually use virtualenv to create dedicated python environment for
supervisord.

- extract virtualenv to /opt
- create supervisord environment by issuing
/opt/virtualenv/virtualenv.py /opt/supervisor
- install supervisord
/opt/supervisor/bin/easy_install supervisor
- configure accordingly (check its documentation)
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110922/a756d7c6/attachment.html>

From nginx-forum at nginx.us  Thu Sep 22 12:08:45 2011
From: nginx-forum at nginx.us (Lintu)
Date: Thu, 22 Sep 2011 08:08:45 -0400
Subject: php-fastcgi always exits.. resulting in 502 bad gateway
In-Reply-To: <CABTjkK=TKgiKm2EjOKqdmGpxrntAM42_w=hr7DvjJ5FLgxYFZQ@mail.gmail.com>
References: <CABTjkK=TKgiKm2EjOKqdmGpxrntAM42_w=hr7DvjJ5FLgxYFZQ@mail.gmail.com>
Message-ID: <a075b5e32a9775f30eaebea7ae5b4ea0.NginxMailingListEnglish@forum.nginx.org>

Hey, thanks for the tip with supervisord.

I found a site
(http://sachachua.com/blog/2010/06/using-supervisord-for-nginxfastcgiphp/)
saying its easy to set up with:

>apt-get install python-setuptools
>easy_install supervisor

And actually it was. No error on that. But then the site says I should
add something to /etc/supervisord.conf, but there is no file like that?
I created it with the content I got told to, but well, nothing happens
now... Or is everything fine now? Is everything automated about
supervisord?

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215576,215593#msg-215593


From jdmls at yahoo.com  Thu Sep 22 12:10:53 2011
From: jdmls at yahoo.com (John Doe)
Date: Thu, 22 Sep 2011 05:10:53 -0700 (PDT)
Subject: php-fastcgi always exits.. resulting in 502 bad gateway
In-Reply-To: <97d9cbc8779183130f5ec6d9cf5ec82e.NginxMailingListEnglish@forum.nginx.org>
References: <97d9cbc8779183130f5ec6d9cf5ec82e.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <1316693453.92570.YahooMailNeo@web114711.mail.gq1.yahoo.com>

From: Lintu <nginx-forum at nginx.us>

> I'm using nginx since some days and I got a little trouble. It seems
> that after some while of working fine nginx exits the fastcgi-daemon and
> whenever I try to call a php site then, it shows me 502 Bad Gateway.
> Then I have to "/usr/bin/php-fastcgi start" and everything runs fine
> again.
> So basically my question: How do I make the php-fastcgi not exit itself
> or how do I make it automatically restart on exit? It's annoying that my
> website everytime turns to "502 Bad Gateway" to everyone and I 
> don't even know about that!

We use spawn-fcgi here.

JD


From edho at myconan.net  Thu Sep 22 12:22:52 2011
From: edho at myconan.net (Edho P Arief)
Date: Thu, 22 Sep 2011 19:22:52 +0700
Subject: php-fastcgi always exits.. resulting in 502 bad gateway
In-Reply-To: <a075b5e32a9775f30eaebea7ae5b4ea0.NginxMailingListEnglish@forum.nginx.org>
References: <CABTjkK=TKgiKm2EjOKqdmGpxrntAM42_w=hr7DvjJ5FLgxYFZQ@mail.gmail.com>
	<a075b5e32a9775f30eaebea7ae5b4ea0.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <CABTjkKn5kKkNGB2Sbq1P6Y-LNq1iAHYUv005T0hu2Bi7tu3uOg@mail.gmail.com>

On Thu, Sep 22, 2011 at 7:08 PM, Lintu <nginx-forum at nginx.us> wrote:
> Hey, thanks for the tip with supervisord.
>
> I found a site
> (http://sachachua.com/blog/2010/06/using-supervisord-for-nginxfastcgiphp/)
> saying its easy to set up with:
>
>>apt-get install python-setuptools
>>easy_install supervisor
>
> And actually it was. No error on that. But then the site says I should
> add something to /etc/supervisord.conf, but there is no file like that?
> I created it with the content I got told to, but well, nothing happens
> now... Or is everything fine now? Is everything automated about
> supervisord?
>

run supervisorctl or just use ps check if the php-cgi processes are
running. Set nginx fastcgi_pass to whatever path you set in
supervisord for php-cgi.

-- 
O< ascii ribbon campaign - stop html mail - www.asciiribbon.org


From edho at myconan.net  Thu Sep 22 12:25:12 2011
From: edho at myconan.net (Edho P Arief)
Date: Thu, 22 Sep 2011 19:25:12 +0700
Subject: php-fastcgi always exits.. resulting in 502 bad gateway
In-Reply-To: <CABTjkKn5kKkNGB2Sbq1P6Y-LNq1iAHYUv005T0hu2Bi7tu3uOg@mail.gmail.com>
References: <CABTjkK=TKgiKm2EjOKqdmGpxrntAM42_w=hr7DvjJ5FLgxYFZQ@mail.gmail.com>
	<a075b5e32a9775f30eaebea7ae5b4ea0.NginxMailingListEnglish@forum.nginx.org>
	<CABTjkKn5kKkNGB2Sbq1P6Y-LNq1iAHYUv005T0hu2Bi7tu3uOg@mail.gmail.com>
Message-ID: <CABTjkKnOjaWs6+M=MDxrNPS+iBr1O=Yth3ydMZmXaCJqJy0fPQ@mail.gmail.com>

On Thu, Sep 22, 2011 at 7:22 PM, Edho P Arief <edho at myconan.net> wrote:
> On Thu, Sep 22, 2011 at 7:08 PM, Lintu <nginx-forum at nginx.us> wrote:
>> Hey, thanks for the tip with supervisord.
>>
>> I found a site
>> (http://sachachua.com/blog/2010/06/using-supervisord-for-nginxfastcgiphp/)
>> saying its easy to set up with:
>>
>>>apt-get install python-setuptools
>>>easy_install supervisor
>>
>> And actually it was. No error on that. But then the site says I should
>> add something to /etc/supervisord.conf, but there is no file like that?


the basic config can be generated by command echo_supervisord_conf
(you need to copy-paste the result or pipe it to file).

-- 
O< ascii ribbon campaign - stop html mail - www.asciiribbon.org


From appa at perusio.net  Thu Sep 22 12:32:17 2011
From: appa at perusio.net (=?UTF-8?B?QW50w7NuaW8=?= P. P. Almeida)
Date: Thu, 22 Sep 2011 13:32:17 +0100
Subject: php-fastcgi always exits.. resulting in 502 bad gateway
In-Reply-To: <97d9cbc8779183130f5ec6d9cf5ec82e.NginxMailingListEnglish@forum.nginx.org>
References: <97d9cbc8779183130f5ec6d9cf5ec82e.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <877h504v1q.wl%appa@perusio.net>

On 22 Set 2011 11h44 WEST, nginx-forum at nginx.us wrote:

> Hello everyone,
>
> I'm using nginx since some days and I got a little trouble. It seems
> that after some while of working fine nginx exits the fastcgi-daemon
> and whenever I try to call a php site then, it shows me 502 Bad
> Gateway.
>
> Then I have to "/usr/bin/php-fastcgi start" and everything runs fine
> again.

> So basically my question: How do I make the php-fastcgi not exit
> itself or how do I make it automatically restart on exit? It's
> annoying that my website everytime turns to "502 Bad Gateway" to
> everyone and I don't even know about that!
>
> Can anyone help me please? :(

Use one of the many init scripts available throughout the
interwebs. Also I suggest monit for monitoring the process. 

I now use php-fpm, but before I used the "regular" php-cgi with this
init script: https://github.com/perusio/php-fastcgi-debian-script and
a monit configuration for monitoring it:
https://github.com/perusio/monit-miscellaneous/blob/master/php-cgi

HTH,
--- appa


From appa at perusio.net  Thu Sep 22 12:41:08 2011
From: appa at perusio.net (=?UTF-8?B?QW50w7NuaW8=?= P. P. Almeida)
Date: Thu, 22 Sep 2011 13:41:08 +0100
Subject: ssl redirect issue
In-Reply-To: <AB9FCA5E-82FA-43A6-9D32-C8AF5EE20B7B@gmail.com>
References: <20A49295-A19F-49A2-8AFB-C4C6145DB2C7@gmail.com>
	<AB9FCA5E-82FA-43A6-9D32-C8AF5EE20B7B@gmail.com>
Message-ID: <8762kk4umz.wl%appa@perusio.net>

On 22 Set 2011 12h23 WEST, atistler at gmail.com wrote:

> Actually, I can work around this issue by adding:
>
> if ($host !~* ^(my.example.net)$ ) {
> rewrite ^(.*) https://my.example.net$1 permanent;
> }
>
> I am curious however this does not work when I put the redirect in
> the default ( '_' ) server config

'_' is a not a valid hostname. The reason it's used in the recommended
configuration for default servers is that there's no risk of colliding
with anything else and you can have individual vhost configs for each
proper hostname. 

Your config should be something like this:

server {
   server_name my.example.net;
   return 301 https://myexample.net$request_uri;
}

Then you should have the regular HTTPs server config that listens on
port 443.

server {
    listen 443 ssl;
    server_name my.example.net;

    # remaining server config...
}

Repeat the above for each domain redirect you want to have.

Additional reading:
http://nginx.org/en/docs/http/configuring_https_servers.html
http://nginx.org/en/docs/http/server_names.html

--- appa


From r at roze.lv  Thu Sep 22 12:45:34 2011
From: r at roze.lv (Reinis Rozitis)
Date: Thu, 22 Sep 2011 15:45:34 +0300
Subject: php-fastcgi always exits.. resulting in 502 bad gateway
In-Reply-To: <2d5f13bd1416db57ed25a2212d34bf94.NginxMailingListEnglish@forum.nginx.org>
References: <3F87DBC5FDD04B4CAA08159D02DF9147@DD21>
	<2d5f13bd1416db57ed25a2212d34bf94.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <5BC8A8A188F3425EB0C8FCD7C8605D8A@DD21>

> I'm running Debian Lenny which has only php 5.2 :(

I'm not a Debian user but you could add the dotdeb repo ( http://www.dotdeb.org/2011/08/30/php-5-3-8-is-available/ ).
Sticking with old and unsupported php release isnt a good thing.


rr


From ian.hobson at ntlworld.com  Thu Sep 22 13:29:38 2011
From: ian.hobson at ntlworld.com (Ian Hobson)
Date: Thu, 22 Sep 2011 14:29:38 +0100
Subject: Config help needed
Message-ID: <4E7B3842.1040708@ntlworld.com>

Hi All,

My site is wordpress, and configured as follows.

# Statements for KindleMonster.com  virtual server
server {
     listen   80;
     #
     server_name kindlemonster.com www.kindlemonster.com;
     root /var/www/kindlemonster.com/htdocs;
     access_log  /var/www/kindlemonster.com/access.log;
     index index.php index.html index.htm;
     #
     # redirect server error pages to the static page /50x.html
     #
     error_page   500 502 503 504  /50x.html;
     # protect usage stats
     location ^~ /usage/ {
       auth_basic "Hello, please login";
       auth_basic_user_file /var/www/kindlemnonster.com/passwords;
     }
     # all  .php requests to fastcgi using uri.
     location ~ \.php$ {
         include /etc/nginx/fastcgi_params;
         fastcgi_pass 127.0.0.1:9000;
         # fastcgi_param SCRIPT_FILENAME $document_root$fastcgi_script_name;
     }
     try_files $uri $uri/ /index.php;
}

Wordpress works just fine. I have just uploaded a new file to /form.php 
and when
I try to browse to it I get the super helpful error message of

No input file specified.

The log says its a 404.

OK - Why do all the wordpress files (mostly php) work just fine, and my 
new file does not?

Thanks

Ian


From appa at perusio.net  Thu Sep 22 13:45:00 2011
From: appa at perusio.net (=?UTF-8?B?QW50w7NuaW8=?= P. P. Almeida)
Date: Thu, 22 Sep 2011 14:45:00 +0100
Subject: Problem with MediaWiki File page
In-Reply-To: <CAGWcB0N1RvaiFLgJJGc7en0FJxbbd0b_k5nSF-JQ=kPN2s+u6A@mail.gmail.com>
References: <CAGWcB0N1RvaiFLgJJGc7en0FJxbbd0b_k5nSF-JQ=kPN2s+u6A@mail.gmail.com>
Message-ID: <871uv84roj.wl%appa@perusio.net>

On 21 Set 2011 21h44 WEST, info at bizhat.com wrote:

> HI,
>
> I have following nginx configuration for a MediaWiki based site
>
>
> server {
> listen 174.36.130.243:80;
> server_name  webhostingneeds.com www.webhostingneeds.com;
>
> location ~* \.(js|css|rdf|xml|ico|txt|gif|jpg|png|jpeg|html)$ {
> root   /home/webhost/public_html/;
> access_log  /var/log/nginx/webhostingneeds.com_static.log;
> }
>
> location / {
> proxy_pass         http://174.36.130.243:81/;
> proxy_redirect     off;
> proxy_set_header   Host             $host;
> proxy_set_header   X-Real-IP        $remote_addr;
> proxy_set_header   X-Forwarded-For  $proxy_add_x_forwarded_for;
> client_max_body_size       1m;
> client_body_buffer_size    128k;
> proxy_connect_timeout      90;
> proxy_send_timeout         90;
> proxy_read_timeout         90;
> proxy_buffer_size          4k;
> proxy_buffers              4 32k;
> proxy_busy_buffers_size    64k;
> proxy_temp_file_write_size 64k;
> }
> }
>
> The problem is when a file is uploaded to MediaWIki, i shows a page
> like
>
> http://webhostingneeds.com/File:FILENAME.exten

This should work:

location / {

    proxy_pass http://174.36.130.243:81/;        
    proxy_redirect     off;
    proxy_set_header   Host             $host;
    proxy_set_header   X-Real-IP        $remote_addr;
    proxy_set_header   X-Forwarded-For  $proxy_add_x_forwarded_for;
    client_max_body_size       1m;
    client_body_buffer_size    128k;
    proxy_connect_timeout      90;
    proxy_send_timeout         90;
    proxy_read_timeout         90;
    proxy_buffer_size          4k;
    proxy_buffers              4 32k;
    proxy_busy_buffers_size    64k;
    proxy_temp_file_write_size 64k;

    location ^~ /File: {
  
        location ~* /File:[^/]*\.(?:jpe?g|png|gif|html)$ {
            proxy_pass http://174.36.130.243:81/;
        }
    }
}

> For example
>
> http://webhostingneeds.com/File:Averttv_device_property.jpg
>
> Since the file extension is .jpg, nginx try to service this as
> static content.
>
> How i can set nginx to serve static files, pass anything starting
> with /File: to apache ?

--- appa


From nginx-forum at nginx.us  Thu Sep 22 15:39:36 2011
From: nginx-forum at nginx.us (etrader)
Date: Thu, 22 Sep 2011 11:39:36 -0400
Subject: How to avoid 404 for images in access log?
Message-ID: <9f2d51afcce11583480dece2d34febee.NginxMailingListEnglish@forum.nginx.org>

My access log is full of 404 logs for images not found; thus, it is
difficult to find real 200 successful visits. The lines include

... GET /css/img/calendar.png HTTP/1.1"  404 31
"http://mydomain.com.page.html ...

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215626,215626#msg-215626


From zzz at zzz.org.ua  Thu Sep 22 15:45:32 2011
From: zzz at zzz.org.ua (Alexandr Gomoliako)
Date: Thu, 22 Sep 2011 18:45:32 +0300
Subject: How to avoid 404 for images in access log?
In-Reply-To: <9f2d51afcce11583480dece2d34febee.NginxMailingListEnglish@forum.nginx.org>
References: <9f2d51afcce11583480dece2d34febee.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <CAKb9G9ix=MeUHvq8z4McZPVPQkFrBcBBCszSHTL6ZKUa545-bQ@mail.gmail.com>

On 9/22/11, etrader <nginx-forum at nginx.us> wrote:
> My access log is full of 404 logs for images not found; thus, it is
> difficult to find real 200 successful visits. The lines include
>
> ... GET /css/img/calendar.png HTTP/1.1"  404 31
> "http://mydomain.com.page.html ...

Try something like
    cat access.log | grep " 200 "


From nginx-forum at nginx.us  Thu Sep 22 15:49:20 2011
From: nginx-forum at nginx.us (openclinica)
Date: Thu, 22 Sep 2011 11:49:20 -0400
Subject: Timeout Issues
Message-ID: <fd18a24dc8055b35599ed3acc4c9428c.NginxMailingListEnglish@forum.nginx.org>

Hi All,

I submitted this originally through the lists but I noticed it did not
create a topic here for it. So I am adding it as a topic, I hope this
acceptable.

Hi All,

Ok I am having issues with NGINX and it timing out. My scenario is that
I am using NGINX as a web server front end for my application. I have
two folder being served as alias?s for static content and one location
doing a proxy_pass to a java server. 

My issue is that on one of the request that goes through the proxy_pass
to the local java web server will time out through nginx within 10
minutes. The timeout is a browser timeout. It takes 11-12 minutes to
load the page without nginx, so I need nginx to not kill the session
until that time has passed. I have tried every timeout setting
imaginable that I could find and I still get a browser timeout at 10
minutes. Note this is an SSL connection, the test without nginx in the
way was also an SSL connection.

Below are my configs, note all the timeout options have been slowly
added over time and lots of testing.....they are not my wanted scenario
but I am trying to figure out how this is done.

NGINX.conf

user www-data;
worker_processes  4;

error_log  /var/log/nginx/error.log;
pid        /var/run/nginx.pid;

events {
    worker_connections  1024;
    # multi_accept on;
}

http {
    include       /etc/nginx/mime.types;

    access_log  /var/log/nginx/access.log;

    sendfile        on;
    #tcp_nopush     on;

    #keepalive_timeout  0;
    keepalive_timeout  900 900;
 client_body_timeout      900;
  client_header_timeout    900;
  send_timeout             900;


    tcp_nodelay        on;

    gzip  on;
    gzip_disable "MSIE [1-6]\.(?!.*SV1)";
    gzip_types text/plain text/html text/css text/javascript image/png
image/x-icon application/x-javascript application/xml image/gif;
    
    ssl_session_cache    shared:SSL:100k;
    ssl_session_timeout  20m;


    include /etc/nginx/conf.d/*.conf;
    include /etc/nginx/sites-enabled/*;
}

Site Conf

server {
        listen       80;
        server_name  localhost;

        #charset koi8-r;

        access_log  /var/log/nginx/80-app.access.log;
        error_log  /var/log/nginx/80-app.error.log;

        location / {
            rewrite ^/(.*)$ https://$host/$1 redirect;
        }

        error_page  404              /404.html;

        # redirect server error pages to the static page /50x.html
        #
        error_page   500 502 503 504  /50x.html;
        location = /50x.html {
            root   html;
        }
}
server {        
        listen       443;
        server_name  localhost; 


        access_log  /var/log/nginx/app.access.log;
        error_log  /var/log/nginx/app.error.log;

        ssl                  on;
        ssl_certificate      /usr/local/oc/install/server.com.crt;
        ssl_certificate_key  /usr/local/oc/install/server.key;

#        fail_timeout 3600s;
    keepalive_timeout  900 900;
 client_body_timeout      900;
  client_header_timeout    900;
  send_timeout             900;
        # directly serve the static files in the `images` directory
        location ~ ^/APP/includes/(.*)$ {
            # add future expiry date to force caching of the file on the
client
            expires max;
            add_header Cache-Control "public";
            alias /usr/local/tomcat/webapps/APP/includes/$1;
        }

        location ~ ^/APP/images/(.*)$ {
            # add future expiry date to force caching of the file on the
client 
            expires max;
            add_header Cache-Control "public";
            alias /usr/local/tomcat/webapps/APP/images/$1;
        }
        # pass all other requests to Tomcat
        location /APP {
            proxy_read_timeout 3600s;
            proxy_connect_timeout 3600s;
            proxy_send_timeout 3600s;

            proxy_pass http://127.0.0.1:8080/APP;
            proxy_set_header Host $host;
            proxy_set_header X-Real-IP $remote_addr;
            proxy_set_header X-Forwarded-For
$proxy_add_x_forwarded_for;
        }
}


Any help is greatly appreciated. I wouldn?t even mind if it is
possible to disable timeout all together just as a test.

Thanks,
Shaun

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215628,215628#msg-215628


From nginx-forum at nginx.us  Thu Sep 22 15:53:10 2011
From: nginx-forum at nginx.us (etrader)
Date: Thu, 22 Sep 2011 11:53:10 -0400
Subject: How to avoid 404 for images in access log?
In-Reply-To: <9f2d51afcce11583480dece2d34febee.NginxMailingListEnglish@forum.nginx.org>
References: <9f2d51afcce11583480dece2d34febee.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <1b2407e8370fcacfce54bc5828f07bad.NginxMailingListEnglish@forum.nginx.org>

You mean to read only 200 requests in ssh or something like that?
I am looking for a way to avoid writing 404 requests in access log file
to keep the file tidy.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215626,215630#msg-215630


From zzz at zzz.org.ua  Thu Sep 22 16:03:09 2011
From: zzz at zzz.org.ua (Alexandr Gomoliako)
Date: Thu, 22 Sep 2011 19:03:09 +0300
Subject: How to avoid 404 for images in access log?
In-Reply-To: <1b2407e8370fcacfce54bc5828f07bad.NginxMailingListEnglish@forum.nginx.org>
References: <9f2d51afcce11583480dece2d34febee.NginxMailingListEnglish@forum.nginx.org>
	<1b2407e8370fcacfce54bc5828f07bad.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <CAKb9G9jK1ODdeN-ijRpcxxL4DpHnSaQVS0pTamNx7E6XVgFJaA@mail.gmail.com>

On 9/22/11, etrader <nginx-forum at nginx.us> wrote:
> You mean to read only 200 requests in ssh or something like that?
> I am looking for a way to avoid writing 404 requests in access log file
> to keep the file tidy.

There is a way. But 404 requests are essential for security, i.e. to
see when someone is trying to hack your website, etc.


From r at roze.lv  Thu Sep 22 16:16:05 2011
From: r at roze.lv (Reinis Rozitis)
Date: Thu, 22 Sep 2011 19:16:05 +0300
Subject: How to avoid 404 for images in access log?
In-Reply-To: <9f2d51afcce11583480dece2d34febee.NginxMailingListEnglish@forum.nginx.org>
References: <9f2d51afcce11583480dece2d34febee.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <49D0103F7210443997EF51AA7BD8EB4B@DD21>

> My access log is full of 404 logs for images not found; thus, it is difficult to find real 200 successful visits. The lines 
> include
> ... GET /css/img/calendar.png HTTP/1.1"  404 31
> "http://mydomain.com.page.html ...


Depending on if you don't want to log 404 requests at all or just for the images can probably do something like this:


location ~* ^.+\.(jpg|jpeg|gif|png)$ {
    error_page 404 @nolog;
}

location @nolog {
   access_log off;
}


If you want it for any request just use the error_page without location or even give a different logfile for the 404s.

rr 


From nginx-forum at nginx.us  Thu Sep 22 16:17:10 2011
From: nginx-forum at nginx.us (etrader)
Date: Thu, 22 Sep 2011 12:17:10 -0400
Subject: How to avoid 404 for images in access log?
In-Reply-To: <CAKb9G9jK1ODdeN-ijRpcxxL4DpHnSaQVS0pTamNx7E6XVgFJaA@mail.gmail.com>
References: <CAKb9G9jK1ODdeN-ijRpcxxL4DpHnSaQVS0pTamNx7E6XVgFJaA@mail.gmail.com>
Message-ID: <aab42e41403850aa79b52ee6019989de.NginxMailingListEnglish@forum.nginx.org>

Isn't it possible to use multiple access_log to split the access log to
traffic and 404 logs ? Or probably writing 404 logs to error_log.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215626,215634#msg-215634


From ian at ianhobson.co.uk  Thu Sep 22 16:33:15 2011
From: ian at ianhobson.co.uk (Ian Hobson)
Date: Thu, 22 Sep 2011 17:33:15 +0100
Subject: Config help needed
In-Reply-To: <4E7B3842.1040708@ntlworld.com>
References: <4E7B3842.1040708@ntlworld.com>
Message-ID: <4E7B634B.2070307@ianhobson.co.uk>

To reply to my own post.

It was a permissions problem.  php could not read the file, and was 
reporting it as not there!

Thanks
Ian

On 22/09/2011 14:29, Ian Hobson wrote:
> Hi All,
>
> My site is wordpress, and configured as follows.
>
> # Statements for KindleMonster.com  virtual server
> server {
>     listen   80;
>     #
>     server_name kindlemonster.com www.kindlemonster.com;
>     root /var/www/kindlemonster.com/htdocs;
>     access_log  /var/www/kindlemonster.com/access.log;
>     index index.php index.html index.htm;
>     #
>     # redirect server error pages to the static page /50x.html
>     #
>     error_page   500 502 503 504  /50x.html;
>     # protect usage stats
>     location ^~ /usage/ {
>       auth_basic "Hello, please login";
>       auth_basic_user_file /var/www/kindlemnonster.com/passwords;
>     }
>     # all  .php requests to fastcgi using uri.
>     location ~ \.php$ {
>         include /etc/nginx/fastcgi_params;
>         fastcgi_pass 127.0.0.1:9000;
>         # fastcgi_param SCRIPT_FILENAME 
> $document_root$fastcgi_script_name;
>     }
>     try_files $uri $uri/ /index.php;
> }
>
> Wordpress works just fine. I have just uploaded a new file to 
> /form.php and when
> I try to browse to it I get the super helpful error message of
>
> No input file specified.
>
> The log says its a 404.
>
> OK - Why do all the wordpress files (mostly php) work just fine, and 
> my new file does not?
>
> Thanks
>
> Ian
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx
>
>
> -----
> No virus found in this message.
> Checked by AVG - www.avg.com
> Version: 10.0.1410 / Virus Database: 1520/3911 - Release Date: 09/21/11
>
>


From nginx-forum at nginx.us  Thu Sep 22 16:54:05 2011
From: nginx-forum at nginx.us (etrader)
Date: Thu, 22 Sep 2011 12:54:05 -0400
Subject: How to avoid 404 for images in access log?
In-Reply-To: <9f2d51afcce11583480dece2d34febee.NginxMailingListEnglish@forum.nginx.org>
References: <9f2d51afcce11583480dece2d34febee.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <520df39faeda9d3f65a3b700f1e0b8dc.NginxMailingListEnglish@forum.nginx.org>

Thanks Reinis! It worked like a charm. But I have another problem (maybe
I should to start a new thread). It lists the access of my own server to
images. When someone visits an html page, an addition to this trick, a
line is added with my server IP for reading the image within the html
page. Since I just want the track of my traffic, this is useless to have
it as my server has loaded the images.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215626,215642#msg-215642


From nginx-forum at nginx.us  Thu Sep 22 17:17:25 2011
From: nginx-forum at nginx.us (etrader)
Date: Thu, 22 Sep 2011 13:17:25 -0400
Subject: $server_name in error_log logfile name
In-Reply-To: <d920539cb3e93201537aebedad3f2e96.NginxMailingListEnglish@forum.nginx.org>
References: <d920539cb3e93201537aebedad3f2e96.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <e7995d2fe8cd4b353151b9af7f566f3c.NginxMailingListEnglish@forum.nginx.org>

As far as I know, Nginx does not support variable in error_log.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215503,215647#msg-215647


From nginx-forum at nginx.us  Thu Sep 22 17:47:47 2011
From: nginx-forum at nginx.us (Lintu)
Date: Thu, 22 Sep 2011 13:47:47 -0400
Subject: php-fastcgi always exits.. resulting in 502 bad gateway
In-Reply-To: <5BC8A8A188F3425EB0C8FCD7C8605D8A@DD21>
References: <5BC8A8A188F3425EB0C8FCD7C8605D8A@DD21>
Message-ID: <cab356ee7c0e06498197bb4c52504d70.NginxMailingListEnglish@forum.nginx.org>

Reinis Rozitis Wrote:
-------------------------------------------------------
> > I'm running Debian Lenny which has only php 5.2
> :(
> 
> I'm not a Debian user but you could add the dotdeb
> repo (
> http://www.dotdeb.org/2011/08/30/php-5-3-8-is-avai
> lable/ ).
> Sticking with old and unsupported php release isnt
> a good thing.
> 
> 
> rr
> 
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx


I used to do this and I hat php 5.3 running with apache. But when I
decided to try something new to go with nginx instead of apache, I had
trouble setting up nginx with php as that repository doesnt have the
php-cgi package. Therefor all attempts to get any phpinfo() file run
failed and I was forced to uninstall php 5.3 to install php 5.2. I
wasn't really happy with that, but at least it works at all... So far
I'm only half-satisfied with nginx, but I'm pretty sure thats just about
the config, nginx sounds fine. Gonne read tons of HowTos and Docs and
stuff I suppose

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215576,215649#msg-215649


From r at roze.lv  Thu Sep 22 18:19:42 2011
From: r at roze.lv (Reinis Rozitis)
Date: Thu, 22 Sep 2011 21:19:42 +0300
Subject: php-fastcgi always exits.. resulting in 502 bad gateway
In-Reply-To: <cab356ee7c0e06498197bb4c52504d70.NginxMailingListEnglish@forum.nginx.org>
References: <5BC8A8A188F3425EB0C8FCD7C8605D8A@DD21>
	<cab356ee7c0e06498197bb4c52504d70.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <029D40CC0F3D4095AEFFEC05BB25DA67@NeiRoze>

> as that repository doesnt have the php-cgi package.

You don't need the -cgi (it is the old naming when fastcgi was built in the 
cgi sapi) package but 'php5-fpm' .
Also I think the packages are prefixed with 'php5' and called php5-fpm / 
php5-cgi ( http://php53.dotdeb.org/dists/lenny/php5/binary-amd64/  )


rr 


From nginx-forum at nginx.us  Thu Sep 22 18:23:08 2011
From: nginx-forum at nginx.us (skyshade)
Date: Thu, 22 Sep 2011 14:23:08 -0400
Subject: New MP4 Module and proxy_cache
Message-ID: <8375196d213446dd1aa144c62daa47d2.NginxMailingListEnglish@forum.nginx.org>

Hello, I test the new mp4 module, but its dont work with proxy_cache.
There is any way to make its works with proxy_cache?

Thanks all

Sky

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215651,215651#msg-215651


From nginx-forum at nginx.us  Thu Sep 22 20:03:51 2011
From: nginx-forum at nginx.us (stonor)
Date: Thu, 22 Sep 2011 16:03:51 -0400
Subject: Serving stale items from cache
In-Reply-To: <4262924B-D501-46F1-95D8-C0DA85D975D3@nginx.com>
References: <4262924B-D501-46F1-95D8-C0DA85D975D3@nginx.com>
Message-ID: <774008bf7b14a3cec6eb468015aedf62.NginxMailingListEnglish@forum.nginx.org>

I would love to see this too in Nginx.

The pattern is described here: http://www.mnot.net/blog/2007/12/12/stale


The only software I've seen this implemented in is Squid 2.7
(stale-while-revalidate).

>From time to time I see backends taking 800ms+  to generate a page at
low traffic. It would be nice to hide most of these requests to the end
user.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,214606,215654#msg-215654


From sjaak23 at gmail.com  Thu Sep 22 20:34:52 2011
From: sjaak23 at gmail.com (Sjaak Pieterse)
Date: Thu, 22 Sep 2011 22:34:52 +0200
Subject: HttpHealthcheckModule server not marked down
Message-ID: <CAP0FwPR78VwaXg4qbftEZbsPFeen-xwrnGQU3wFgdd8rPu9XOg@mail.gmail.com>

Hi there,

i'm trying to use the HttpHealthcheckModule for nginx, but i have some
troubles with it.

i have two servers in my upstream, when sabotaging the health for one
server i see in the status view of healthcheck that the server is
down(1), but if i go to the website i'm checking i still come out on
it and see a broken page.

how can i arrange that the server automatically is marked as down when
the check fails?

sorry for my bad english and maybe noob questions.

config:
?upstream www-health{
??????? server x.x.x.1 ;
??????? server x.x.x.2 ;
??? healthcheck_enabled;
??? healthcheck_delay 10000 ;
??? healthcheck_timeout 1000;
??? healthcheck_failcount 2;
??? #healthcheck_expected 'I_AM_ALIVE';
??? #Important: HTTP/1.0
??? healthcheck_send "GET / HTTP/1.0" 'Host: health.test.x.com'
'Conection: close' ;
}

nginx: nginx version: nginx/1.0.6
nginx: built by gcc 4.4.3 (Ubuntu 4.4.3-4ubuntu5)
nginx: TLS SNI support enabled
nginx: configure arguments: --with-http_ssl_module
--add-module=/gnosek-nginx-upstream-fair-2131c73
--with-http_stub_status_module
--add-module=/cep21-healthcheck_nginx_upstreams-b33a846
--prefix=/usr/local/nginx-1.0.6 --with-debug

used:
peckhardt at test-nginx:~/nginx-1.0.6$patch -p1 <
/cep21-healthcheck_nginx_upstreams-5fa4bff/nginx.patch

hope someone would help me.

greetings


From nginx-forum at nginx.us  Thu Sep 22 20:50:41 2011
From: nginx-forum at nginx.us (Lintu)
Date: Thu, 22 Sep 2011 16:50:41 -0400
Subject: php-fastcgi always exits.. resulting in 502 bad gateway
In-Reply-To: <029D40CC0F3D4095AEFFEC05BB25DA67@NeiRoze>
References: <029D40CC0F3D4095AEFFEC05BB25DA67@NeiRoze>
Message-ID: <fadf6ec71516b6c98dc6a7af56240f53.NginxMailingListEnglish@forum.nginx.org>

So I removed php again, putting dotdeb into sources and commenting
debian out, trying to reinstall php5-common, but this appears:

Package php5-common is not available, but is referred to by another
package.
This may mean that the package is missing, has been obsoleted, or
is only available from another source


What am I doing wrong now? I did update apt-get after editing sources

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215576,215659#msg-215659


From nginx-forum at nginx.us  Thu Sep 22 21:02:18 2011
From: nginx-forum at nginx.us (danieljoleary)
Date: Thu, 22 Sep 2011 17:02:18 -0400
Subject: SSL_write() failed - v1.0.4
Message-ID: <1a69ebe0ef8c4aba5a2fe9ad3e1098b3.NginxMailingListEnglish@forum.nginx.org>

Hello,

I am getting the following in my error.log file:


2011/09/22 14:09:19 [crit] 29589#0: *27891 SSL_write() failed (SSL:)
while sending to client, client: 187.32.96.60, server:
innovationgames.com, request: "GET
/game_view/instant_play/SECVSRHNYL3G5UZGCFSOP2455TURX25Q HTTP/1.1",
upstream:
"http://127.0.0.1:9911/game_view/instant_play/SECVSRHNYL3G5UZGCFSOP2455TURX25Q",
host: "innovationgames.com", referrer:
"http://translate.googleusercontent.com/translate_c?hl=pt-BR&rurl=translate.google.com.br&sl=en&tl=pt&u=http://innovationgames.com/&usg=ALkJrhjEddqgpHPPquhnwql-3tdLHC5nDQ"

I am running nginx v1.0.4.  I noticed another user (dryang) posted a
similar problem agains v0.7.67, but there was no further information on
that post.

I also notice the following post in the development forum
(http://mailman.nginx.org/pipermail/nginx-devel/2010-February/000118.html)
which indicated a patch but that was over a year and a half ago.

Any help very much appreciated.

Thanks for allowing me to post here.


Dan O'Leary
The Innovation Games Company

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215662,215662#msg-215662


From nginx-forum at nginx.us  Thu Sep 22 21:03:24 2011
From: nginx-forum at nginx.us (Lintu)
Date: Thu, 22 Sep 2011 17:03:24 -0400
Subject: php-fastcgi always exits.. resulting in 502 bad gateway
In-Reply-To: <fadf6ec71516b6c98dc6a7af56240f53.NginxMailingListEnglish@forum.nginx.org>
References: <029D40CC0F3D4095AEFFEC05BB25DA67@NeiRoze>
	<fadf6ec71516b6c98dc6a7af56240f53.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <dd75fbb626a9c9775935fbfcf39ce7ce.NginxMailingListEnglish@forum.nginx.org>

Disregard that, I just directly downloaded the debs with wget. For the
moment its working, thanks alot! I hope it won't crash down every now
and then anymore.

Thanks alot again to anyone who spent his time in this thread!

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215576,215663#msg-215663


From erik at erikwickstrom.com  Thu Sep 22 21:09:45 2011
From: erik at erikwickstrom.com (Erik Wickstrom)
Date: Thu, 22 Sep 2011 14:09:45 -0700
Subject: Serving stale items from cache
In-Reply-To: <774008bf7b14a3cec6eb468015aedf62.NginxMailingListEnglish@forum.nginx.org>
References: <4262924B-D501-46F1-95D8-C0DA85D975D3@nginx.com>
	<774008bf7b14a3cec6eb468015aedf62.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <CAHfS1hMSJDkVvdca+MSnYZmSPD_9ff5DqO_u+GunONRgwz29QQ@mail.gmail.com>

I ended up using Varnish for this.  I set a long TTL for the cache,
and then another long "grace" period for those expired objects.  Then
I use a script to loop through all of my URLs and rewarm the cache
every few days.

https://www.varnish-cache.org/trac/wiki/VCLExampleHashAlwaysMiss

Seems to be working.

On Thu, Sep 22, 2011 at 1:03 PM, stonor <nginx-forum at nginx.us> wrote:
> I would love to see this too in Nginx.
>
> The pattern is described here: http://www.mnot.net/blog/2007/12/12/stale
>
>
> The only software I've seen this implemented in is Squid 2.7
> (stale-while-revalidate).
>
> From time to time I see backends taking 800ms+ ?to generate a page at
> low traffic. It would be nice to hide most of these requests to the end
> user.
>
> Posted at Nginx Forum: http://forum.nginx.org/read.php?2,214606,215654#msg-215654
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx
>


From brian at akins.org  Thu Sep 22 23:05:01 2011
From: brian at akins.org (Brian Akins)
Date: Thu, 22 Sep 2011 19:05:01 -0400
Subject: Serving stale items from cache
In-Reply-To: <774008bf7b14a3cec6eb468015aedf62.NginxMailingListEnglish@forum.nginx.org>
References: <4262924B-D501-46F1-95D8-C0DA85D975D3@nginx.com>
	<774008bf7b14a3cec6eb468015aedf62.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <A46C46A4-6116-423B-B58F-74B3A43A2AE8@akins.org>


proxy_cache_use_stale updating does soemthing similar.

You may could do something like (from the docs)

 proxy_cache_bypass $http_my_secret_header;

without a corresponding proxy_no_cache and that should make nginx go to the origin server but cache the results.

--Brian

From mdounin at mdounin.ru  Thu Sep 22 23:18:23 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Fri, 23 Sep 2011 03:18:23 +0400
Subject: SSL_write() failed - v1.0.4
In-Reply-To: <1a69ebe0ef8c4aba5a2fe9ad3e1098b3.NginxMailingListEnglish@forum.nginx.org>
References: <1a69ebe0ef8c4aba5a2fe9ad3e1098b3.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20110922231822.GK1137@mdounin.ru>

Hello!

On Thu, Sep 22, 2011 at 05:02:18PM -0400, danieljoleary wrote:

> Hello,
> 
> I am getting the following in my error.log file:
> 
> 
> 2011/09/22 14:09:19 [crit] 29589#0: *27891 SSL_write() failed (SSL:)
> while sending to client, client: 187.32.96.60, server:
> innovationgames.com, request: "GET
> /game_view/instant_play/SECVSRHNYL3G5UZGCFSOP2455TURX25Q HTTP/1.1",
> upstream:
> "http://127.0.0.1:9911/game_view/instant_play/SECVSRHNYL3G5UZGCFSOP2455TURX25Q",
> host: "innovationgames.com", referrer:
> "http://translate.googleusercontent.com/translate_c?hl=pt-BR&rurl=translate.google.com.br&sl=en&tl=pt&u=http://innovationgames.com/&usg=ALkJrhjEddqgpHPPquhnwql-3tdLHC5nDQ"
> 
> I am running nginx v1.0.4.  I noticed another user (dryang) posted a
> similar problem agains v0.7.67, but there was no further information on
> that post.

Are you able to reproduce this?  If yes, please provide 
debug log.  See http://wiki.nginx.org/Debugging for details.

> I also notice the following post in the development forum
> (http://mailman.nginx.org/pipermail/nginx-devel/2010-February/000118.html)
> which indicated a patch but that was over a year and a half ago.

This post just mentions such errors as something 
not-yet-identified, patch is unrelated and tunes logging level for 
other common SSL errors.

Maxim Dounin


From nginx-forum at nginx.us  Thu Sep 22 23:26:57 2011
From: nginx-forum at nginx.us (danieljoleary)
Date: Thu, 22 Sep 2011 19:26:57 -0400
Subject: SSL_write() failed - v1.0.4
In-Reply-To: <20110922231822.GK1137@mdounin.ru>
References: <20110922231822.GK1137@mdounin.ru>
Message-ID: <17379b62945337b8bfdbd270abb82999.NginxMailingListEnglish@forum.nginx.org>

Hi Maxim,

I currently only have one instance of the error but expect there will be
more.  To prepare for this, I will build a debug version of nginx and
set the log level to debug per the directions you have indicated.  It
may be a while before I come back on this thread.

Thanks for your response!


Dan

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215662,215671#msg-215671


From amitkulz at gmail.com  Thu Sep 22 23:54:48 2011
From: amitkulz at gmail.com (Amit Kulkarni)
Date: Thu, 22 Sep 2011 18:54:48 -0500
Subject: nginx imported into OpenBSD base
Message-ID: <CAOS-L3gzWVNHE09SsWqpKTZDgtHC=5ApuaWOUW4wVFGTt=a7ag@mail.gmail.com>

Hi all,

You guys will get that chroot stuff and the httpd man page due to the
hard work of OpenBSD guys... Eventually it will replace Apache-1.3 in
base. It will be mutually beneficial relationship.

thanks!!!


From nginx-forum at nginx.us  Fri Sep 23 01:01:25 2011
From: nginx-forum at nginx.us (ahu)
Date: Thu, 22 Sep 2011 21:01:25 -0400
Subject: mp4 module
In-Reply-To: <76c3136923f6db464986dd2fd3c4fa01.NginxMailingListEnglish@forum.nginx.org>
References: <20110909115107.GH97288@nginx.com>
	<76c3136923f6db464986dd2fd3c4fa01.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <6deffe35747aebe8fddcc59d0a973221.NginxMailingListEnglish@forum.nginx.org>

for me,"If you use 3rd-party mp4 module, it should be removed." just
means they are different.I want to know the detail :)
Should I use the official mp4 module instead of the 3rd-party one?

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215011,215678#msg-215678


From ianevans at digitalhit.com  Fri Sep 23 06:45:09 2011
From: ianevans at digitalhit.com (Ian Evans)
Date: Fri, 23 Sep 2011 02:45:09 -0400
Subject: CSS file in Facebook app being served up as text/html
Message-ID: <4E7C2AF5.1060403@digitalhit.com>

I'm hosting a Facebook canvas app on my server and a CSS file is not 
getting loaded because it's apparently being served up as text/html and 
not text/css.

I have no problems with the css files on my web site hosted on the same 
server and my nginx conf has the standard mime types file loaded in the 
conf.

Not sure why calling it in a facebook iframe is causing the error. 
Here's the dump from Live HTTP Headers:

GET /myappname/multifriend.css HTTP/1.1
Host: www.example.com
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:6.0.2) Gecko/20100101 
Firefox/6.0.2
Accept: text/css,*/*;q=0.1
Accept-Language: en-gb,en;q=0.5
Accept-Encoding: gzip, deflate
Accept-Charset: ISO-8859-1,utf-8;q=0.7,*;q=0.7
Connection: keep-alive
Referer: https://www.example.com/myappname/invite3.php


HTTP/1.1 200 OK
Date: Fri, 23 Sep 2011 06:28:49 GMT
Content-Type: text/html
Transfer-Encoding: chunked
Connection: keep-alive
X-Powered-By: PHP/5.3.6

Thanks!


From nginx-forum at nginx.us  Fri Sep 23 07:24:26 2011
From: nginx-forum at nginx.us (yamanyar)
Date: Fri, 23 Sep 2011 03:24:26 -0400
Subject: Accept all server_name's except empty
Message-ID: <411ab344243934ff710696ca9a21ed6d.NginxMailingListEnglish@forum.nginx.org>

Dear all,

How can I set up my nginx configuration file so that it will accept all
request; however it will deny if "host" header is not added?

Also currently I am logging bytes_send per "host" and then take sum of
them with a cron job monthly. Is there a better way to achieve this?

Cheers,
Kaan Yamanyar

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215685,215685#msg-215685


From igor at sysoev.ru  Fri Sep 23 07:47:56 2011
From: igor at sysoev.ru (Igor Sysoev)
Date: Fri, 23 Sep 2011 11:47:56 +0400
Subject: Accept all server_name's except empty
In-Reply-To: <411ab344243934ff710696ca9a21ed6d.NginxMailingListEnglish@forum.nginx.org>
References: <411ab344243934ff710696ca9a21ed6d.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <0D73320C-46C0-42F4-9029-40DD8A0C83CC@sysoev.ru>

On Sep 23, 2011, at 11:24 , yamanyar wrote:

> Dear all,
> 
> How can I set up my nginx configuration file so that it will accept all
> request; however it will deny if "host" header is not added?

server {
    server_name  "";  # no host header field in request
    return 444;
}

server {
    server_name  _;
    ...
}

> Also currently I am logging bytes_send per "host" and then take sum of
> them with a cron job monthly. Is there a better way to achieve this?

Currently no.


--
Igor Sysoev
http://sysoev.ru/en/


From osa at FreeBSD.org.ru  Fri Sep 23 08:31:25 2011
From: osa at FreeBSD.org.ru (Sergey A. Osokin)
Date: Fri, 23 Sep 2011 12:31:25 +0400
Subject: [ANN] ngx_http_redis-0.3.5 released
Message-ID: <20110923083125.GD81654@FreeBSD.org.ru>

ngx_http_redis module version 0.3.5 released, available for immediate 
download at http://people.freebsd.org/~osa/ngx_http_redis-0.3.5.tar.gz

SHA256 (ngx_http_redis-0.3.5.tar.gz) = 5398805bbc790b8ae5ddb5079cf726df63c8c206bbf9a3a4b035412194aac82e
SIZE (ngx_http_redis-0.3.5.tar.gz) = 11373
 
<ChangeLog>

*) Feature: add test for not set $redis_db directive.

*) Feature: keep-alive support merged from original
   memcached module 1.1.4.
 
</ChangeLog>

Note.  Keep-alive support works only with nginx 1.1.4 and later.
For previous versions of nginx use ngx_http_upstream_keepalive module
from Maxim Dounin.

-- 
Sergey A. Osokin
osa at FreeBSD.ORG.ru
osa at FreeBSD.ORG


From nginx-forum at nginx.us  Fri Sep 23 10:52:50 2011
From: nginx-forum at nginx.us (Hety)
Date: Fri, 23 Sep 2011 06:52:50 -0400
Subject: $server_name in error_log logfile name
In-Reply-To: <e7995d2fe8cd4b353151b9af7f566f3c.NginxMailingListEnglish@forum.nginx.org>
References: <d920539cb3e93201537aebedad3f2e96.NginxMailingListEnglish@forum.nginx.org>
	<e7995d2fe8cd4b353151b9af7f566f3c.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <6416e397fc9e97304d5da10be69a1ebe.NginxMailingListEnglish@forum.nginx.org>

Why its supports it in access_log then? I fail to see logic here.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215503,215702#msg-215702


From r at roze.lv  Fri Sep 23 11:09:45 2011
From: r at roze.lv (Reinis Rozitis)
Date: Fri, 23 Sep 2011 14:09:45 +0300
Subject: CSS file in Facebook app being served up as text/html
In-Reply-To: <4E7C2AF5.1060403@digitalhit.com>
References: <4E7C2AF5.1060403@digitalhit.com>
Message-ID: <CA44284EBA8D4A759134558546A23C35@DD21>

> I have no problems with the css files on my web site hosted on the same server and my nginx conf has the standard mime types file 
> loaded in the conf.

Post your nginx config since you may have a misconfiguration somewhere and css files are being served through php - header line' 
X-Powered-By: PHP/5.3.6' indicates that (it shouldn't be there for static content).


p.s. If that is intended though you will have to modify your php and add <? header('Content-type: text/css'); ?> at the top.

rr 


From mdounin at mdounin.ru  Fri Sep 23 11:19:09 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Fri, 23 Sep 2011 15:19:09 +0400
Subject: $server_name in error_log logfile name
In-Reply-To: <6416e397fc9e97304d5da10be69a1ebe.NginxMailingListEnglish@forum.nginx.org>
References: <d920539cb3e93201537aebedad3f2e96.NginxMailingListEnglish@forum.nginx.org>
	<e7995d2fe8cd4b353151b9af7f566f3c.NginxMailingListEnglish@forum.nginx.org>
	<6416e397fc9e97304d5da10be69a1ebe.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20110923111908.GR1137@mdounin.ru>

Hello!

On Fri, Sep 23, 2011 at 06:52:50AM -0400, Hety wrote:

> Why its supports it in access_log then? I fail to see logic here.

Access log is quite different as it's only needed/used during 
http request processing.  On the other hand, error_log is global thing 
and used even if no http requests exists, e.g. to log signals and 
various other internal actions.  It's not possible to use 
variables which are only exists in http request context to log 
internal actions.

And again (this tends to be a FAQ):

You shouldn't use variables to shorten configs by storing static 
strings in them.  Variables are evaluated at run-time during 
request process, they are costly compared to static strings.  You 
need some macro expansion instead.  Right now it may be done via 
external tools, e.g. sed + make will do (as well as any other 
template mechanism).

Maxim Dounin


From nginx-forum at nginx.us  Fri Sep 23 11:58:23 2011
From: nginx-forum at nginx.us (Hety)
Date: Fri, 23 Sep 2011 07:58:23 -0400
Subject: $server_name in error_log logfile name
In-Reply-To: <20110923111908.GR1137@mdounin.ru>
References: <20110923111908.GR1137@mdounin.ru>
Message-ID: <a2a242186cddf69411048557e62191a2.NginxMailingListEnglish@forum.nginx.org>

Thanks. Point taken. Will go RTFMmore and will try to avoid stupid
questions in future :)

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215503,215713#msg-215713


From igor at sysoev.ru  Fri Sep 23 15:22:58 2011
From: igor at sysoev.ru (Igor Sysoev)
Date: Fri, 23 Sep 2011 19:22:58 +0400
Subject: mp4 module
In-Reply-To: <6deffe35747aebe8fddcc59d0a973221.NginxMailingListEnglish@forum.nginx.org>
References: <20110909115107.GH97288@nginx.com>
	<76c3136923f6db464986dd2fd3c4fa01.NginxMailingListEnglish@forum.nginx.org>
	<6deffe35747aebe8fddcc59d0a973221.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <F606CB2F-DD75-434A-92B3-C2D0DCB7A3E2@sysoev.ru>

On Sep 23, 2011, at 5:01 , ahu wrote:

> for me,"If you use 3rd-party mp4 module, it should be removed." just
> means they are different.I want to know the detail :)
> Should I use the official mp4 module instead of the 3rd-party one?

You should use only one of them. If you have issues with 3rd-party,
you can try the official mp4 module. If you have issues with the
official module, you should report them and I wll try to fix them.


--
Igor Sysoev
http://sysoev.ru/en/


From contact at jpluscplusm.com  Fri Sep 23 16:37:29 2011
From: contact at jpluscplusm.com (Jonathan Matthews)
Date: Fri, 23 Sep 2011 17:37:29 +0100
Subject: Override proxy's incorrect Content-Type via mime settings
Message-ID: <CAKsTx7Abwp3e+BpLOXtXMPe-oO2n_HdBpLOPoCqj6ymw5qs-WQ@mail.gmail.com>

Hi all -

I have an upstream proxy that sends the wrong Content-Type header for
a load of images. Sometime it doesn't send any header at all.

I'm trying to get Nginx not only to remove the upstream's Content-Type
header ("proxy_hide_header"), but also to insert the correct header.
Unfortunately, I can't hard-code the value as there are a variety of
image formats being used. "proxy_hide_header" doesn't seem to do
anything other than block the incorrect value, and
"proxy_ignore_headers" won't accept "Content-Type".

How can I both remote upstream's incorrect header, and also add the
header than Nginx *would* have used if it were serving the content
from local disk?
In other words, how do I add the header with a value derived from the
MIME settings currently configured, via the requested file's
extension?

Version 0.7.67, but can upgrade if required.

Many thanks,
Jonathan
-- 
Jonathan Matthews
London, UK
http://www.jpluscplusm.com/contact.html


From contact at jpluscplusm.com  Fri Sep 23 16:45:33 2011
From: contact at jpluscplusm.com (Jonathan Matthews)
Date: Fri, 23 Sep 2011 17:45:33 +0100
Subject: Override proxy's incorrect Content-Type via mime settings
In-Reply-To: <CAKsTx7Abwp3e+BpLOXtXMPe-oO2n_HdBpLOPoCqj6ymw5qs-WQ@mail.gmail.com>
References: <CAKsTx7Abwp3e+BpLOXtXMPe-oO2n_HdBpLOPoCqj6ymw5qs-WQ@mail.gmail.com>
Message-ID: <CAKsTx7Ach=J_mv97rpwdNbUcMZ0O3RNYb_Ju0KFNVzmsgonXJQ@mail.gmail.com>

On 23 September 2011 17:37, Jonathan Matthews <contact at jpluscplusm.com> wrote:
> How can I both remote upstream's incorrect header, and also add the
                *remove*


From ianevans at digitalhit.com  Fri Sep 23 19:12:03 2011
From: ianevans at digitalhit.com (Ian Evans)
Date: Fri, 23 Sep 2011 15:12:03 -0400
Subject: CSS file in Facebook app being served up as text/html
In-Reply-To: <CA44284EBA8D4A759134558546A23C35@DD21>
References: <4E7C2AF5.1060403@digitalhit.com>
	<CA44284EBA8D4A759134558546A23C35@DD21>
Message-ID: <4E7CDA03.2020308@digitalhit.com>

On 23/09/2011 7:09 AM, Reinis Rozitis wrote:
> Post your nginx config since you may have a misconfiguration somewhere
> and css files are being served through php - header line' X-Powered-By:
> PHP/5.3.6' indicates that (it shouldn't be there for static content).

You hit the nail on the head...I don't do much work on the SSL side of 
my server and forgot to allow it to be served directly.

Thanks.


From nginx-forum at nginx.us  Sat Sep 24 01:15:11 2011
From: nginx-forum at nginx.us (ahu)
Date: Fri, 23 Sep 2011 21:15:11 -0400
Subject: mp4 module
In-Reply-To: <F606CB2F-DD75-434A-92B3-C2D0DCB7A3E2@sysoev.ru>
References: <F606CB2F-DD75-434A-92B3-C2D0DCB7A3E2@sysoev.ru>
Message-ID: <ca0fba55e1e181e1e75331f73ce07307.NginxMailingListEnglish@forum.nginx.org>

Ok,I get it.Thanks very much.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215011,215751#msg-215751


From luke at visionlaunchers.com  Sat Sep 24 07:12:35 2011
From: luke at visionlaunchers.com (Luke Scott)
Date: Sat, 24 Sep 2011 07:12:35 +0000
Subject: $https variable?
Message-ID: <283FEC93C84E1248B931166DC960A8AA1A647C3D@MBX205.domain.local>

It would be great if there was an $https variable that was "on" or "off" (or "") so you could do something like this:

fastcgi_param  HTTPS $https;

Because currently telling PHP https is enabled is much harder, especially for those that prefer to have a single server{} block for both HTTP and HTTPS.

Typically PHP uses $_SERVER['HTTPS'] === 'on' to detect whether or not HTTPS is enabled.

http://us3.php.net/manual/en/reserved.variables.server.php

Luke

From ano at bestmx.ru  Sat Sep 24 07:24:48 2011
From: ano at bestmx.ru (Andrey N. Oktyabrski)
Date: Sat, 24 Sep 2011 11:24:48 +0400
Subject: $https variable?
In-Reply-To: <283FEC93C84E1248B931166DC960A8AA1A647C3D@MBX205.domain.local>
References: <283FEC93C84E1248B931166DC960A8AA1A647C3D@MBX205.domain.local>
Message-ID: <4E7D85C0.3080103@bestmx.ru>

On 24.09.11 11:12, Luke Scott wrote:
> It would be great if there was an $https variable that was "on" or
> "off" (or "") so you could do something like this:
> fastcgi_param  HTTPS $https;
map $scheme $https {
   https    on;
   default  off;
}


From mike503 at gmail.com  Sat Sep 24 07:25:55 2011
From: mike503 at gmail.com (Michael Shadle)
Date: Sat, 24 Sep 2011 00:25:55 -0700
Subject: $https variable?
In-Reply-To: <283FEC93C84E1248B931166DC960A8AA1A647C3D@MBX205.domain.local>
References: <283FEC93C84E1248B931166DC960A8AA1A647C3D@MBX205.domain.local>
Message-ID: <CAKa35F5GpaaoVupeK9rB3YmxjGLsGsYO5NHC=+5Xp2ntWxBYzg@mail.gmail.com>

Igor proposed this a while ago:

on the global level (in http{} block)
map $scheme $php_https { https on; http off; }

then in fastcgi_params
fastcgi_param HTTPS $php_https;



On Sat, Sep 24, 2011 at 12:12 AM, Luke Scott <luke at visionlaunchers.com> wrote:
> It would be great if there was an $https variable that was "on" or "off" (or "") so you could do something like this:
>
> fastcgi_param ?HTTPS $https;
>
> Because currently telling PHP https is enabled is much harder, especially for those that prefer to have a single server{} block for both HTTP and HTTPS.
>
> Typically PHP uses $_SERVER['HTTPS'] === 'on' to detect whether or not HTTPS is enabled.
>
> http://us3.php.net/manual/en/reserved.variables.server.php
>
> Luke
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx
>


From luc.juggery at gmail.com  Sat Sep 24 11:43:12 2011
From: luc.juggery at gmail.com (Luc Juggery)
Date: Sat, 24 Sep 2011 13:43:12 +0200
Subject: Cannot remove Content Length header
Message-ID: <CAEVF9DX4MNPp5Vh7opoX7-Ub4e7CP564jtXNwRoB_3YidTU-wg@mail.gmail.com>

Hello,

I'm using Nginx as a front end to my node.js application, nginx proxying
https request to my app.
When the browser gets the response back, instead of pure json
({"error":"query error","message":"Parameter(s) missing: user,password"}), I
receive something like:

47
{"error":"query error","message":"Parameter(s) missing: user,password"}
0

I use 'more_clear_headers' module but do not have any luck.

This is the config I use:

server {
    listen      443;
    server_name my.server.com;
    ssl                 on;
    ssl_certificate    certificate.pem;
    ssl_certificate_key privatekey.pem;

    keepalive_timeout 70;

    location / {
      more_clear_headers 'Content-Length';   # Does not work as I would
expect
      proxy_pass https://localhost:8443;
      proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
      proxy_set_header Host $http_host;
      proxy_buffering off;
      proxy_buffers 8 8k;
    }
}

Would you have any idea ?
Thanks a lot for your help,

-- 
Luc
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110924/a124f349/attachment.html>

From Richard.Kearsley at m247.com  Sat Sep 24 11:49:34 2011
From: Richard.Kearsley at m247.com (Richard Kearsley)
Date: Sat, 24 Sep 2011 11:49:34 +0000
Subject: Cannot remove Content Length header
In-Reply-To: <CAEVF9DX4MNPp5Vh7opoX7-Ub4e7CP564jtXNwRoB_3YidTU-wg@mail.gmail.com>
References: <CAEVF9DX4MNPp5Vh7opoX7-Ub4e7CP564jtXNwRoB_3YidTU-wg@mail.gmail.com>
Message-ID: <DE9B192E-2CCA-40B8-BA01-982B1A696CF1@m247.com>

How are you testing it?
That looks like results I saw using telnet but those numbers do not appear with a http client such as curl


On 24 Sep 2011, at 12:43, "Luc Juggery" <luc.juggery at gmail.com<mailto:luc.juggery at gmail.com>> wrote:

Hello,

I'm using Nginx as a front end to my node.js application, nginx proxying https request to my app.
When the browser gets the response back, instead of pure json ({"error":"query error","message":"Parameter(s) missing: user,password"}), I receive something like:

47
{"error":"query error","message":"Parameter(s) missing: user,password"}
0

I use 'more_clear_headers' module but do not have any luck.

This is the config I use:

server {
    listen      443;
    server_name <http://my.server.com> my.server.com<http://my.server.com>;
    ssl                 on;
    ssl_certificate    certificate.pem;
    ssl_certificate_key privatekey.pem;

    keepalive_timeout 70;

    location / {
      more_clear_headers 'Content-Length';   # Does not work as I would expect
      proxy_pass <https://localhost:8443> https://localhost:8443;
      proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
      proxy_set_header Host $http_host;
      proxy_buffering off;
      proxy_buffers 8 8k;
    }
}

Would you have any idea ?
Thanks a lot for your help,

--
Luc
_______________________________________________
nginx mailing list
nginx at nginx.org<mailto:nginx at nginx.org>
http://mailman.nginx.org/mailman/listinfo/nginx
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110924/c001af77/attachment-0001.html>

From mdounin at mdounin.ru  Sat Sep 24 12:05:31 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Sat, 24 Sep 2011 16:05:31 +0400
Subject: Cannot remove Content Length header
In-Reply-To: <CAEVF9DX4MNPp5Vh7opoX7-Ub4e7CP564jtXNwRoB_3YidTU-wg@mail.gmail.com>
References: <CAEVF9DX4MNPp5Vh7opoX7-Ub4e7CP564jtXNwRoB_3YidTU-wg@mail.gmail.com>
Message-ID: <20110924120530.GA1137@mdounin.ru>

Hello!

On Sat, Sep 24, 2011 at 01:43:12PM +0200, Luc Juggery wrote:

> Hello,
> 
> I'm using Nginx as a front end to my node.js application, nginx proxying
> https request to my app.
> When the browser gets the response back, instead of pure json
> ({"error":"query error","message":"Parameter(s) missing: user,password"}), I
> receive something like:
> 
> 47
> {"error":"query error","message":"Parameter(s) missing: user,password"}
> 0

This is chunked encoding, likely sent by your backend in violation 
of HTTP specification.

Right now there are two options available:

1. Fix your backend.

2. Upgrade to nginx 1.1.4, it has code needed to talk to such 
broken backends.

Maxim Dounin


From luc.juggery at gmail.com  Sat Sep 24 12:58:59 2011
From: luc.juggery at gmail.com (Luc Juggery)
Date: Sat, 24 Sep 2011 14:58:59 +0200
Subject: Cannot remove Content Length header
In-Reply-To: <20110924120530.GA1137@mdounin.ru>
References: <CAEVF9DX4MNPp5Vh7opoX7-Ub4e7CP564jtXNwRoB_3YidTU-wg@mail.gmail.com>
	<20110924120530.GA1137@mdounin.ru>
Message-ID: <B0CB97C9-FE33-4F41-BAC2-E52CD5CD21C3@gmail.com>

Hello,

thanks for your answers.
Maxim, what would you recommend to fix the backend ? I do not really see where could be the violation in fact.

--
Luc

Le Sep 24, 2011 ? 14:05, Maxim Dounin <mdounin at mdounin.ru> a ?crit :

> Hello!
> 
> On Sat, Sep 24, 2011 at 01:43:12PM +0200, Luc Juggery wrote:
> 
>> Hello,
>> 
>> I'm using Nginx as a front end to my node.js application, nginx proxying
>> https request to my app.
>> When the browser gets the response back, instead of pure json
>> ({"error":"query error","message":"Parameter(s) missing: user,password"}), I
>> receive something like:
>> 
>> 47
>> {"error":"query error","message":"Parameter(s) missing: user,password"}
>> 0
> 
> This is chunked encoding, likely sent by your backend in violation 
> of HTTP specification.
> 
> Right now there are two options available:
> 
> 1. Fix your backend.
> 
> 2. Upgrade to nginx 1.1.4, it has code needed to talk to such 
> broken backends.
> 
> Maxim Dounin
> 
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx


From luc.juggery at gmail.com  Sat Sep 24 13:06:32 2011
From: luc.juggery at gmail.com (Luc Juggery)
Date: Sat, 24 Sep 2011 15:06:32 +0200
Subject: Cannot remove Content Length header
In-Reply-To: <DE9B192E-2CCA-40B8-BA01-982B1A696CF1@m247.com>
References: <CAEVF9DX4MNPp5Vh7opoX7-Ub4e7CP564jtXNwRoB_3YidTU-wg@mail.gmail.com>
	<DE9B192E-2CCA-40B8-BA01-982B1A696CF1@m247.com>
Message-ID: <E20A512C-A775-4336-A48C-5CF9F9817AF4@gmail.com>

Hello,

I'm testing it from a browser and with curl as well and I got the same.
When I bypass nginx the response is correct. When I request from nginx, the lenght is added at the beginning and 0 at the end of the json I would expect.

--
Luc

Le Sep 24, 2011 ? 13:49, Richard Kearsley <Richard.Kearsley at m247.com> a ?crit :

> How are you testing it?
> That looks like results I saw using telnet but those numbers do not appear with a http client such as curl
> 
> 
> On 24 Sep 2011, at 12:43, "Luc Juggery" <luc.juggery at gmail.com> wrote:
> 
>> Hello,
>> 
>> I'm using Nginx as a front end to my node.js application, nginx proxying https request to my app.
>> When the browser gets the response back, instead of pure json ({"error":"query error","message":"Parameter(s) missing: user,password"}), I receive something like:
>> 
>> 47
>> {"error":"query error","message":"Parameter(s) missing: user,password"}
>> 0
>> 
>> I use 'more_clear_headers' module but do not have any luck.
>> 
>> This is the config I use:
>> 
>> server {
>>     listen      443;
>>     server_name my.server.com;
>>     ssl                 on;
>>     ssl_certificate    certificate.pem;
>>     ssl_certificate_key privatekey.pem;
>> 
>>     keepalive_timeout 70;
>> 
>>     location / {
>>       more_clear_headers 'Content-Length';   # Does not work as I would expect
>>       proxy_pass https://localhost:8443;
>>       proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
>>       proxy_set_header Host $http_host;
>>       proxy_buffering off;
>>       proxy_buffers 8 8k;
>>     } 
>> }
>> 
>> Would you have any idea ?
>> Thanks a lot for your help,
>> 
>> -- 
>> Luc
> 
>> _______________________________________________
>> nginx mailing list
>> nginx at nginx.org
>> http://mailman.nginx.org/mailman/listinfo/nginx
> 
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110924/890524cf/attachment.html>

From abioy.sun at gmail.com  Sat Sep 24 14:47:50 2011
From: abioy.sun at gmail.com (Abioy Sun)
Date: Sat, 24 Sep 2011 22:47:50 +0800
Subject: will NGX_AGAIN kill performance?
Message-ID: <CAM6b-vko_0rQELdeJH=ED9rxEvJTAhHo0UtHc+eqTzSm-0-D6A@mail.gmail.com>

hi,

i currently dev a module which makes multi subrequest sequently in the
access phase handler, via returning NGX_AGAIN.

my question is, will this kill the performance? and how far could it go in
killing performance.

thx!

Abioy
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110924/cb276af9/attachment.html>

From mdounin at mdounin.ru  Sat Sep 24 14:54:47 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Sat, 24 Sep 2011 18:54:47 +0400
Subject: Cannot remove Content Length header
In-Reply-To: <B0CB97C9-FE33-4F41-BAC2-E52CD5CD21C3@gmail.com>
References: <CAEVF9DX4MNPp5Vh7opoX7-Ub4e7CP564jtXNwRoB_3YidTU-wg@mail.gmail.com>
	<20110924120530.GA1137@mdounin.ru>
	<B0CB97C9-FE33-4F41-BAC2-E52CD5CD21C3@gmail.com>
Message-ID: <20110924145447.GB1137@mdounin.ru>

Hello!

On Sat, Sep 24, 2011 at 02:58:59PM +0200, Luc Juggery wrote:

> Hello,
> 
> thanks for your answers.
> Maxim, what would you recommend to fix the backend ? I do not really see where could be the violation in fact.

Backend MUST NOT (as per RFC 2616) return chunked to HTTP/1.0 
clients, while nginx uses HTTP/1.0 to talk to backends.

See RFC 2616, "3.6 Transfer Codings"
http://tools.ietf.org/html/rfc2616#section-3.6

   ... A server MUST NOT send transfer-codings to an HTTP/1.0
   client.

Maxim Dounin

> 
> --
> Luc
> 
> Le Sep 24, 2011 ? 14:05, Maxim Dounin <mdounin at mdounin.ru> a ?crit :
> 
> > Hello!
> > 
> > On Sat, Sep 24, 2011 at 01:43:12PM +0200, Luc Juggery wrote:
> > 
> >> Hello,
> >> 
> >> I'm using Nginx as a front end to my node.js application, nginx proxying
> >> https request to my app.
> >> When the browser gets the response back, instead of pure json
> >> ({"error":"query error","message":"Parameter(s) missing: user,password"}), I
> >> receive something like:
> >> 
> >> 47
> >> {"error":"query error","message":"Parameter(s) missing: user,password"}
> >> 0
> > 
> > This is chunked encoding, likely sent by your backend in violation 
> > of HTTP specification.
> > 
> > Right now there are two options available:
> > 
> > 1. Fix your backend.
> > 
> > 2. Upgrade to nginx 1.1.4, it has code needed to talk to such 
> > broken backends.
> > 
> > Maxim Dounin
> > 
> > _______________________________________________
> > nginx mailing list
> > nginx at nginx.org
> > http://mailman.nginx.org/mailman/listinfo/nginx
> 
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx


From luc.juggery at gmail.com  Sat Sep 24 17:24:08 2011
From: luc.juggery at gmail.com (Luc Juggery)
Date: Sat, 24 Sep 2011 19:24:08 +0200
Subject: Cannot remove Content Length header
In-Reply-To: <20110924145447.GB1137@mdounin.ru>
References: <CAEVF9DX4MNPp5Vh7opoX7-Ub4e7CP564jtXNwRoB_3YidTU-wg@mail.gmail.com>
	<20110924120530.GA1137@mdounin.ru>
	<B0CB97C9-FE33-4F41-BAC2-E52CD5CD21C3@gmail.com>
	<20110924145447.GB1137@mdounin.ru>
Message-ID: <B9C24323-96B9-4162-A730-C76D457F7AEE@gmail.com>

Does it mean something (like a parameter or something) needs to be modified on my node.js application ? Do I need to send additionnal stuff ? I did not really get the "chunked" part.

--
Luc

Le Sep 24, 2011 ? 16:54, Maxim Dounin <mdounin at mdounin.ru> a ?crit :

> Hello!
> 
> On Sat, Sep 24, 2011 at 02:58:59PM +0200, Luc Juggery wrote:
> 
>> Hello,
>> 
>> thanks for your answers.
>> Maxim, what would you recommend to fix the backend ? I do not really see where could be the violation in fact.
> 
> Backend MUST NOT (as per RFC 2616) return chunked to HTTP/1.0 
> clients, while nginx uses HTTP/1.0 to talk to backends.
> 
> See RFC 2616, "3.6 Transfer Codings"
> http://tools.ietf.org/html/rfc2616#section-3.6
> 
>   ... A server MUST NOT send transfer-codings to an HTTP/1.0
>   client.
> 
> Maxim Dounin
> 
>> 
>> --
>> Luc
>> 
>> Le Sep 24, 2011 ? 14:05, Maxim Dounin <mdounin at mdounin.ru> a ?crit :
>> 
>>> Hello!
>>> 
>>> On Sat, Sep 24, 2011 at 01:43:12PM +0200, Luc Juggery wrote:
>>> 
>>>> Hello,
>>>> 
>>>> I'm using Nginx as a front end to my node.js application, nginx proxying
>>>> https request to my app.
>>>> When the browser gets the response back, instead of pure json
>>>> ({"error":"query error","message":"Parameter(s) missing: user,password"}), I
>>>> receive something like:
>>>> 
>>>> 47
>>>> {"error":"query error","message":"Parameter(s) missing: user,password"}
>>>> 0
>>> 
>>> This is chunked encoding, likely sent by your backend in violation 
>>> of HTTP specification.
>>> 
>>> Right now there are two options available:
>>> 
>>> 1. Fix your backend.
>>> 
>>> 2. Upgrade to nginx 1.1.4, it has code needed to talk to such 
>>> broken backends.
>>> 
>>> Maxim Dounin
>>> 
>>> _______________________________________________
>>> nginx mailing list
>>> nginx at nginx.org
>>> http://mailman.nginx.org/mailman/listinfo/nginx
>> 
>> _______________________________________________
>> nginx mailing list
>> nginx at nginx.org
>> http://mailman.nginx.org/mailman/listinfo/nginx
> 
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx


From savages at mozapps.com  Sun Sep 25 00:39:32 2011
From: savages at mozapps.com (Shaun savage)
Date: Sun, 25 Sep 2011 08:39:32 +0800
Subject: $http_HEADER
Message-ID: <4E7E7844.7060005@mozapps.com>

Does the $http_HEADER have custom headers?

example
X-Auth-Token=0123456789

does $http_x_auth_token  is it defined and have a value?



From contact at jpluscplusm.com  Sun Sep 25 00:42:59 2011
From: contact at jpluscplusm.com (Jonathan Matthews)
Date: Sun, 25 Sep 2011 01:42:59 +0100
Subject: $http_HEADER
In-Reply-To: <4E7E7844.7060005@mozapps.com>
References: <4E7E7844.7060005@mozapps.com>
Message-ID: <CAKsTx7B_dkGp2D3toqLwtPdQVV90Cyx0uBeH8=ruFfKN2O4iOw@mail.gmail.com>

On 25 September 2011 01:39, Shaun savage <savages at mozapps.com> wrote:
> Does the $http_HEADER have custom headers?
>
> example
> X-Auth-Token=0123456789
>
> does $http_x_auth_token ?is it defined and have a value?

http://wiki.nginx.org/HttpCoreModule#.24http_HEADER

HTH,
Jonathan
-- 
Jonathan Matthews
London, UK
http://www.jpluscplusm.com/contact.html


From savages at savages.com  Sun Sep 25 01:38:31 2011
From: savages at savages.com (Shaun savage)
Date: Sun, 25 Sep 2011 09:38:31 +0800
Subject: $http_HEADER
In-Reply-To: <4E7E7844.7060005@mozapps.com>
References: <4E7E7844.7060005@mozapps.com>
Message-ID: <4E7E8617.2090000@savages.com>

I was using curl and the header was formated wrong, my bad.
It works as expected.

On 09/25/2011 08:39 AM, Shaun savage wrote:
> Does the $http_HEADER have custom headers?
>
> example
> X-Auth-Token=0123456789
>
> does $http_x_auth_token  is it defined and have a value?
>
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx


From nginx-forum at nginx.us  Sun Sep 25 03:14:03 2011
From: nginx-forum at nginx.us (khizar)
Date: Sat, 24 Sep 2011 23:14:03 -0400
Subject: nginx download resume is not working plzzz help...
Message-ID: <96279e852d47e36832439223e565e66a.NginxMailingListEnglish@forum.nginx.org>

hello guys...
hope you all are doing well

download resume for any file is not working

here is nginx configuration
##########################################################
user  www www;

worker_processes 1;

error_log  /home/wwwlogs/nginx_error.log  crit;

pid        /usr/local/nginx/logs/nginx.pid;

#Specifies the value for maximum file descriptors that can be opened by
this process.
worker_rlimit_nofile 51200;

events
	{
		use epoll;
		worker_connections 51200;
	}

http
	{
		include       mime.types;
		default_type  application/octet-stream;

		server_names_hash_bucket_size 32;
		client_header_buffer_size 4k;
		large_client_header_buffers 256 4k;
		client_max_body_size 20m;

		sendfile on;
		tcp_nopush     on;

		keepalive_timeout 45;

		tcp_nodelay on;

		fastcgi_connect_timeout 300;
		fastcgi_send_timeout 300;
		fastcgi_read_timeout 300;
		fastcgi_buffer_size 8k;
		fastcgi_buffers 256 8k;
		fastcgi_busy_buffers_size 128k;
		fastcgi_temp_file_write_size 16m;

		gzip off;
		gzip_min_length  1k;
		gzip_buffers     16 8k;
		gzip_http_version 1.0;
		gzip_comp_level 9;
		gzip_types       text/plain application/x-javascript text/css
application/xml;
		gzip_vary on;

		#limit_zone  crawler  $binary_remote_addr  10m;

server
	{
		listen       80;
		server_name xyz.com;
		index index.html index.htm index.php;
		root  /home/wwwroot/m/;
access_log   off;
		location ~ .*\.(php|php5)?$
			{
				fastcgi_pass  unix:/tmp/php-cgi.sock;
				fastcgi_index index.php;
				include fcgi.conf;
			}

		location /status {
			stub_status on;
			access_log   off;
		}

		location ~ .*\.(gif|jpg|jpeg|png|bmp|swf)$
			{
				expires      30d;
			}

		location ~ .*\.(js|css)?$
			{
				expires      7d;
			}

error_page 404 500 502 504 503 /error.htm;

rewrite ^/menu/all_files/page(.*)\.html$ /index.php?a=all&p=$1 last;
rewrite ^/menu/all_files/$ /index.php?a=all last;
rewrite ^/file-extension/(.*)\.html$ /index.php?a=ft&ft=$1 last;
rewrite ^/file/(.*)/(.*)/warn\.html$
/index.php?a=fid&fid=$2&warn=adult&k=$1 last;


#		log_format  access  '$remote_addr - $remote_user [$time_local]
"$request" '
#             '$status $body_bytes_sent "$http_referer" '
#             '"$http_user_agent" $http_x_forwarded_for';
#		access_log  /home/wwwlogs/access.log  access;

	}
include vhost/*.conf;
}
###############################################################

hey guys please tell me why download resume is not working and how can i
solve this issue...

thank you

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215775,215775#msg-215775


From nginx-forum at nginx.us  Sun Sep 25 05:03:18 2011
From: nginx-forum at nginx.us (ohjyops)
Date: Sun, 25 Sep 2011 01:03:18 -0400
Subject: Help: My rewriting does not work.
Message-ID: <2cfc08de7f692d76366169377e136fee.NginxMailingListEnglish@forum.nginx.org>

I want to rewrite an old page to a new location, I use this code:

rewrite ^/%C8%CB%CC%E5%D1%A8%CE%BB%CD%BC.html /w/??? permanent;

But it does not work. I still can open the old page from
http://mysite.com/%C8%CB%CC%E5%D1%A8%CE%BB%CD%BC.html, not being
redirected to the new location. 

Can you help me correct my rewriting code? Thanks.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215776,215776#msg-215776


From avishai at fewbytes.com  Sat Sep 24 20:36:25 2011
From: avishai at fewbytes.com (Avishai Ish-Shalom)
Date: Sat, 24 Sep 2011 23:36:25 +0300
Subject: process corruption on CentOS 5.6
In-Reply-To: <20110921201810.GA1137@mdounin.ru>
References: <4E7A4313.9090703@mail.huji.ac.il>
	<20110921201810.GA1137@mdounin.ru>
Message-ID: <4E7E3F49.6070707@fewbytes.com>

Thanks! this patch seems to have fixed it.

Regards,
Avishai



On -10/01/37 21:59, Maxim Dounin wrote:
> Hello!
>
> On Wed, Sep 21, 2011 at 11:03:31PM +0300, Avishai Ish-Shalom wrote:
>
>> On CentOS 5.6, nginx 1.1.4 configured to work with php-fpm using fastcgi
>> port. nginx was compiled with: configure --prefix=/usr/local/nginx
>> --with-http_ssl_module --with-http_realip_module
>> --with-http_geoip_module --with-http_flv_module
>> --with-http_gzip_static_module --with-http_stub_status_module
>> --with-http_addition_module --with-http_image_filter_module
>> --with-http_sub_module --with-http_degradation_module
>> --add-module=addons/ngx_upstream_fair --with-cc-opt="-D
>> NGX_UPSTREAM_KEEPALIVE_PATCHED"
> Just a side note: you don't need --with-cc-opt="-D 
> NGX_UPSTREAM_KEEPALIVE_PATCHED"'.
>
>> every few seconds, i get:
>>
>> *** glibc detected *** nginx: worker process: double free or corruption
>> (!prev): 0x000000001d069800 ***
> Could you please provide backtrace?  See here for details:
> http://wiki.nginx.org/Debugging
>
> Alternatively, please try the patch from here:
> http://mailman.nginx.org/pipermail/nginx-devel/2011-September/001233.html
>
> (And provide backtrace if it doesn't help.)
>
> Maxim Dounin
>
>


From agentzh at gmail.com  Sun Sep 25 07:23:36 2011
From: agentzh at gmail.com (agentzh)
Date: Sun, 25 Sep 2011 15:23:36 +0800
Subject: will NGX_AGAIN kill performance?
In-Reply-To: <CAM6b-vko_0rQELdeJH=ED9rxEvJTAhHo0UtHc+eqTzSm-0-D6A@mail.gmail.com>
References: <CAM6b-vko_0rQELdeJH=ED9rxEvJTAhHo0UtHc+eqTzSm-0-D6A@mail.gmail.com>
Message-ID: <CAB4Tn6Me+AXq7eWcXjE3fbGkt1yGAksBpyjz2f-yn0ndTETkdg@mail.gmail.com>

On Sat, Sep 24, 2011 at 10:47 PM, Abioy Sun <abioy.sun at gmail.com> wrote:
> hi,
>
> i currently dev a module which makes multi subrequest sequently in the
> access phase handler, via returning NGX_AGAIN.
>
> my question is, will this kill the performance? and how far could it go in
> killing performance.

NGX_AGAIN is usually for non-blocking I/O operations, it's key for
good performance under high concurrency levels.

Regards,
-agentzh


From nginx-forum at nginx.us  Sun Sep 25 09:13:54 2011
From: nginx-forum at nginx.us (orensol)
Date: Sun, 25 Sep 2011 05:13:54 -0400
Subject: Weird behavior on SSL, and corruption on reload
Message-ID: <6eab22ff27053776e27ee385bb82062d.NginxMailingListEnglish@forum.nginx.org>

Hello,

I had a weird behavior in nginx, in which SSL connections were suddenly
denied. It happened with no apparent reason. A reload of nginx seemed to
have solved it, but on reload, error.log had this stack trace:

*** glibc detected *** nginx: worker process is shutting down: double
free or corruption (out): 0x080ca518 ***
======= Backtrace: =========
/lib/i686/nosegneg/libc.so.6[0xb7d0ccfd]
/lib/i686/nosegneg/libc.so.6(cfree+0x90)[0xb7d103b0]
/lib/libcrypto.so.6(CRYPTO_free+0x3a)[0xb7e3046a]
/lib/libcrypto.so.6(BN_free+0x68)[0xb7e5a1d8]
/lib/libcrypto.so.6(BN_MONT_CTX_free+0x29)[0xb7e63d99]
/lib/libcrypto.so.6[0xb7e683e9]
/lib/libcrypto.so.6(RSA_free+0x61)[0xb7e6afc1]
/lib/libcrypto.so.6[0xb7e8cacd]
/lib/libcrypto.so.6(EVP_PKEY_free+0x67)[0xb7e8cb47]
/lib/libssl.so.6(ssl_cert_free+0xa0)[0xb7f5e5c0]
/lib/libssl.so.6(SSL_CTX_free+0xe3)[0xb7f5bb73]
nginx: worker process is shutting down[0x8066bbb]
nginx: worker process is shutting down[0x804d1d0]
nginx: worker process is shutting down[0x80645fe]
nginx: worker process is shutting down[0x80646eb]
nginx: worker process is shutting down[0x8062ff1]
nginx: worker process is shutting down[0x8064f2c]
nginx: worker process is shutting down[0x804c891]
/lib/i686/nosegneg/libc.so.6(__libc_start_main+0xdc)[0xb7cbbdec]
nginx: worker process is shutting down[0x804b141]
======= Memory map: ========
08048000-080ab000 r-xp 00000000 08:01 298520    
/usr/local/nginx/sbin/nginx
080ab000-080b3000 rw-p 00062000 08:01 298520    
/usr/local/nginx/sbin/nginx
080b3000-0829e000 rw-p 080b3000 00:00 0          [heap]
b4600000-b4621000 rw-p b4600000 00:00 0
b4621000-b4700000 ---p b4621000 00:00 0
b479b000-b47a6000 r-xp 00000000 08:01 262401    
/lib/libgcc_s-4.1.1-20070105.so.1
b47a6000-b47a7000 rw-p 0000a000 08:01 262401    
/lib/libgcc_s-4.1.1-20070105.so.1
b47a7000-b47ab000 r-xp 00000000 08:01 262411     /lib/libnss_dns-2.5.so
b47ab000-b47ac000 r--p 00003000 08:01 262411     /lib/libnss_dns-2.5.so
b47ac000-b47ad000 rw-p 00004000 08:01 262411     /lib/libnss_dns-2.5.so
b47b3000-b49a0000 rw-p b47b3000 00:00 0
b49a0000-b49a1000 rw-s 00000000 00:08 1620412881  /dev/zero (deleted)
b49a1000-b7ba1000 rw-s 00000000 00:08 1620412878  /dev/zero (deleted)
b7ba1000-b7baa000 r-xp 00000000 08:01 262412    
/lib/libnss_files-2.5.so
b7baa000-b7bab000 r--p 00008000 08:01 262412    
/lib/libnss_files-2.5.so
b7bab000-b7bac000 rw-p 00009000 08:01 262412    
/lib/libnss_files-2.5.so
b7bac000-b7bae000 rw-p b7bac000 00:00 0 
b7bae000-b7bb5000 r-xp 00000000 08:01 65580     
/usr/lib/libkrb5support.so.0.1
b7bb5000-b7bb6000 rw-p 00006000 08:01 65580     
/usr/lib/libkrb5support.so.0.1
b7bb6000-b7bc5000 r-xp 00000000 08:01 262423     /lib/libresolv-2.5.so
b7bc5000-b7bc6000 r--p 0000e000 08:01 262423     /lib/libresolv-2.5.so
b7bc6000-b7bc7000 rw-p 0000f000 08:01 262423     /lib/libresolv-2.5.so
b7bc7000-b7bc9000 rw-p b7bc7000 00:00 0 
b7bc9000-b7bee000 r-xp 00000000 08:01 65574     
/usr/lib/libk5crypto.so.3.0
b7bee000-b7bef000 rw-p 00025000 08:01 65574     
/usr/lib/libk5crypto.so.3.0
b7bef000-b7bf1000 r-xp 00000000 08:01 262387     /lib/libcom_err.so.2.1
b7bf1000-b7bf2000 rw-p 00001000 08:01 262387     /lib/libcom_err.so.2.1
b7bf2000-b7bf3000 rw-p b7bf2000 00:00 0 
b7bf3000-b7c79000 r-xp 00000000 08:01 65579     
/usr/lib/libkrb5.so.3.2
b7c79000-b7c7b000 rw-p 00086000 08:01 65579     
/usr/lib/libkrb5.so.3.2
b7c7b000-b7ca5000 r-xp 00000000 08:01 65561     
/usr/lib/libgssapi_krb5.so.2.2
b7ca5000-b7ca6000 rw-p 00029000 08:01 65561     
/usr/lib/libgssapi_krb5.so.2.2
b7ca6000-b7de1000 r-xp 00000000 08:01 262438    
/lib/i686/nosegneg/libc-2.5.so
b7de1000-b7de3000 r--p 0013a000 08:01 262438    
/lib/i686/nosegneg/libc-2.5.so
b7de3000-b7de4000 rw-p 0013c000 08:01 262438    
/lib/i686/nosegneg/libc-2.5.so
b7de4000-b7de7000 rw-p b7de4000 00:00 0
b7de7000-b7df9000 r-xp 00000000 08:01 65693      /usr/lib/libz.so.1.2.3
b7df9000-b7dfa000 rw-p 00011000 08:01 65693      /usr/lib/libz.so.1.2.3
b7dfa000-b7dfc000 r-xp 00000000 08:01 262397     /lib/libdl-2.5.so
b7dfc000-b7dfd000 r--p 00001000 08:01 262397     /lib/libdl-2.5.so
b7dfd000-b7dfe000 rw-p 00002000 08:01 262397     /lib/libdl-2.5.so
b7dfe000-b7f1b000 r-xp 00000000 08:01 262389    
/lib/libcrypto.so.0.9.8b
b7f1b000-b7f2e000 rw-p 0011c000 08:01 262389    
/lib/libcrypto.so.0.9.8b
b7f2e000-b7f32000 rw-p b7f2e000 00:00 0
b7f32000-b7f73000 r-xp 00000000 08:01 262428     /lib/libssl.so.0.9.8b
b7f73000-b7f77000 rw-p 00040000 08:01 262428     /lib/libssl.so.0.9.8b
b7f77000-b7f93000 r-xp 00000000 08:01 262420     /lib/libpcre.so.0.0.1
b7f93000-b7f94000 rw-p 0001b000 08:01 262420     /lib/libpcre.so.0.0.1
b7f94000-b7f99000 r-xp 00000000 08:01 262388     /lib/libcrypt-2.5.so
b7f99000-b7f9a000 r--p 00004000 08:01 262388     /lib/libcrypt-2.5.so
b7f9a000-b7f9b000 rw-p 00005000 08:01 262388     /lib/libcrypt-2.5.so
b7f9b000-b7fc2000 rw-p b7f9b000 00:00 0
b7fc8000-b7fc9000 rw-p b7fc8000 00:00 0
b7fc9000-b7fca000 r-xp b7fc9000 00:00 0          [vdso]
b7fca000-b7fe3000 r-xp 00000000 08:01 262375     /lib/ld-2.5.so
b7fe3000-b7fe4000 r--p 00018000 08:01 262375     /lib/ld-2.5.so
b7fe4000-b7fe5000 rw-p 00019000 08:01 262375     /lib/ld-2.5.so
bfdae000-bfdd2000 rw-p bfdae000 00:00 0          [stack]
2011/09/25 05:01:43 [alert] 21233#0: worker process 1870 exited on
signal 6


Any ideas on what happened, and what can be done to prevent it in the
future?

Thanks,
Oren

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215785,215785#msg-215785


From nginx-forum at nginx.us  Sun Sep 25 09:50:27 2011
From: nginx-forum at nginx.us (Ensiferous)
Date: Sun, 25 Sep 2011 05:50:27 -0400
Subject: nginx download resume is not working plzzz help...
In-Reply-To: <96279e852d47e36832439223e565e66a.NginxMailingListEnglish@forum.nginx.org>
References: <96279e852d47e36832439223e565e66a.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <ab1c283976fe0a66153e4072f076c8fa.NginxMailingListEnglish@forum.nginx.org>

If you are having PHP read the file directly then Nginx is not
responsible for the file sending and as such your PHP application needs
to implementing range requests.

Ideally, you should be using x-accel-redirect to have Nginx handle the
file sending once PHP is done.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215775,215789#msg-215789


From mdounin at mdounin.ru  Sun Sep 25 11:01:13 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Sun, 25 Sep 2011 15:01:13 +0400
Subject: Weird behavior on SSL, and corruption on reload
In-Reply-To: <6eab22ff27053776e27ee385bb82062d.NginxMailingListEnglish@forum.nginx.org>
References: <6eab22ff27053776e27ee385bb82062d.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20110925110112.GD1137@mdounin.ru>

Hello!

On Sun, Sep 25, 2011 at 05:13:54AM -0400, orensol wrote:

> Hello,
> 
> I had a weird behavior in nginx, in which SSL connections were suddenly
> denied. It happened with no apparent reason. A reload of nginx seemed to
> have solved it, but on reload, error.log had this stack trace:
> 
> *** glibc detected *** nginx: worker process is shutting down: double
> free or corruption (out): 0x080ca518 ***
> ======= Backtrace: =========
> /lib/i686/nosegneg/libc.so.6[0xb7d0ccfd]
> /lib/i686/nosegneg/libc.so.6(cfree+0x90)[0xb7d103b0]
> /lib/libcrypto.so.6(CRYPTO_free+0x3a)[0xb7e3046a]
> /lib/libcrypto.so.6(BN_free+0x68)[0xb7e5a1d8]
> /lib/libcrypto.so.6(BN_MONT_CTX_free+0x29)[0xb7e63d99]
> /lib/libcrypto.so.6[0xb7e683e9]
> /lib/libcrypto.so.6(RSA_free+0x61)[0xb7e6afc1]
> /lib/libcrypto.so.6[0xb7e8cacd]
> /lib/libcrypto.so.6(EVP_PKEY_free+0x67)[0xb7e8cb47]
> /lib/libssl.so.6(ssl_cert_free+0xa0)[0xb7f5e5c0]
> /lib/libssl.so.6(SSL_CTX_free+0xe3)[0xb7f5bb73]
> nginx: worker process is shutting down[0x8066bbb]
> nginx: worker process is shutting down[0x804d1d0]
> nginx: worker process is shutting down[0x80645fe]
> nginx: worker process is shutting down[0x80646eb]
> nginx: worker process is shutting down[0x8062ff1]
> nginx: worker process is shutting down[0x8064f2c]
> nginx: worker process is shutting down[0x804c891]
> /lib/i686/nosegneg/libc.so.6(__libc_start_main+0xdc)[0xb7cbbdec]
> nginx: worker process is shutting down[0x804b141]

[...]

> 2011/09/25 05:01:43 [alert] 21233#0: worker process 1870 exited on
> signal 6
> 
> 
> Any ideas on what happened, and what can be done to prevent it in the
> future?

Are you able to reproduce the problem?  If yes, please follow 
instructions here:

http://wiki.nginx.org/Debugging

At least proper backtrace is needed to debug this further.  And 
you may need to make sure your nginx binary isn't stripped.

If not, please at least provide "nginx -V" output, config and 
OpenSSL version details (output of "openssl version -a").

Maxim Dounin


From nginx-forum at nginx.us  Sun Sep 25 11:14:40 2011
From: nginx-forum at nginx.us (orensol)
Date: Sun, 25 Sep 2011 07:14:40 -0400
Subject: Weird behavior on SSL, and corruption on reload
In-Reply-To: <6eab22ff27053776e27ee385bb82062d.NginxMailingListEnglish@forum.nginx.org>
References: <6eab22ff27053776e27ee385bb82062d.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <0df21e21f71559a9b9d85dba3a785ef2.NginxMailingListEnglish@forum.nginx.org>

Hello Maxim,

For now I can't reproduce the problem, if it happens again i'll try to
catch a real backtrace.

Here are the other details. Thanks!



nginx:
nginx version: nginx/0.8.35
built by gcc 4.1.1 20070105 (Red Hat 4.1.1-52)
TLS SNI support disabled
configure arguments: --with-http_ssl_module
--add-module=/root/ngx_cache_purge-1.0



openssl:
OpenSSL 0.9.8b 04 May 2006
built on: Wed Oct 17 18:15:17 EDT 2007
platform: linux-elf
options:  bn(64,32) md2(int) rc4(idx,int) des(ptr,risc1,16,long)
blowfish(idx) 
compiler: gcc -fPIC -DOPENSSL_PIC -DZLIB -DOPENSSL_THREADS -D_REENTRANT
-DDSO_DLFCN -DHAVE_DLFCN_H -DKRB5_MIT -I/usr/kerberos/include -DL_ENDIAN
-DTERMIO -Wall -O2 -g -pipe -Wall -Wp,-D_FORTIFY_SOURCE=2 -fexceptions
-fstack-protector --param=ssp-buffer-size=4 -m32 -march=i686
-mtune=generic -fasynchronous-unwind-tables -Wa,--noexecstack
-DOPENSSL_BN_ASM_PART_WORDS -DOPENSSL_IA32_SSE2 -DSHA1_ASM -DMD5_ASM
-DRMD160_ASM -DAES_ASM
OPENSSLDIR: "/etc/pki/tls"
engines:  dynamic padlock



nginx conf:

user              nginx;
worker_processes  1;
worker_rlimit_nofile 10240;
error_log         /var/log/nginx/error.log warn;
pid               /var/run/nginx.pid;

events {
    worker_connections  10240;
}
 
http {
    include       /usr/local/nginx/conf/mime.types;
    default_type  application/octet-stream;

    log_format  main  '$remote_addr - $remote_user [$time_local]
$request '
                      '"$status" $body_bytes_sent "$http_referer" '
                      '"$http_user_agent" "$http_x_forwarded_for"';

    access_log  /var/log/nginx/access.log  main;

    log_format cache '***$time_local '
                     '$upstream_cache_status '
                     'Cache-Control: $upstream_http_cache_control '
                     'Expires: $upstream_http_expires '
                     '"$host" "$request" ($status) '
                     '"$http_user_agent" ';
    access_log  /var/log/nginx/cache.log cache;

    sendfile        on;

    keepalive_timeout  65;

    gzip  on;
    gzip_types  text/css text/javascript application/x-javascript;


        proxy_cache_path /mnt/nginx/cache levels=1:2
keys_zone=myzone:50m inactive=2h max_size=3000m;
        proxy_temp_path /mnt/nginx/temp;


	upstream  backend_ssl  {
		        server 10.255.10.255:443;
	}

	server {
	    listen       443;
	    server_name  _;

	    ssl                  on;
	    ssl_certificate      /usr/local/nginx/conf/cert;
	    ssl_certificate_key  /usr/local/nginx/conf/key;

	    ssl_session_timeout  5m;

	    ssl_protocols  SSLv2 SSLv3 TLSv1;
	    ssl_ciphers 
ALL:!ADH:!EXPORT56:RC4+RSA:+HIGH:+MEDIUM:+LOW:+SSLv2:+EXP;
	    ssl_prefer_server_ciphers   on;

		location / {

		        proxy_pass http://backend_ssl;
		        proxy_set_header Host $host;
		        proxy_set_header X-Real-IP $remote_addr;
		        proxy_set_header X-Forwarded-For $remote_addr;
		}

		# static files served locally
		location ~* \.(css|js)$ {
		    root   /home/static;
		}

		# static images served locally and get Expires header
		location ~* \.(jpg|jpeg|gif|png|ico|bmp)$ {
		    root   /home/static;
		    expires 30d;
		}
	}


}

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215785,215795#msg-215795


From mdounin at mdounin.ru  Sun Sep 25 12:13:26 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Sun, 25 Sep 2011 16:13:26 +0400
Subject: Weird behavior on SSL, and corruption on reload
In-Reply-To: <0df21e21f71559a9b9d85dba3a785ef2.NginxMailingListEnglish@forum.nginx.org>
References: <6eab22ff27053776e27ee385bb82062d.NginxMailingListEnglish@forum.nginx.org>
	<0df21e21f71559a9b9d85dba3a785ef2.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20110925121326.GE1137@mdounin.ru>

Hello!

On Sun, Sep 25, 2011 at 07:14:40AM -0400, orensol wrote:

> Hello Maxim,
> 
> For now I can't reproduce the problem, if it happens again i'll try to
> catch a real backtrace.
> 
> Here are the other details. Thanks!
> 
> 
> 
> nginx:
> nginx version: nginx/0.8.35
> built by gcc 4.1.1 20070105 (Red Hat 4.1.1-52)
> TLS SNI support disabled
> configure arguments: --with-http_ssl_module
> --add-module=/root/ngx_cache_purge-1.0
> 
> 
> 
> openssl:
> OpenSSL 0.9.8b 04 May 2006
> built on: Wed Oct 17 18:15:17 EDT 2007
> platform: linux-elf
> options:  bn(64,32) md2(int) rc4(idx,int) des(ptr,risc1,16,long)
> blowfish(idx) 
> compiler: gcc -fPIC -DOPENSSL_PIC -DZLIB -DOPENSSL_THREADS -D_REENTRANT
> -DDSO_DLFCN -DHAVE_DLFCN_H -DKRB5_MIT -I/usr/kerberos/include -DL_ENDIAN
> -DTERMIO -Wall -O2 -g -pipe -Wall -Wp,-D_FORTIFY_SOURCE=2 -fexceptions
> -fstack-protector --param=ssp-buffer-size=4 -m32 -march=i686
> -mtune=generic -fasynchronous-unwind-tables -Wa,--noexecstack
> -DOPENSSL_BN_ASM_PART_WORDS -DOPENSSL_IA32_SSE2 -DSHA1_ASM -DMD5_ASM
> -DRMD160_ASM -DAES_ASM
> OPENSSLDIR: "/etc/pki/tls"
> engines:  dynamic padlock

Both nginx 0.8.35 and openssl 0.9.8b are rather old and have known 
memory corruption issues.  You may want to upgrade before doing 
anything else.

Maxim Dounin


From nginx-forum at nginx.us  Sun Sep 25 13:02:37 2011
From: nginx-forum at nginx.us (ahu)
Date: Sun, 25 Sep 2011 09:02:37 -0400
Subject: Trouble getting UploadProgress module to work
In-Reply-To: <4BF8482A.1050208@daysofwonder.com>
References: <4BF8482A.1050208@daysofwonder.com>
Message-ID: <0352c082063ce9c35bf75e551259733e.NginxMailingListEnglish@forum.nginx.org>

I got the uploadprogress module worked yestoday and I have not change
any configuration today,while I find it out of working then.
Who can help me? Here are some debug logs I find:
-----------------------------------------------------------------
2011/09/25 21:04:53 [debug] 9445#0: *10 test location: "/progress"
2011/09/25 21:04:53 [debug] 9445#0: *10 using configuration "/progress"
2011/09/25 21:04:53 [debug] 9445#0: *10 http cl:-1 max:2097152000
2011/09/25 21:04:53 [debug] 9445#0: *10 rewrite phase: 2
2011/09/25 21:04:53 [debug] 9445#0: *10 rewrite phase: 3
2011/09/25 21:04:53 [debug] 9445#0: *10 post rewrite phase: 4
2011/09/25 21:04:53 [debug] 9445#0: *10 generic phase: 5
2011/09/25 21:04:53 [debug] 9445#0: *10 generic phase: 6
2011/09/25 21:04:53 [debug] 9445#0: *10 access phase: 7
2011/09/25 21:04:53 [debug] 9445#0: *10 access phase: 8
2011/09/25 21:04:53 [debug] 9445#0: *10 post access phase: 9
2011/09/25 21:04:53 [debug] 9445#0: *10 http set discard body
2011/09/25 21:04:53 [debug] 9445#0: *10 upload-progress:
get_tracking_id
2011/09/25 21:04:53 [debug] 9445#0: *10 malloc: 0000000007D1D2C0:16
2011/09/25 21:04:53 [debug] 9445#0: *10 upload-progress: get_tracking_id
found header: d3182603d117b2a2505d00183485feb5
2011/09/25 21:04:53 [debug] 9445#0: *10 reportuploads handler found id:
d3182603d117b2a2505d00183485feb5
2011/09/25 21:04:53 [debug] 9445#0: shmtx lock
2011/09/25 21:04:53 [debug] 9445#0: *10 upload-progress: find_node
d3182603d117b2a2505d00183485feb5
2011/09/25 21:04:53 [debug] 9445#0: *10 upload-progress: can't find
node
2011/09/25 21:04:53 [debug] 9445#0: *10 reportuploads not found:
d3182603d117b2a2505d00183485feb5
2011/09/25 21:04:53 [debug] 9445#0: shmtx unlock
2011/09/25 21:04:53 [debug] 9445#0: *10 http script copy: "new Object({
'state' : 'starting' })

"
2011/09/25 21:04:53 [debug] 9445#0: *10 upload progress: state=0,
err_status=0, remaining=0, length=0
2011/09/25 21:04:53 [debug] 9445#0: *10 uploadprogress error-tracker
error: 0
2011/09/25 21:04:53 [debug] 9445#0: *10 HTTP/1.1 200 OK

Server: nginx/1.1.4

Date: Sun, 25 Sep 2011 13:04:53 GMT

Content-Type: text/javascript

Content-Length: 38

Connection: keep-alive

Expires: Thu, 01 Jan 1970 00:00:01 GMT

Cache-Control: no-cache


2011/09/25 21:04:53 [debug] 9445#0: *10 write new buf t:1 f:0
0000000007D8C178, pos 0000000007D8C178, size: 217 file: 0, size: 0
2011/09/25 21:04:53 [debug] 9445#0: *10 http write filter: l:0 f:0
s:217
2011/09/25 21:04:53 [debug] 9445#0: *10 http output filter "/progress?"
2011/09/25 21:04:53 [debug] 9445#0: *10 http copy filter: "/progress?"
2011/09/25 21:04:53 [debug] 9445#0: *10 http postpone filter
"/progress?" 00007FFFB0EF7DB0
2011/09/25 21:04:53 [debug] 9445#0: *10 write old buf t:1 f:0
0000000007D8C178, pos 0000000007D8C178, size: 217 file: 0, size: 0
2011/09/25 21:04:53 [debug] 9445#0: *10 write new buf t:1 f:0
0000000007D8C0B0, pos 0000000007D8C0B0, size: 38 file: 0, size: 0
2011/09/25 21:04:53 [debug] 9445#0: *10 http write filter: l:1 f:0
s:255
2011/09/25 21:04:53 [debug] 9445#0: *10 http write filter limit 0
2011/09/25 21:04:53 [debug] 9445#0: *10 writev: 255
2011/09/25 21:04:53 [debug] 9445#0: *10 http write filter
0000000000000000
2011/09/25 21:04:53 [debug] 9445#0: *10 http copy filter: 0
"/progress?"
2011/09/25 21:04:53 [debug] 9445#0: *10 http finalize request: 0,
"/progress?" a:1, c:1
2011/09/25 21:04:53 [debug] 9445#0: *10 set http keepalive handler
2011/09/25 21:04:53 [debug] 9445#0: *10 http close request
2011/09/25 21:04:53 [debug] 9445#0: *10 http log handler
2011/09/25 21:04:53 [debug] 9445#0: *10 free: 0000000007D8B400, unused:
42
2011/09/25 21:04:53 [debug] 9445#0: *10 event timer add: 14:
60000:1316955953219
2011/09/25 21:04:53 [debug] 9445#0: *10 free: 0000000007D2A600
2011/09/25 21:04:53 [debug] 9445#0: *10 free: 0000000007D803C0
2011/09/25 21:04:53 [debug] 9445#0: *10 hc free: 0000000000000000 0
2011/09/25 21:04:53 [debug] 9445#0: *10 hc busy: 0000000000000000 0
2011/09/25 21:04:53 [debug] 9445#0: *10 reusable connection: 1
2011/09/25 21:04:53 [debug] 9445#0: *10 post event 00002B2063134148
2011/09/25 21:04:53 [debug] 9445#0: timer delta: 495
2011/09/25 21:04:53 [debug] 9445#0: posted events 00002B2063134148
2011/09/25 21:04:53 [debug] 9445#0: posted event 00002B2063134148
2011/09/25 21:04:53 [debug] 9445#0: *10 delete posted event
00002B2063134148
2011/09/25 21:04:53 [debug] 9445#0: *10 http keepalive handler
2011/09/25 21:04:53 [debug] 9445#0: *10 malloc: 0000000007D803C0:32768
2011/09/25 21:04:53 [debug] 9445#0: *10 recv: fd:14 -1 of 32768
2011/09/25 21:04:53 [debug] 9445#0: *10 recv() not ready (11: Resource
temporarily unavailable)
2011/09/25 21:04:53 [debug] 9445#0: posted event 0000000000000000
2011/09/25 21:04:53 [debug] 9445#0: worker cycle
2011/09/25 21:04:53 [debug] 9445#0: accept mutex lock failed: 0
2011/09/25 21:04:53 [debug] 9445#0: epoll timer: 500
2011/09/25 21:04:53 [debug] 9448#0: timer delta: 500
2011/09/25 21:04:53 [debug] 9448#0: posted events 0000000000000000
2011/09/25 21:04:53 [debug] 9448#0: worker cycle
2011/09/25 21:04:53 [debug] 9448#0: accept mutex lock failed: 0
2011/09/25 21:04:53 [debug] 9448#0: epoll timer: 500
2011/09/25 21:04:53 [debug] 9449#0: timer delta: 500
2011/09/25 21:04:53 [debug] 9449#0: posted events 0000000000000000
2011/09/25 21:04:53 [debug] 9449#0: worker cycle
2011/09/25 21:04:53 [debug] 9449#0: accept mutex lock failed: 0
2011/09/25 21:04:53 [debug] 9449#0: epoll timer: 500
2011/09/25 21:04:53 [debug] 9445#0: timer delta: 510
2011/09/25 21:04:53 [debug] 9445#0: posted events 0000000000000000
2011/09/25 21:04:53 [debug] 9445#0: worker cycle
2011/09/25 21:04:53 [debug] 9445#0: accept mutex lock failed: 0
2011/09/25 21:04:53 [debug] 9445#0: epoll timer: 500
2011/09/25 21:04:53 [debug] 9448#0: timer delta: 500
2011/09/25 21:04:53 [debug] 9448#0: posted events 0000000000000000
2011/09/25 21:04:53 [debug] 9448#0: worker cycle
2011/09/25 21:04:53 [debug] 9448#0: accept mutex lock failed: 0
2011/09/25 21:04:53 [debug] 9448#0: epoll timer: 500
2011/09/25 21:04:54 [debug] 9449#0: timer delta: 500
2011/09/25 21:04:54 [debug] 9449#0: posted events 0000000000000000
2011/09/25 21:04:54 [debug] 9449#0: worker cycle
2011/09/25 21:04:54 [debug] 9449#0: accept mutex lock failed: 0
2011/09/25 21:04:54 [debug] 9449#0: epoll timer: 500
2011/09/25 21:04:54 [debug] 9445#0: epoll: fd:14 ev:0001
d:00002B20627D3250

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,88713,215798#msg-215798


From hagaia at qwilt.com  Sun Sep 25 14:42:11 2011
From: hagaia at qwilt.com (Hagai Avrahami)
Date: Sun, 25 Sep 2011 16:42:11 +0200
Subject: FLV
Message-ID: <CAN-ZXpvz=wQ9h+JiEgX9zrmify2kkszN5LKYG0iFPAAu2OvS9w@mail.gmail.com>

Hi

My question is about FLV Module

NGINX documentation mentions that FLV module supports time-based offsets*
"This module provides the ability to seek within FLV (Flash) files using
time-based offsets."*
http://wiki.nginx.org/HttpFlvStreamModule

When looking on the source code and some testing I could see that its
supports byte-based offset

Can you help me fill the gap
Thanks a lot
*Hagai Avrahami*

Qwilt | Work: +972-72-2221644| Mobile: +972-54-4895656 |
hagaia at qwilt.com<yoav at qwilt.com>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110925/ccb57695/attachment.html>

From luky-37 at hotmail.com  Sun Sep 25 15:16:46 2011
From: luky-37 at hotmail.com (Lukas Tribus)
Date: Sun, 25 Sep 2011 17:16:46 +0200
Subject: FLV
In-Reply-To: <CAN-ZXpvz=wQ9h+JiEgX9zrmify2kkszN5LKYG0iFPAAu2OvS9w@mail.gmail.com>
References: <CAN-ZXpvz=wQ9h+JiEgX9zrmify2kkszN5LKYG0iFPAAu2OvS9w@mail.gmail.com>
Message-ID: <DUB107-W294EF013659595FFA71513EDF20@phx.gbl>


FLV pseudostreaming is always in bytes, not seconds (like MP4).



Please see http://www.longtailvideo.com/support/jw-player/jw-player-for-flash-v5/12534/video-delivery-http-pseudo-streaming#mechanism





Lukas



From: hagaia at qwilt.com
Date: Sun, 25 Sep 2011 16:42:11 +0200
Subject: FLV
To: nginx at nginx.org

Hi

My question is about FLV Module
 
NGINX documentation mentions that FLV module supports time-based offsets
"This module provides the ability to seek within FLV (Flash) files using time-based offsets."


http://wiki.nginx.org/HttpFlvStreamModule

When looking on the source code and some testing I could see that its supports byte-based offset

Can you help me fill the gap


Thanks a lot
Hagai Avrahami

Qwilt | Work: +972-72-2221644| Mobile: +972-54-4895656 | hagaia at qwilt.com



_______________________________________________
nginx mailing list
nginx at nginx.org
http://mailman.nginx.org/mailman/listinfo/nginx 		 	   		  
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110925/6600ca74/attachment.html>

From luc.juggery at gmail.com  Sun Sep 25 16:06:19 2011
From: luc.juggery at gmail.com (Luc Juggery)
Date: Sun, 25 Sep 2011 18:06:19 +0200
Subject: Cannot remove Content Length header
In-Reply-To: <B9C24323-96B9-4162-A730-C76D457F7AEE@gmail.com>
References: <CAEVF9DX4MNPp5Vh7opoX7-Ub4e7CP564jtXNwRoB_3YidTU-wg@mail.gmail.com>
	<20110924120530.GA1137@mdounin.ru>
	<B0CB97C9-FE33-4F41-BAC2-E52CD5CD21C3@gmail.com>
	<20110924145447.GB1137@mdounin.ru>
	<B9C24323-96B9-4162-A730-C76D457F7AEE@gmail.com>
Message-ID: <CAEVF9DUF_Rht7SPZYHEgYJKya9mKQFsgGyD8GkZEjevEtcC4sw@mail.gmail.com>

Thanks, I corrected this on back-end side (adding a content-length).
Luc

2011/9/24 Luc Juggery <luc.juggery at gmail.com>

> Does it mean something (like a parameter or something) needs to be modified
> on my node.js application ? Do I need to send additionnal stuff ? I did not
> really get the "chunked" part.
>
> --
> Luc
>
> Le Sep 24, 2011 ? 16:54, Maxim Dounin <mdounin at mdounin.ru> a ?crit :
>
> > Hello!
> >
> > On Sat, Sep 24, 2011 at 02:58:59PM +0200, Luc Juggery wrote:
> >
> >> Hello,
> >>
> >> thanks for your answers.
> >> Maxim, what would you recommend to fix the backend ? I do not really see
> where could be the violation in fact.
> >
> > Backend MUST NOT (as per RFC 2616) return chunked to HTTP/1.0
> > clients, while nginx uses HTTP/1.0 to talk to backends.
> >
> > See RFC 2616, "3.6 Transfer Codings"
> > http://tools.ietf.org/html/rfc2616#section-3.6
> >
> >   ... A server MUST NOT send transfer-codings to an HTTP/1.0
> >   client.
> >
> > Maxim Dounin
> >
> >>
> >> --
> >> Luc
> >>
> >> Le Sep 24, 2011 ? 14:05, Maxim Dounin <mdounin at mdounin.ru> a ?crit :
> >>
> >>> Hello!
> >>>
> >>> On Sat, Sep 24, 2011 at 01:43:12PM +0200, Luc Juggery wrote:
> >>>
> >>>> Hello,
> >>>>
> >>>> I'm using Nginx as a front end to my node.js application, nginx
> proxying
> >>>> https request to my app.
> >>>> When the browser gets the response back, instead of pure json
> >>>> ({"error":"query error","message":"Parameter(s) missing:
> user,password"}), I
> >>>> receive something like:
> >>>>
> >>>> 47
> >>>> {"error":"query error","message":"Parameter(s) missing:
> user,password"}
> >>>> 0
> >>>
> >>> This is chunked encoding, likely sent by your backend in violation
> >>> of HTTP specification.
> >>>
> >>> Right now there are two options available:
> >>>
> >>> 1. Fix your backend.
> >>>
> >>> 2. Upgrade to nginx 1.1.4, it has code needed to talk to such
> >>> broken backends.
> >>>
> >>> Maxim Dounin
> >>>
> >>> _______________________________________________
> >>> nginx mailing list
> >>> nginx at nginx.org
> >>> http://mailman.nginx.org/mailman/listinfo/nginx
> >>
> >> _______________________________________________
> >> nginx mailing list
> >> nginx at nginx.org
> >> http://mailman.nginx.org/mailman/listinfo/nginx
> >
> > _______________________________________________
> > nginx mailing list
> > nginx at nginx.org
> > http://mailman.nginx.org/mailman/listinfo/nginx
>



-- 
Luc
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110925/b23c7744/attachment.html>

From nginx-forum at nginx.us  Sun Sep 25 16:45:20 2011
From: nginx-forum at nginx.us (tsaavik)
Date: Sun, 25 Sep 2011 12:45:20 -0400
Subject: How to tune nginx for 800-1200 concurrent connections?
In-Reply-To: <ef98ae9bdd90878e490c36035da42860.NginxMailingListEnglish@forum.nginx.org>
References: <ef98ae9bdd90878e490c36035da42860.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <e0caee1576cdf5a6d612f11b88d33cf5.NginxMailingListEnglish@forum.nginx.org>

Lots of directions to go with this question, here are 2.

1. The IO bound questions: 
   Your client_max_body_size has me wondering how big these images
actually are. Whats the average size of these images?

   A few lines of output from "dstat" or "vmstat 1" during peak output
would be useful (or even "sar" if you run it).

   If you are IO bound like I suspect, a reverse proxy via nginx's
proxy_cache might be beneficial. You could just toss some 'hot' files
into a ramdrive   (/dev/shm) and symlink them (ln -s) to test.


2. Your PHP app is inefficent:
I run a php gallery (gallery2). Its pretty damn slow/cpu hungry. In
order to speed things up make sure your php has something like
http://eaccelerator.net running. It increased my performance more then
moving from apace1 (yes 1) to nginx.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215383,215807#msg-215807


From mdounin at mdounin.ru  Sun Sep 25 18:33:52 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Sun, 25 Sep 2011 22:33:52 +0400
Subject: FLV
In-Reply-To: <CAN-ZXpvz=wQ9h+JiEgX9zrmify2kkszN5LKYG0iFPAAu2OvS9w@mail.gmail.com>
References: <CAN-ZXpvz=wQ9h+JiEgX9zrmify2kkszN5LKYG0iFPAAu2OvS9w@mail.gmail.com>
Message-ID: <20110925183352.GG1137@mdounin.ru>

Hello!

On Sun, Sep 25, 2011 at 04:42:11PM +0200, Hagai Avrahami wrote:

> Hi
> 
> My question is about FLV Module
> 
> NGINX documentation mentions that FLV module supports time-based offsets*
> "This module provides the ability to seek within FLV (Flash) files using
> time-based offsets."*
> http://wiki.nginx.org/HttpFlvStreamModule

Looks like wiki is wrong, feel free to fix it.  Original 
documentation (in Russian) may be found here:

http://sysoev.ru/nginx/docs/http/ngx_http_flv_module.html

> When looking on the source code and some testing I could see that its
> supports byte-based offset

Yes, for flv offset is in bytes.

Maxim Dounin


From dennisml at conversis.de  Sun Sep 25 19:20:30 2011
From: dennisml at conversis.de (Dennis Jacobfeuerborn)
Date: Sun, 25 Sep 2011 21:20:30 +0200
Subject: How to tune nginx for 800-1200 concurrent connections?
In-Reply-To: <e0caee1576cdf5a6d612f11b88d33cf5.NginxMailingListEnglish@forum.nginx.org>
References: <ef98ae9bdd90878e490c36035da42860.NginxMailingListEnglish@forum.nginx.org>
	<e0caee1576cdf5a6d612f11b88d33cf5.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <4E7F7EFE.80704@conversis.de>

On 09/25/2011 06:45 PM, tsaavik wrote:
> Lots of directions to go with this question, here are 2.
>
> 1. The IO bound questions:
>     Your client_max_body_size has me wondering how big these images
> actually are. Whats the average size of these images?
>
>     A few lines of output from "dstat" or "vmstat 1" during peak output
> would be useful (or even "sar" if you run it).
>
>     If you are IO bound like I suspect, a reverse proxy via nginx's
> proxy_cache might be beneficial. You could just toss some 'hot' files
> into a ramdrive   (/dev/shm) and symlink them (ln -s) to test.

The page cache will do something like this transparently. If there's not 
enough ram left for the page cache to keep the files in memory then you 
also don't have enough ram to add a ram drive.

Regards,
   Dennis


From nginx-forum at nginx.us  Sun Sep 25 21:50:58 2011
From: nginx-forum at nginx.us (ptimo3)
Date: Sun, 25 Sep 2011 17:50:58 -0400
Subject: nginx redirect loop under ssl
In-Reply-To: <AANLkTi=4gTAW+kUO_DAVmQnt0+sSDYnKOrdp27zR7F8W@mail.gmail.com>
References: <AANLkTi=4gTAW+kUO_DAVmQnt0+sSDYnKOrdp27zR7F8W@mail.gmail.com>
Message-ID: <128fac09f98f0bff96508a5f0faa8a24.NginxMailingListEnglish@forum.nginx.org>

similar to this ? http://forum.nginx.org/read.php?9,200930

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,165266,215812#msg-215812


From magicbearmo at gmail.com  Mon Sep 26 00:16:03 2011
From: magicbearmo at gmail.com (MagicBear)
Date: Mon, 26 Sep 2011 08:16:03 +0800
Subject: Help: My rewriting does not work.
In-Reply-To: <2cfc08de7f692d76366169377e136fee.NginxMailingListEnglish@forum.nginx.org>
References: <2cfc08de7f692d76366169377e136fee.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <CAENo=s0at_W6YC5tuwzGfbO=86Tg3ywJpG+8fUZ6par5zq6CjA@mail.gmail.com>

you may need to using ^/?????.html /w/??? permanent;

note: You need save the file at ASCII mode for match GBK charset. But
I am not share may it work for this.

2011/9/25 ohjyops <nginx-forum at nginx.us>:
> I want to rewrite an old page to a new location, I use this code:
>
> rewrite ^/%C8%CB%CC%E5%D1%A8%CE%BB%CD%BC.html /w/??? permanent;
>
> But it does not work. I still can open the old page from
> http://mysite.com/%C8%CB%CC%E5%D1%A8%CE%BB%CD%BC.html, not being
> redirected to the new location.
>
> Can you help me correct my rewriting code? Thanks.
>
> Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215776,215776#msg-215776
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx



-- 
MagicBear


From abioy.sun at gmail.com  Mon Sep 26 02:59:14 2011
From: abioy.sun at gmail.com (Abioy Sun)
Date: Mon, 26 Sep 2011 10:59:14 +0800
Subject: will NGX_AGAIN kill performance?
In-Reply-To: <CAB4Tn6Me+AXq7eWcXjE3fbGkt1yGAksBpyjz2f-yn0ndTETkdg@mail.gmail.com>
References: <CAM6b-vko_0rQELdeJH=ED9rxEvJTAhHo0UtHc+eqTzSm-0-D6A@mail.gmail.com>
	<CAB4Tn6Me+AXq7eWcXjE3fbGkt1yGAksBpyjz2f-yn0ndTETkdg@mail.gmail.com>
Message-ID: <CAM6b-vkEdtRagHvxyHNsF7fRXrKAT2EG_jDWBwZmxLshrKzp5w@mail.gmail.com>

hi,

>
> NGX_AGAIN is usually for non-blocking I/O operations, it's key for
> good performance under high concurrency level
in fact, that's what i'm doing. but i also doubt if the main request wake up
every time an event arrived(made by not only subrequest), will it spent most
of the time doing sth useless?

thx

Abioy
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110926/05e150e8/attachment.html>

From nginx-forum at nginx.us  Mon Sep 26 04:19:21 2011
From: nginx-forum at nginx.us (ohjyops)
Date: Mon, 26 Sep 2011 00:19:21 -0400
Subject: Help: My rewriting does not work.
In-Reply-To: <CAENo=s0at_W6YC5tuwzGfbO=86Tg3ywJpG+8fUZ6par5zq6CjA@mail.gmail.com>
References: <CAENo=s0at_W6YC5tuwzGfbO=86Tg3ywJpG+8fUZ6par5zq6CjA@mail.gmail.com>
Message-ID: <48fa1a1e3528583681e75c927b640e2b.NginxMailingListEnglish@forum.nginx.org>

Thank you for your reply.
I tried:
^/?????.html /w/??? permanent;
But it did not work. ( I did not covert the file to ASCII mode, It is
UTF8)
 
My nginx configuration file was saved in UTF8 encoding. The name of the
old page file is a unescaped form of "?????.html" in GBK
encoding. I converted my old website system from GBK encoding to a new
UTF8 encoding one, and I use a new location  /w/??? to replace the
old "/%C8%CB%CC%E5%D1%A8%CE%BB%CD%BC.html".

Here is the related part of my conf file:
server {
...
  	charset utf-8;
  	source_charset utf-8;
...
location ~ .html$ {
  rewrite ^/%C8%CB%CC%E5%D1%A8%CE%BB%CD%BC.html /w/??? permanent;
}
}

What words will "/%C8%CB%CC%E5%D1%A8%CE%BB%CD%BC.html" be translated by
nginx? How can I catch the path and rewrite it to the new location?

If I covert my nginx conf file to ASCII encoding, will my new site
system (in UTF8 encoding) still work?

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215776,215824#msg-215824


From hagaia at qwilt.com  Mon Sep 26 07:05:45 2011
From: hagaia at qwilt.com (Hagai Avrahami)
Date: Mon, 26 Sep 2011 09:05:45 +0200
Subject: FLV
In-Reply-To: <DUB107-W294EF013659595FFA71513EDF20@phx.gbl>
References: <CAN-ZXpvz=wQ9h+JiEgX9zrmify2kkszN5LKYG0iFPAAu2OvS9w@mail.gmail.com>
	<DUB107-W294EF013659595FFA71513EDF20@phx.gbl>
Message-ID: <CAN-ZXpsc2XNt90fR__e9wOBUNhXLb3LpjvcYpSP7p6pJXB6GgA@mail.gmail.com>

Hi
You Tube Videos is in FLV format and use begin=X in millisecond for Advance
seeking
Hagai



On Sun, Sep 25, 2011 at 5:16 PM, Lukas Tribus <luky-37 at hotmail.com> wrote:

>  FLV pseudostreaming is always in bytes, not seconds (like MP4).
>
> Please see
> http://www.longtailvideo.com/support/jw-player/jw-player-for-flash-v5/12534/video-delivery-http-pseudo-streaming#mechanism
>
>
> Lukas
>
>
> ------------------------------
> From: hagaia at qwilt.com
> Date: Sun, 25 Sep 2011 16:42:11 +0200
> Subject: FLV
> To: nginx at nginx.org
>
>
> Hi
>
> My question is about FLV Module
>
> NGINX documentation mentions that FLV module supports time-based offsets*
> "This module provides the ability to seek within FLV (Flash) files using
> time-based offsets."*
> http://wiki.nginx.org/HttpFlvStreamModule
>
> When looking on the source code and some testing I could see that its
> supports byte-based offset
>
> Can you help me fill the gap
> Thanks a lot
> *Hagai Avrahami*
> Qwilt | Work: +972-72-2221644| Mobile: +972-54-4895656 | hagaia at qwilt.com<yoav at qwilt.com>
>
>
> _______________________________________________ nginx mailing list
> nginx at nginx.org http://mailman.nginx.org/mailman/listinfo/nginx
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx
>



-- 

*Hagai Avrahami*
Qwilt | Work: +972-72-2221644| Mobile: +972-54-4895656 |
hagaia at qwilt.com<yoav at qwilt.com>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110926/ad7e6ead/attachment.html>

From hagaia at qwilt.com  Mon Sep 26 07:09:19 2011
From: hagaia at qwilt.com (Hagai Avrahami)
Date: Mon, 26 Sep 2011 09:09:19 +0200
Subject: FLV
In-Reply-To: <20110925183352.GG1137@mdounin.ru>
References: <CAN-ZXpvz=wQ9h+JiEgX9zrmify2kkszN5LKYG0iFPAAu2OvS9w@mail.gmail.com>
	<20110925183352.GG1137@mdounin.ru>
Message-ID: <CAN-ZXps6r-HhkqNChOYSDbD3FD_dO643VgiWErynnCuREV-qXg@mail.gmail.com>

Thanks

On Sun, Sep 25, 2011 at 8:33 PM, Maxim Dounin <mdounin at mdounin.ru> wrote:

> Hello!
>
> On Sun, Sep 25, 2011 at 04:42:11PM +0200, Hagai Avrahami wrote:
>
> > Hi
> >
> > My question is about FLV Module
> >
> > NGINX documentation mentions that FLV module supports time-based offsets*
> > "This module provides the ability to seek within FLV (Flash) files using
> > time-based offsets."*
> > http://wiki.nginx.org/HttpFlvStreamModule
>
> Looks like wiki is wrong, feel free to fix it.  Original
> documentation (in Russian) may be found here:
>
> http://sysoev.ru/nginx/docs/http/ngx_http_flv_module.html
>
> > When looking on the source code and some testing I could see that its
> > supports byte-based offset
>
> Yes, for flv offset is in bytes.
>
> Maxim Dounin
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx
>



-- 

*Hagai Avrahami*
Qwilt | Work: +972-72-2221644| Mobile: +972-54-4895656 |
hagaia at qwilt.com<yoav at qwilt.com>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110926/fc663d39/attachment.html>

From lists at ruby-forum.com  Mon Sep 26 08:16:00 2011
From: lists at ruby-forum.com (Sandesh K.)
Date: Mon, 26 Sep 2011 10:16:00 +0200
Subject: Cannot remove Content Length header
In-Reply-To: <CAEVF9DUF_Rht7SPZYHEgYJKya9mKQFsgGyD8GkZEjevEtcC4sw@mail.gmail.com>
References: <CAEVF9DX4MNPp5Vh7opoX7-Ub4e7CP564jtXNwRoB_3YidTU-wg@mail.gmail.com>
	<20110924120530.GA1137@mdounin.ru>
	<B0CB97C9-FE33-4F41-BAC2-E52CD5CD21C3@gmail.com>
	<20110924145447.GB1137@mdounin.ru>
	<B9C24323-96B9-4162-A730-C76D457F7AEE@gmail.com>
	<CAEVF9DUF_Rht7SPZYHEgYJKya9mKQFsgGyD8GkZEjevEtcC4sw@mail.gmail.com>
Message-ID: <b42ecf46a45632e1f70fafa0376dbd7c@ruby-forum.com>

Luc Juggery wrote in post #1023663:
> Thanks, I corrected this on back-end side (adding a content-length).
> Luc
>
> 2011/9/24 Luc Juggery <luc.juggery at gmail.com>

Hi Maxim,

We are facing issues with certain phones which post data with chunked 
headers and no content length. These clients get error as 411.

i have tried this w/ apache it works.

Have we fixed this in any of current stable release we have any patch 
for it to work.

Issue is similar to the one discussed in this link:
http://www.ruby-forum.com/topic/162976


Regards,
Sandesh K

-- 
Posted via http://www.ruby-forum.com/.


From nginx-forum at nginx.us  Mon Sep 26 08:42:07 2011
From: nginx-forum at nginx.us (csg)
Date: Mon, 26 Sep 2011 04:42:07 -0400
Subject: Does Nginx honor DNS TTLs for proxy upstreams?
Message-ID: <848a831e65a88f0624cdf43d741d0cda.NginxMailingListEnglish@forum.nginx.org>

I have a short question on Nginx's proxy module. In our setup we reverse
proxy specific requests to the server of an external partner via
proxy_pass. Over the weekend the partner updated DNS but Nginx wasn't
catching up the change and still sent traffic to the old server despite
the fact the TTL of the record was little enough.

An additional obstacle might be that in our case we have for example

proxy_pass http://foo.example.com;

where foo.example.com is a CNAME to bar.example.com (TTL of 1 hour)
which is a A record (TTL of 60 seconds).

Does Nginx honor DNS changes for upstream proxies or are lookups only
done once on startup? If not, will the TTL being honor even if it has to
traverse a list of CNAME records until it hits an A record?

Both configuration directives, resolver and resolver_timeout are not
set, therefore the default of 30 seconds should apply (if that is
relevant here).

$ nginx -V
nginx: nginx version: nginx/1.0.3
nginx: configure arguments: --prefix=/usr/share/nginx
--conf-path=/etc/nginx/nginx.conf --sbin-path=/usr/sbin/nginx
--http-log-path=/var/log/nginx/access.log
--error-log-path=/var/log/nginx/error.log --pid-path=/var/run/nginx.pid
--user=nginx --group=nginx --with-http_geoip_module
--with-http_gzip_static_module --with-http_realip_module
--with-http_secure_link_module --with-http_stub_status_module
--add-module=../agentzh-echo-nginx-module-1c4e116
--add-module=../ngx-fancyindex-0.3.1
--add-module=../agentzh-headers-more-nginx-module-2cbbc15

Thanks in advance!

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215830,215830#msg-215830


From nginx-forum at nginx.us  Mon Sep 26 09:30:49 2011
From: nginx-forum at nginx.us (dullnicker)
Date: Mon, 26 Sep 2011 05:30:49 -0400
Subject: How to tune nginx for 800-1200 concurrent connections?
In-Reply-To: <ef98ae9bdd90878e490c36035da42860.NginxMailingListEnglish@forum.nginx.org>
References: <ef98ae9bdd90878e490c36035da42860.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <12107d44f2052e1fa0e179d4319587b1.NginxMailingListEnglish@forum.nginx.org>

Dear all,

thank you very much for your answers. What I did some days ago was
installing APC (Alternative PHP Cache) on the server. I could not
believe the results, but still - days after - they are as good as they
were from the start. Server Load dropped like crazy. The momentary
average server load is < 0.60, while it was >2.00 before. Therefore the
need for any further optimization has vanished. Again, thank you both
for taking the time to answer to my question.

Kind regards
-Amitz

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215383,215831#msg-215831


From mdounin at mdounin.ru  Mon Sep 26 09:37:47 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Mon, 26 Sep 2011 13:37:47 +0400
Subject: Does Nginx honor DNS TTLs for proxy upstreams?
In-Reply-To: <848a831e65a88f0624cdf43d741d0cda.NginxMailingListEnglish@forum.nginx.org>
References: <848a831e65a88f0624cdf43d741d0cda.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20110926093747.GL1137@mdounin.ru>

Hello!

On Mon, Sep 26, 2011 at 04:42:07AM -0400, csg wrote:

> I have a short question on Nginx's proxy module. In our setup we reverse
> proxy specific requests to the server of an external partner via
> proxy_pass. Over the weekend the partner updated DNS but Nginx wasn't
> catching up the change and still sent traffic to the old server despite
> the fact the TTL of the record was little enough.
> 
> An additional obstacle might be that in our case we have for example
> 
> proxy_pass http://foo.example.com;
> 
> where foo.example.com is a CNAME to bar.example.com (TTL of 1 hour)
> which is a A record (TTL of 60 seconds).
> 
> Does Nginx honor DNS changes for upstream proxies or are lookups only
> done once on startup? If not, will the TTL being honor even if it has to
> traverse a list of CNAME records until it hits an A record?

No, domain names statically configured in config are only looked 
up once on startup (or configuration reload).

> Both configuration directives, resolver and resolver_timeout are not
> set, therefore the default of 30 seconds should apply (if that is
> relevant here).

These are not relevant for "proxy_pass http://foo.example.com;".  
Resolver is only used for proxy_pass with variables, i.e. 
something like

    resolver 127.0.0.1;
    set $backend "foo.example.com";
    proxy_pass http://$backend;

In such setup ip address of "foo.example.com" will be looked up 
dynamically and result will be cached for 5 minutes.

Maxim Dounin


From nginx-forum at nginx.us  Mon Sep 26 10:11:44 2011
From: nginx-forum at nginx.us (tcbarrett)
Date: Mon, 26 Sep 2011 06:11:44 -0400
Subject: Best way to burst static file cache
Message-ID: <0668d2360623a0b4f9f9c84c0ab39f30.NginxMailingListEnglish@forum.nginx.org>

Hi

Loving nginx, thanks Igor!


I'm caching static files in a very traditional way, I hope:
location ~* \.(js|css|png|jpg|jpeg|gif|ico)$ {
    #expires       max;
    expires       30m;
    access_log    off;
    log_not_found on;
  }

I would like to have longer caching. Sometimes there is a fast turn
around on live sites and clients have an issue bursting the cache. These
tend to be Windows based clients. No matter what they or I do, they
continue to retrieve cached versions of css and pngs.

Is there a better configuration, or anything that I am not doing at my
end to ensure they get served the fresh files?

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215835,215835#msg-215835


From calin.don at gmail.com  Mon Sep 26 10:57:14 2011
From: calin.don at gmail.com (Calin Don)
Date: Mon, 26 Sep 2011 13:57:14 +0300
Subject: Log response errors to a separate file
Message-ID: <CAEOe2Jwo6iUoEt3Omj8bNG=w9etM-gJUFERr=vpNQWbHoxWUyQ@mail.gmail.com>

Hello, I would like to log to the general access log and also to a separate
log 404, 403, 500, 501, 502, 503 responses. Is there a way of doing this? Is
anyone doing this?

Currently nginx is running in front of fastcgi and I would like 404 to be
handled by the fastcgi backend (front controller pattern).

Thanks.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110926/e6953b4d/attachment.html>

From nginx-forum at nginx.us  Mon Sep 26 13:40:53 2011
From: nginx-forum at nginx.us (csg)
Date: Mon, 26 Sep 2011 09:40:53 -0400
Subject: Does Nginx honor DNS TTLs for proxy upstreams?
In-Reply-To: <20110926093747.GL1137@mdounin.ru>
References: <20110926093747.GL1137@mdounin.ru>
Message-ID: <800929933ba3c1edd4f169e2a7fb2319.NginxMailingListEnglish@forum.nginx.org>

Maxim Dounin Wrote:
-------------------------------------------------------
> Resolver is only used for proxy_pass with
> variables, i.e. 
> something like
> 
>     resolver 127.0.0.1;
>     set $backend "foo.example.com";
>     proxy_pass http://$backend;
> 
> In such setup ip address of "foo.example.com" will
> be looked up 
> dynamically and result will be cached for 5
> minutes.

Thanks Maxim, that would be great but is it really supposed to work (in
the stable branch)? The 1.0.3 which was deployed up to now gave me

no host in upstream ""

errors, with 1.0.6 I am getting

invalid URL prefix in "http://"

errors. If it helps I can try to setup a simplified test case, because
posting the full configuration would certainly exceed some acceptable
message limit of this list/forum.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215830,215844#msg-215844


From mdounin at mdounin.ru  Mon Sep 26 13:56:25 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Mon, 26 Sep 2011 17:56:25 +0400
Subject: Does Nginx honor DNS TTLs for proxy upstreams?
In-Reply-To: <800929933ba3c1edd4f169e2a7fb2319.NginxMailingListEnglish@forum.nginx.org>
References: <20110926093747.GL1137@mdounin.ru>
	<800929933ba3c1edd4f169e2a7fb2319.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20110926135624.GO1137@mdounin.ru>

Hello!

On Mon, Sep 26, 2011 at 09:40:53AM -0400, csg wrote:

> Maxim Dounin Wrote:
> -------------------------------------------------------
> > Resolver is only used for proxy_pass with
> > variables, i.e. 
> > something like
> > 
> >     resolver 127.0.0.1;
> >     set $backend "foo.example.com";
> >     proxy_pass http://$backend;
> > 
> > In such setup ip address of "foo.example.com" will
> > be looked up 
> > dynamically and result will be cached for 5
> > minutes.
> 
> Thanks Maxim, that would be great but is it really supposed to work (in
> the stable branch)? The 1.0.3 which was deployed up to now gave me
> 
> no host in upstream ""
> 
> errors, with 1.0.6 I am getting
> 
> invalid URL prefix in "http://"
> 
> errors. If it helps I can try to setup a simplified test case, because
> posting the full configuration would certainly exceed some acceptable
> message limit of this list/forum.

Errors suggest you don't have $backend set, and resulting 
proxy_pass tries to go to "http://" which is invalid.

Maxim Dounin


From ian.hobson at ntlworld.com  Mon Sep 26 20:33:11 2011
From: ian.hobson at ntlworld.com (Ian Hobson)
Date: Mon, 26 Sep 2011 21:33:11 +0100
Subject: Does the long polling module work with nginx 1.0.6?
Message-ID: <4E80E187.6030807@ntlworld.com>

Hi,

I am using nginx with the long polling module.  When 0.7 something was 
new, I compiled nginx with the long polling module and wrote my 
application.  It ran under Ubuntu for 18 months or so, without problem.

During this time I have not changed the code that handles messages or 
startup. In fact, there are only two commits to the VCS.

About a month ago, I recompiled 1.0.6 with the long polling module using 
the same script as before.  And the application has started to show 
intermittent errors.

Can anyone confirm that they are using the long-polling module with 
1.0.6 without problems?

The symptom appears to be that if the second person to join the chat 
room, does not send a message for 5 minutes and they do not receive one 
in this time either, then they get timed out, and never receive a reply 
to the read issued after the start-up phase.

Regards

Ian





From tomlove at gmail.com  Mon Sep 26 20:58:56 2011
From: tomlove at gmail.com (Thomas Love)
Date: Mon, 26 Sep 2011 22:58:56 +0200
Subject: Does the long polling module work with nginx 1.0.6?
In-Reply-To: <4E80E187.6030807@ntlworld.com>
References: <4E80E187.6030807@ntlworld.com>
Message-ID: <CAJCGDi8ksNcSHyf-iLm5vZevAKBMew_M2NvA5s62_wyGRLGtGw@mail.gmail.com>

Are you referring to the push module? (http://pushmodule.slact.net/)

Thomas



On 26 September 2011 22:33, Ian Hobson <ian.hobson at ntlworld.com> wrote:
> Hi,
>
> I am using nginx with the long polling module. ?When 0.7 something was new,
> I compiled nginx with the long polling module and wrote my application. ?It
> ran under Ubuntu for 18 months or so, without problem.
>
> During this time I have not changed the code that handles messages or
> startup. In fact, there are only two commits to the VCS.
>
> About a month ago, I recompiled 1.0.6 with the long polling module using the
> same script as before. ?And the application has started to show intermittent
> errors.
>
> Can anyone confirm that they are using the long-polling module with 1.0.6
> without problems?
>
> The symptom appears to be that if the second person to join the chat room,
> does not send a message for 5 minutes and they do not receive one in this
> time either, then they get timed out, and never receive a reply to the read
> issued after the start-up phase.
>
> Regards
>
> Ian
>
>
>
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx
>


From ian at ianhobson.co.uk  Mon Sep 26 21:14:49 2011
From: ian at ianhobson.co.uk (Ian Hobson)
Date: Mon, 26 Sep 2011 22:14:49 +0100
Subject: Does the long polling module work with nginx 1.0.6?
In-Reply-To: <CAJCGDi8ksNcSHyf-iLm5vZevAKBMew_M2NvA5s62_wyGRLGtGw@mail.gmail.com>
References: <4E80E187.6030807@ntlworld.com>
	<CAJCGDi8ksNcSHyf-iLm5vZevAKBMew_M2NvA5s62_wyGRLGtGw@mail.gmail.com>
Message-ID: <4E80EB49.9000108@ianhobson.co.uk>

On 26/09/2011 21:58, Thomas Love wrote:
> Are you referring to the push module? (http://pushmodule.slact.net/)
>
> Thomas
>
Yes - Version 0.69


From tomlove at gmail.com  Mon Sep 26 22:10:40 2011
From: tomlove at gmail.com (Thomas Love)
Date: Tue, 27 Sep 2011 00:10:40 +0200
Subject: Does the long polling module work with nginx 1.0.6?
In-Reply-To: <4E80EB49.9000108@ianhobson.co.uk>
References: <4E80E187.6030807@ntlworld.com>
	<CAJCGDi8ksNcSHyf-iLm5vZevAKBMew_M2NvA5s62_wyGRLGtGw@mail.gmail.com>
	<4E80EB49.9000108@ianhobson.co.uk>
Message-ID: <CAJCGDi_jWABeVRNZUEeKmZ9q3cpdFOsRWgH9ASiU_=GNSycdBQ@mail.gmail.com>

I put the push module into production for a while a few months ago and
suffered intermittent segfaults. It also has a couple of design
limitations that were problematic in my case. It doesn't seem to be
under active development, so I doubt these issues will be fixed any
time soon.

But I wonder whether the module isn't a poor solution architecturally,
in any case. I eventually dropped it and implemented long-polling in a
Node.js script. This turned out to be almost trivial -- far faster
than trying to fix the module. Nginx is set up as a reverse proxy to
this server, so it's insulated from any bad behaviour. I found the
stability and ease of maintenance of this design easily worth the
extra ~50 MB and CPU cycles.

Cheers
Thomas




On 26 September 2011 23:14, Ian Hobson <ian at ianhobson.co.uk> wrote:
> On 26/09/2011 21:58, Thomas Love wrote:
>>
>> Are you referring to the push module? (http://pushmodule.slact.net/)
>>
>> Thomas
>>
> Yes - Version 0.69
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx
>


From ezwetkow at gmx.de  Tue Sep 27 04:52:03 2011
From: ezwetkow at gmx.de (Elena Zwetkow)
Date: Tue, 27 Sep 2011 06:52:03 +0200
Subject: Does the long polling module work with nginx 1.0.6?
In-Reply-To: <CAJCGDi_jWABeVRNZUEeKmZ9q3cpdFOsRWgH9ASiU_=GNSycdBQ@mail.gmail.com>
References: <4E80E187.6030807@ntlworld.com>
	<CAJCGDi8ksNcSHyf-iLm5vZevAKBMew_M2NvA5s62_wyGRLGtGw@mail.gmail.com>
	<4E80EB49.9000108@ianhobson.co.uk>
	<CAJCGDi_jWABeVRNZUEeKmZ9q3cpdFOsRWgH9ASiU_=GNSycdBQ@mail.gmail.com>
Message-ID: <20110927045203.20830@gmx.net>


Hi,

-------- Original-Nachricht --------
> Datum: Tue, 27 Sep 2011 00:10:40 +0200
> Von: Thomas Love <tomlove at gmail.com>
> An: nginx at nginx.org
> Betreff: Re: Does the long polling module work with nginx 1.0.6?

> I put the push module into production for a while a few months ago and
> suffered intermittent segfaults. It also has a couple of design
> limitations that were problematic in my case. It doesn't seem to be
> under active development, so I doubt these issues will be fixed any
> time soon.
> 
> But I wonder whether the module isn't a poor solution architecturally,
> in any case. I eventually dropped it and implemented long-polling in a
> Node.js script. This turned out to be almost trivial -- far faster
> than trying to fix the module. Nginx is set up as a reverse proxy to
> this server, so it's insulated from any bad behaviour. I found the
> stability and ease of maintenance of this design easily worth the
> extra ~50 MB and CPU cycles.
> 
> Cheers
> Thomas
> 
> 
> 
> 
> On 26 September 2011 23:14, Ian Hobson <ian at ianhobson.co.uk> wrote:
> > On 26/09/2011 21:58, Thomas Love wrote:
> >>
> >> Are you referring to the push module? (http://pushmodule.slact.net/)
> >>
> >> Thomas
> >>
> > Yes - Version 0.69
> >
> > _______________________________________________
> > nginx mailing list
> > nginx at nginx.org
> > http://mailman.nginx.org/mailman/listinfo/nginx
> >
> 
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx

or have a look at wandenbergs nginx-push-module, since version 0.3.0 (master branch) it also support long-polling and polling and its under active develpment.

https://github.com/wandenberg/nginx-push-stream-module

Cheers,

Elena

-- 
NEU: FreePhone - 0ct/min Handyspartarif mit Geld-zur?ck-Garantie!		
Jetzt informieren: http://www.gmx.net/de/go/freephone


From martronic at martronic.ch  Tue Sep 27 06:41:33 2011
From: martronic at martronic.ch (Martronic SA)
Date: Tue, 27 Sep 2011 08:41:33 +0200
Subject: Nginx Substitutions Filter problem
In-Reply-To: <20110927045203.20830@gmx.net>
References: <4E80E187.6030807@ntlworld.com>
	<CAJCGDi8ksNcSHyf-iLm5vZevAKBMew_M2NvA5s62_wyGRLGtGw@mail.gmail.com>
	<4E80EB49.9000108@ianhobson.co.uk>
	<CAJCGDi_jWABeVRNZUEeKmZ9q3cpdFOsRWgH9ASiU_=GNSycdBQ@mail.gmail.com>
	<20110927045203.20830@gmx.net>
Message-ID: <4E81701D.408@martronic.ch>

Hello,

We are using the "Substitutions Filter module"
to replace html content in our pages.

Our configuration is :

location /{
subs_filter_types text/html text/xml;
subs_filter 'http://bale.hopla.net/bons-plans' 
'http://bons-plans.hopla.net';
subs_filter 'id="portal-logo" href="http://bale.hopla.net"' 
'id="portal-logo" href="http://www.hopla.net"';
}

The first substitution is working, but the second not.
Are we missing something? Is the "space" supported in substitution module?

Thank you for your help!

Martronic SA

nginx -V
nginx: nginx version: nginx/1.1.4
nginx: built by gcc 4.3.2 (Debian 4.3.2-1.1)
nginx: TLS SNI support enabled
nginx: configure arguments: --conf-path=/etc/nginx/nginx.conf
--error-log-path=/var/log/nginx/error.log --pid-path=/var/run/nginx.pid
--lock-path=/var/lock/nginx.lock --http-log-path=/var/log/nginx/access.log
--with-http_dav_module --http-client-body-temp-path=/var/lib/nginx/body --with-http_ssl_module
--http-proxy-temp-path=/var/lib/nginx/proxy --with-http_stub_status_module
--http-fastcgi-temp-path=/var/lib/nginx/fastcgi
--add-module=/var/lib/substitutions4nginx-read-only




From nginx-forum at nginx.us  Tue Sep 27 07:54:07 2011
From: nginx-forum at nginx.us (springbok)
Date: Tue, 27 Sep 2011 03:54:07 -0400
Subject: Nginx and/or Passenger failing
Message-ID: <e60b22e2ac38e31eb0ad369f42ccb4e7.NginxMailingListEnglish@forum.nginx.org>

Hi,

Envirionment: 
Passenger 3.0.9 
nginx 1.0.6 
Suse SLE 11 
Rails 3.0.7 
Ruby 1.9.2 p136 
Quad core 
8 GB 
Single Rails application 

nginx settings:

worker_processes 3;
worker_rlimit_nofile 8192;
worker_priority 0;
events {
  multi_accept off;
  worker_connections 4096;
}

pid /var/run/nginx.pid;
error_log  logs/error.log debug;
http {
    passenger_root
/usr/local/rvm/gems/ruby-1.9.2-p136/gems/passenger-3.0.9;
    passenger_ruby /usr/local/rvm/wrappers/ruby-1.9.2-p136/ruby;
    passenger_max_pool_size 4; 
    passenger_min_instances 1;
    keepalive_timeout 180;
    client_body_buffer_size 128k;
    client_header_buffer_size 2k;
    large_client_header_buffers 32 64k;
    passenger_pool_idle_time 300;
    passenger_log_level 2;
    passenger_use_global_queue on;
    include       /opt/nginx/conf/mime.types;
    default_type  application/octet-stream;
    log_format main      '$remote_addr - $remote_user [$time_local]  '
      '"$request" $status $bytes_sent '
      '"$http_referer" "$http_user_agent" '
      '"$gzip_ratio"';
    access_log  logs/access.log  main;
    sendfile        on;
    tcp_nopush        on;
    tcp_nodelay       off;
    gzip            on;
    gzip_http_version 1.0;
    gzip_comp_level 2;
    gzip_proxied any;
    gzip_min_length  1000;
    gzip_types      text/plain text/css application/x-javascript
text/xml application/xml application/xml+rss text/javascript;
...

I'm not sure if the problem lies with passenger or nginx, after a period
of time the passenger as well as the nginx master process shuts down.
When I look in the log file I have the following:

2011/09/27 16:45:36 [info] 6408#0: *204 client 61.88.141.194 closed
keepalive connection
[ pid=6381 thr=140088026498848
file=ext/common/LoggingAgent/LoggingServer.h:829 time=2011-09-27
16:45:46.849 ]: Flushing all sinks (periodic action)
[ pid=6381 thr=140088026498848
file=ext/common/LoggingAgent/LoggingServer.h:829 time=2011-09-27
16:46:01.848 ]: Flushing all sinks (periodic action)
[ pid=6381 thr=140088026498848
file=ext/common/LoggingAgent/LoggingServer.h:829 time=2011-09-27
16:46:16.848 ]: Flushing all sinks (periodic action)
[ pid=6381 thr=140088026498848
file=ext/common/LoggingAgent/LoggingServer.h:829 time=2011-09-27
16:46:31.848 ]: Flushing all sinks (periodic action)
[ pid=6381 thr=140088026498848
file=ext/common/LoggingAgent/LoggingServer.h:829 time=2011-09-27
16:46:46.849 ]: Flushing all sinks (periodic action)
[ pid=6381 thr=140088026498848
file=ext/common/LoggingAgent/LoggingServer.h:829 time=2011-09-27
16:47:01.849 ]: Flushing all sinks (periodic action)
[ pid=6381 thr=140088026498848
file=ext/common/LoggingAgent/LoggingServer.h:829 time=2011-09-27
16:47:16.848 ]: Flushing all sinks (periodic action)
[ pid=6381 thr=140088026498848
file=ext/common/LoggingAgent/LoggingServer.h:829 time=2011-09-27
16:47:31.848 ]: Flushing all sinks (periodic action)
[ pid=6371 thr=139975134312192 file=ext/common/Watchdog.cpp:1057
time=2011-09-27 16:47:38.756 ]: Web server did not exit gracefully,
forcing shutdown of all service processes...
2011/09/27 16:49:00 [crit] 6410#0: *250 connect() to
unix:/passenger_helper_server failed (2: No such file or directory)
while connecting to upstream, client: 120.146.194.176, serv
er: localhost, request: "GET /candidates/new HTTP/1.1", upstream:
"passenger:unix:/passenger_helper_server:", host: "61.88.141.221",
referrer: "http://61.88.141.221/home/view_repor
ts"

When I look at the running processes the nginx master is no longer
running but the workers are still running, I have to kill them before I
can restart. All passenger processes are not longer running.

As I said I'm not sure which process is causing the other to close and I
have no idea what the problem may be, so any help greatly appreciated.

Etienne.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215871,215871#msg-215871


From wangxichun at 300.cn  Tue Sep 27 07:55:18 2011
From: wangxichun at 300.cn (wangxc)
Date: Tue, 27 Sep 2011 15:55:18 +0800
Subject: how can i run/debug nginx in eclipse?
Message-ID: <4E818166.1070705@300.cn>



hi, all:

I want to run/debug the nginx in Eclipse, how can i configure it ?

thanks in advance!

-- 

best regards

rmn190


From david.yu.ftw at gmail.com  Tue Sep 27 08:02:00 2011
From: david.yu.ftw at gmail.com (David Yu)
Date: Tue, 27 Sep 2011 16:02:00 +0800
Subject: how can i run/debug nginx in eclipse?
In-Reply-To: <4E818166.1070705@300.cn>
References: <4E818166.1070705@300.cn>
Message-ID: <CAOkdov+1+hdULOC47S6ArD0q2kUYjagG0fD4-mACWmV7Aqp4ag@mail.gmail.com>

Run via External tools and point to your local nginx bin dir.

On Tue, Sep 27, 2011 at 3:55 PM, wangxc <wangxichun at 300.cn> wrote:

>
>
> hi, all:
>
> I want to run/debug the nginx in Eclipse, how can i configure it ?
>
> thanks in advance!
>
> --
>
> best regards
>
> rmn190
>
> ______________________________**_________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/**mailman/listinfo/nginx<http://mailman.nginx.org/mailman/listinfo/nginx>
>



-- 
When the cat is away, the mouse is alone.
- David Yu
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110927/32c15bac/attachment.html>

From osa at FreeBSD.org.ru  Tue Sep 27 08:21:35 2011
From: osa at FreeBSD.org.ru (Sergey A. Osokin)
Date: Tue, 27 Sep 2011 12:21:35 +0400
Subject: Nginx and/or Passenger failing
In-Reply-To: <e60b22e2ac38e31eb0ad369f42ccb4e7.NginxMailingListEnglish@forum.nginx.org>
References: <e60b22e2ac38e31eb0ad369f42ccb4e7.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20110927082135.GF81654@FreeBSD.org.ru>

On Tue, Sep 27, 2011 at 03:54:07AM -0400, springbok wrote:
> Hi,
> 
> Envirionment: 
> Passenger 3.0.9 
> nginx 1.0.6 
> Suse SLE 11 
> Rails 3.0.7 
> Ruby 1.9.2 p136 
> Quad core 
> 8 GB 
> Single Rails application 
> 
> nginx settings:
> 
> worker_processes 3;
> worker_rlimit_nofile 8192;
> worker_priority 0;
> events {
>   multi_accept off;
>   worker_connections 4096;
> }
> 
> pid /var/run/nginx.pid;
> error_log  logs/error.log debug;
> http {
>     passenger_root
> /usr/local/rvm/gems/ruby-1.9.2-p136/gems/passenger-3.0.9;
>     passenger_ruby /usr/local/rvm/wrappers/ruby-1.9.2-p136/ruby;
>     passenger_max_pool_size 4; 
>     passenger_min_instances 1;
>     keepalive_timeout 180;
>     client_body_buffer_size 128k;
>     client_header_buffer_size 2k;
>     large_client_header_buffers 32 64k;
>     passenger_pool_idle_time 300;
>     passenger_log_level 2;
>     passenger_use_global_queue on;
>     include       /opt/nginx/conf/mime.types;
>     default_type  application/octet-stream;
>     log_format main      '$remote_addr - $remote_user [$time_local]  '
>       '"$request" $status $bytes_sent '
>       '"$http_referer" "$http_user_agent" '
>       '"$gzip_ratio"';
>     access_log  logs/access.log  main;
>     sendfile        on;
>     tcp_nopush        on;
>     tcp_nodelay       off;
>     gzip            on;
>     gzip_http_version 1.0;
>     gzip_comp_level 2;
>     gzip_proxied any;
>     gzip_min_length  1000;
>     gzip_types      text/plain text/css application/x-javascript
> text/xml application/xml application/xml+rss text/javascript;
> ...
> 
> I'm not sure if the problem lies with passenger or nginx, after a period
> of time the passenger as well as the nginx master process shuts down.
> When I look in the log file I have the following:
> 
> 2011/09/27 16:45:36 [info] 6408#0: *204 client 61.88.141.194 closed
> keepalive connection
> [ pid=6381 thr=140088026498848
> file=ext/common/LoggingAgent/LoggingServer.h:829 time=2011-09-27
> 16:45:46.849 ]: Flushing all sinks (periodic action)
> [ pid=6381 thr=140088026498848
> file=ext/common/LoggingAgent/LoggingServer.h:829 time=2011-09-27
> 16:46:01.848 ]: Flushing all sinks (periodic action)
> [ pid=6381 thr=140088026498848
> file=ext/common/LoggingAgent/LoggingServer.h:829 time=2011-09-27
> 16:46:16.848 ]: Flushing all sinks (periodic action)
> [ pid=6381 thr=140088026498848
> file=ext/common/LoggingAgent/LoggingServer.h:829 time=2011-09-27
> 16:46:31.848 ]: Flushing all sinks (periodic action)
> [ pid=6381 thr=140088026498848
> file=ext/common/LoggingAgent/LoggingServer.h:829 time=2011-09-27
> 16:46:46.849 ]: Flushing all sinks (periodic action)
> [ pid=6381 thr=140088026498848
> file=ext/common/LoggingAgent/LoggingServer.h:829 time=2011-09-27
> 16:47:01.849 ]: Flushing all sinks (periodic action)
> [ pid=6381 thr=140088026498848
> file=ext/common/LoggingAgent/LoggingServer.h:829 time=2011-09-27
> 16:47:16.848 ]: Flushing all sinks (periodic action)
> [ pid=6381 thr=140088026498848
> file=ext/common/LoggingAgent/LoggingServer.h:829 time=2011-09-27
> 16:47:31.848 ]: Flushing all sinks (periodic action)
> [ pid=6371 thr=139975134312192 file=ext/common/Watchdog.cpp:1057
> time=2011-09-27 16:47:38.756 ]: Web server did not exit gracefully,
> forcing shutdown of all service processes...
> 2011/09/27 16:49:00 [crit] 6410#0: *250 connect() to
> unix:/passenger_helper_server failed (2: No such file or directory)
> while connecting to upstream, client: 120.146.194.176, serv
> er: localhost, request: "GET /candidates/new HTTP/1.1", upstream:
> "passenger:unix:/passenger_helper_server:", host: "61.88.141.221",
> referrer: "http://61.88.141.221/home/view_repor
> ts"
> 
> When I look at the running processes the nginx master is no longer
> running but the workers are still running, I have to kill them before I
> can restart. All passenger processes are not longer running.
> 
> As I said I'm not sure which process is causing the other to close and I
> have no idea what the problem may be, so any help greatly appreciated.

Could you try very simple config with small application from passenger
distro, i.e. something like this.

http {
    include       mime.types;
    default_type  application/octet-stream;

    passenger_root /usr/local/rvm/gems/ruby-1.9.2-p136/gems/passenger-3.0.9;
    passenger_ruby /usr/local/rvm/wrappers/ruby-1.9.2-p136/ruby;
    passenger_default_user www;

    server {
        listen       127.0.0.1:80;
        server_name  localhost;

        root /usr/local/rvm/gems/ruby-1.9.2-p136/gems/passenger-3.0.9/test/stub/rack/public;
        passenger_enabled on;
        passenger_use_global_queue on;
    }
}

After restart your nginx with passenger you should find nginx and passenger
processes.

% ps auxww | grep nginx | grep -v nginx
root   18932  4,0 15,9 44464 39656  ??  Ss    8:46     0:00,01 nginx: master process /usr/local/sbin/nginx
www    18933  4,0 16,0 44464 39904  ??  S     8:46     0:00,02 nginx: worker process (nginx)
% ps uaxww | grep -i passenger | grep -v grep
root   18923  0,0  1,0  5760  2508  ??  Is    8:46     0:00,07 PassengerWatchdog
root   18926  0,0  1,5 15052  3820  ??  I     8:46     0:00,09 PassengerHelperAgent
root   18929  0,0  2,7 13176  6620  ??  I     8:46     0:01,75 ruby: Passenger spawn server (ruby)
nobody 18930  0,0  1,4  6688  3600  ??  I     8:46     0:00,07 PassengerLoggingAgent

% telnet 127.0.0.1 80
Trying 127.0.0.1...
Connected to localhost.
Escape character is '^]'.
GET / HTTP/1.0

HTTP/1.1 200 OK
Content-Type: text/html
Connection: close
Status: 200
X-Powered-By: Phusion Passenger (mod_rails/mod_rack) 3.0.9
Server: nginx/1.0.6 + Phusion Passenger 3.0.9 (mod_rails/mod_rack)

hello <b>world</b>Connection closed by foreign host.

Does it possible reproduce your problem for simple configuration?

-- 
Sergey A. Osokin
osa at FreeBSD.ORG.ru
osa at FreeBSD.ORG


From nginx-forum at nginx.us  Tue Sep 27 08:56:27 2011
From: nginx-forum at nginx.us (springbok)
Date: Tue, 27 Sep 2011 04:56:27 -0400
Subject: Nginx and/or Passenger failing
In-Reply-To: <20110927082135.GF81654@FreeBSD.org.ru>
References: <20110927082135.GF81654@FreeBSD.org.ru>
Message-ID: <95a734f49d7c5f4544e1e759ff8e57e2.NginxMailingListEnglish@forum.nginx.org>

Sergey A. Osokin Wrote:
-------------------------------------------------------
> On Tue, Sep 27, 2011 at 03:54:07AM -0400,
> 
> Could you try very simple config with small
> application from passenger
> distro, i.e. something like this.
> 
> http {
>     include       mime.types;
>     default_type  application/octet-stream;
> 
>     passenger_root
> /usr/local/rvm/gems/ruby-1.9.2-p136/gems/passenger
> -3.0.9;
>     passenger_ruby
> /usr/local/rvm/wrappers/ruby-1.9.2-p136/ruby;
>     passenger_default_user www;
> 
>     server {
>         listen       127.0.0.1:80;
>         server_name  localhost;
> 
>         root
> /usr/local/rvm/gems/ruby-1.9.2-p136/gems/passenger
> -3.0.9/test/stub/rack/public;
>         passenger_enabled on;
>         passenger_use_global_queue on;
>     }
> }
> 
> After restart your nginx with passenger you should
> find nginx and passenger
> processes.
> 
> % ps auxww | grep nginx | grep -v nginx
> root   18932  4,0 15,9 44464 39656  ??  Ss    8:46
>     0:00,01 nginx: master process
> /usr/local/sbin/nginx
> www    18933  4,0 16,0 44464 39904  ??  S     8:46
>     0:00,02 nginx: worker process (nginx)
> % ps uaxww | grep -i passenger | grep -v grep
> root   18923  0,0  1,0  5760  2508  ??  Is    8:46
>     0:00,07 PassengerWatchdog
> root   18926  0,0  1,5 15052  3820  ??  I     8:46
>     0:00,09 PassengerHelperAgent
> root   18929  0,0  2,7 13176  6620  ??  I     8:46
>     0:01,75 ruby: Passenger spawn server (ruby)
> nobody 18930  0,0  1,4  6688  3600  ??  I     8:46
>     0:00,07 PassengerLoggingAgent
> 
> % telnet 127.0.0.1 80
> Trying 127.0.0.1...
> Connected to localhost.
> Escape character is '^]'.
> GET / HTTP/1.0
> 
> HTTP/1.1 200 OK
> Content-Type: text/html
> Connection: close
> Status: 200
> X-Powered-By: Phusion Passenger
> (mod_rails/mod_rack) 3.0.9
> Server: nginx/1.0.6 + Phusion Passenger 3.0.9
> (mod_rails/mod_rack)
> 
> hello <b>world</b>Connection closed by foreign
> host.
> 
> Does it possible reproduce your problem for simple
> configuration?
> 
> -- 
> Sergey A. Osokin
> osa at FreeBSD.ORG.ru
> osa at FreeBSD.ORG
> 
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx

Hi Sergey,

Thank you so much for the prompt reply.

Sorry I just can't get this working, nginx fires up ok but nothing
happens when I telnet. I checked passenger using passenger-status
--verbose
and it shows the details for passenger but when I check for running
passenger processes it's not running.

When I do GET / HTTP/1.0 nothing happens.

Am I missing something?


user  ....;
events {
  multi_accept off;
  worker_connections 4096;
}

http {
include mime.types;
default_type application/octet-stream;

passenger_root
/usr/local/rvm/gems/ruby-1.9.2-p136/gems/passenger-3.0.9;
passenger_ruby /usr/local/rvm/wrappers/ruby-1.9.2-p136/ruby;

passenger_default_user ...;

server {
listen 80;
server_name localhost;

root /home/etienne/test/stub/rack/public;
passenger_enabled on;
passenger_use_global_queue on;
}
}

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215871,215876#msg-215876


From nginx-forum at nginx.us  Tue Sep 27 09:10:48 2011
From: nginx-forum at nginx.us (springbok)
Date: Tue, 27 Sep 2011 05:10:48 -0400
Subject: Nginx and/or Passenger failing
In-Reply-To: <95a734f49d7c5f4544e1e759ff8e57e2.NginxMailingListEnglish@forum.nginx.org>
References: <20110927082135.GF81654@FreeBSD.org.ru>
	<95a734f49d7c5f4544e1e759ff8e57e2.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <1dbda862d38fd600df56430d84b13c56.NginxMailingListEnglish@forum.nginx.org>

springbok Wrote:
-------------------------------------------------------
> Sergey A. Osokin Wrote:
> --------------------------------------------------
> -----
> > On Tue, Sep 27, 2011 at 03:54:07AM -0400,
> > 
> > Could you try very simple config with small
> > application from passenger
> > distro, i.e. something like this.
> > 
> > http {
> >     include       mime.types;
> >     default_type  application/octet-stream;
> > 
> >     passenger_root
> >
> /usr/local/rvm/gems/ruby-1.9.2-p136/gems/passenger
> 
> > -3.0.9;
> >     passenger_ruby
> > /usr/local/rvm/wrappers/ruby-1.9.2-p136/ruby;
> >     passenger_default_user www;
> > 
> >     server {
> >         listen       127.0.0.1:80;
> >         server_name  localhost;
> > 
> >         root
> >
> /usr/local/rvm/gems/ruby-1.9.2-p136/gems/passenger
> 
> > -3.0.9/test/stub/rack/public;
> >         passenger_enabled on;
> >         passenger_use_global_queue on;
> >     }
> > }
> > 
> > After restart your nginx with passenger you
> should
> > find nginx and passenger
> > processes.
> > 
> > % ps auxww | grep nginx | grep -v nginx
> > root   18932  4,0 15,9 44464 39656  ??  Ss   
> 8:46
> >     0:00,01 nginx: master process
> > /usr/local/sbin/nginx
> > www    18933  4,0 16,0 44464 39904  ??  S    
> 8:46
> >     0:00,02 nginx: worker process (nginx)
> > % ps uaxww | grep -i passenger | grep -v grep
> > root   18923  0,0  1,0  5760  2508  ??  Is   
> 8:46
> >     0:00,07 PassengerWatchdog
> > root   18926  0,0  1,5 15052  3820  ??  I    
> 8:46
> >     0:00,09 PassengerHelperAgent
> > root   18929  0,0  2,7 13176  6620  ??  I    
> 8:46
> >     0:01,75 ruby: Passenger spawn server (ruby)
> > nobody 18930  0,0  1,4  6688  3600  ??  I    
> 8:46
> >     0:00,07 PassengerLoggingAgent
> > 
> > % telnet 127.0.0.1 80
> > Trying 127.0.0.1...
> > Connected to localhost.
> > Escape character is '^]'.
> > GET / HTTP/1.0
> > 
> > HTTP/1.1 200 OK
> > Content-Type: text/html
> > Connection: close
> > Status: 200
> > X-Powered-By: Phusion Passenger
> > (mod_rails/mod_rack) 3.0.9
> > Server: nginx/1.0.6 + Phusion Passenger 3.0.9
> > (mod_rails/mod_rack)
> > 
> > hello <b>world</b>Connection closed by foreign
> > host.
> > 
> > Does it possible reproduce your problem for
> simple
> > configuration?
> > 
> > -- 
> > Sergey A. Osokin
> > osa at FreeBSD.ORG.ru
> > osa at FreeBSD.ORG
> > 
> > _______________________________________________
> > nginx mailing list
> > nginx at nginx.org
> > http://mailman.nginx.org/mailman/listinfo/nginx
> 
> Hi Sergey,
> 
> Thank you so much for the prompt reply.
> 
> Sorry I just can't get this working, nginx fires
> up ok but nothing happens when I telnet. I checked
> passenger using passenger-status --verbose
> and it shows the details for passenger but when I
> check for running passenger processes it's not
> running.
> 
> When I do GET / HTTP/1.0 nothing happens.
> 
> Am I missing something?
> 
> 
> user  ....;
> events {
>   multi_accept off;
>   worker_connections 4096;
> }
> 
> http {
> include mime.types;
> default_type application/octet-stream;
> 
> passenger_root
> /usr/local/rvm/gems/ruby-1.9.2-p136/gems/passenger
> -3.0.9;
> passenger_ruby
> /usr/local/rvm/wrappers/ruby-1.9.2-p136/ruby;
> 
> passenger_default_user ...;
> 
> server {
> listen 80;
> server_name localhost;
> 
> root /home/etienne/test/stub/rack/public;
> passenger_enabled on;
> passenger_use_global_queue on;
> }
> }

Hi Sergey,

One other thing I'd mention is that our application works great for most
of the time but intermittently it fails with the symptoms I describe so
I'm not sure I can reproduce it with the simple setup if I did get it
going.

Regards,

Etienne.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215871,215877#msg-215877


From nginx-forum at nginx.us  Tue Sep 27 09:35:57 2011
From: nginx-forum at nginx.us (springbok)
Date: Tue, 27 Sep 2011 05:35:57 -0400
Subject: Nginx and/or Passenger failing
In-Reply-To: <1dbda862d38fd600df56430d84b13c56.NginxMailingListEnglish@forum.nginx.org>
References: <20110927082135.GF81654@FreeBSD.org.ru>
	<95a734f49d7c5f4544e1e759ff8e57e2.NginxMailingListEnglish@forum.nginx.org>
	<1dbda862d38fd600df56430d84b13c56.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <5d87c4291809984407c6d4ce1679c712.NginxMailingListEnglish@forum.nginx.org>

Ok, I've got passenger loading and appearing in the process list, I
created a index.html file and was able to display it doing a GET / but
when I try GET / HTTP/1.0 it just sits there and nothing happens. If I
remove the index.html I get "Hello world" returned, no mention of
passenger.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215871,215879#msg-215879


From osa at FreeBSD.org.ru  Tue Sep 27 11:01:46 2011
From: osa at FreeBSD.org.ru (Sergey A. Osokin)
Date: Tue, 27 Sep 2011 15:01:46 +0400
Subject: Nginx and/or Passenger failing
In-Reply-To: <5d87c4291809984407c6d4ce1679c712.NginxMailingListEnglish@forum.nginx.org>
References: <20110927082135.GF81654@FreeBSD.org.ru>
	<95a734f49d7c5f4544e1e759ff8e57e2.NginxMailingListEnglish@forum.nginx.org>
	<1dbda862d38fd600df56430d84b13c56.NginxMailingListEnglish@forum.nginx.org>
	<5d87c4291809984407c6d4ce1679c712.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20110927110146.GA90093@FreeBSD.org.ru>

On Tue, Sep 27, 2011 at 05:35:57AM -0400, springbok wrote:
> Ok, I've got passenger loading and appearing in the process list, I
> created a index.html file and was able to display it doing a GET / but
> when I try GET / HTTP/1.0 it just sits there and nothing happens. If I
> remove the index.html I get "Hello world" returned, no mention of
> passenger.

So, if you got "Hello, <b>world</b> this means passenger works well with
default installation.  Could you please check your application or try to use
previous version of nginx and/or passenger.

-- 
Sergey A. Osokin
osa at FreeBSD.ORG.ru
osa at FreeBSD.ORG


From nginx-forum at nginx.us  Tue Sep 27 11:08:00 2011
From: nginx-forum at nginx.us (liushaobo)
Date: Tue, 27 Sep 2011 07:08:00 -0400
Subject: php-fpm  Input/output error
Message-ID: <2b6a402bd983a82820bbf0286a825073.NginxMailingListEnglish@forum.nginx.org>

hi all ,please help me

php-fpm error log

[27-Sep-2011 11:42:20] ERROR: ptrace(PEEKDATA) failed: Input/output
error (5)

 

php-fpm.slow.log

[27-Sep-2011 13:49:50] [pool www] pid 24386
script_filename = /www/nmarket/index.php
[0x000000001adaa0e8] execute()
/data0/www/yii-1.1.8.r3324/framework/yiilite.php:8599
[0x000000001ada9f48] queryInternal()
/data0/www/yii-1.1.8.r3324/framework/yiilite.php:8550
[0x000000001ada95f0] queryAll()
/data0/www/nmarket/protected/modules/web/components/GuessWidget.php:36
[0x000000001ada9480] getGuessList()
/data0/www/nmarket/protected/modules/web/components/GuessWidget.php:62
[0x000000001ada9198] run()
/data0/www/yii-1.1.8.r3324/framework/yiilite.php:3097
[0x000000001ada6fd8] widget()
/data0/www/nmarket/protected/modules/web/views/soft/show.php:157
[0x000000001ada6dc0] +++ dump failed
[27-Sep-2011 13:49:50] [pool www] pid 25608
script_filename = /www/nmarket/index.php
[0x000000001adbd6c0] execute()
/data0/www/yii-1.1.8.r3324/framework/yiilite.php:8599
[0x000000001adbd520] queryInternal()
/data0/www/yii-1.1.8.r3324/framework/yiilite.php:8550
[0x000000001adbc7a8] queryAll()
/data0/www/nmarket/protected/modules/web/components/JustDownloadedByCat.php:45
[0x000000001adbc4c0] run()
/data0/www/yii-1.1.8.r3324/framework/yiilite.php:3097
[0x000000001adbb6e0] widget()
/data0/www/nmarket/protected/modules/web/views/soft/index.php:56
[0x000000001adbb4c8] +++ dump failed

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215882,215882#msg-215882


From hagaia at qwilt.com  Tue Sep 27 11:36:45 2011
From: hagaia at qwilt.com (Hagai Avrahami)
Date: Tue, 27 Sep 2011 13:36:45 +0200
Subject: Static Files
Message-ID: <CAN-ZXpt10N9m_garKuxC35yFo5JHTgNSkKa0yMZnis2KYfu+Pw@mail.gmail.com>

Hi

I am intend to use NGINX (LINUX)  for serving static files with high
concurrency and bandwidth.
Some of the files is served from the disk and some from cache.
I am trying to find the configuration for best performance.

I have some questions regarding NGINX configuration.

1. *output_buffers 1 32k*
----------------------------------------
This directive excepts 2 inputs
can someone explain me the meaning of each input
I can see that 32K is size of data (per fetch) when reading from a DIsk -
what else it influences?
When serving big files, best Disk utilization is in blocks of 1MB, is it OK
to change this directive to 1024k?

2. *aio*
----------------------------------------
**I read in Wiki the following
*" This directive is usable as of Linux kernel 2.6.22. For Linux it is
required to use directio, this automatically disables sendfile support."*
when using AIO I must use *directio on*? (when testing I could see the file
is served from the cache - which means it is not using O_DIRECT)


3. *sendfile
*----------------------------------------
Is NGINX try to send a file using 1 sendfile call (count = file size)?
If so when trying to send file residing only on the Disk, the server is
blocked till end of transmition?
In this model serving multiple files some on the disk and some cached is
inefficient?

Will be happy to get any answer or opinion

Thanks
Hagai





-- 

*Hagai Avrahami*
Qwilt | Work: +972-72-2221644| Mobile: +972-54-4895656 |
hagaia at qwilt.com<yoav at qwilt.com>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110927/0eb36533/attachment.html>

From nginx-forum at nginx.us  Tue Sep 27 14:03:30 2011
From: nginx-forum at nginx.us (springbok)
Date: Tue, 27 Sep 2011 10:03:30 -0400
Subject: Nginx and/or Passenger failing
In-Reply-To: <20110927110146.GA90093@FreeBSD.org.ru>
References: <20110927110146.GA90093@FreeBSD.org.ru>
Message-ID: <e8e7dbea348df2d17f59dc0533a667b6.NginxMailingListEnglish@forum.nginx.org>

Sergey A. Osokin Wrote:
-------------------------------------------------------
> On Tue, Sep 27, 2011 at 05:35:57AM -0400,
> springbok wrote:
> > Ok, I've got passenger loading and appearing in
> the process list, I
> > created a index.html file and was able to
> display it doing a GET / but
> > when I try GET / HTTP/1.0 it just sits there and
> nothing happens. If I
> > remove the index.html I get "Hello world"
> returned, no mention of
> > passenger.
> 
> So, if you got "Hello, <b>world</b> this means
> passenger works well with
> default installation.  Could you please check your
> application or try to use
> previous version of nginx and/or passenger.
> 
> -- 
> Sergey A. Osokin
> osa at FreeBSD.ORG.ru
> osa at FreeBSD.ORG
> 
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx

Hi Sergey,

I was using 0.8.53 when I started getting the errors and decided to
upgrade before posting my question.

Do you believe this is an nginx or passenger problem, is nginx causing
passenger to shutdown or the other way around? To me it looks like the
watchdog process is getting an error from nginx and closing the
passenger processes.

We have 8GB of memory and 4 processors, do you think it could be memory
related/CPU load related, I know it's difficult to say without seeing
our app and the load it places on the system, but if you had to guess?
It has plenty of memory available so I don't think that's a problem, the
IT manager for the client which  I developed the app is convinced the
shutdown is caused by CPU load. When monitoring the load the passenger
processes spike up to around 50-80% but settle down to 10% of less when
idling, at the moment they are < 1% as the load is low. The shutdown has
happened when there is not a lot of load so I'm not convinced.

I just really don't know where to start looking, could it be something
external for example the DB the app uses is running on a different
server maybe it's a network problem, what about rogue messages is that
possible, the IT manager did say he removed the firewall to the machine.
I'm sure he has something else in place, but this app has been running
fine for the last 6 months, it only started happening after the weekend
when we went live and the DB was moved from the same machine to
another.

Sorry for all the questions but I'm really stumped and don't know what
to try next.

Thanks again for your help.

Etienne.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215871,215902#msg-215902


From osa at FreeBSD.org.ru  Tue Sep 27 15:38:59 2011
From: osa at FreeBSD.org.ru (Sergey A. Osokin)
Date: Tue, 27 Sep 2011 19:38:59 +0400
Subject: Nginx and/or Passenger failing
In-Reply-To: <e8e7dbea348df2d17f59dc0533a667b6.NginxMailingListEnglish@forum.nginx.org>
References: <20110927110146.GA90093@FreeBSD.org.ru>
	<e8e7dbea348df2d17f59dc0533a667b6.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20110927153859.GB90093@FreeBSD.org.ru>

On Tue, Sep 27, 2011 at 10:03:30AM -0400, springbok wrote:
> Sergey A. Osokin Wrote:
> -------------------------------------------------------
> > On Tue, Sep 27, 2011 at 05:35:57AM -0400,
> > springbok wrote:
> > > Ok, I've got passenger loading and appearing in
> > the process list, I
> > > created a index.html file and was able to
> > display it doing a GET / but
> > > when I try GET / HTTP/1.0 it just sits there and
> > nothing happens. If I
> > > remove the index.html I get "Hello world"
> > returned, no mention of
> > > passenger.
> > 
> > So, if you got "Hello, <b>world</b> this means
> > passenger works well with
> > default installation.  Could you please check your
> > application or try to use
> > previous version of nginx and/or passenger.
> > 
> > -- 
> > Sergey A. Osokin
> > osa at FreeBSD.ORG.ru
> > osa at FreeBSD.ORG
> > 
> > _______________________________________________
> > nginx mailing list
> > nginx at nginx.org
> > http://mailman.nginx.org/mailman/listinfo/nginx
> 
> Hi Sergey,
> 
> I was using 0.8.53 when I started getting the errors and decided to
> upgrade before posting my question.
> 
> Do you believe this is an nginx or passenger problem, is nginx causing
> passenger to shutdown or the other way around?

I don't know.

> To me it looks like the
> watchdog process is getting an error from nginx and closing the
> passenger processes.

I think you should ask about your problem authors of passenger at first.

> We have 8GB of memory and 4 processors, do you think it could be memory
> related/CPU load related, I know it's difficult to say without seeing
> our app and the load it places on the system, but if you had to guess?
> It has plenty of memory available so I don't think that's a problem, the
> IT manager for the client which  I developed the app is convinced the
> shutdown is caused by CPU load. When monitoring the load the passenger
> processes spike up to around 50-80% but settle down to 10% of less when
> idling, at the moment they are < 1% as the load is low. The shutdown has
> happened when there is not a lot of load so I'm not convinced.
> 
> I just really don't know where to start looking, could it be something
> external for example the DB the app uses is running on a different
> server maybe it's a network problem, what about rogue messages is that
> possible, the IT manager did say he removed the firewall to the machine.
> I'm sure he has something else in place, but this app has been running
> fine for the last 6 months, it only started happening after the weekend
> when we went live and the DB was moved from the same machine to
> another.
> 
> Sorry for all the questions but I'm really stumped and don't know what
> to try next.

-- 
Sergey A. Osokin
osa at FreeBSD.ORG
osa at FreeBSD.ORG.ru


From antoine.bonavita at gmail.com  Tue Sep 27 15:52:44 2011
From: antoine.bonavita at gmail.com (Antoine Bonavita)
Date: Tue, 27 Sep 2011 17:52:44 +0200
Subject: how can i run/debug nginx in eclipse?
In-Reply-To: <CAOkdov+1+hdULOC47S6ArD0q2kUYjagG0fD4-mACWmV7Aqp4ag@mail.gmail.com>
References: <4E818166.1070705@300.cn>
	<CAOkdov+1+hdULOC47S6ArD0q2kUYjagG0fD4-mACWmV7Aqp4ag@mail.gmail.com>
Message-ID: <CA+FsSoKdMXKsB0=KZ39svYGGZZkJe03mKftkhY7Nb=J_Y8pj1g@mail.gmail.com>

A few extra things might be needed for debugging
(http://www.nginx-discovery.com/2011/03/day-36-nginx-debug-and-valgrind-from.html)
but you just need to import nginx as a C/C++ project with Makefile.

Hope this helps,

A.

On Tue, Sep 27, 2011 at 10:02 AM, David Yu <david.yu.ftw at gmail.com> wrote:
> Run via External tools and point to your local nginx bin dir.
>
> On Tue, Sep 27, 2011 at 3:55 PM, wangxc <wangxichun at 300.cn> wrote:
>>
>>
>> hi, all:
>>
>> I want to run/debug the nginx in Eclipse, how can i configure it ?
>>
>> thanks in advance!
>>
>> --
>>
>> best regards
>>
>> rmn190
>>
>> _______________________________________________
>> nginx mailing list
>> nginx at nginx.org
>> http://mailman.nginx.org/mailman/listinfo/nginx
>
>
>
> --
> When the cat is away, the mouse is alone.
> - David Yu
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx
>


From contact at jpluscplusm.com  Tue Sep 27 22:23:47 2011
From: contact at jpluscplusm.com (Jonathan Matthews)
Date: Tue, 27 Sep 2011 23:23:47 +0100
Subject: Override proxy's incorrect Content-Type via mime settings
In-Reply-To: <CAKsTx7Abwp3e+BpLOXtXMPe-oO2n_HdBpLOPoCqj6ymw5qs-WQ@mail.gmail.com>
References: <CAKsTx7Abwp3e+BpLOXtXMPe-oO2n_HdBpLOPoCqj6ymw5qs-WQ@mail.gmail.com>
Message-ID: <CAKsTx7BK1r-qhAXfni7Tg3+h7XR9PGqPLLnOVnPA=P5yGT_gWQ@mail.gmail.com>

Just a quick thread bump to see if anyone has any ideas ...

On 23 September 2011 17:37, Jonathan Matthews <contact at jpluscplusm.com> wrote:
> Hi all -
>
> I have an upstream proxy that sends the wrong Content-Type header for
> a load of images. Sometime it doesn't send any header at all.
>
> I'm trying to get Nginx not only to remove the upstream's Content-Type
> header ("proxy_hide_header"), but also to insert the correct header.
> Unfortunately, I can't hard-code the value as there are a variety of
> image formats being used. "proxy_hide_header" doesn't seem to do
> anything other than block the incorrect value, and
> "proxy_ignore_headers" won't accept "Content-Type".
>
> How can I both remove upstream's incorrect header, and also add the
> header than Nginx *would* have used if it were serving the content
> from local disk?
> In other words, how do I add the header with a value derived from the
> MIME settings currently configured, via the requested file's
> extension?
>
> Version 0.7.67, but can upgrade if required.
>
> Many thanks,
> Jonathan
> --
> Jonathan Matthews
> London, UK
> http://www.jpluscplusm.com/contact.html
>


From abioy.sun at gmail.com  Tue Sep 27 22:42:08 2011
From: abioy.sun at gmail.com (Abioy Sun)
Date: Wed, 28 Sep 2011 06:42:08 +0800
Subject: how can i run/debug nginx in eclipse?
In-Reply-To: <4E818166.1070705@300.cn>
References: <4E818166.1070705@300.cn>
Message-ID: <CAM6b-vk1yYzZZkVr9-wnDoNMzOwmRDofArOFB4PAzmHmBBjxgA@mail.gmail.com>

in gdb, you may need to 'set follow-fork-mode child' or detach the worker's
pid
>
>
> hi, all:
>
> I want to run/debug the nginx in Eclipse, how can i configure it ?
>
> thanks in advance!
>
> --
>
> best regards
>
> rmn190
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110928/7d5d5e65/attachment.html>

From abioy.sun at gmail.com  Tue Sep 27 22:48:21 2011
From: abioy.sun at gmail.com (Abioy Sun)
Date: Wed, 28 Sep 2011 06:48:21 +0800
Subject: how can i run/debug nginx in eclipse?
In-Reply-To: <CAM6b-vk1yYzZZkVr9-wnDoNMzOwmRDofArOFB4PAzmHmBBjxgA@mail.gmail.com>
References: <4E818166.1070705@300.cn>
	<CAM6b-vk1yYzZZkVr9-wnDoNMzOwmRDofArOFB4PAzmHmBBjxgA@mail.gmail.com>
Message-ID: <CAM6b-v=N3kPRPs4vA3TZCHAOfeiLJTW8QwOXwoSuR2aYDr65yw@mail.gmail.com>

> in gdb, you may need to 'set follow-fork-mode child' or detach the
worker's pid
sorry, should be 'attach the worker'
>>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110928/08635afd/attachment.html>

From peter.a.portante at gmail.com  Wed Sep 28 01:30:40 2011
From: peter.a.portante at gmail.com (Peter Portante)
Date: Tue, 27 Sep 2011 21:30:40 -0400
Subject: Small bug in rewrite rules using if (ugh, yes, if is bad ...)
In-Reply-To: <20110831155233.GZ1137@mdounin.ru>
References: <CAPzLFkWx7NmYP1rayJZcrRc9yGmtqUCTB7q_N2TybM0j=RDZ_w@mail.gmail.com>
	<20110831155233.GZ1137@mdounin.ru>
Message-ID: <CAPzLFkWUxDKF0_K=RadZjgkzGjrw8BGXOFHnG2A5yLgdg44wkA@mail.gmail.com>

On Wed, Aug 31, 2011 at 11:52 AM, Maxim Dounin <mdounin at mdounin.ru> wrote:

> Hello!
>
> On Wed, Aug 31, 2011 at 11:26:29AM -0400, Peter Portante wrote:
>
> > Hello folks,
> >
> > Just thought I would report a small bug with rewrite interacting with if.
> >
> > Here is my simple setup:
> >
> > server  {
> >     server_name     example.dev.com;
> >     location    ^~  /salad/add {
> >         if ($arg_paramThree = "foobar") {
> >             rewrite     ^
> > https://example.this.com?PARAM_ONE=$arg_paramOne&PARAM_TWO=$arg_paramTwo
> ?
> > redirect;
> >         }
> >         proxy_pass      http://127.0.0.1:8080$request_uri;
> >     }
> > }
> >
> > So when I hit the server with this URL:
> >
> >
> http://example.dev.com/salad/add?paramThree=foobar&paramTwo=two&paramOne=one
> >
> >
> > It is rewritten to be:
> >
> > https://example.this.com?PARAM_ONE=one&PARAM_TWO=two?paramThree=foobar
> >
> >
> > If I use the following if conditional, the behavior is as expected:
> >
> > if ($args ~ "paramThree=foobar") {
>
> Works ok here.  Which nginx version do you use?
>

1.0.4

thanks for taking the time to check it out.


>
> Maxim Dounin
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110927/1992fb6e/attachment.html>

From nginx-forum at nginx.us  Wed Sep 28 02:56:40 2011
From: nginx-forum at nginx.us (baalchina)
Date: Tue, 27 Sep 2011 22:56:40 -0400
Subject: =?UTF-8?Q?why_my_php_file_in_subdirectory_doesn=27t_work=EF=BC=9F?=
Message-ID: <b7f93170ac1488440681bdfe7111f351.NginxMailingListEnglish@forum.nginx.org>

Hello everyone?

I am working on an centos6 with nginx 1.0.6, php 5.3.6, and fpm.

I found that php file in my subdirectory returns a 404,but in nginx home
direcory works fine.

For example , a phpinfo.php file works fine in /data/web/, where
/data/web is my nginx document root. Buy the same file in /data/web/sub,
the browser and nginx both tell a 404 error.

Here is my nginx config file:


    server {
        listen       80;
        server_name  abc.example.com;

        location / {
            root   /data/web/;
            index  index.html index.htm index.php;

    location ~ .*\.(php|php5)?$
        {
        fastcgi_pass  127.0.0.1:9000;
        fastcgi_index index.php;
        include fastcgi.conf;
        fastcgi_connect_timeout 300;
        fastcgi_send_timeout 300;
        fastcgi_read_timeout 300;
        fastcgi_buffer_size 64k;
        fastcgi_buffers 4 64;
        fastcgi_busy_buffers_size 128k;
        fastcgi_temp_file_write_size 128k;
        }
        }
        }


btw, static file such as html works fine.

Thans a lot.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215921,215921#msg-215921


From simohayha.bobo at gmail.com  Wed Sep 28 03:03:10 2011
From: simohayha.bobo at gmail.com (Simon Liu)
Date: Wed, 28 Sep 2011 11:03:10 +0800
Subject: =?UTF-8?Q?Re=3A_why_my_php_file_in_subdirectory_doesn=27t_work=EF=BC=9F?=
In-Reply-To: <b7f93170ac1488440681bdfe7111f351.NginxMailingListEnglish@forum.nginx.org>
References: <b7f93170ac1488440681bdfe7111f351.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <CAB6pCs8A61N3vGp4ga5=7KvioxS4S54EXs-0Fc307ceL1nNOWA@mail.gmail.com>

Hello.

You shoude set (fastcgi_param  SCRIPT_FILENAME) .

example:
 location ~ .*\.(php|php5)?$
       {
  fastcgi_param  SCRIPT_FILENAME /data/web/$1;

       }


On Wed, Sep 28, 2011 at 10:56 AM, baalchina <nginx-forum at nginx.us> wrote:

> Hello everyone?
>
> I am working on an centos6 with nginx 1.0.6, php 5.3.6, and fpm.
>
> I found that php file in my subdirectory returns a 404,but in nginx home
> direcory works fine.
>
> For example , a phpinfo.php file works fine in /data/web/, where
> /data/web is my nginx document root. Buy the same file in /data/web/sub,
> the browser and nginx both tell a 404 error.
>
> Here is my nginx config file:
>
>
>    server {
>        listen       80;
>        server_name  abc.example.com;
>
>        location / {
>            root   /data/web/;
>            index  index.html index.htm index.php;
>
>    location ~ .*\.(php|php5)?$
>        {
>        fastcgi_pass  127.0.0.1:9000;
>        fastcgi_index index.php;
>        include fastcgi.conf;
>        fastcgi_connect_timeout 300;
>        fastcgi_send_timeout 300;
>        fastcgi_read_timeout 300;
>        fastcgi_buffer_size 64k;
>        fastcgi_buffers 4 64;
>        fastcgi_busy_buffers_size 128k;
>        fastcgi_temp_file_write_size 128k;
>        }
>        }
>        }
>
>
> btw, static file such as html works fine.
>
> Thans a lot.
>
> Posted at Nginx Forum:
> http://forum.nginx.org/read.php?2,215921,215921#msg-215921
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx




-- 
douban:www.douban.com/people/mustang/

blog: www.pagefault.info

twitter: www.twitter.com/minibobo

weibo:  www.weibo.com/diaoliang
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110928/3046b9fe/attachment.html>

From simohayha.bobo at gmail.com  Wed Sep 28 03:48:28 2011
From: simohayha.bobo at gmail.com (Simon Liu)
Date: Wed, 28 Sep 2011 11:48:28 +0800
Subject: Static Files
In-Reply-To: <CAN-ZXpt10N9m_garKuxC35yFo5JHTgNSkKa0yMZnis2KYfu+Pw@mail.gmail.com>
References: <CAN-ZXpt10N9m_garKuxC35yFo5JHTgNSkKa0yMZnis2KYfu+Pw@mail.gmail.com>
Message-ID: <CAB6pCs8cHaZXfyMewGAYE0BM9M8E1ECZeg6syXLuOiPKa2NRcQ@mail.gmail.com>

1 output_buffers

because static file used sendfile in nginx, and  if you need modify file
content, you need read file content to memory(example gzip filter), this
memory size is output_buffers.

first arg is buf number, second is buf size. if allocated memory greater
than number*size, nginx will send buf  and  then continue read;

2 aio

yes.

3  sendfile

first,  sendfile have limit(2g - 1) in nginx, if file size is  greater than
limit, nginx will send partially everytime. second, because fd is NONBLOCK,
so server is not block.

On Tue, Sep 27, 2011 at 7:36 PM, Hagai Avrahami <hagaia at qwilt.com> wrote:

> Hi
>
> I am intend to use NGINX (LINUX)  for serving static files with high
> concurrency and bandwidth.
> Some of the files is served from the disk and some from cache.
> I am trying to find the configuration for best performance.
>
> I have some questions regarding NGINX configuration.
>
> 1. *output_buffers 1 32k*
> ----------------------------------------
> This directive excepts 2 inputs
> can someone explain me the meaning of each input
> I can see that 32K is size of data (per fetch) when reading from a DIsk -
> what else it influences?
> When serving big files, best Disk utilization is in blocks of 1MB, is it OK
> to change this directive to 1024k?
>
> 2. *aio*
> ----------------------------------------
> **I read in Wiki the following
> *" This directive is usable as of Linux kernel 2.6.22. For Linux it is
> required to use directio, this automatically disables sendfile support."*
> when using AIO I must use *directio on*? (when testing I could see the
> file is served from the cache - which means it is not using O_DIRECT)
>
>
> 3. *sendfile
> *----------------------------------------
> Is NGINX try to send a file using 1 sendfile call (count = file size)?
> If so when trying to send file residing only on the Disk, the server is
> blocked till end of transmition?
> In this model serving multiple files some on the disk and some cached is
> inefficient?
>
> Will be happy to get any answer or opinion
>
> Thanks
> Hagai
>
>
>
>
>
> --
>
> *Hagai Avrahami*
> Qwilt | Work: +972-72-2221644| Mobile: +972-54-4895656 | hagaia at qwilt.com<yoav at qwilt.com>
>
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx
>



-- 
douban:www.douban.com/people/mustang/

blog: www.pagefault.info

twitter: www.twitter.com/minibobo

weibo:  www.weibo.com/diaoliang
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110928/3df958c0/attachment.html>

From nbubingo at gmail.com  Wed Sep 28 08:11:54 2011
From: nbubingo at gmail.com (=?GB2312?B?0qbOsLHz?=)
Date: Wed, 28 Sep 2011 16:11:54 +0800
Subject: Nginx Substitutions Filter problem
In-Reply-To: <4E81701D.408@martronic.ch>
References: <4E80E187.6030807@ntlworld.com>
	<CAJCGDi8ksNcSHyf-iLm5vZevAKBMew_M2NvA5s62_wyGRLGtGw@mail.gmail.com>
	<4E80EB49.9000108@ianhobson.co.uk>
	<CAJCGDi_jWABeVRNZUEeKmZ9q3cpdFOsRWgH9ASiU_=GNSycdBQ@mail.gmail.com>
	<20110927045203.20830@gmx.net> <4E81701D.408@martronic.ch>
Message-ID: <CAGxuLK3+sVfD8WC=9gL6z0daokOB=qc8HaE3drqsxSgbocS6=A@mail.gmail.com>

Can you show me your source page?

Thanks.

2011/9/27 Martronic SA <martronic at martronic.ch>

> Hello,
>
> We are using the "Substitutions Filter module"
> to replace html content in our pages.
>
> Our configuration is :
>
> location /{
> subs_filter_types text/html text/xml;
> subs_filter 'http://bale.hopla.net/bons-**plans<http://bale.hopla.net/bons-plans>'
> 'http://bons-plans.hopla.net';
> subs_filter 'id="portal-logo" href="http://bale.hopla.net"'
> 'id="portal-logo" href="http://www.hopla.net"';
> }
>
> The first substitution is working, but the second not.
> Are we missing something? Is the "space" supported in substitution module?
>
> Thank you for your help!
>
> Martronic SA
>
> nginx -V
> nginx: nginx version: nginx/1.1.4
> nginx: built by gcc 4.3.2 (Debian 4.3.2-1.1)
> nginx: TLS SNI support enabled
> nginx: configure arguments: --conf-path=/etc/nginx/nginx.**conf
> --error-log-path=/var/log/**nginx/error.log --pid-path=/var/run/nginx.pid
> --lock-path=/var/lock/nginx.**lock --http-log-path=/var/log/**
> nginx/access.log
> --with-http_dav_module --http-client-body-temp-path=/**var/lib/nginx/body
> --with-http_ssl_module
> --http-proxy-temp-path=/var/**lib/nginx/proxy
> --with-http_stub_status_module
> --http-fastcgi-temp-path=/var/**lib/nginx/fastcgi
> --add-module=/var/lib/**substitutions4nginx-read-only
>
>
>
> ______________________________**_________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/**mailman/listinfo/nginx<http://mailman.nginx.org/mailman/listinfo/nginx>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110928/f5cd1241/attachment.html>

From mdounin at mdounin.ru  Wed Sep 28 08:25:39 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Wed, 28 Sep 2011 12:25:39 +0400
Subject: Small bug in rewrite rules using if (ugh, yes, if is bad ...)
In-Reply-To: <CAPzLFkWUxDKF0_K=RadZjgkzGjrw8BGXOFHnG2A5yLgdg44wkA@mail.gmail.com>
References: <CAPzLFkWx7NmYP1rayJZcrRc9yGmtqUCTB7q_N2TybM0j=RDZ_w@mail.gmail.com>
	<20110831155233.GZ1137@mdounin.ru>
	<CAPzLFkWUxDKF0_K=RadZjgkzGjrw8BGXOFHnG2A5yLgdg44wkA@mail.gmail.com>
Message-ID: <20110928082539.GV1137@mdounin.ru>

Hello!

On Tue, Sep 27, 2011 at 09:30:40PM -0400, Peter Portante wrote:

> On Wed, Aug 31, 2011 at 11:52 AM, Maxim Dounin <mdounin at mdounin.ru> wrote:
> 
> > Hello!
> >
> > On Wed, Aug 31, 2011 at 11:26:29AM -0400, Peter Portante wrote:
> >
> > > Hello folks,
> > >
> > > Just thought I would report a small bug with rewrite interacting with if.
> > >
> > > Here is my simple setup:
> > >
> > > server  {
> > >     server_name     example.dev.com;
> > >     location    ^~  /salad/add {
> > >         if ($arg_paramThree = "foobar") {
> > >             rewrite     ^
> > > https://example.this.com?PARAM_ONE=$arg_paramOne&PARAM_TWO=$arg_paramTwo
> > ?
> > > redirect;
> > >         }
> > >         proxy_pass      http://127.0.0.1:8080$request_uri;
> > >     }
> > > }
> > >
> > > So when I hit the server with this URL:
> > >
> > >
> > http://example.dev.com/salad/add?paramThree=foobar&paramTwo=two&paramOne=one
> > >
> > >
> > > It is rewritten to be:
> > >
> > > https://example.this.com?PARAM_ONE=one&PARAM_TWO=two?paramThree=foobar
> > >
> > >
> > > If I use the following if conditional, the behavior is as expected:
> > >
> > > if ($args ~ "paramThree=foobar") {
> >
> > Works ok here.  Which nginx version do you use?
> >
> 
> 1.0.4
> 
> thanks for taking the time to check it out.

Could you please provide debug log?  Please see 
http://wiki.nginx.org/Debugging for details.

Maxim Dounin


From cyril.lavier at davromaniak.eu  Wed Sep 28 08:39:26 2011
From: cyril.lavier at davromaniak.eu (Cyril LAVIER)
Date: Wed, 28 Sep 2011 10:39:26 +0200
Subject: Nginx build under Debian Hurd i386
Message-ID: <bc6dd4f929b101b45190afc5d657d2f1@davromaniak.eu>

  

Hi nginx lovers. 

As it's my first mail on this mailing list, I
will introduce myself a little bit. 

I'm Cyril Lavier, a 25 years old
System Administrator from France. I use nginx for barely a year now. I
use it for my own website, and also for the company I'm working. 

Now,
let's work :). 

For the Debian packaging purposes, we try to make nginx
build properly under the hurd-i386 kernel. 

But it never built well,
and Kartik sent a mail in January about this. 

Now, as I'm contributing
to this packaging, I tried to figure out why nginx build failed. 

For
this, I opened a ticket (http://trac.nginx.org/nginx/ticket/22), which
provide a patch for this. With this patch it's building well and working
well. 

After, we add the MP4 module to the nginx-extras package, but
the module failed to compile under hurd-i386 (the build works like a
charm on other architectures). 

I opened the ticket #23
(http://trac.nginx.org/nginx/ticket/23), with a first draft of a patch,
which resolves one issue. By the way, my C coding knowledge is too low
to completely correct this issue. 

We would like the hurd package being
the same as the others, so if the developers (or other contributors with
sufficient knowledge) can take a look at both tickets, and judging about
a patch inclusion for #22 and a more complete patch #23, this would be
very appreciated. 

Thanks. 

-- 
Cyril "Davromaniak" Lavier
  
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110928/5df82189/attachment-0001.html>

From martronic at martronic.ch  Wed Sep 28 09:42:29 2011
From: martronic at martronic.ch (Martronic SA)
Date: Wed, 28 Sep 2011 11:42:29 +0200 (CEST)
Subject: Nginx Substitutions Filter problem
In-Reply-To: <CAGxuLK3+sVfD8WC=9gL6z0daokOB=qc8HaE3drqsxSgbocS6=A@mail.gmail.com>
References: <4E80E187.6030807@ntlworld.com>
	<CAJCGDi8ksNcSHyf-iLm5vZevAKBMew_M2NvA5s62_wyGRLGtGw@mail.gmail.com>
	<4E80EB49.9000108@ianhobson.co.uk>
	<CAJCGDi_jWABeVRNZUEeKmZ9q3cpdFOsRWgH9ASiU_=GNSycdBQ@mail.gmail.com>
	<20110927045203.20830@gmx.net> <4E81701D.408@martronic.ch>
	<CAGxuLK3+sVfD8WC=9gL6z0daokOB=qc8HaE3drqsxSgbocS6=A@mail.gmail.com>
Message-ID: <7c0b111efe8055c61d650f7706356751.squirrel@webmail.martronic.ch>


full source attachment
Thanks.


<a id="portal-logo" href="http://bale.hopla.net" accesskey="1" class="visualNoPrint"><img
src="http://bale.hopla.net/portal_logo/image" alt="Logo" title="Logo Hopla-Net" height="175"
width="230" /></a>

<div id="solgemabandeau" class="visualNoPrint">

  <div id="portal-bandeau">
    <div id="inner-portal-bandeau" class="bannersContainer">

          <div style="position:relative; display: block" id="carousel-banner-0"
class="bando.jpg carousel-banner selected">
            <div style="height:200px; width:100%; background:transparent
url(http://bale.hopla.net/bando.jpg) no-repeat scroll left top;" title="Une maison
dans un jardin, nouveau bandeau Hopla pour le printemps."
class="bandeau_image"></div>
          </div>


          <div style="display: none" id="carousel-banner-1" class="bandeauhoplabale.jpg
carousel-banner">
            <div style="height:200px; width:100%; background:transparent
url(http://bale.hopla.net/bandeauhoplabale.jpg) no-repeat scroll left top;"
title="Le printemps est de retour sur St Louis et B?le"
class="bandeau_image"></div>
          </div>

        <span class="bandeauTopLeft"></span>

        <span class="bandeauTopRight"></span>
        <span class="bandeauBottomLeft"></span>
        <span class="bandeauBottomRight"></span>
    </div>
</div>

</div>


> Can you show me your source page?
>
> Thanks.
>
> 2011/9/27 Martronic SA <martronic at martronic.ch>
>
>> Hello,
>>
>> We are using the "Substitutions Filter module"
>> to replace html content in our pages.
>>
>> Our configuration is :
>>
>> location /{
>> subs_filter_types text/html text/xml;
>> subs_filter 'http://bale.hopla.net/bons-**plans<http://bale.hopla.net/bons-plans>'
>> 'http://bons-plans.hopla.net';
>> subs_filter 'id="portal-logo" href="http://bale.hopla.net"'
>> 'id="portal-logo" href="http://www.hopla.net"';
>> }
>>
>> The first substitution is working, but the second not.
>> Are we missing something? Is the "space" supported in substitution module?
>>
>> Thank you for your help!
>>
>> Martronic SA
>>
>> nginx -V
>> nginx: nginx version: nginx/1.1.4
>> nginx: built by gcc 4.3.2 (Debian 4.3.2-1.1)
>> nginx: TLS SNI support enabled
>> nginx: configure arguments: --conf-path=/etc/nginx/nginx.**conf
>> --error-log-path=/var/log/**nginx/error.log --pid-path=/var/run/nginx.pid
>> --lock-path=/var/lock/nginx.**lock --http-log-path=/var/log/**
>> nginx/access.log
>> --with-http_dav_module --http-client-body-temp-path=/**var/lib/nginx/body
>> --with-http_ssl_module
>> --http-proxy-temp-path=/var/**lib/nginx/proxy
>> --with-http_stub_status_module
>> --http-fastcgi-temp-path=/var/**lib/nginx/fastcgi
>> --add-module=/var/lib/**substitutions4nginx-read-only
>>
>>
>>
>> ______________________________**_________________
>> nginx mailing list
>> nginx at nginx.org
>> http://mailman.nginx.org/**mailman/listinfo/nginx<http://mailman.nginx.org/mailman/listinfo/nginx>
>>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx


Avec nos meilleures salutations,

MARTRONIC SA

De votre site internet

? votre serveur d?applications
Venez visiter notre nouvelle plateforme collaborative Web 2 de Martronic
http://www.solgema.com
http://www.solgema.ch
http://www.martronic.ch

Rue du Ch?teau-Vieux 12
Case postale 186
1870 Monthey VS
Suisse / Switzerland

T?l +41 24 471 51 07
Fax +41 24 471 58 47
e-mail : martronic at martronic.ch


From martronic at martronic.ch  Wed Sep 28 09:43:54 2011
From: martronic at martronic.ch (Martronic SA)
Date: Wed, 28 Sep 2011 11:43:54 +0200 (CEST)
Subject: Nginx Substitutions Filter problem
In-Reply-To: <CAGxuLK3+sVfD8WC=9gL6z0daokOB=qc8HaE3drqsxSgbocS6=A@mail.gmail.com>
References: <4E80E187.6030807@ntlworld.com>
	<CAJCGDi8ksNcSHyf-iLm5vZevAKBMew_M2NvA5s62_wyGRLGtGw@mail.gmail.com>
	<4E80EB49.9000108@ianhobson.co.uk>
	<CAJCGDi_jWABeVRNZUEeKmZ9q3cpdFOsRWgH9ASiU_=GNSycdBQ@mail.gmail.com>
	<20110927045203.20830@gmx.net> <4E81701D.408@martronic.ch>
	<CAGxuLK3+sVfD8WC=9gL6z0daokOB=qc8HaE3drqsxSgbocS6=A@mail.gmail.com>
Message-ID: <2a4c6102eba7e97f63ad316476db06b8.squirrel@webmail.martronic.ch>

full source attachment
Thanks.


<a id="portal-logo" href="http://bale.hopla.net" accesskey="1" class="visualNoPrint"><img
src="http://bale.hopla.net/portal_logo/image" alt="Logo" title="Logo Hopla-Net" height="175"
width="230" /></a>

<div id="solgemabandeau" class="visualNoPrint">

  <div id="portal-bandeau">
    <div id="inner-portal-bandeau" class="bannersContainer">

          <div style="position:relative; display: block" id="carousel-banner-0"
class="bando.jpg carousel-banner selected">
            <div style="height:200px; width:100%; background:transparent
url(http://bale.hopla.net/bando.jpg) no-repeat scroll left top;" title="Une maison
dans un jardin, nouveau bandeau Hopla pour le printemps."
class="bandeau_image"></div>
          </div>


          <div style="display: none" id="carousel-banner-1" class="bandeauhoplabale.jpg
carousel-banner">
            <div style="height:200px; width:100%; background:transparent
url(http://bale.hopla.net/bandeauhoplabale.jpg) no-repeat scroll left top;"
title="Le printemps est de retour sur St Louis et B?le"
class="bandeau_image"></div>
          </div>

        <span class="bandeauTopLeft"></span>

        <span class="bandeauTopRight"></span>
        <span class="bandeauBottomLeft"></span>
        <span class="bandeauBottomRight"></span>
    </div>
</div>

</div>

> Can you show me your source page?
>
> Thanks.
>
> 2011/9/27 Martronic SA <martronic at martronic.ch>
>
>> Hello,
>>
>> We are using the "Substitutions Filter module"
>> to replace html content in our pages.
>>
>> Our configuration is :
>>
>> location /{
>> subs_filter_types text/html text/xml;
>> subs_filter 'http://bale.hopla.net/bons-**plans<http://bale.hopla.net/bons-plans>'
>> 'http://bons-plans.hopla.net';
>> subs_filter 'id="portal-logo" href="http://bale.hopla.net"'
>> 'id="portal-logo" href="http://www.hopla.net"';
>> }
>>
>> The first substitution is working, but the second not.
>> Are we missing something? Is the "space" supported in substitution module?
>>
>> Thank you for your help!
>>
>> Martronic SA
>>
>> nginx -V
>> nginx: nginx version: nginx/1.1.4
>> nginx: built by gcc 4.3.2 (Debian 4.3.2-1.1)
>> nginx: TLS SNI support enabled
>> nginx: configure arguments: --conf-path=/etc/nginx/nginx.**conf
>> --error-log-path=/var/log/**nginx/error.log --pid-path=/var/run/nginx.pid
>> --lock-path=/var/lock/nginx.**lock --http-log-path=/var/log/**
>> nginx/access.log
>> --with-http_dav_module --http-client-body-temp-path=/**var/lib/nginx/body
>> --with-http_ssl_module
>> --http-proxy-temp-path=/var/**lib/nginx/proxy
>> --with-http_stub_status_module
>> --http-fastcgi-temp-path=/var/**lib/nginx/fastcgi
>> --add-module=/var/lib/**substitutions4nginx-read-only
>>
>>
>>
>> ______________________________**_________________
>> nginx mailing list
>> nginx at nginx.org
>> http://mailman.nginx.org/**mailman/listinfo/nginx<http://mailman.nginx.org/mailman/listinfo/nginx>
>>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx


Avec nos meilleures salutations,

MARTRONIC SA

De votre site internet

? votre serveur d?applications
Venez visiter notre nouvelle plateforme collaborative Web 2 de Martronic
http://www.solgema.com
http://www.solgema.ch
http://www.martronic.ch

Rue du Ch?teau-Vieux 12
Case postale 186
1870 Monthey VS
Suisse / Switzerland

T?l +41 24 471 51 07
Fax +41 24 471 58 47
e-mail : martronic at martronic.ch
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110928/62a7d918/attachment-0001.htm>

From antoine.bonavita at gmail.com  Wed Sep 28 10:20:21 2011
From: antoine.bonavita at gmail.com (Antoine Bonavita)
Date: Wed, 28 Sep 2011 12:20:21 +0200
Subject: Happy birthday, Igor ?
Message-ID: <CA+FsSoLqNX9j8SqjWa6-1tY-_-nCak1CfoMzkZDPd8RR-QsnBA@mail.gmail.com>

Given the amount of times Sep 28th 1970 is mentioned in the nginx code
(see below), I am putting a rough guess that this is your birthday.
So, in this very special day, I wanted to thank you for your amazing
little kid and I would have loved to buy you a beer but you're a bit
far away. If you ever get near the south of France, I'll buy you one.

Of course, I could be completely wrong.

A.

nginx-HEAD]$ grep '1970' -R src
src/http/ngx_http_variables.c:
sizeof("Last-Modified: Mon, 28 Sep 1970 06:00:00 GMT") - 1);
src/http/ngx_http_parse_time.c:             * 719527 days were between
March 1, 1 BC and March 1, 1970,
src/http/ngx_http_parse_time.c:             * 31 and 28 days were in
January and February 1970
src/http/ngx_http_header_filter_module.c:        len += sizeof("Date:
Mon, 28 Sep 1970 06:00:00 GMT" CRLF) - 1;
src/http/ngx_http_header_filter_module.c:        len +=
sizeof("Last-Modified: Mon, 28 Sep 1970 06:00:00 GMT" CRLF) - 1;
src/http/modules/ngx_http_headers_filter_module.c:    len =
sizeof("Mon, 28 Sep 1970 06:00:00 GMT");
src/http/modules/ngx_http_headers_filter_module.c:
expires->value.data = (u_char *) "Thu, 01 Jan 1970 00:00:01 GMT";
src/http/modules/ngx_http_log_module.c:    { ngx_string("time_local"),
sizeof("28/Sep/1970:12:00:00 +0600") - 1,
src/http/modules/ngx_http_log_module.c:    {
ngx_string("time_iso8601"), sizeof("1970-09-28T12:00:00+06:00") - 1,
src/http/modules/ngx_http_autoindex_module.c:            + sizeof("
28-Sep-1970 12:00 ") - 1
src/core/ngx_times.c:
[sizeof("1970/09/28 12:00:00")];
src/core/ngx_times.c:                                    [sizeof("Mon,
28 Sep 1970 06:00:00 GMT")];
src/core/ngx_times.c:
[sizeof("28/Sep/1970:12:00:00 +0600")];
src/core/ngx_times.c:
[sizeof("1970-09-28T12:00:00+06:00")];
src/core/ngx_times.c:    ngx_cached_err_log_time.len =
sizeof("1970/09/28 12:00:00") - 1;
src/core/ngx_times.c:    ngx_cached_http_time.len = sizeof("Mon, 28
Sep 1970 06:00:00 GMT") - 1;
src/core/ngx_times.c:    ngx_cached_http_log_time.len =
sizeof("28/Sep/1970:12:00:00 +0600") - 1;
src/core/ngx_times.c:    ngx_cached_http_log_iso8601.len =
sizeof("1970-09-28T12:00:00+06:00") - 1;
src/core/ngx_times.c:    /* Jaunary 1, 1970 was Thursday */
src/os/win32/ngx_time.c:     * Between January 1, 1970 (Epoch) and
January 1, 1601 there were


From nginx-forum at nginx.us  Wed Sep 28 11:25:58 2011
From: nginx-forum at nginx.us (locojohn)
Date: Wed, 28 Sep 2011 07:25:58 -0400
Subject: $https variable?
In-Reply-To: <CAKa35F5GpaaoVupeK9rB3YmxjGLsGsYO5NHC=+5Xp2ntWxBYzg@mail.gmail.com>
References: <CAKa35F5GpaaoVupeK9rB3YmxjGLsGsYO5NHC=+5Xp2ntWxBYzg@mail.gmail.com>
Message-ID: <7473606372f371777d0c7915f4012213.NginxMailingListEnglish@forum.nginx.org>

Unfortunately, PHP code in many known packages (such as Moodle)
sometimes relies on the wrong assumption:

$HTTPS = isset($HTTP_SERVER_VARS['HTTPS']) ? $HTTP_SERVER_VARS['HTTPS']
: 'off';

Because in Apache HTTPS variable is never submitted at all in http mode
and coders believe this is enough information to check.  So, is there a
way to never submit HTTPS=off at all if HTTPS is not present?

Andrejs

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215756,215941#msg-215941


From igor at sysoev.ru  Wed Sep 28 11:35:54 2011
From: igor at sysoev.ru (Igor Sysoev)
Date: Wed, 28 Sep 2011 15:35:54 +0400
Subject: Happy birthday, Igor ?
In-Reply-To: <CA+FsSoLqNX9j8SqjWa6-1tY-_-nCak1CfoMzkZDPd8RR-QsnBA@mail.gmail.com>
References: <CA+FsSoLqNX9j8SqjWa6-1tY-_-nCak1CfoMzkZDPd8RR-QsnBA@mail.gmail.com>
Message-ID: <20110928113554.GF80782@nginx.com>

On Wed, Sep 28, 2011 at 12:20:21PM +0200, Antoine Bonavita wrote:
> Given the amount of times Sep 28th 1970 is mentioned in the nginx code
> (see below), I am putting a rough guess that this is your birthday.
> So, in this very special day, I wanted to thank you for your amazing
> little kid and I would have loved to buy you a beer but you're a bit
> far away. If you ever get near the south of France, I'll buy you one.
> 
> Of course, I could be completely wrong.

You are right, this is my birthday. Thank you.


-- 
Igor Sysoev


From mike503 at gmail.com  Wed Sep 28 11:41:05 2011
From: mike503 at gmail.com (Michael Shadle)
Date: Wed, 28 Sep 2011 04:41:05 -0700
Subject: Happy birthday, Igor ?
In-Reply-To: <20110928113554.GF80782@nginx.com>
References: <CA+FsSoLqNX9j8SqjWa6-1tY-_-nCak1CfoMzkZDPd8RR-QsnBA@mail.gmail.com>
	<20110928113554.GF80782@nginx.com>
Message-ID: <58847010-7129-4E97-A293-FA95987BE07D@gmail.com>

Happy birthday Igor :) nice find Antoine!

On Sep 28, 2011, at 4:35 AM, Igor Sysoev <igor at sysoev.ru> wrote:

> On Wed, Sep 28, 2011 at 12:20:21PM +0200, Antoine Bonavita wrote:
>> Given the amount of times Sep 28th 1970 is mentioned in the nginx code
>> (see below), I am putting a rough guess that this is your birthday.
>> So, in this very special day, I wanted to thank you for your amazing
>> little kid and I would have loved to buy you a beer but you're a bit
>> far away. If you ever get near the south of France, I'll buy you one.
>> 
>> Of course, I could be completely wrong.
> 
> You are right, this is my birthday. Thank you.
> 
> 
> -- 
> Igor Sysoev
> 
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx


From smallfish.xy at gmail.com  Wed Sep 28 11:43:08 2011
From: smallfish.xy at gmail.com (smallfish)
Date: Wed, 28 Sep 2011 19:43:08 +0800
Subject: Happy birthday, Igor ?
In-Reply-To: <58847010-7129-4E97-A293-FA95987BE07D@gmail.com>
References: <CA+FsSoLqNX9j8SqjWa6-1tY-_-nCak1CfoMzkZDPd8RR-QsnBA@mail.gmail.com>
	<20110928113554.GF80782@nginx.com>
	<58847010-7129-4E97-A293-FA95987BE07D@gmail.com>
Message-ID: <CACPpB2wU5ZfF5egtc2g2g1COWP3FLK6La8Nwsw0eWN70i44DKg@mail.gmail.com>

happy birthday~~~
--
blog: http://chenxiaoyu.org



On Wed, Sep 28, 2011 at 7:41 PM, Michael Shadle <mike503 at gmail.com> wrote:

> Happy birthday Igor :) nice find Antoine!
>
> On Sep 28, 2011, at 4:35 AM, Igor Sysoev <igor at sysoev.ru> wrote:
>
> > On Wed, Sep 28, 2011 at 12:20:21PM +0200, Antoine Bonavita wrote:
> >> Given the amount of times Sep 28th 1970 is mentioned in the nginx code
> >> (see below), I am putting a rough guess that this is your birthday.
> >> So, in this very special day, I wanted to thank you for your amazing
> >> little kid and I would have loved to buy you a beer but you're a bit
> >> far away. If you ever get near the south of France, I'll buy you one.
> >>
> >> Of course, I could be completely wrong.
> >
> > You are right, this is my birthday. Thank you.
> >
> >
> > --
> > Igor Sysoev
> >
> > _______________________________________________
> > nginx mailing list
> > nginx at nginx.org
> > http://mailman.nginx.org/mailman/listinfo/nginx
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110928/1fc25a5b/attachment.html>

From nbubingo at gmail.com  Wed Sep 28 11:48:00 2011
From: nbubingo at gmail.com (=?GB2312?B?0qbOsLHz?=)
Date: Wed, 28 Sep 2011 19:48:00 +0800
Subject: Happy birthday, Igor ?
In-Reply-To: <CACPpB2wU5ZfF5egtc2g2g1COWP3FLK6La8Nwsw0eWN70i44DKg@mail.gmail.com>
References: <CA+FsSoLqNX9j8SqjWa6-1tY-_-nCak1CfoMzkZDPd8RR-QsnBA@mail.gmail.com>
	<20110928113554.GF80782@nginx.com>
	<58847010-7129-4E97-A293-FA95987BE07D@gmail.com>
	<CACPpB2wU5ZfF5egtc2g2g1COWP3FLK6La8Nwsw0eWN70i44DKg@mail.gmail.com>
Message-ID: <CAGxuLK2=SzbkZG0qd9VYUNDVXjhvtCWMiyVGsMvPdEmhOXWkLA@mail.gmail.com>

Best wishes from China[?][?].

2011/9/28 smallfish <smallfish.xy at gmail.com>

> happy birthday~~~
> --
> blog: http://chenxiaoyu.org
>
>
>
> On Wed, Sep 28, 2011 at 7:41 PM, Michael Shadle <mike503 at gmail.com> wrote:
>
>> Happy birthday Igor :) nice find Antoine!
>>
>> On Sep 28, 2011, at 4:35 AM, Igor Sysoev <igor at sysoev.ru> wrote:
>>
>> > On Wed, Sep 28, 2011 at 12:20:21PM +0200, Antoine Bonavita wrote:
>> >> Given the amount of times Sep 28th 1970 is mentioned in the nginx code
>> >> (see below), I am putting a rough guess that this is your birthday.
>> >> So, in this very special day, I wanted to thank you for your amazing
>> >> little kid and I would have loved to buy you a beer but you're a bit
>> >> far away. If you ever get near the south of France, I'll buy you one.
>> >>
>> >> Of course, I could be completely wrong.
>> >
>> > You are right, this is my birthday. Thank you.
>> >
>> >
>> > --
>> > Igor Sysoev
>> >
>> > _______________________________________________
>> > nginx mailing list
>> > nginx at nginx.org
>> > http://mailman.nginx.org/mailman/listinfo/nginx
>>
>> _______________________________________________
>> nginx mailing list
>> nginx at nginx.org
>> http://mailman.nginx.org/mailman/listinfo/nginx
>>
>
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110928/b13b0e37/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 330.gif
Type: image/gif
Size: 96 bytes
Desc: not available
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110928/b13b0e37/attachment.gif>

From mat999 at gmail.com  Wed Sep 28 11:50:42 2011
From: mat999 at gmail.com (SplitIce)
Date: Wed, 28 Sep 2011 21:50:42 +1000
Subject: Happy birthday, Igor ?
In-Reply-To: <CAGxuLK2=SzbkZG0qd9VYUNDVXjhvtCWMiyVGsMvPdEmhOXWkLA@mail.gmail.com>
References: <CA+FsSoLqNX9j8SqjWa6-1tY-_-nCak1CfoMzkZDPd8RR-QsnBA@mail.gmail.com>
	<20110928113554.GF80782@nginx.com>
	<58847010-7129-4E97-A293-FA95987BE07D@gmail.com>
	<CACPpB2wU5ZfF5egtc2g2g1COWP3FLK6La8Nwsw0eWN70i44DKg@mail.gmail.com>
	<CAGxuLK2=SzbkZG0qd9VYUNDVXjhvtCWMiyVGsMvPdEmhOXWkLA@mail.gmail.com>
Message-ID: <CALFfGYbFG2hL=aKxMJaeHyYi2sb3Mi=PvuYeqppy+DWH9NHuFA@mail.gmail.com>

Happy birthday man today and many more in the future.

On Wed, Sep 28, 2011 at 9:48 PM, ??? <nbubingo at gmail.com> wrote:

> Best wishes from China[?][?].
>
> 2011/9/28 smallfish <smallfish.xy at gmail.com>
>
>> happy birthday~~~
>> --
>> blog: http://chenxiaoyu.org
>>
>>
>>
>> On Wed, Sep 28, 2011 at 7:41 PM, Michael Shadle <mike503 at gmail.com>wrote:
>>
>>> Happy birthday Igor :) nice find Antoine!
>>>
>>> On Sep 28, 2011, at 4:35 AM, Igor Sysoev <igor at sysoev.ru> wrote:
>>>
>>> > On Wed, Sep 28, 2011 at 12:20:21PM +0200, Antoine Bonavita wrote:
>>> >> Given the amount of times Sep 28th 1970 is mentioned in the nginx code
>>> >> (see below), I am putting a rough guess that this is your birthday.
>>> >> So, in this very special day, I wanted to thank you for your amazing
>>> >> little kid and I would have loved to buy you a beer but you're a bit
>>> >> far away. If you ever get near the south of France, I'll buy you one.
>>> >>
>>> >> Of course, I could be completely wrong.
>>> >
>>> > You are right, this is my birthday. Thank you.
>>> >
>>> >
>>> > --
>>> > Igor Sysoev
>>> >
>>> > _______________________________________________
>>> > nginx mailing list
>>> > nginx at nginx.org
>>> > http://mailman.nginx.org/mailman/listinfo/nginx
>>>
>>> _______________________________________________
>>> nginx mailing list
>>> nginx at nginx.org
>>> http://mailman.nginx.org/mailman/listinfo/nginx
>>>
>>
>>
>> _______________________________________________
>> nginx mailing list
>> nginx at nginx.org
>> http://mailman.nginx.org/mailman/listinfo/nginx
>>
>
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110928/2fe26eff/attachment-0001.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 330.gif
Type: image/gif
Size: 96 bytes
Desc: not available
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110928/2fe26eff/attachment-0001.gif>

From petite.abeille at gmail.com  Wed Sep 28 11:57:57 2011
From: petite.abeille at gmail.com (Petite Abeille)
Date: Wed, 28 Sep 2011 13:57:57 +0200
Subject: Happy birthday, Igor ?
In-Reply-To: <20110928113554.GF80782@nginx.com>
References: <CA+FsSoLqNX9j8SqjWa6-1tY-_-nCak1CfoMzkZDPd8RR-QsnBA@mail.gmail.com>
	<20110928113554.GF80782@nginx.com>
Message-ID: <51A736D4-C827-4E3E-A58E-83EED5EE8BF2@gmail.com>


On Sep 28, 2011, at 1:35 PM, Igor Sysoev wrote:

> You are right, this is my birthday. Thank you.

Happy birthday! Or should we call it Igor Epoch Day! :)


From appa at perusio.net  Wed Sep 28 12:08:26 2011
From: appa at perusio.net (=?UTF-8?B?QW50w7NuaW8=?= P. P. Almeida)
Date: Wed, 28 Sep 2011 13:08:26 +0100
Subject: Happy birthday, Igor ?
In-Reply-To: <20110928113554.GF80782@nginx.com>
References: <CA+FsSoLqNX9j8SqjWa6-1tY-_-nCak1CfoMzkZDPd8RR-QsnBA@mail.gmail.com>
	<20110928113554.GF80782@nginx.com>
Message-ID: <8762kcna2t.wl%appa@perusio.net>

On 28 Set 2011 12h35 WEST, igor at sysoev.ru wrote:

Congratulations Igor, from Lisbon. Thank you for your work on Nginx. 

All the best,
--- appa


From frumentius at gmail.com  Wed Sep 28 12:11:15 2011
From: frumentius at gmail.com (Joe)
Date: Wed, 28 Sep 2011 19:11:15 +0700
Subject: Happy birthday, Igor ?
In-Reply-To: <8762kcna2t.wl%appa@perusio.net>
References: <CA+FsSoLqNX9j8SqjWa6-1tY-_-nCak1CfoMzkZDPd8RR-QsnBA@mail.gmail.com>
	<20110928113554.GF80782@nginx.com> <8762kcna2t.wl%appa@perusio.net>
Message-ID: <CAPihCJWB4YmBZvFW6CYzK8Am8pB_X0d9caM_yRQMMSWzm_tVrQ@mail.gmail.com>

We are here from Indonesia, want to say Happy Birthday to Igor.
All the Best... Always...


Regards,
Joe


On Wed, Sep 28, 2011 at 7:08 PM, Ant?nio P. P. Almeida <appa at perusio.net>wrote:

> On 28 Set 2011 12h35 WEST, igor at sysoev.ru wrote:
>
> Congratulations Igor, from Lisbon. Thank you for your work on Nginx.
>
> All the best,
> --- appa
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110928/3a0d5afa/attachment.html>

From 191919 at gmail.com  Wed Sep 28 12:12:13 2011
From: 191919 at gmail.com (191919)
Date: Wed, 28 Sep 2011 20:12:13 +0800
Subject: Happy birthday, Igor ?
In-Reply-To: <20110928113554.GF80782@nginx.com>
References: <CA+FsSoLqNX9j8SqjWa6-1tY-_-nCak1CfoMzkZDPd8RR-QsnBA@mail.gmail.com>
	<20110928113554.GF80782@nginx.com>
Message-ID: <4E830F1D.8020102@gmail.com>

On 9/28/11 7:35 PM, Igor Sysoev wrote:

Happy birthday to Igor, from Beijing, China.
> On Wed, Sep 28, 2011 at 12:20:21PM +0200, Antoine Bonavita wrote:
>> Given the amount of times Sep 28th 1970 is mentioned in the nginx code
>> (see below), I am putting a rough guess that this is your birthday.
>> So, in this very special day, I wanted to thank you for your amazing
>> little kid and I would have loved to buy you a beer but you're a bit
>> far away. If you ever get near the south of France, I'll buy you one.
>>
>> Of course, I could be completely wrong.
> You are right, this is my birthday. Thank you.
>
>
-- 
Regards,
191919


From pascal.nitsche at fansubcode.org  Wed Sep 28 12:26:31 2011
From: pascal.nitsche at fansubcode.org (Pascal Nitsche)
Date: Wed, 28 Sep 2011 14:26:31 +0200
Subject: Happy birthday, Igor ?
In-Reply-To: <20110928113554.GF80782@nginx.com>
References: <CA+FsSoLqNX9j8SqjWa6-1tY-_-nCak1CfoMzkZDPd8RR-QsnBA@mail.gmail.com>
	<20110928113554.GF80782@nginx.com>
Message-ID: <4E831277.7020400@fansubcode.org>

Happy Birthday Igor from M?lheim an der Ruhr (Germany).

Am 28.09.2011 13:35, schrieb Igor Sysoev:
> On Wed, Sep 28, 2011 at 12:20:21PM +0200, Antoine Bonavita wrote:
>> Given the amount of times Sep 28th 1970 is mentioned in the nginx code
>> (see below), I am putting a rough guess that this is your birthday.
>> So, in this very special day, I wanted to thank you for your amazing
>> little kid and I would have loved to buy you a beer but you're a bit
>> far away. If you ever get near the south of France, I'll buy you one.
>>
>> Of course, I could be completely wrong.
> You are right, this is my birthday. Thank you.
>
>


From varia at e-healthexpert.org  Wed Sep 28 12:29:05 2011
From: varia at e-healthexpert.org (Mark Alan)
Date: Wed, 28 Sep 2011 13:29:05 +0100
Subject: Happy birthday, Igor ?
In-Reply-To: <20110928113554.GF80782@nginx.com>
References: <CA+FsSoLqNX9j8SqjWa6-1tY-_-nCak1CfoMzkZDPd8RR-QsnBA@mail.gmail.com>
	<20110928113554.GF80782@nginx.com>
Message-ID: <20110928132905.75d20ee5@e-healthexpert.org>

On Wed, 28 Sep 2011 15:35:54 +0400, Igor Sysoev <igor at sysoev.ru> wrote:
> Happy birthday, Igor

Happy birthday Igor.
Keep up the good work.

M.


From usirsiwal at verivue.com  Wed Sep 28 12:44:23 2011
From: usirsiwal at verivue.com (Sirsiwal, Umesh)
Date: Wed, 28 Sep 2011 08:44:23 -0400
Subject: Happy birthday, Igor ?
In-Reply-To: <20110928132905.75d20ee5@e-healthexpert.org>
References: <CA+FsSoLqNX9j8SqjWa6-1tY-_-nCak1CfoMzkZDPd8RR-QsnBA@mail.gmail.com>
	<20110928113554.GF80782@nginx.com>,
	<20110928132905.75d20ee5@e-healthexpert.org>
Message-ID: <B293D35E9E20694A8D9DFFE1BC655C556AB265E81E@vvexch.verivue.com>

Happy birthday Igor.

-Umesh
________________________________________
From: nginx-bounces at nginx.org [nginx-bounces at nginx.org] On Behalf Of Mark Alan [varia at e-healthexpert.org]
Sent: Wednesday, September 28, 2011 8:29 AM
To: nginx at nginx.org
Subject: Re: Happy birthday, Igor ?

On Wed, 28 Sep 2011 15:35:54 +0400, Igor Sysoev <igor at sysoev.ru> wrote:
> Happy birthday, Igor

Happy birthday Igor.
Keep up the good work.

M.

_______________________________________________
nginx mailing list
nginx at nginx.org
http://mailman.nginx.org/mailman/listinfo/nginx


From gerjoo at gmail.com  Wed Sep 28 12:49:38 2011
From: gerjoo at gmail.com (Gerard Meier)
Date: Wed, 28 Sep 2011 14:49:38 +0200
Subject: Happy birthday, Igor ?
In-Reply-To: <B293D35E9E20694A8D9DFFE1BC655C556AB265E81E@vvexch.verivue.com>
References: <CA+FsSoLqNX9j8SqjWa6-1tY-_-nCak1CfoMzkZDPd8RR-QsnBA@mail.gmail.com>
	<20110928113554.GF80782@nginx.com>
	<20110928132905.75d20ee5@e-healthexpert.org>
	<B293D35E9E20694A8D9DFFE1BC655C556AB265E81E@vvexch.verivue.com>
Message-ID: <CAKjeDB+fd_0sNn-uhKREkWk0piLwStRoQQ5x_nDwJso994GeQw@mail.gmail.com>

Happy birthday from Amsterdam!

Gerard

On Wed, Sep 28, 2011 at 2:44 PM, Sirsiwal, Umesh <usirsiwal at verivue.com>wrote:

> Happy birthday Igor.
>
> -Umesh
> ________________________________________
> From: nginx-bounces at nginx.org [nginx-bounces at nginx.org] On Behalf Of Mark
> Alan [varia at e-healthexpert.org]
> Sent: Wednesday, September 28, 2011 8:29 AM
> To: nginx at nginx.org
> Subject: Re: Happy birthday, Igor ?
>
> On Wed, 28 Sep 2011 15:35:54 +0400, Igor Sysoev <igor at sysoev.ru> wrote:
> > Happy birthday, Igor
>
> Happy birthday Igor.
> Keep up the good work.
>
> M.
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110928/2b1508c1/attachment.html>

From martinloy.uy at gmail.com  Wed Sep 28 12:52:24 2011
From: martinloy.uy at gmail.com (Martin Loy)
Date: Wed, 28 Sep 2011 09:52:24 -0300
Subject: Happy birthday, Igor ?
In-Reply-To: <CAKjeDB+fd_0sNn-uhKREkWk0piLwStRoQQ5x_nDwJso994GeQw@mail.gmail.com>
References: <CA+FsSoLqNX9j8SqjWa6-1tY-_-nCak1CfoMzkZDPd8RR-QsnBA@mail.gmail.com>
	<20110928113554.GF80782@nginx.com>
	<20110928132905.75d20ee5@e-healthexpert.org>
	<B293D35E9E20694A8D9DFFE1BC655C556AB265E81E@vvexch.verivue.com>
	<CAKjeDB+fd_0sNn-uhKREkWk0piLwStRoQQ5x_nDwJso994GeQw@mail.gmail.com>
Message-ID: <CAHbGeW63TkH23zc1LX9W3JwY1aVREt1jno8kMkTRF2V05ESsvA@mail.gmail.com>

Congratulations Igor, from uruguay southern cone.

Keep the hard work coming :P

Regards

M


On Wed, Sep 28, 2011 at 9:49 AM, Gerard Meier <gerjoo at gmail.com> wrote:
>
> Happy birthday from Amsterdam!
> Gerard
>
> On Wed, Sep 28, 2011 at 2:44 PM, Sirsiwal, Umesh <usirsiwal at verivue.com> wrote:
>>
>> Happy birthday Igor.
>>
>> -Umesh
>> ________________________________________
>> From: nginx-bounces at nginx.org [nginx-bounces at nginx.org] On Behalf Of Mark Alan [varia at e-healthexpert.org]
>> Sent: Wednesday, September 28, 2011 8:29 AM
>> To: nginx at nginx.org
>> Subject: Re: Happy birthday, Igor ?
>>
>> On Wed, 28 Sep 2011 15:35:54 +0400, Igor Sysoev <igor at sysoev.ru> wrote:
>> > Happy birthday, Igor
>>
>> Happy birthday Igor.
>> Keep up the good work.
>>
>> M.
>>
>> _______________________________________________
>> nginx mailing list
>> nginx at nginx.org
>> http://mailman.nginx.org/mailman/listinfo/nginx
>>
>> _______________________________________________
>> nginx mailing list
>> nginx at nginx.org
>> http://mailman.nginx.org/mailman/listinfo/nginx
>
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx



--
Nunca hubo un amigo que hiciese un favor a un enano, ni un enemigo que
le hiciese un mal, que no se viese recompensado por entero.


From rami.essaid at gmail.com  Wed Sep 28 13:01:01 2011
From: rami.essaid at gmail.com (Rami Essaid)
Date: Wed, 28 Sep 2011 09:01:01 -0400
Subject: Happy birthday, Igor ?
In-Reply-To: <4E831277.7020400@fansubcode.org>
References: <CA+FsSoLqNX9j8SqjWa6-1tY-_-nCak1CfoMzkZDPd8RR-QsnBA@mail.gmail.com>
	<20110928113554.GF80782@nginx.com>
	<4E831277.7020400@fansubcode.org>
Message-ID: <CAPMF7wiFAG2j2FqKFy+feoCohhPaHUumv9dF4cw4KhdA4y7UvA@mail.gmail.com>

Happy birthday Igor.


On Wed, Sep 28, 2011 at 8:26 AM, Pascal Nitsche <
pascal.nitsche at fansubcode.org> wrote:

> Happy Birthday Igor from M?lheim an der Ruhr (Germany).
>
> Am 28.09.2011 13:35, schrieb Igor Sysoev:
>
>  On Wed, Sep 28, 2011 at 12:20:21PM +0200, Antoine Bonavita wrote:
>>
>>> Given the amount of times Sep 28th 1970 is mentioned in the nginx code
>>> (see below), I am putting a rough guess that this is your birthday.
>>> So, in this very special day, I wanted to thank you for your amazing
>>> little kid and I would have loved to buy you a beer but you're a bit
>>> far away. If you ever get near the south of France, I'll buy you one.
>>>
>>> Of course, I could be completely wrong.
>>>
>> You are right, this is my birthday. Thank you.
>>
>>
>>
> ______________________________**_________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/**mailman/listinfo/nginx<http://mailman.nginx.org/mailman/listinfo/nginx>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110928/71df970c/attachment-0001.html>

From mdounin at mdounin.ru  Wed Sep 28 13:13:38 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Wed, 28 Sep 2011 17:13:38 +0400
Subject: Nginx build under Debian Hurd i386
In-Reply-To: <bc6dd4f929b101b45190afc5d657d2f1@davromaniak.eu>
References: <bc6dd4f929b101b45190afc5d657d2f1@davromaniak.eu>
Message-ID: <20110928131338.GX1137@mdounin.ru>

Hello!

On Wed, Sep 28, 2011 at 10:39:26AM +0200, Cyril LAVIER wrote:

>   
> 
> Hi nginx lovers. 
> 
> As it's my first mail on this mailing list, I
> will introduce myself a little bit. 
> 
> I'm Cyril Lavier, a 25 years old
> System Administrator from France. I use nginx for barely a year now. I
> use it for my own website, and also for the company I'm working. 
> 
> Now,
> let's work :). 
> 
> For the Debian packaging purposes, we try to make nginx
> build properly under the hurd-i386 kernel. 
> 
> But it never built well,
> and Kartik sent a mail in January about this. 
> 
> Now, as I'm contributing
> to this packaging, I tried to figure out why nginx build failed. 
> 
> For
> this, I opened a ticket (http://trac.nginx.org/nginx/ticket/22), which
> provide a patch for this. With this patch it's building well and working
> well. 

There is a problem with this patch: nginx uses PATH_MAX to 
allocate buffer used in realpath(), and POSIX claims:

: If resolved_name is not a null pointer and {PATH_MAX} is not 
: defined as a constant in the <limits.h> header, the behavior is 
: undefined.

http://pubs.opengroup.org/onlinepubs/9699919799/functions/realpath.html

Just using arbitrary value instead of PATH_MAX will likely lead to 
a problem.  If we want to avoid usage of PATH_MAX the relevant 
code should be converted to use realpath() with NULL as 
resolved_name.

Other current use of PATH_MAX is for getcwd() buffer which should 
be ok with any buffer size as getcwd() interface includes "size" 
argument.

> After, we add the MP4 module to the nginx-extras package, but
> the module failed to compile under hurd-i386 (the build works like a
> charm on other architectures). 
> 
> I opened the ticket #23
> (http://trac.nginx.org/nginx/ticket/23), with a first draft of a patch,
> which resolves one issue. By the way, my C coding knowledge is too low
> to completely correct this issue. 
> 
> We would like the hurd package being
> the same as the others, so if the developers (or other contributors with
> sufficient knowledge) can take a look at both tickets, and judging about
> a patch inclusion for #22 and a more complete patch #23, this would be
> very appreciated. 

This looks like result of the fact that off_t used is 32-bit wide.  
I believe correct patch to support 32-bit off_t would require a 
bit more work.

Meanwhile, you may want to focus on using 64-bit off_t instead.  
Under Linux it's typically done with "#define _FILE_OFFSET_BITS 
64".  Is it possible to do the same on Debian GNU/Hurd?

Maxim Dounin


From phanquochien at gmail.com  Wed Sep 28 13:20:30 2011
From: phanquochien at gmail.com (Hien P)
Date: Wed, 28 Sep 2011 20:20:30 +0700
Subject: Happy birthday, Igor ?
In-Reply-To: <20110928113554.GF80782@nginx.com>
References: <CA+FsSoLqNX9j8SqjWa6-1tY-_-nCak1CfoMzkZDPd8RR-QsnBA@mail.gmail.com>
	<20110928113554.GF80782@nginx.com>
Message-ID: <CAMwoMPSA2=vozfmaQxGJ6vyiQ-8kjEVdEO6SMeEvMkfmZMCgiQ@mail.gmail.com>

Happy birthday Igor ~~~ from Vietnam

On Wed, Sep 28, 2011 at 6:35 PM, Igor Sysoev <igor at sysoev.ru> wrote:

> On Wed, Sep 28, 2011 at 12:20:21PM +0200, Antoine Bonavita wrote:
> > Given the amount of times Sep 28th 1970 is mentioned in the nginx code
> > (see below), I am putting a rough guess that this is your birthday.
> > So, in this very special day, I wanted to thank you for your amazing
> > little kid and I would have loved to buy you a beer but you're a bit
> > far away. If you ever get near the south of France, I'll buy you one.
> >
> > Of course, I could be completely wrong.
>
> You are right, this is my birthday. Thank you.
>
>
> --
> Igor Sysoev
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx
>



-- 
Best regards,
Mr.Hien

<http://www.mrhien.info>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110928/994ebad7/attachment.html>

From cyril.lavier at davromaniak.eu  Wed Sep 28 13:32:54 2011
From: cyril.lavier at davromaniak.eu (Cyril LAVIER)
Date: Wed, 28 Sep 2011 15:32:54 +0200
Subject: Nginx build under Debian Hurd i386
In-Reply-To: <20110928131338.GX1137@mdounin.ru>
References: <bc6dd4f929b101b45190afc5d657d2f1@davromaniak.eu>
	<20110928131338.GX1137@mdounin.ru>
Message-ID: <a7fd6403b948e5eb88d929bf4bbf264a@davromaniak.eu>

 On Wed, 28 Sep 2011 17:13:38 +0400, Maxim Dounin wrote:
> Hello!
>
> On Wed, Sep 28, 2011 at 10:39:26AM +0200, Cyril LAVIER wrote:
>
>>
>>
>> Hi nginx lovers.
>>
>> As it's my first mail on this mailing list, I
>> will introduce myself a little bit.
>>
>> I'm Cyril Lavier, a 25 years old
>> System Administrator from France. I use nginx for barely a year now. 
>> I
>> use it for my own website, and also for the company I'm working.
>>
>> Now,
>> let's work :).
>>
>> For the Debian packaging purposes, we try to make nginx
>> build properly under the hurd-i386 kernel.
>>
>> But it never built well,
>> and Kartik sent a mail in January about this.
>>
>> Now, as I'm contributing
>> to this packaging, I tried to figure out why nginx build failed.
>>
>> For
>> this, I opened a ticket (http://trac.nginx.org/nginx/ticket/22), 
>> which
>> provide a patch for this. With this patch it's building well and 
>> working
>> well.
>
> There is a problem with this patch: nginx uses PATH_MAX to
> allocate buffer used in realpath(), and POSIX claims:
>
> : If resolved_name is not a null pointer and {PATH_MAX} is not
> : defined as a constant in the <limits.h> header, the behavior is
> : undefined.
>
> 
> http://pubs.opengroup.org/onlinepubs/9699919799/functions/realpath.html
>
> Just using arbitrary value instead of PATH_MAX will likely lead to
> a problem.  If we want to avoid usage of PATH_MAX the relevant
> code should be converted to use realpath() with NULL as
> resolved_name.
>

 Thanks for this information.

 I was not aware of this, and also I don't have a big knowledge in C 
 coding, I patched the file the easy way, by choosing an arbitrary value 
 for PATH_MAX, but as it may cause problems, we may see a more accurate 
 fix.

 Maybe Igor or any other developer can help providing a solution for 
 this issue, because I know I won't be able to do a clean patch.

> Other current use of PATH_MAX is for getcwd() buffer which should
> be ok with any buffer size as getcwd() interface includes "size"
> argument.
>
>> After, we add the MP4 module to the nginx-extras package, but
>> the module failed to compile under hurd-i386 (the build works like a
>> charm on other architectures).
>>
>> I opened the ticket #23
>> (http://trac.nginx.org/nginx/ticket/23), with a first draft of a 
>> patch,
>> which resolves one issue. By the way, my C coding knowledge is too 
>> low
>> to completely correct this issue.
>>
>> We would like the hurd package being
>> the same as the others, so if the developers (or other contributors 
>> with
>> sufficient knowledge) can take a look at both tickets, and judging 
>> about
>> a patch inclusion for #22 and a more complete patch #23, this would 
>> be
>> very appreciated.
>
> This looks like result of the fact that off_t used is 32-bit wide.
> I believe correct patch to support 32-bit off_t would require a
> bit more work.
>
> Meanwhile, you may want to focus on using 64-bit off_t instead.
> Under Linux it's typically done with "#define _FILE_OFFSET_BITS
> 64".  Is it possible to do the same on Debian GNU/Hurd?
>

 I don't know, as I said earlier, my knowledge in this domain is 
 limited, but I just wanted to research a bit to try to help resolve this 
 issue.

 Maybe Kartik will have this information, or maybe any other person in 
 this mailing list.

 Thanks.

> Maxim Dounin
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx

-- 
 Cyril "Davromaniak" Lavier


From tillere at gmail.com  Wed Sep 28 13:50:41 2011
From: tillere at gmail.com (Ed Tiller)
Date: Wed, 28 Sep 2011 09:50:41 -0400
Subject: Happy birthday, Igor ?
In-Reply-To: <20110928113554.GF80782@nginx.com>
References: <CA+FsSoLqNX9j8SqjWa6-1tY-_-nCak1CfoMzkZDPd8RR-QsnBA@mail.gmail.com>
	<20110928113554.GF80782@nginx.com>
Message-ID: <C529C7BD-5326-4552-972C-72AC67A4CE34@gmail.com>

Happy Birthday Igor, from Washington DC. Thank you for all of your work on Nginx.

Ed Tiller



On Sep 28, 2011, at 7:35 AM, Igor Sysoev <igor at sysoev.ru> wrote:

> On Wed, Sep 28, 2011 at 12:20:21PM +0200, Antoine Bonavita wrote:
>> Given the amount of times Sep 28th 1970 is mentioned in the nginx code
>> (see below), I am putting a rough guess that this is your birthday.
>> So, in this very special day, I wanted to thank you for your amazing
>> little kid and I would have loved to buy you a beer but you're a bit
>> far away. If you ever get near the south of France, I'll buy you one.
>> 
>> Of course, I could be completely wrong.
> 
> You are right, this is my birthday. Thank you.
> 
> 
> -- 
> Igor Sysoev
> 
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx


From www at lc365.net  Wed Sep 28 13:53:42 2011
From: www at lc365.net (=?gb2312?B?ufnV8cGi?=)
Date: Wed, 28 Sep 2011 21:53:42 +0800
Subject: Happy birthday, Igor ?
In-Reply-To: <20110928113554.GF80782@nginx.com>
References: <CA+FsSoLqNX9j8SqjWa6-1tY-_-nCak1CfoMzkZDPd8RR-QsnBA@mail.gmail.com>,
	<20110928113554.GF80782@nginx.com>
Message-ID: <BLU162-W11E141D9DFA16D491DDE60E3F10@phx.gbl>


Hi Igor,


          Happy birthday to you, and best wishes for you.



          PS. the OP is really a intelligent man.

> Date: Wed, 28 Sep 2011 15:35:54 +0400
> From: igor at sysoev.ru
> To: nginx at nginx.org
> Subject: Re: Happy birthday, Igor ?
> 
> On Wed, Sep 28, 2011 at 12:20:21PM +0200, Antoine Bonavita wrote:
> > Given the amount of times Sep 28th 1970 is mentioned in the nginx code
> > (see below), I am putting a rough guess that this is your birthday.
> > So, in this very special day, I wanted to thank you for your amazing
> > little kid and I would have loved to buy you a beer but you're a bit
> > far away. If you ever get near the south of France, I'll buy you one.
> > 
> > Of course, I could be completely wrong.
> 
> You are right, this is my birthday. Thank you.
> 
> 
> -- 
> Igor Sysoev
> 
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx
 		 	   		  
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110928/a6713374/attachment.html>

From igor at sysoev.ru  Wed Sep 28 14:01:25 2011
From: igor at sysoev.ru (Igor Sysoev)
Date: Wed, 28 Sep 2011 18:01:25 +0400
Subject: Happy birthday, Igor ?
In-Reply-To: <20110928113554.GF80782@nginx.com>
References: <CA+FsSoLqNX9j8SqjWa6-1tY-_-nCak1CfoMzkZDPd8RR-QsnBA@mail.gmail.com>
	<20110928113554.GF80782@nginx.com>
Message-ID: <20110928140124.GG80782@nginx.com>

Thank you for congratulations, but there are more than thousand
subscribers in this list and if even tenth of them will write
congratulation, this will result in a lot of emails.

Thank you.


-- 
Igor Sysoev


From kartik.mistry at gmail.com  Wed Sep 28 14:38:51 2011
From: kartik.mistry at gmail.com (Kartik Mistry)
Date: Wed, 28 Sep 2011 20:08:51 +0530
Subject: Happy birthday, Igor ?
Message-ID: <CAPDygezogboOX8M7RO4KbApxNtP2ge1LP_=D7DW082+RoHYv_w@mail.gmail.com>

Hi Igor,

Many many returns of the day! Happy Birthday!!

-- with Debian package maintainer hat on. Proud to be working with one
of the nicest and quick upstream I've seen :)

-- 
Kartik Mistry
Debian GNU/Linux Developer
IRC: kart_ | Identica: @kartikm


From saimonmoore at gmail.com  Wed Sep 28 15:14:06 2011
From: saimonmoore at gmail.com (Saimon Moore)
Date: Wed, 28 Sep 2011 17:14:06 +0200
Subject: Happy birthday, Igor ?
In-Reply-To: <CAPDygezogboOX8M7RO4KbApxNtP2ge1LP_=D7DW082+RoHYv_w@mail.gmail.com>
References: <CAPDygezogboOX8M7RO4KbApxNtP2ge1LP_=D7DW082+RoHYv_w@mail.gmail.com>
Message-ID: <CAEo1HV6KwbFCZpJ1KN5RaMYSVJsYrLZVMYTuNzSPDyBtXvx2EA@mail.gmail.com>

Happy birthday Igor...from Barcelona

On Wed, Sep 28, 2011 at 4:38 PM, Kartik Mistry <kartik.mistry at gmail.com>wrote:

> Hi Igor,
>
> Many many returns of the day! Happy Birthday!!
>
> -- with Debian package maintainer hat on. Proud to be working with one
> of the nicest and quick upstream I've seen :)
>
> --
> Kartik Mistry
> Debian GNU/Linux Developer
> IRC: kart_ | Identica: @kartikm
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx
>



-- 
Saimon Moore
Web Developer @ Teambox.com
(http://teambox.com)

Skype: saimonmoore
Yahoo IM: saimonmoore
Google IM: saimonmoore
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110928/fa2e9b15/attachment.html>

From huseyinin at gmail.com  Wed Sep 28 15:35:45 2011
From: huseyinin at gmail.com (=?ISO-8859-1?Q?H=FCseyin_Mert?=)
Date: Wed, 28 Sep 2011 18:35:45 +0300
Subject: Happy birthday, Igor ?
In-Reply-To: <CAEo1HV6KwbFCZpJ1KN5RaMYSVJsYrLZVMYTuNzSPDyBtXvx2EA@mail.gmail.com>
References: <CAPDygezogboOX8M7RO4KbApxNtP2ge1LP_=D7DW082+RoHYv_w@mail.gmail.com>
	<CAEo1HV6KwbFCZpJ1KN5RaMYSVJsYrLZVMYTuNzSPDyBtXvx2EA@mail.gmail.com>
Message-ID: <CABAk0iqq_dcbjBDj1yF_zO9h+tAPGq_hkkCCG6qeamo_+rD8vw@mail.gmail.com>

Happy birthday Igor...from Ankara
--
H?seyin Mert
http://www.hmert.com


On Wed, Sep 28, 2011 at 6:14 PM, Saimon Moore <saimonmoore at gmail.com> wrote:

> Happy birthday Igor...from Barcelona
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110928/d704deec/attachment-0001.html>

From nginx-forum at nginx.us  Wed Sep 28 16:05:07 2011
From: nginx-forum at nginx.us (csg)
Date: Wed, 28 Sep 2011 12:05:07 -0400
Subject: Does Nginx honor DNS TTLs for proxy upstreams?
In-Reply-To: <20110926135624.GO1137@mdounin.ru>
References: <20110926135624.GO1137@mdounin.ru>
Message-ID: <0ae2bc8b7ee889ae2c8b3841914e7d40.NginxMailingListEnglish@forum.nginx.org>

Maxim Dounin Wrote:
-------------------------------------------------------
> Errors suggest you don't have $backend set, and
> resulting 
> proxy_pass tries to go to "http://" which is
> invalid.

Yup, I am just curious how to accomplish that if you want to perform
some rewriting before, for example to strip off some leading parts of
the URL path. I came up with following test case:

worker_processes                        3;

events {
    worker_connections                  1024;
}

http {
    server_tokens                       off;

    include                             nginx.mime.types;
    default_type                        application/octet-stream;

    sendfile                            on;
    keepalive_timeout                   15;

    server {
        listen                          0.0.0.0:90;
        server_name                     _;

        location                        / {
            root                        html;
            index                       index.html;
        }
    }

    server {
        listen                          0.0.0.0:91;
        server_name                     _;

        location                        / {
            rewrite                     ^/foo/(.*)$  /$1  break;

            proxy_pass                  http://localhost:90;
            proxy_redirect              off;
        }
    }

    server {
        listen                          0.0.0.0:92;
        server_name                     _;

        location                        / {
            rewrite                     ^/foo/(.*)$  /$1  break;

            resolver                    127.0.0.1;

            set                         $backend  "localhost:90";
            proxy_pass                  http://$backend;
            proxy_redirect              off;
        }

        error_log                       /var/log/nginx/error.log 
debug;
    }
}

The port 91 server is a reverse proxy to the port 90 server with static
upstream (DNS lookup only on startup/reload). I delivers happily the
index.html of the port 90 server, regardless if I request
http://localhost:91/ of http://localhost:91/foo/.

Bringing the resolver into play doesn't seem to be that straight forward
if you look at the port 92 reverse proxy. It works when no rewriting has
to be performed, i.e., http://localhost:92/ works. But if I want to
strip off the leading /foo/ URL path segment of the request
http://localhost:92/foo/.

2011/09/28 17:55:51 [notice] 1490#0: *2 "^/foo/(.*)$" matches "/foo/",
client: 127.0.0.1, server: _, request: "GET /foo/ HTTP/1.1", host:
"localhost:92"
2011/09/28 17:55:51 [notice] 1490#0: *2 rewritten data: "/", args: "",
client: 127.0.0.1, server: _, request: "GET /foo/ HTTP/1.1", host:
"localhost:92"
2011/09/28 17:55:51 [warn] 1490#0: *2 using uninitialized "backend"
variable, client: 127.0.0.1, server: _, request: "GET /foo/ HTTP/1.1",
host: "localhost:92"
2011/09/28 17:55:51 [error] 1490#0: *2 invalid URL prefix in "http://",
client: 127.0.0.1, server: _, request: "GET /foo/ HTTP/1.1", host:
"localhost:92"

Any hints how to solve this the easiest way? Basically me requirement is
to reverse proxy http://proxy.example.com/foo/bar/quux to
http://upstream.example.com/bar/quux  (so strip off "foo/") while the IP
of upstream.example.com can change over time without prior announcement.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215830,215979#msg-215979


From mdounin at mdounin.ru  Wed Sep 28 16:13:55 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Wed, 28 Sep 2011 20:13:55 +0400
Subject: Does Nginx honor DNS TTLs for proxy upstreams?
In-Reply-To: <0ae2bc8b7ee889ae2c8b3841914e7d40.NginxMailingListEnglish@forum.nginx.org>
References: <20110926135624.GO1137@mdounin.ru>
	<0ae2bc8b7ee889ae2c8b3841914e7d40.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20110928161355.GZ1137@mdounin.ru>

Hello!

On Wed, Sep 28, 2011 at 12:05:07PM -0400, csg wrote:

> Maxim Dounin Wrote:
> -------------------------------------------------------
> > Errors suggest you don't have $backend set, and
> > resulting 
> > proxy_pass tries to go to "http://" which is
> > invalid.
> 
> Yup, I am just curious how to accomplish that if you want to perform
> some rewriting before, for example to strip off some leading parts of
> the URL path. I came up with following test case:

[...]

>         location                        / {
>             rewrite                     ^/foo/(.*)$  /$1  break;

The "break" here will stop execution of rewrite module directives 
(including "set" below) on match.

> 
>             resolver                    127.0.0.1;
> 
>             set                         $backend  "localhost:90";

This "set" won't be executed due to break above...

>             proxy_pass                  http://$backend;

... and $backend won't be set as a result.  Moving "set" before 
"rewrite .. break" will do the trick.

Maxim Dounin


From nginx-forum at nginx.us  Wed Sep 28 19:35:50 2011
From: nginx-forum at nginx.us (csg)
Date: Wed, 28 Sep 2011 15:35:50 -0400
Subject: Does Nginx honor DNS TTLs for proxy upstreams?
In-Reply-To: <20110928161355.GZ1137@mdounin.ru>
References: <20110928161355.GZ1137@mdounin.ru>
Message-ID: <03e18959a4108577d38690656ce65b0f.NginxMailingListEnglish@forum.nginx.org>

Maxim Dounin Wrote:
-------------------------------------------------------
Hello again!
 
> .... and $backend won't be set as a result. 
> Moving "set" before 
> "rewrite .. break" will do the trick.

Right, in this case the $backend applies to proxy_pass and the request
gets proxied but unfortunatelly in this case the rewrite rule does not
have effect anymore. It still gets applied according to the error logs,
but instead of the rewritten URL path to original one gets proxied. To
make it more clear I changed the example configuration I posted above by
having each virtual host its own access and error log.

When I run following test ...

$ curl -s -o - -D - http://localhost:92/foo/
HTTP/1.1 404 Not Found
Server: nginx
Date: Wed, 28 Sep 2011 19:26:04 GMT
Content-Type: text/html
Connection: keep-alive
Content-Length: 162

<html>
<head><title>404 Not Found</title></head>
<body bgcolor="white">
<center><h1>404 Not Found</h1></center>
<hr><center>nginx</center>
</body>
</html>

I will get this in the error logs

# tail -F /var/log/nginx/{access,error}9[012].log 
==> /var/log/nginx/access90.log <==
127.0.0.1 - - [28/Sep/2011:21:26:04 +0200] "GET /foo/ HTTP/1.0" 404 162
"-" "curl/7.15.5 (i686-redhat-linux-gnu) libcurl/7.15.5 OpenSSL/0.9.8b
zlib/1.2.3 libidn/0.6.5"

==> /var/log/nginx/access92.log <==
127.0.0.1 - - [28/Sep/2011:21:26:04 +0200] "GET /foo/ HTTP/1.1" 404 162
"-" "curl/7.15.5 (i686-redhat-linux-gnu) libcurl/7.15.5 OpenSSL/0.9.8b
zlib/1.2.3 libidn/0.6.5"

==> /var/log/nginx/error90.log <==
2011/09/28 21:26:04 [error] 3078#0: *15
"/usr/share/nginx/html/foo/index.html" is not found (2: No such file or
directory), client: 127.0.0.1, server: _, request: "GET /foo/ HTTP/1.0",
host: "localhost:90"

==> /var/log/nginx/error92.log <==
2011/09/28 21:26:04 [notice] 3078#0: *13 "^/foo/(.*)$" matches "/foo/",
client: 127.0.0.1, server: _, request: "GET /foo/ HTTP/1.1", host:
"localhost:92"
2011/09/28 21:26:04 [notice] 3078#0: *13 rewritten data: "/", args: "",
client: 127.0.0.1, server: _, request: "GET /foo/ HTTP/1.1", host:
"localhost:92"
2011/09/28 21:26:04 [notice] 3078#0: *13 "^/foo/(.*)$" does not match
"/", client: 127.0.0.1, server: _, request: "GET /foo/ HTTP/1.1", host:
"localhost:92"
2011/09/28 21:26:04 [info] 3078#0: *13 client 127.0.0.1 closed keepalive
connection

If this in not what you would expect, I can rebuild Nginx with
--with-debug tomorrow and repeat the test.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215830,215983#msg-215983


From paul at itools.com  Wed Sep 28 20:44:30 2011
From: paul at itools.com (Paul Sarena)
Date: Wed, 28 Sep 2011 13:44:30 -0700
Subject: Birthday gift for Igor
In-Reply-To: <20110928140124.GG80782@nginx.com>
References: <CA+FsSoLqNX9j8SqjWa6-1tY-_-nCak1CfoMzkZDPd8RR-QsnBA@mail.gmail.com>
	<20110928113554.GF80782@nginx.com>
	<20110928140124.GG80782@nginx.com>
Message-ID: <4E83872E.1030800@itools.com>

nginx has been such a treat to use.  To mark Igor's birthday I made a 
donation to his PayPal address: paypal at nginx.net

If it's time for you to show some love, please consider making it an 
extra special day for him.

Paul


From nginx-forum at nginx.us  Wed Sep 28 23:06:56 2011
From: nginx-forum at nginx.us (mojiz)
Date: Wed, 28 Sep 2011 19:06:56 -0400
Subject: nginx won't honor limit_rate
Message-ID: <830f3c7b2c19582deb96f252f5bb927e.NginxMailingListEnglish@forum.nginx.org>

Hi

I have an nginx server 
 nginx version: nginx/1.1.2
I'm using limit_rate to limt download speed of my users however after an
upgrade(I can't remember the version)
it doesn't work anymore.
Do you know why or how can I debug this?
The debug log doesn't show anything.
BTW I'm using nginx_proxy to send a file from my backends to my users.
I have used X-Accel-Limit-Rate, limit_rate and $limit_rate but neither
worked


Regards

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215989,215989#msg-215989


From mdounin at mdounin.ru  Wed Sep 28 23:23:28 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Thu, 29 Sep 2011 03:23:28 +0400
Subject: nginx won't honor limit_rate
In-Reply-To: <830f3c7b2c19582deb96f252f5bb927e.NginxMailingListEnglish@forum.nginx.org>
References: <830f3c7b2c19582deb96f252f5bb927e.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20110928232327.GA1137@mdounin.ru>

Hello!

On Wed, Sep 28, 2011 at 07:06:56PM -0400, mojiz wrote:

> Hi
> 
> I have an nginx server 
>  nginx version: nginx/1.1.2
> I'm using limit_rate to limt download speed of my users however after an
> upgrade(I can't remember the version)
> it doesn't work anymore.
> Do you know why or how can I debug this?
> The debug log doesn't show anything.
> BTW I'm using nginx_proxy to send a file from my backends to my users.
> I have used X-Accel-Limit-Rate, limit_rate and $limit_rate but neither
> worked

Most likely you are using "proxy_buffering off;", it disables any 
buffering including limit_rate one.

If in doubt, please follow http://wiki.nginx.org/Debugging.

Maxim Dounin


From nginx-forum at nginx.us  Wed Sep 28 23:31:35 2011
From: nginx-forum at nginx.us (mojiz)
Date: Wed, 28 Sep 2011 19:31:35 -0400
Subject: nginx won't honor limit_rate
In-Reply-To: <20110928232327.GA1137@mdounin.ru>
References: <20110928232327.GA1137@mdounin.ru>
Message-ID: <5ac045ab558aa2e1c5c87c2eadb4a37a.NginxMailingListEnglish@forum.nginx.org>

Thanks Max.
It worked!

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215989,215991#msg-215991


From pengli.hust at qq.com  Thu Sep 29 00:46:41 2011
From: pengli.hust at qq.com (lipeng)
Date: Thu, 29 Sep 2011 08:46:41 +0800
Subject: Happy birthday, Igor ?
In-Reply-To: <20110928113554.GF80782@nginx.com>
References: <CA+FsSoLqNX9j8SqjWa6-1tY-_-nCak1CfoMzkZDPd8RR-QsnBA@mail.gmail.com>
	<20110928113554.GF80782@nginx.com>
Message-ID: <4E83BFF1.5090600@qq.com>

Happy birthday Igor!

Thanks for your amazing work.


On 2011?09?28? 19:35, Igor Sysoev wrote:
> On Wed, Sep 28, 2011 at 12:20:21PM +0200, Antoine Bonavita wrote:
>> Given the amount of times Sep 28th 1970 is mentioned in the nginx code
>> (see below), I am putting a rough guess that this is your birthday.
>> So, in this very special day, I wanted to thank you for your amazing
>> little kid and I would have loved to buy you a beer but you're a bit
>> far away. If you ever get near the south of France, I'll buy you one.
>>
>> Of course, I could be completely wrong.
> You are right, this is my birthday. Thank you.
>
>


From delphij at delphij.net  Thu Sep 29 00:50:22 2011
From: delphij at delphij.net (Xin LI)
Date: Wed, 28 Sep 2011 17:50:22 -0700
Subject: Happy birthday, Igor ?
In-Reply-To: <20110928113554.GF80782@nginx.com>
References: <CA+FsSoLqNX9j8SqjWa6-1tY-_-nCak1CfoMzkZDPd8RR-QsnBA@mail.gmail.com>
	<20110928113554.GF80782@nginx.com>
Message-ID: <4E83C0CE.2030401@delphij.net>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA256

On 09/28/11 04:35, Igor Sysoev wrote:
> On Wed, Sep 28, 2011 at 12:20:21PM +0200, Antoine Bonavita wrote:
>> Given the amount of times Sep 28th 1970 is mentioned in the nginx
>> code (see below), I am putting a rough guess that this is your
>> birthday. So, in this very special day, I wanted to thank you for
>> your amazing little kid and I would have loved to buy you a beer
>> but you're a bit far away. If you ever get near the south of
>> France, I'll buy you one.
>> 
>> Of course, I could be completely wrong.
> 
> You are right, this is my birthday. Thank you.

Wish you enjoy your birthday and thanks for your work on nginx and
 FreeBSD! :)

Cheers,
- -- 
Xin LI <delphij at delphij.net>	https://www.delphij.net/
FreeBSD - The Power to Serve!		Live free or die
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.18 (FreeBSD)

iQEcBAEBCAAGBQJOg8DOAAoJEATO+BI/yjfBVg8H/ju3VAWb31JP/z+lXecjApEt
m/r7Z1VoYSEBfxkHUZwBw9mnWyU7ppxZ16YWV3+T93Autd4uCPo2XnY9uiHS9Bx/
/BYBMrpR7rhUljP7oahrrUW5x+UHv5TGL7QCh5Huppmr/RreV2eZrbgRLwryru4d
CPQO67u44W6swibJAisbCBm7XLTpXbPAH2mjpnF9cXDHoYNXP1qsp+ceWohXiF7c
NJSVF6qL+3bmTiAWHI1DMFhLyv9dV8pK/WmAXxk5WaMeIuecb8TsC8gDus1By66u
7pISt0Fs9knQIGuZ6N9lDgTTMj4knDmJiST50V73es5vBk3PEKMl+gRM2V6rbaI=
=8b2Z
-----END PGP SIGNATURE-----


From delta.yeh at gmail.com  Thu Sep 29 00:57:35 2011
From: delta.yeh at gmail.com (Delta Yeh)
Date: Thu, 29 Sep 2011 08:57:35 +0800
Subject: Happy birthday, Igor ?
In-Reply-To: <20110928113554.GF80782@nginx.com>
References: <CA+FsSoLqNX9j8SqjWa6-1tY-_-nCak1CfoMzkZDPd8RR-QsnBA@mail.gmail.com>
	<20110928113554.GF80782@nginx.com>
Message-ID: <CADKMG52QFJxb3-_i2mhs5ZsKGbHvPbSHjRT_Lcj5SuppNxtPTw@mail.gmail.com>

Congratulations  from Shanghai, China!


2011/9/28 Igor Sysoev <igor at sysoev.ru>:
> On Wed, Sep 28, 2011 at 12:20:21PM +0200, Antoine Bonavita wrote:
>> Given the amount of times Sep 28th 1970 is mentioned in the nginx code
>> (see below), I am putting a rough guess that this is your birthday.
>> So, in this very special day, I wanted to thank you for your amazing
>> little kid and I would have loved to buy you a beer but you're a bit
>> far away. If you ever get near the south of France, I'll buy you one.
>>
>> Of course, I could be completely wrong.
>
> You are right, this is my birthday. Thank you.
>
>
> --
> Igor Sysoev
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx
>


From rkferreira at gmail.com  Thu Sep 29 02:12:13 2011
From: rkferreira at gmail.com (Rodrigo K. Ferreira)
Date: Wed, 28 Sep 2011 23:12:13 -0300
Subject: Happy birthday, Igor ?
In-Reply-To: <CADKMG52QFJxb3-_i2mhs5ZsKGbHvPbSHjRT_Lcj5SuppNxtPTw@mail.gmail.com>
References: <CA+FsSoLqNX9j8SqjWa6-1tY-_-nCak1CfoMzkZDPd8RR-QsnBA@mail.gmail.com>
	<20110928113554.GF80782@nginx.com>
	<CADKMG52QFJxb3-_i2mhs5ZsKGbHvPbSHjRT_Lcj5SuppNxtPTw@mail.gmail.com>
Message-ID: <CAFP_Vh1UyDq_XtSRSFaqVVJQrP7XEXVV5gaTfkhRMK3CVdtC4Q@mail.gmail.com>

Happy birthday from Brazil!

On 9/28/11, Delta Yeh <delta.yeh at gmail.com> wrote:
> Congratulations  from Shanghai, China!
>
>
> 2011/9/28 Igor Sysoev <igor at sysoev.ru>:
>> On Wed, Sep 28, 2011 at 12:20:21PM +0200, Antoine Bonavita wrote:
>>> Given the amount of times Sep 28th 1970 is mentioned in the nginx code
>>> (see below), I am putting a rough guess that this is your birthday.
>>> So, in this very special day, I wanted to thank you for your amazing
>>> little kid and I would have loved to buy you a beer but you're a bit
>>> far away. If you ever get near the south of France, I'll buy you one.
>>>
>>> Of course, I could be completely wrong.
>>
>> You are right, this is my birthday. Thank you.
>>
>>
>> --
>> Igor Sysoev
>>
>> _______________________________________________
>> nginx mailing list
>> nginx at nginx.org
>> http://mailman.nginx.org/mailman/listinfo/nginx
>>
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx
>


From amoiz.shine at gmail.com  Thu Sep 29 03:19:38 2011
From: amoiz.shine at gmail.com (Sharl.Jimh.Tsin)
Date: Thu, 29 Sep 2011 11:19:38 +0800
Subject: Happy birthday, Igor ?
In-Reply-To: <CABAk0iqq_dcbjBDj1yF_zO9h+tAPGq_hkkCCG6qeamo_+rD8vw@mail.gmail.com>
References: <CAPDygezogboOX8M7RO4KbApxNtP2ge1LP_=D7DW082+RoHYv_w@mail.gmail.com>
	<CAEo1HV6KwbFCZpJ1KN5RaMYSVJsYrLZVMYTuNzSPDyBtXvx2EA@mail.gmail.com>
	<CABAk0iqq_dcbjBDj1yF_zO9h+tAPGq_hkkCCG6qeamo_+rD8vw@mail.gmail.com>
Message-ID: <1317266378.1993.3.camel@sharl-desktop>

? 2011-09-28?? 18:35 +0300?H?seyin Mert???
> Happy birthday Igor...from Ankara 
> --
> H?seyin Mert
> http://www.hmert.com
> 
> 
> On Wed, Sep 28, 2011 at 6:14 PM, Saimon Moore <saimonmoore at gmail.com>
> wrote:
>         Happy birthday Igor...from Barcelona
> 
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx

Happy birthday Igor +1 from Shanghai,China.

-- 
Best regards,
Sharl.Jimh.Tsin (From China **Obviously Taiwan INCLUDED**)

Using Gmail? Please read this important notice:
http://www.fsf.org/campaigns/jstrap/gmail?10073.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 490 bytes
Desc: This is a digitally signed message part
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110929/6b0e3098/attachment.bin>

From nbubingo at gmail.com  Thu Sep 29 04:42:39 2011
From: nbubingo at gmail.com (=?GB2312?B?0qbOsLHz?=)
Date: Thu, 29 Sep 2011 12:42:39 +0800
Subject: Nginx Substitutions Filter problem
In-Reply-To: <2a4c6102eba7e97f63ad316476db06b8.squirrel@webmail.martronic.ch>
References: <4E80E187.6030807@ntlworld.com>
	<CAJCGDi8ksNcSHyf-iLm5vZevAKBMew_M2NvA5s62_wyGRLGtGw@mail.gmail.com>
	<4E80EB49.9000108@ianhobson.co.uk>
	<CAJCGDi_jWABeVRNZUEeKmZ9q3cpdFOsRWgH9ASiU_=GNSycdBQ@mail.gmail.com>
	<20110927045203.20830@gmx.net> <4E81701D.408@martronic.ch>
	<CAGxuLK3+sVfD8WC=9gL6z0daokOB=qc8HaE3drqsxSgbocS6=A@mail.gmail.com>
	<2a4c6102eba7e97f63ad316476db06b8.squirrel@webmail.martronic.ch>
Message-ID: <CAGxuLK1VRVX_6rsXDSomzcnpndg5L0Rd4CqCTfnU3PfJrJyvkw@mail.gmail.com>

It works in my box. Can you show me your debug log?

http://wiki.nginx.org/Debugging

Do't use this in the production server. The debug log will be very large.


2011/9/28 Martronic SA <martronic at martronic.ch>

> full source attachment
> Thanks.
>
>
> <a id="portal-logo" href="http://bale.hopla.net" accesskey="1"
> class="visualNoPrint"><img
> src="http://bale.hopla.net/portal_logo/image" alt="Logo" title="Logo
> Hopla-Net" height="175"
> width="230" /></a>
>
> <div id="solgemabandeau" class="visualNoPrint">
>
>  <div id="portal-bandeau">
>    <div id="inner-portal-bandeau" class="bannersContainer">
>
>          <div style="position:relative; display: block"
> id="carousel-banner-0"
> class="bando.jpg carousel-banner selected">
>            <div style="height:200px; width:100%; background:transparent
> url(http://bale.hopla.net/bando.jpg) no-repeat scroll left top;"
> title="Une maison
> dans un jardin, nouveau bandeau Hopla pour le printemps."
> class="bandeau_image"></div>
>          </div>
>
>
>          <div style="display: none" id="carousel-banner-1"
> class="bandeauhoplabale.jpg
> carousel-banner">
>            <div style="height:200px; width:100%; background:transparent
> url(http://bale.hopla.net/bandeauhoplabale.jpg) no-repeat scroll left
> top;"
> title="Le printemps est de retour sur St Louis et B?le"
> class="bandeau_image"></div>
>          </div>
>
>        <span class="bandeauTopLeft"></span>
>
>        <span class="bandeauTopRight"></span>
>        <span class="bandeauBottomLeft"></span>
>        <span class="bandeauBottomRight"></span>
>    </div>
> </div>
>
> </div>
>
> > Can you show me your source page?
> >
> > Thanks.
> >
> > 2011/9/27 Martronic SA <martronic at martronic.ch>
> >
> >> Hello,
> >>
> >> We are using the "Substitutions Filter module"
> >> to replace html content in our pages.
> >>
> >> Our configuration is :
> >>
> >> location /{
> >> subs_filter_types text/html text/xml;
> >> subs_filter 'http://bale.hopla.net/bons-**plans<
> http://bale.hopla.net/bons-plans>'
> >> 'http://bons-plans.hopla.net';
> >> subs_filter 'id="portal-logo" href="http://bale.hopla.net"'
> >> 'id="portal-logo" href="http://www.hopla.net"';
> >> }
> >>
> >> The first substitution is working, but the second not.
> >> Are we missing something? Is the "space" supported in substitution
> module?
> >>
> >> Thank you for your help!
> >>
> >> Martronic SA
> >>
> >> nginx -V
> >> nginx: nginx version: nginx/1.1.4
> >> nginx: built by gcc 4.3.2 (Debian 4.3.2-1.1)
> >> nginx: TLS SNI support enabled
> >> nginx: configure arguments: --conf-path=/etc/nginx/nginx.**conf
> >> --error-log-path=/var/log/**nginx/error.log
> --pid-path=/var/run/nginx.pid
> >> --lock-path=/var/lock/nginx.**lock --http-log-path=/var/log/**
> >> nginx/access.log
> >> --with-http_dav_module
> --http-client-body-temp-path=/**var/lib/nginx/body
> >> --with-http_ssl_module
> >> --http-proxy-temp-path=/var/**lib/nginx/proxy
> >> --with-http_stub_status_module
> >> --http-fastcgi-temp-path=/var/**lib/nginx/fastcgi
> >> --add-module=/var/lib/**substitutions4nginx-read-only
> >>
> >>
> >>
> >> ______________________________**_________________
> >> nginx mailing list
> >> nginx at nginx.org
> >> http://mailman.nginx.org/**mailman/listinfo/nginx<
> http://mailman.nginx.org/mailman/listinfo/nginx>
> >>
> > _______________________________________________
> > nginx mailing list
> > nginx at nginx.org
> > http://mailman.nginx.org/mailman/listinfo/nginx
>
>
> Avec nos meilleures salutations,
>
> MARTRONIC SA
>
> De votre site internet?
> ? votre serveur d?applications
> Venez visiter notre nouvelle plateforme collaborative Web 2 de Martronic
> http://www.solgema.com
> http://www.solgema.ch
> http://www.martronic.ch
>
> Rue du Ch?teau-Vieux 12
> Case postale 186
> 1870 Monthey VS
> Suisse / Switzerland
>
> T?l +41 24 471 51 07
> Fax +41 24 471 58 47
> e-mail : martronic at martronic.ch
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110929/fa605e72/attachment.html>

From savages at mozapps.com  Thu Sep 29 07:15:55 2011
From: savages at mozapps.com (Shaun savage)
Date: Thu, 29 Sep 2011 15:15:55 +0800
Subject: fastcgi msg_id always 1?
Message-ID: <4E841B2B.5060704@mozapps.com>

I was looking at the fastcgi, I found out that all fcgi request have a
msg_id of 1? 

What happens when more than 1 fcgi requests comes in and first one is
still being handled. how does the fcgi server know  the the difference
in requests?  Usually each request has it own connection, but if
keepalive is on the then many request can come in on the same connection. 

I am use "ab" to test the nginx+ fastcgi server response time.



From mdounin at mdounin.ru  Thu Sep 29 07:28:33 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Thu, 29 Sep 2011 11:28:33 +0400
Subject: Does Nginx honor DNS TTLs for proxy upstreams?
In-Reply-To: <03e18959a4108577d38690656ce65b0f.NginxMailingListEnglish@forum.nginx.org>
References: <20110928161355.GZ1137@mdounin.ru>
	<03e18959a4108577d38690656ce65b0f.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20110929072833.GB1137@mdounin.ru>

Hello!

On Wed, Sep 28, 2011 at 03:35:50PM -0400, csg wrote:

> Maxim Dounin Wrote:
> -------------------------------------------------------
> Hello again!
>  
> > .... and $backend won't be set as a result. 
> > Moving "set" before 
> > "rewrite .. break" will do the trick.
> 
> Right, in this case the $backend applies to proxy_pass and the request
> gets proxied but unfortunatelly in this case the rewrite rule does not
> have effect anymore. It still gets applied according to the error logs,
> but instead of the rewritten URL path to original one gets proxied. To
> make it more clear I changed the example configuration I posted above by
> having each virtual host its own access and error log.

This looks like a bug (or at least misfeature).  Please try the 
attached patch.

Maxim Dounin
-------------- next part --------------
# HG changeset patch
# User Maxim Dounin <mdounin at mdounin.ru>
# Date 1317281138 -14400
# Node ID d6a0787307868c06073d92bb39eb2f73acf19d7c
# Parent  29d417d419d946dce33023c71cce0e586ef3547b
Proxy: made proxy_pass with variables more consitent.

If proxy_pass were used with variables and there were no URI component,
nginx always used unparsed URI.  This isn't consistent with "no variables"
case, where e.g. rewrites are applied even if there are no URI component.

Fix is to use same logic in both cases, i.e. only use unparsed URI if
it's valid and request is main one.

diff --git a/src/http/modules/ngx_http_proxy_module.c b/src/http/modules/ngx_http_proxy_module.c
--- a/src/http/modules/ngx_http_proxy_module.c
+++ b/src/http/modules/ngx_http_proxy_module.c
@@ -738,9 +738,6 @@ ngx_http_proxy_eval(ngx_http_request_t *
             url.uri.len++;
             url.uri.data = p - 1;
         }
-
-    } else {
-        url.uri = r->unparsed_uri;
     }
 
     ctx->vars.key_start = u->schema;
@@ -808,7 +805,7 @@ ngx_http_proxy_create_key(ngx_http_reque
         return NGX_ERROR;
     }
 
-    if (plcf->proxy_lengths) {
+    if (plcf->proxy_lengths && ctx->vars.uri.len) {
 
         *key = ctx->vars.uri;
         u->uri = ctx->vars.uri;
@@ -918,7 +915,7 @@ ngx_http_proxy_create_request(ngx_http_r
     loc_len = 0;
     unparsed_uri = 0;
 
-    if (plcf->proxy_lengths) {
+    if (plcf->proxy_lengths && ctx->vars.uri.len) {
         uri_len = ctx->vars.uri.len;
 
     } else if (ctx->vars.uri.len == 0 && r->valid_unparsed_uri && r == r->main)
@@ -1024,7 +1021,7 @@ ngx_http_proxy_create_request(ngx_http_r
 
     u->uri.data = b->last;
 
-    if (plcf->proxy_lengths) {
+    if (plcf->proxy_lengths && ctx->vars.uri.len) {
         b->last = ngx_copy(b->last, ctx->vars.uri.data, ctx->vars.uri.len);
 
     } else if (unparsed_uri) {

From nginx-forum at nginx.us  Thu Sep 29 07:31:07 2011
From: nginx-forum at nginx.us (csg)
Date: Thu, 29 Sep 2011 03:31:07 -0400
Subject: Does Nginx honor DNS TTLs for proxy upstreams?
In-Reply-To: <03e18959a4108577d38690656ce65b0f.NginxMailingListEnglish@forum.nginx.org>
References: <20110928161355.GZ1137@mdounin.ru>
	<03e18959a4108577d38690656ce65b0f.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <826ac94aad7afd513c2d51d50fe1601f.NginxMailingListEnglish@forum.nginx.org>

Here as a follow up the debug output plus additional information:

# nginx -V
nginx: nginx version: nginx/1.0.6
nginx: configure arguments: --with-debug --prefix=/usr/share/nginx
--conf-path=/etc/nginx/nginx.conf --sbin-path=/usr/sbin/nginx
--http-log-path=/var/log/nginx/access.log
--error-log-path=/var/log/nginx/error.log --pid-path=/var/run/nginx.pid
--user=nginx --group=nginx --with-http_geoip_module
--with-http_gzip_static_module --with-http_realip_module
--with-http_secure_link_module --with-http_stub_status_module
--add-module=../chaoslawful-drizzle-nginx-module-1fc3af0
--add-module=../agentzh-echo-nginx-module-13dd12f
--add-module=../ngx-fancyindex-0.3.1
--add-module=../agentzh-headers-more-nginx-module-137855d
--add-module=../ngx_http_upstream_keepalive-2ee28064a04a
--add-module=../agentzh-memc-nginx-module-4f1952c
--add-module=../simpl-ngx_devel_kit-bc97eea
--add-module=../chaoslawful-lua-nginx-module-dd0d4d0
--add-module=../agentzh-rds-csv-nginx-module-eb697d1
--add-module=../agentzh-rds-json-nginx-module-376db1f

# cat /etc/nginx/nginx.conf
##
##  /etc/nginx/nginx.conf
##

worker_processes                        3;

events {
    worker_connections                  1024;
}

http {
    server_tokens                       off;

    include                             nginx.mime.types;
    default_type                        application/octet-stream;

    sendfile                            on;
    keepalive_timeout                   15;

    server {
        listen                          0.0.0.0:90;
        server_name                     _;

        location                        / {
            root                        html;
            index                       index.html;
        }

        access_log                      /var/log/nginx/access90.log;
        error_log                       /var/log/nginx/error90.log 
debug;
    }

    server {
        listen                          0.0.0.0:91;
        server_name                     _;

        location                        / {
            rewrite                     ^/foo/(.*)$  /$1  break;

            proxy_pass                  http://localhost:90;
            proxy_redirect              off;
        }

        access_log                      /var/log/nginx/access91.log;
        error_log                       /var/log/nginx/error91.log 
debug;
    }

    server {
        listen                          0.0.0.0:92;
        server_name                     _;

        location                        / {
            set                         $backend  "localhost:90";

            rewrite                     ^/foo/(.*)$  /$1  break;

            resolver                    127.0.0.1;

            proxy_pass                  http://$backend;
            proxy_redirect              off;
        }

        access_log                      /var/log/nginx/access92.log;
        error_log                       /var/log/nginx/error92.log 
debug;
    }
}

$ curl -s -o - -D - http://localhost:92/foo/
HTTP/1.1 404 Not Found
Server: nginx
Date: Thu, 29 Sep 2011 07:21:07 GMT
Content-Type: text/html
Connection: keep-alive
Content-Length: 162

<html>
<head><title>404 Not Found</title></head>
<body bgcolor="white">
<center><h1>404 Not Found</h1></center>
<hr><center>nginx</center>
</body>
</html>

# cat access92.log
127.0.0.1 - - [29/Sep/2011:09:21:07 +0200] "GET /foo/ HTTP/1.1" 404 162
"-" "curl/7.15.5 (i686-redhat-linux-gnu) libcurl/7.15.5 OpenSSL/0.9.8b
zlib/1.2.3 libidn/0.6.5"

# cat access90.log
127.0.0.1 - - [29/Sep/2011:09:21:07 +0200] "GET /foo/ HTTP/1.0" 404 162
"-" "curl/7.15.5 (i686-redhat-linux-gnu) libcurl/7.15.5 OpenSSL/0.9.8b
zlib/1.2.3 libidn/0.6.5"

# cat error92.log
2011/09/29 09:21:05 [debug] 3565#0: epoll add event: fd:13 op:1
ev:00000001
2011/09/29 09:21:07 [debug] 3565#0: post event 0A15AC50
2011/09/29 09:21:07 [debug] 3565#0: delete posted event 0A15AC50
2011/09/29 09:21:07 [debug] 3565#0: accept on 0.0.0.0:92, ready: 0
2011/09/29 09:21:07 [debug] 3565#0: posix_memalign: 0A11BC80:256 @16
2011/09/29 09:21:07 [debug] 3565#0: *1 accept: 127.0.0.1 fd:17
2011/09/29 09:21:07 [debug] 3565#0: *1 event timer add: 17:
60000:3020934938
2011/09/29 09:21:07 [debug] 3565#0: *1 epoll add event: fd:17 op:1
ev:80000001
2011/09/29 09:21:07 [debug] 3565#0: *1 post event 0A15ACB8
2011/09/29 09:21:07 [debug] 3565#0: *1 delete posted event 0A15ACB8
2011/09/29 09:21:07 [debug] 3565#0: *1 malloc: 0A124890:664
2011/09/29 09:21:07 [debug] 3565#0: *1 malloc: 0A124B30:1024
2011/09/29 09:21:07 [debug] 3565#0: *1 posix_memalign: 0A121C90:4096
@16
2011/09/29 09:21:07 [debug] 3565#0: *1 http process request line
2011/09/29 09:21:07 [debug] 3565#0: *1 recv: fd:17 158 of 1024
2011/09/29 09:21:07 [debug] 3565#0: *1 http request line: "GET /foo/
HTTP/1.1"
2011/09/29 09:21:07 [debug] 3565#0: *1 http uri: "/foo/"
2011/09/29 09:21:07 [debug] 3565#0: *1 http args: ""
2011/09/29 09:21:07 [debug] 3565#0: *1 http exten: ""
2011/09/29 09:21:07 [debug] 3565#0: *1 http process request header line
2011/09/29 09:21:07 [debug] 3565#0: *1 http header: "User-Agent:
curl/7.15.5 (i686-redhat-linux-gnu) libcurl/7.15.5 OpenSSL/0.9.8b
zlib/1.2.3 libidn/0.6.5"
2011/09/29 09:21:07 [debug] 3565#0: *1 http header: "Host:
localhost:92"
2011/09/29 09:21:07 [debug] 3565#0: *1 http header: "Accept: */*"
2011/09/29 09:21:07 [debug] 3565#0: *1 http header done
2011/09/29 09:21:07 [debug] 3565#0: *1 event timer del: 17: 3020934938
2011/09/29 09:21:07 [debug] 3565#0: *1 generic phase: 0
2011/09/29 09:21:07 [debug] 3565#0: *1 rewrite phase: 1
2011/09/29 09:21:07 [debug] 3565#0: *1 test location: "/"
2011/09/29 09:21:07 [debug] 3565#0: *1 using configuration "/"
2011/09/29 09:21:07 [debug] 3565#0: *1 http cl:-1 max:1048576
2011/09/29 09:21:07 [debug] 3565#0: *1 rewrite phase: 3
2011/09/29 09:21:07 [debug] 3565#0: *1 http script value:
"localhost:90"
2011/09/29 09:21:07 [debug] 3565#0: *1 http script set $backend
2011/09/29 09:21:07 [debug] 3565#0: *1 http script regex: "^/foo/(.*)$"
2011/09/29 09:21:07 [notice] 3565#0: *1 "^/foo/(.*)$" matches "/foo/",
client: 127.0.0.1, server: _, request: "GET /foo/ HTTP/1.1", host:
"localhost:92"
2011/09/29 09:21:07 [debug] 3565#0: *1 http script copy: "/"
2011/09/29 09:21:07 [debug] 3565#0: *1 http script capture: ""
2011/09/29 09:21:07 [debug] 3565#0: *1 http script regex end
2011/09/29 09:21:07 [notice] 3565#0: *1 rewritten data: "/", args: "",
client: 127.0.0.1, server: _, request: "GET /foo/ HTTP/1.1", host:
"localhost:92"
2011/09/29 09:21:07 [debug] 3565#0: *1 post rewrite phase: 4
2011/09/29 09:21:07 [debug] 3565#0: *1 generic phase: 5
2011/09/29 09:21:07 [debug] 3565#0: *1 generic phase: 6
2011/09/29 09:21:07 [debug] 3565#0: *1 generic phase: 7
2011/09/29 09:21:07 [debug] 3565#0: *1 access phase: 8
2011/09/29 09:21:07 [debug] 3565#0: *1 access phase: 9
2011/09/29 09:21:07 [debug] 3565#0: *1 post access phase: 10
2011/09/29 09:21:07 [debug] 3565#0: *1 http script copy: "http://"
2011/09/29 09:21:07 [debug] 3565#0: *1 http script var: "localhost:90"
2011/09/29 09:21:07 [debug] 3565#0: *1 http init upstream, client timer:
0
2011/09/29 09:21:07 [debug] 3565#0: *1 epoll add event: fd:17 op:3
ev:80000005
2011/09/29 09:21:07 [debug] 3565#0: *1 http script copy: "Host: "
2011/09/29 09:21:07 [debug] 3565#0: *1 http script var: "localhost:90"
2011/09/29 09:21:07 [debug] 3565#0: *1 http script copy: "
"
2011/09/29 09:21:07 [debug] 3565#0: *1 http script copy: "Connection:
close
"
2011/09/29 09:21:07 [debug] 3565#0: *1 http proxy header: "User-Agent:
curl/7.15.5 (i686-redhat-linux-gnu) libcurl/7.15.5 OpenSSL/0.9.8b
zlib/1.2.3 libidn/0.6.5"
2011/09/29 09:21:07 [debug] 3565#0: *1 http proxy header: "Accept: */*"
2011/09/29 09:21:07 [debug] 3565#0: *1 http proxy header:
"GET /foo/ HTTP/1.0
Host: localhost:90
Connection: close
User-Agent: curl/7.15.5 (i686-redhat-linux-gnu) libcurl/7.15.5
OpenSSL/0.9.8b zlib/1.2.3 libidn/0.6.5
Accept: */*

"
2011/09/29 09:21:07 [debug] 3565#0: *1 http cleanup add: 0A122630
2011/09/29 09:21:07 [debug] 3565#0: *1 get rr peer, try: 1
2011/09/29 09:21:07 [debug] 3565#0: *1 socket 18
2011/09/29 09:21:07 [debug] 3565#0: *1 epoll add connection: fd:18
ev:80000005
2011/09/29 09:21:07 [debug] 3565#0: *1 connect to 127.0.0.1:90, fd:18
#2
2011/09/29 09:21:07 [debug] 3565#0: *1 http upstream connect: -2
2011/09/29 09:21:07 [debug] 3565#0: *1 event timer add: 18:
60000:3020934939
2011/09/29 09:21:07 [debug] 3565#0: *1 http finalize request: -4, "/?"
a:1, c:2
2011/09/29 09:21:07 [debug] 3565#0: *1 http request count:2 blk:0
2011/09/29 09:21:07 [debug] 3565#0: *1 post event 0A167CF4
2011/09/29 09:21:07 [debug] 3565#0: *1 post event 0A167CC0
2011/09/29 09:21:07 [debug] 3565#0: *1 delete posted event 0A167CC0
2011/09/29 09:21:07 [debug] 3565#0: *1 http run request: "/?"
2011/09/29 09:21:07 [debug] 3565#0: *1 http upstream check client, write
event:1, "/"
2011/09/29 09:21:07 [debug] 3565#0: *1 http upstream recv(): -1 (11:
Resource temporarily unavailable)
2011/09/29 09:21:07 [debug] 3565#0: *1 delete posted event 0A167CF4
2011/09/29 09:21:07 [debug] 3565#0: *1 http upstream request: "/?"
2011/09/29 09:21:07 [debug] 3565#0: *1 http upstream send request
handler
2011/09/29 09:21:07 [debug] 3565#0: *1 http upstream send request
2011/09/29 09:21:07 [debug] 3565#0: *1 chain writer buf fl:1 s:177
2011/09/29 09:21:07 [debug] 3565#0: *1 chain writer in: 0A12264C
2011/09/29 09:21:07 [debug] 3565#0: *1 writev: 177
2011/09/29 09:21:07 [debug] 3565#0: *1 chain writer out: 00000000
2011/09/29 09:21:07 [debug] 3565#0: *1 event timer del: 18: 3020934939
2011/09/29 09:21:07 [debug] 3565#0: *1 event timer add: 18:
60000:3020934939
2011/09/29 09:21:07 [debug] 3565#0: *1 post event 0A15ACEC
2011/09/29 09:21:07 [debug] 3565#0: *1 post event 0A167CF4
2011/09/29 09:21:07 [debug] 3565#0: *1 delete posted event 0A167CF4
2011/09/29 09:21:07 [debug] 3565#0: *1 http upstream request: "/?"
2011/09/29 09:21:07 [debug] 3565#0: *1 http upstream dummy handler
2011/09/29 09:21:07 [debug] 3565#0: *1 delete posted event 0A15ACEC
2011/09/29 09:21:07 [debug] 3565#0: *1 http upstream request: "/?"
2011/09/29 09:21:07 [debug] 3565#0: *1 http upstream process header
2011/09/29 09:21:07 [debug] 3565#0: *1 malloc: 0A174BF8:4096
2011/09/29 09:21:07 [debug] 3565#0: *1 recv: fd:18 305 of 4096
2011/09/29 09:21:07 [debug] 3565#0: *1 http proxy status 404 "404 Not
Found"
2011/09/29 09:21:07 [debug] 3565#0: *1 http proxy header: "Server:
nginx"
2011/09/29 09:21:07 [debug] 3565#0: *1 http proxy header: "Date: Thu, 29
Sep 2011 07:21:07 GMT"
2011/09/29 09:21:07 [debug] 3565#0: *1 http proxy header: "Content-Type:
text/html"
2011/09/29 09:21:07 [debug] 3565#0: *1 http proxy header:
"Content-Length: 162"
2011/09/29 09:21:07 [debug] 3565#0: *1 http proxy header: "Connection:
close"
2011/09/29 09:21:07 [debug] 3565#0: *1 http proxy header done
2011/09/29 09:21:07 [debug] 3565#0: *1 lua capture header filter, uri
"/"
2011/09/29 09:21:07 [debug] 3565#0: *1 HTTP/1.1 404 Not Found
Server: nginx
Date: Thu, 29 Sep 2011 07:21:07 GMT
Content-Type: text/html
Connection: keep-alive
Content-Length: 162

2011/09/29 09:21:07 [debug] 3565#0: *1 write new buf t:1 f:0 0A1227F8,
pos 0A1227F8, size: 148 file: 0, size: 0
2011/09/29 09:21:07 [debug] 3565#0: *1 http write filter: l:0 f:0 s:148
2011/09/29 09:21:07 [debug] 3565#0: *1 http cacheable: 0
2011/09/29 09:21:07 [debug] 3565#0: *1 http upstream process upstream
2011/09/29 09:21:07 [debug] 3565#0: *1 pipe read upstream: 1
2011/09/29 09:21:07 [debug] 3565#0: *1 pipe preread: 162
2011/09/29 09:21:07 [debug] 3565#0: *1 readv: 1:3791
2011/09/29 09:21:07 [debug] 3565#0: *1 pipe recv chain: 0
2011/09/29 09:21:07 [debug] 3565#0: *1 pipe buf free s:0 t:1 f:0
0A174BF8, pos 0A174C87, size: 162 file: 0, size: 0
2011/09/29 09:21:07 [debug] 3565#0: *1 input buf #0
2011/09/29 09:21:07 [debug] 3565#0: *1 pipe write downstream: 1
2011/09/29 09:21:07 [debug] 3565#0: *1 pipe write downstream flush in
2011/09/29 09:21:07 [debug] 3565#0: *1 http output filter "/?"
2011/09/29 09:21:07 [debug] 3565#0: *1 http copy filter: "/?"
2011/09/29 09:21:07 [debug] 3565#0: *1 lua capture body filter, uri "/"
2011/09/29 09:21:07 [debug] 3565#0: *1 http postpone filter "/?"
0A122970
2011/09/29 09:21:07 [debug] 3565#0: *1 write old buf t:1 f:0 0A1227F8,
pos 0A1227F8, size: 148 file: 0, size: 0
2011/09/29 09:21:07 [debug] 3565#0: *1 write new buf t:1 f:0 0A174BF8,
pos 0A174C87, size: 162 file: 0, size: 0
2011/09/29 09:21:07 [debug] 3565#0: *1 http write filter: l:0 f:0 s:310
2011/09/29 09:21:07 [debug] 3565#0: *1 http copy filter: 0 "/?"
2011/09/29 09:21:07 [debug] 3565#0: *1 pipe write downstream done
2011/09/29 09:21:07 [debug] 3565#0: *1 event timer: 18, old: 3020934939,
new: 3020934939
2011/09/29 09:21:07 [debug] 3565#0: *1 http upstream exit: 00000000
2011/09/29 09:21:07 [debug] 3565#0: *1 finalize http upstream request:
0
2011/09/29 09:21:07 [debug] 3565#0: *1 finalize http proxy request
2011/09/29 09:21:07 [debug] 3565#0: *1 free rr peer 1 0
2011/09/29 09:21:07 [debug] 3565#0: *1 close http upstream connection:
18
2011/09/29 09:21:07 [debug] 3565#0: *1 event timer del: 18: 3020934939
2011/09/29 09:21:07 [debug] 3565#0: *1 reusable connection: 0
2011/09/29 09:21:07 [debug] 3565#0: *1 http upstream temp fd: -1
2011/09/29 09:21:07 [debug] 3565#0: *1 http output filter "/?"
2011/09/29 09:21:07 [debug] 3565#0: *1 http copy filter: "/?"
2011/09/29 09:21:07 [debug] 3565#0: *1 lua capture body filter, uri "/"
2011/09/29 09:21:07 [debug] 3565#0: *1 http postpone filter "/?"
BF8FAFD8
2011/09/29 09:21:07 [debug] 3565#0: *1 write old buf t:1 f:0 0A1227F8,
pos 0A1227F8, size: 148 file: 0, size: 0
2011/09/29 09:21:07 [debug] 3565#0: *1 write old buf t:1 f:0 0A174BF8,
pos 0A174C87, size: 162 file: 0, size: 0
2011/09/29 09:21:07 [debug] 3565#0: *1 write new buf t:0 f:0 00000000,
pos 00000000, size: 0 file: 0, size: 0
2011/09/29 09:21:07 [debug] 3565#0: *1 http write filter: l:1 f:0 s:310
2011/09/29 09:21:07 [debug] 3565#0: *1 http write filter limit 0
2011/09/29 09:21:07 [debug] 3565#0: *1 writev: 310
2011/09/29 09:21:07 [debug] 3565#0: *1 http write filter 00000000
2011/09/29 09:21:07 [debug] 3565#0: *1 http copy filter: 0 "/?"
2011/09/29 09:21:07 [debug] 3565#0: *1 http finalize request: 0, "/?"
a:1, c:1
2011/09/29 09:21:07 [debug] 3565#0: *1 set http keepalive handler
2011/09/29 09:21:07 [debug] 3565#0: *1 http close request
2011/09/29 09:21:07 [debug] 3565#0: *1 http log handler
2011/09/29 09:21:07 [debug] 3565#0: *1 free: 0A174BF8
2011/09/29 09:21:07 [debug] 3565#0: *1 free: 0A121C90, unused: 483
2011/09/29 09:21:07 [debug] 3565#0: *1 event timer add: 17:
15000:3020889939
2011/09/29 09:21:07 [debug] 3565#0: *1 free: 0A124890
2011/09/29 09:21:07 [debug] 3565#0: *1 free: 0A124B30
2011/09/29 09:21:07 [debug] 3565#0: *1 hc free: 00000000 0
2011/09/29 09:21:07 [debug] 3565#0: *1 hc busy: 00000000 0
2011/09/29 09:21:07 [debug] 3565#0: *1 tcp_nodelay
2011/09/29 09:21:07 [debug] 3565#0: *1 reusable connection: 1
2011/09/29 09:21:07 [debug] 3565#0: *1 post event 0A15ACB8
2011/09/29 09:21:07 [debug] 3565#0: *1 delete posted event 0A15ACB8
2011/09/29 09:21:07 [debug] 3565#0: *1 http keepalive handler
2011/09/29 09:21:07 [debug] 3565#0: *1 malloc: 0A124890:1024
2011/09/29 09:21:07 [debug] 3565#0: *1 recv: fd:17 0 of 1024
2011/09/29 09:21:07 [info] 3565#0: *1 client 127.0.0.1 closed keepalive
connection
2011/09/29 09:21:07 [debug] 3565#0: *1 close http connection: 17
2011/09/29 09:21:07 [debug] 3565#0: *1 event timer del: 17: 3020889939
2011/09/29 09:21:07 [debug] 3565#0: *1 reusable connection: 0
2011/09/29 09:21:07 [debug] 3565#0: *1 free: 0A124890
2011/09/29 09:21:07 [debug] 3565#0: *1 free: 00000000
2011/09/29 09:21:07 [debug] 3565#0: *1 free: 0A11BC80, unused: 56

# cat error90.log
2011/09/29 09:21:05 [debug] 3565#0: epoll add event: fd:11 op:1
ev:00000001
2011/09/29 09:21:07 [debug] 3565#0: post event 0A15ABE8
2011/09/29 09:21:07 [debug] 3565#0: delete posted event 0A15ABE8
2011/09/29 09:21:07 [debug] 3565#0: accept on 0.0.0.0:90, ready: 0
2011/09/29 09:21:07 [debug] 3565#0: posix_memalign: 0A11BE10:256 @16
2011/09/29 09:21:07 [debug] 3565#0: *3 accept: 127.0.0.1 fd:19
2011/09/29 09:21:07 [debug] 3565#0: *3 event timer add: 19:
60000:3020934939
2011/09/29 09:21:07 [debug] 3565#0: *3 epoll add event: fd:19 op:1
ev:80000001
2011/09/29 09:21:07 [debug] 3565#0: *3 post event 0A15AD20
2011/09/29 09:21:07 [debug] 3565#0: *3 delete posted event 0A15AD20
2011/09/29 09:21:07 [debug] 3565#0: *3 malloc: 0A124F38:664
2011/09/29 09:21:07 [debug] 3565#0: *3 malloc: 0A122CA8:1024
2011/09/29 09:21:07 [debug] 3565#0: *3 posix_memalign: 0A1230B0:4096
@16
2011/09/29 09:21:07 [debug] 3565#0: *3 http process request line
2011/09/29 09:21:07 [debug] 3565#0: *3 recv: fd:19 177 of 1024
2011/09/29 09:21:07 [debug] 3565#0: *3 http request line: "GET /foo/
HTTP/1.0"
2011/09/29 09:21:07 [debug] 3565#0: *3 http uri: "/foo/"
2011/09/29 09:21:07 [debug] 3565#0: *3 http args: ""
2011/09/29 09:21:07 [debug] 3565#0: *3 http exten: ""
2011/09/29 09:21:07 [debug] 3565#0: *3 http process request header line
2011/09/29 09:21:07 [debug] 3565#0: *3 http header: "Host:
localhost:90"
2011/09/29 09:21:07 [debug] 3565#0: *3 http header: "Connection: close"
2011/09/29 09:21:07 [debug] 3565#0: *3 http header: "User-Agent:
curl/7.15.5 (i686-redhat-linux-gnu) libcurl/7.15.5 OpenSSL/0.9.8b
zlib/1.2.3 libidn/0.6.5"
2011/09/29 09:21:07 [debug] 3565#0: *3 http header: "Accept: */*"
2011/09/29 09:21:07 [debug] 3565#0: *3 http header done
2011/09/29 09:21:07 [debug] 3565#0: *3 event timer del: 19: 3020934939
2011/09/29 09:21:07 [debug] 3565#0: *3 generic phase: 0
2011/09/29 09:21:07 [debug] 3565#0: *3 rewrite phase: 1
2011/09/29 09:21:07 [debug] 3565#0: *3 test location: "/"
2011/09/29 09:21:07 [debug] 3565#0: *3 using configuration "/"
2011/09/29 09:21:07 [debug] 3565#0: *3 http cl:-1 max:1048576
2011/09/29 09:21:07 [debug] 3565#0: *3 rewrite phase: 3
2011/09/29 09:21:07 [debug] 3565#0: *3 post rewrite phase: 4
2011/09/29 09:21:07 [debug] 3565#0: *3 generic phase: 5
2011/09/29 09:21:07 [debug] 3565#0: *3 generic phase: 6
2011/09/29 09:21:07 [debug] 3565#0: *3 generic phase: 7
2011/09/29 09:21:07 [debug] 3565#0: *3 access phase: 8
2011/09/29 09:21:07 [debug] 3565#0: *3 access phase: 9
2011/09/29 09:21:07 [debug] 3565#0: *3 post access phase: 10
2011/09/29 09:21:07 [debug] 3565#0: *3 content phase: 11
2011/09/29 09:21:07 [debug] 3565#0: *3 open index
"/usr/share/nginx/html/foo/index.html"
2011/09/29 09:21:07 [debug] 3565#0: *3 stat()
"/usr/share/nginx/html/foo/index.html" failed (2: No such file or
directory)
2011/09/29 09:21:07 [debug] 3565#0: *3 http index check dir:
"/usr/share/nginx/html/foo"
2011/09/29 09:21:07 [error] 3565#0: *3
"/usr/share/nginx/html/foo/index.html" is not found (2: No such file or
directory), client: 127.0.0.1, server: _, request: "GET /foo/ HTTP/1.0",
host: "localhost:90"
2011/09/29 09:21:07 [debug] 3565#0: *3 http finalize request: 404,
"/foo/?" a:1, c:1
2011/09/29 09:21:07 [debug] 3565#0: *3 http special response: 404,
"/foo/?"
2011/09/29 09:21:07 [debug] 3565#0: *3 http set discard body
2011/09/29 09:21:07 [debug] 3565#0: *3 lua capture header filter, uri
"/foo/"
2011/09/29 09:21:07 [debug] 3565#0: *3 HTTP/1.1 404 Not Found
Server: nginx
Date: Thu, 29 Sep 2011 07:21:07 GMT
Content-Type: text/html
Content-Length: 162
Connection: close

2011/09/29 09:21:07 [debug] 3565#0: *3 write new buf t:1 f:0 0A12362C,
pos 0A12362C, size: 143 file: 0, size: 0
2011/09/29 09:21:07 [debug] 3565#0: *3 http write filter: l:0 f:0 s:143
2011/09/29 09:21:07 [debug] 3565#0: *3 http output filter "/foo/?"
2011/09/29 09:21:07 [debug] 3565#0: *3 http copy filter: "/foo/?"
2011/09/29 09:21:07 [debug] 3565#0: *3 lua capture body filter, uri
"/foo/"
2011/09/29 09:21:07 [debug] 3565#0: *3 http postpone filter "/foo/?"
0A123778
2011/09/29 09:21:07 [debug] 3565#0: *3 write old buf t:1 f:0 0A12362C,
pos 0A12362C, size: 143 file: 0, size: 0
2011/09/29 09:21:07 [debug] 3565#0: *3 write new buf t:0 f:0 00000000,
pos 08116EE0, size: 116 file: 0, size: 0
2011/09/29 09:21:07 [debug] 3565#0: *3 write new buf t:0 f:0 00000000,
pos 08116920, size: 46 file: 0, size: 0
2011/09/29 09:21:07 [debug] 3565#0: *3 http write filter: l:1 f:0 s:305
2011/09/29 09:21:07 [debug] 3565#0: *3 http write filter limit 0
2011/09/29 09:21:07 [debug] 3565#0: *3 writev: 305
2011/09/29 09:21:07 [debug] 3565#0: *3 http write filter 00000000
2011/09/29 09:21:07 [debug] 3565#0: *3 http copy filter: 0 "/foo/?"
2011/09/29 09:21:07 [debug] 3565#0: *3 http finalize request: 0,
"/foo/?" a:1, c:1
2011/09/29 09:21:07 [debug] 3565#0: *3 event timer add: 19:
5000:3020879939
2011/09/29 09:21:07 [debug] 3565#0: *3 http lingering close handler
2011/09/29 09:21:07 [debug] 3565#0: *3 recv: fd:19 -1 of 4096
2011/09/29 09:21:07 [debug] 3565#0: *3 recv() not ready (11: Resource
temporarily unavailable)
2011/09/29 09:21:07 [debug] 3565#0: *3 lingering read: -2
2011/09/29 09:21:07 [debug] 3565#0: *3 event timer: 19, old: 3020879939,
new: 3020879939
2011/09/29 09:21:07 [debug] 3565#0: *3 post event 0A15AD20
2011/09/29 09:21:07 [debug] 3565#0: *3 delete posted event 0A15AD20
2011/09/29 09:21:07 [debug] 3565#0: *3 http lingering close handler
2011/09/29 09:21:07 [debug] 3565#0: *3 recv: fd:19 0 of 4096
2011/09/29 09:21:07 [debug] 3565#0: *3 lingering read: 0
2011/09/29 09:21:07 [debug] 3565#0: *3 http request count:1 blk:0
2011/09/29 09:21:07 [debug] 3565#0: *3 http close request
2011/09/29 09:21:07 [debug] 3565#0: *3 http log handler
2011/09/29 09:21:07 [debug] 3565#0: *3 free: 0A1230B0, unused: 2143
2011/09/29 09:21:07 [debug] 3565#0: *3 close http connection: 19
2011/09/29 09:21:07 [debug] 3565#0: *3 event timer del: 19: 3020879939
2011/09/29 09:21:07 [debug] 3565#0: *3 reusable connection: 0
2011/09/29 09:21:07 [debug] 3565#0: *3 free: 0A122CA8
2011/09/29 09:21:07 [debug] 3565#0: *3 free: 0A124F38
2011/09/29 09:21:07 [debug] 3565#0: *3 free: 0A11BE10, unused: 56

Any advice how to get rid of the 'foo/" in the proxied request would be
highly welcome.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215830,216005#msg-216005


From nginx-forum at nginx.us  Thu Sep 29 07:32:10 2011
From: nginx-forum at nginx.us (csg)
Date: Thu, 29 Sep 2011 03:32:10 -0400
Subject: Does Nginx honor DNS TTLs for proxy upstreams?
In-Reply-To: <826ac94aad7afd513c2d51d50fe1601f.NginxMailingListEnglish@forum.nginx.org>
References: <20110928161355.GZ1137@mdounin.ru>
	<03e18959a4108577d38690656ce65b0f.NginxMailingListEnglish@forum.nginx.org>
	<826ac94aad7afd513c2d51d50fe1601f.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <5f94f17a77abf763d5f2bbcca03f036f.NginxMailingListEnglish@forum.nginx.org>

Ah sorry, just saw your patch ... will give that a try! Many thanks :-)

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215830,216006#msg-216006


From mdounin at mdounin.ru  Thu Sep 29 07:35:52 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Thu, 29 Sep 2011 11:35:52 +0400
Subject: fastcgi msg_id always 1?
In-Reply-To: <4E841B2B.5060704@mozapps.com>
References: <4E841B2B.5060704@mozapps.com>
Message-ID: <20110929073552.GC1137@mdounin.ru>

Hello!

On Thu, Sep 29, 2011 at 03:15:55PM +0800, Shaun savage wrote:

> I was looking at the fastcgi, I found out that all fcgi request have a
> msg_id of 1? 

Yes.

> What happens when more than 1 fcgi requests comes in and first one is
> still being handled. how does the fcgi server know  the the difference
> in requests?  Usually each request has it own connection, but if
> keepalive is on the then many request can come in on the same connection. 

The id is only need to be unique when multiple request exists in 
parallel within a single connection.  This is not the case even 
with keepalive used.

Maxim Dounin


From nginx-forum at nginx.us  Thu Sep 29 07:53:47 2011
From: nginx-forum at nginx.us (csg)
Date: Thu, 29 Sep 2011 03:53:47 -0400
Subject: Does Nginx honor DNS TTLs for proxy upstreams?
In-Reply-To: <5f94f17a77abf763d5f2bbcca03f036f.NginxMailingListEnglish@forum.nginx.org>
References: <20110928161355.GZ1137@mdounin.ru>
	<03e18959a4108577d38690656ce65b0f.NginxMailingListEnglish@forum.nginx.org>
	<826ac94aad7afd513c2d51d50fe1601f.NginxMailingListEnglish@forum.nginx.org>
	<5f94f17a77abf763d5f2bbcca03f036f.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <d9f1d01da63587bdabd2b46319ececb8.NginxMailingListEnglish@forum.nginx.org>

I can confirm that the patch fixes the issue on my test setup :-) Will
add it to my Nginx package and deploy it on production.

Many thanks Maxim for your fantastic support on this issue (and all the
work you do on this amazing and very helpful list)!

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215830,216008#msg-216008


From david.yu.ftw at gmail.com  Thu Sep 29 07:58:12 2011
From: david.yu.ftw at gmail.com (David Yu)
Date: Thu, 29 Sep 2011 15:58:12 +0800
Subject: fastcgi msg_id always 1?
In-Reply-To: <20110929073552.GC1137@mdounin.ru>
References: <4E841B2B.5060704@mozapps.com> <20110929073552.GC1137@mdounin.ru>
Message-ID: <CAOkdovLJzi+HDc6RMKfW+WVSOD1C_NpLG3z7wtpyaJH7t-fSkQ@mail.gmail.com>

On Thu, Sep 29, 2011 at 3:35 PM, Maxim Dounin <mdounin at mdounin.ru> wrote:

> Hello!
>
> On Thu, Sep 29, 2011 at 03:15:55PM +0800, Shaun savage wrote:
>
> > I was looking at the fastcgi, I found out that all fcgi request have a
> > msg_id of 1?
>
> Yes.
>
> > What happens when more than 1 fcgi requests comes in and first one is
> > still being handled. how does the fcgi server know  the the difference
> > in requests?  Usually each request has it own connection, but if
> > keepalive is on the then many request can come in on the same connection.
>
> The id is only need to be unique when multiple request exists in
> parallel within a single connection.  This is not the case even
> with keepalive used.
>
E.g multiplexing not supported.

>
> Maxim Dounin
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx
>



-- 
When the cat is away, the mouse is alone.
- David Yu
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110929/b972dab2/attachment.html>

From martronic at martronic.ch  Thu Sep 29 09:34:01 2011
From: martronic at martronic.ch (Martronic SA)
Date: Thu, 29 Sep 2011 11:34:01 +0200 (CEST)
Subject: Nginx Substitutions Filter problem
In-Reply-To: <CAGxuLK1VRVX_6rsXDSomzcnpndg5L0Rd4CqCTfnU3PfJrJyvkw@mail.gmail.com>
References: <4E80E187.6030807@ntlworld.com>
	<CAJCGDi8ksNcSHyf-iLm5vZevAKBMew_M2NvA5s62_wyGRLGtGw@mail.gmail.com>
	<4E80EB49.9000108@ianhobson.co.uk>
	<CAJCGDi_jWABeVRNZUEeKmZ9q3cpdFOsRWgH9ASiU_=GNSycdBQ@mail.gmail.com>
	<20110927045203.20830@gmx.net> <4E81701D.408@martronic.ch>
	<CAGxuLK3+sVfD8WC=9gL6z0daokOB=qc8HaE3drqsxSgbocS6=A@mail.gmail.com>
	<2a4c6102eba7e97f63ad316476db06b8.squirrel@webmail.martronic.ch>
	<CAGxuLK1VRVX_6rsXDSomzcnpndg5L0Rd4CqCTfnU3PfJrJyvkw@mail.gmail.com>
Message-ID: <744053bb97f08749b6230bfa2202e47e.squirrel@webmail.martronic.ch>

hi

log file and configuration file for nginx hopla.net are attached

Thanks
Martial


> It works in my box. Can you show me your debug log?
>
> http://wiki.nginx.org/Debugging
>
> Do't use this in the production server. The debug log will be very large.
>
>
> 2011/9/28 Martronic SA <martronic at martronic.ch>
>
>> full source attachment
>> Thanks.
>>
>>
>> <a id="portal-logo" href="http://bale.hopla.net" accesskey="1"
>> class="visualNoPrint"><img
>> src="http://bale.hopla.net/portal_logo/image" alt="Logo" title="Logo
>> Hopla-Net" height="175"
>> width="230" /></a>
>>
>> <div id="solgemabandeau" class="visualNoPrint">
>>
>>  <div id="portal-bandeau">
>>    <div id="inner-portal-bandeau" class="bannersContainer">
>>
>>          <div style="position:relative; display: block"
>> id="carousel-banner-0"
>> class="bando.jpg carousel-banner selected">
>>            <div style="height:200px; width:100%; background:transparent
>> url(http://bale.hopla.net/bando.jpg) no-repeat scroll left top;"
>> title="Une maison
>> dans un jardin, nouveau bandeau Hopla pour le printemps."
>> class="bandeau_image"></div>
>>          </div>
>>
>>
>>          <div style="display: none" id="carousel-banner-1"
>> class="bandeauhoplabale.jpg
>> carousel-banner">
>>            <div style="height:200px; width:100%; background:transparent
>> url(http://bale.hopla.net/bandeauhoplabale.jpg) no-repeat scroll left
>> top;"
>> title="Le printemps est de retour sur St Louis et B?le"
>> class="bandeau_image"></div>
>>          </div>
>>
>>        <span class="bandeauTopLeft"></span>
>>
>>        <span class="bandeauTopRight"></span>
>>        <span class="bandeauBottomLeft"></span>
>>        <span class="bandeauBottomRight"></span>
>>    </div>
>> </div>
>>
>> </div>
>>
>> > Can you show me your source page?
>> >
>> > Thanks.
>> >
>> > 2011/9/27 Martronic SA <martronic at martronic.ch>
>> >
>> >> Hello,
>> >>
>> >> We are using the "Substitutions Filter module"
>> >> to replace html content in our pages.
>> >>
>> >> Our configuration is :
>> >>
>> >> location /{
>> >> subs_filter_types text/html text/xml;
>> >> subs_filter 'http://bale.hopla.net/bons-**plans<
>> http://bale.hopla.net/bons-plans>'
>> >> 'http://bons-plans.hopla.net';
>> >> subs_filter 'id="portal-logo" href="http://bale.hopla.net"'
>> >> 'id="portal-logo" href="http://www.hopla.net"';
>> >> }
>> >>
>> >> The first substitution is working, but the second not.
>> >> Are we missing something? Is the "space" supported in substitution
>> module?
>> >>
>> >> Thank you for your help!
>> >>
>> >> Martronic SA
>> >>
>> >> nginx -V
>> >> nginx: nginx version: nginx/1.1.4
>> >> nginx: built by gcc 4.3.2 (Debian 4.3.2-1.1)
>> >> nginx: TLS SNI support enabled
>> >> nginx: configure arguments: --conf-path=/etc/nginx/nginx.**conf
>> >> --error-log-path=/var/log/**nginx/error.log
>> --pid-path=/var/run/nginx.pid
>> >> --lock-path=/var/lock/nginx.**lock --http-log-path=/var/log/**
>> >> nginx/access.log
>> >> --with-http_dav_module
>> --http-client-body-temp-path=/**var/lib/nginx/body
>> >> --with-http_ssl_module
>> >> --http-proxy-temp-path=/var/**lib/nginx/proxy
>> >> --with-http_stub_status_module
>> >> --http-fastcgi-temp-path=/var/**lib/nginx/fastcgi
>> >> --add-module=/var/lib/**substitutions4nginx-read-only
>> >>
>> >>
>> >>
>> >> ______________________________**_________________
>> >> nginx mailing list
>> >> nginx at nginx.org
>> >> http://mailman.nginx.org/**mailman/listinfo/nginx<
>> http://mailman.nginx.org/mailman/listinfo/nginx>
>> >>
>> > _______________________________________________
>> > nginx mailing list
>> > nginx at nginx.org
>> > http://mailman.nginx.org/mailman/listinfo/nginx
>>
>>
>>
>> _______________________________________________
>> nginx mailing list
>> nginx at nginx.org
>> http://mailman.nginx.org/mailman/listinfo/nginx
>>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx
-------------- next part --------------
A non-text attachment was scrubbed...
Name: hopla.net.gz
Type: application/gzip
Size: 1176 bytes
Desc: not available
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110929/5ba5ba28/attachment-0002.bin>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: debug_nginx.log.gz
Type: application/gzip
Size: 106603 bytes
Desc: not available
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110929/5ba5ba28/attachment-0003.bin>

From nginx-forum at nginx.us  Thu Sep 29 10:18:51 2011
From: nginx-forum at nginx.us (sajt)
Date: Thu, 29 Sep 2011 06:18:51 -0400
Subject: conflicting server name "" on 0.0.0.0:80, ignored
Message-ID: <c7d3f9f51b967e0f29a6ed91c6a502d7.NginxMailingListEnglish@forum.nginx.org>

Hello,

I have an Ubuntu for drupal development with 2+1 local domain. I putted
2 domain name to my hosts file, I have the default and two similar
configuration files in my sites-enabled folder:

--
server {
  if ($request_method !~ ^(GET|HEAD|POST)$ ) {
    return 444;
  }

  root /home/sajt/Projectek/eupharma/htdocs;
  index index.php
  server_name eupharmaltd;
  location / {
    # First attempt to serve request as file, then
    # as directory, then fall back to index.html
    try_files $uri $uri/ /index.html;
  }
  location ~ .php$ {
    fastcgi_pass   127.0.0.1:9000;  # By all means use a different
server for the fcgi processes if you need to
    fastcgi_index  index.php;

    #fastcgi_param  SCRIPT_FILENAME 
/path/to/drupal$fastcgi_script_name;   # !! <--- Another path reference
for you.
    include /etc/nginx/fastcgi_params;
 }
 if (!-e $request_filename) {
    rewrite ^/(.*)$ /index.php?q=$1 last;
    break;
 }
}
--

The difference is only the root and the server_name. I get the following
error:

Restarting nginx: [warn]: conflicting server name "" on 0.0.0.0:80,
ignored

And when I try to access the all domain is showing the same page. The
default is working fine.

Thanks for answers

Tamas

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,216017,216017#msg-216017


From nbubingo at gmail.com  Thu Sep 29 10:23:57 2011
From: nbubingo at gmail.com (=?GB2312?B?0qbOsLHz?=)
Date: Thu, 29 Sep 2011 18:23:57 +0800
Subject: Nginx Substitutions Filter problem
In-Reply-To: <744053bb97f08749b6230bfa2202e47e.squirrel@webmail.martronic.ch>
References: <4E80E187.6030807@ntlworld.com>
	<CAJCGDi8ksNcSHyf-iLm5vZevAKBMew_M2NvA5s62_wyGRLGtGw@mail.gmail.com>
	<4E80EB49.9000108@ianhobson.co.uk>
	<CAJCGDi_jWABeVRNZUEeKmZ9q3cpdFOsRWgH9ASiU_=GNSycdBQ@mail.gmail.com>
	<20110927045203.20830@gmx.net> <4E81701D.408@martronic.ch>
	<CAGxuLK3+sVfD8WC=9gL6z0daokOB=qc8HaE3drqsxSgbocS6=A@mail.gmail.com>
	<2a4c6102eba7e97f63ad316476db06b8.squirrel@webmail.martronic.ch>
	<CAGxuLK1VRVX_6rsXDSomzcnpndg5L0Rd4CqCTfnU3PfJrJyvkw@mail.gmail.com>
	<744053bb97f08749b6230bfa2202e47e.squirrel@webmail.martronic.ch>
Message-ID: <CAGxuLK2+3Z8M=UTfXvhEuyRgpJskXEX4EncQckNNRsRzbrJ-9A@mail.gmail.com>

Sorry. I can't find any log from my substitutions module in your debug log.

And I find the possible problem. The response from backend  server is
gzipped. You can compress the response with Nginx's gzip module.

You should disable the gzipped response from backend server, like this:

proxy_set_header Accept-Encoding "";


2011/9/29 Martronic SA <martronic at martronic.ch>

> hi
>
> log file and configuration file for nginx hopla.net are attached
>
> Thanks
> Martial
>
>
> > It works in my box. Can you show me your debug log?
> >
> > http://wiki.nginx.org/Debugging
> >
> > Do't use this in the production server. The debug log will be very large.
> >
> >
> > 2011/9/28 Martronic SA <martronic at martronic.ch>
> >
> >> full source attachment
> >> Thanks.
> >>
> >>
> >> <a id="portal-logo" href="http://bale.hopla.net" accesskey="1"
> >> class="visualNoPrint"><img
> >> src="http://bale.hopla.net/portal_logo/image" alt="Logo" title="Logo
> >> Hopla-Net" height="175"
> >> width="230" /></a>
> >>
> >> <div id="solgemabandeau" class="visualNoPrint">
> >>
> >>  <div id="portal-bandeau">
> >>    <div id="inner-portal-bandeau" class="bannersContainer">
> >>
> >>          <div style="position:relative; display: block"
> >> id="carousel-banner-0"
> >> class="bando.jpg carousel-banner selected">
> >>            <div style="height:200px; width:100%; background:transparent
> >> url(http://bale.hopla.net/bando.jpg) no-repeat scroll left top;"
> >> title="Une maison
> >> dans un jardin, nouveau bandeau Hopla pour le printemps."
> >> class="bandeau_image"></div>
> >>          </div>
> >>
> >>
> >>          <div style="display: none" id="carousel-banner-1"
> >> class="bandeauhoplabale.jpg
> >> carousel-banner">
> >>            <div style="height:200px; width:100%; background:transparent
> >> url(http://bale.hopla.net/bandeauhoplabale.jpg) no-repeat scroll left
> >> top;"
> >> title="Le printemps est de retour sur St Louis et B?le"
> >> class="bandeau_image"></div>
> >>          </div>
> >>
> >>        <span class="bandeauTopLeft"></span>
> >>
> >>        <span class="bandeauTopRight"></span>
> >>        <span class="bandeauBottomLeft"></span>
> >>        <span class="bandeauBottomRight"></span>
> >>    </div>
> >> </div>
> >>
> >> </div>
> >>
> >> > Can you show me your source page?
> >> >
> >> > Thanks.
> >> >
> >> > 2011/9/27 Martronic SA <martronic at martronic.ch>
> >> >
> >> >> Hello,
> >> >>
> >> >> We are using the "Substitutions Filter module"
> >> >> to replace html content in our pages.
> >> >>
> >> >> Our configuration is :
> >> >>
> >> >> location /{
> >> >> subs_filter_types text/html text/xml;
> >> >> subs_filter 'http://bale.hopla.net/bons-**plans<
> >> http://bale.hopla.net/bons-plans>'
> >> >> 'http://bons-plans.hopla.net';
> >> >> subs_filter 'id="portal-logo" href="http://bale.hopla.net"'
> >> >> 'id="portal-logo" href="http://www.hopla.net"';
> >> >> }
> >> >>
> >> >> The first substitution is working, but the second not.
> >> >> Are we missing something? Is the "space" supported in substitution
> >> module?
> >> >>
> >> >> Thank you for your help!
> >> >>
> >> >> Martronic SA
> >> >>
> >> >> nginx -V
> >> >> nginx: nginx version: nginx/1.1.4
> >> >> nginx: built by gcc 4.3.2 (Debian 4.3.2-1.1)
> >> >> nginx: TLS SNI support enabled
> >> >> nginx: configure arguments: --conf-path=/etc/nginx/nginx.**conf
> >> >> --error-log-path=/var/log/**nginx/error.log
> >> --pid-path=/var/run/nginx.pid
> >> >> --lock-path=/var/lock/nginx.**lock --http-log-path=/var/log/**
> >> >> nginx/access.log
> >> >> --with-http_dav_module
> >> --http-client-body-temp-path=/**var/lib/nginx/body
> >> >> --with-http_ssl_module
> >> >> --http-proxy-temp-path=/var/**lib/nginx/proxy
> >> >> --with-http_stub_status_module
> >> >> --http-fastcgi-temp-path=/var/**lib/nginx/fastcgi
> >> >> --add-module=/var/lib/**substitutions4nginx-read-only
> >> >>
> >> >>
> >> >>
> >> >> ______________________________**_________________
> >> >> nginx mailing list
> >> >> nginx at nginx.org
> >> >> http://mailman.nginx.org/**mailman/listinfo/nginx<
> >> http://mailman.nginx.org/mailman/listinfo/nginx>
> >> >>
> >> > _______________________________________________
> >> > nginx mailing list
> >> > nginx at nginx.org
> >> > http://mailman.nginx.org/mailman/listinfo/nginx
> >>
> >>
> >>
> >> _______________________________________________
> >> nginx mailing list
> >> nginx at nginx.org
> >> http://mailman.nginx.org/mailman/listinfo/nginx
> >>
> > _______________________________________________
> > nginx mailing list
> > nginx at nginx.org
> > http://mailman.nginx.org/mailman/listinfo/nginx
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110929/abae9449/attachment.html>

From ggarron at alketech.com  Thu Sep 29 10:26:21 2011
From: ggarron at alketech.com (Guillermo Garron)
Date: Thu, 29 Sep 2011 06:26:21 -0400
Subject: Happy birthday, Igor ?
In-Reply-To: <20110928113554.GF80782@nginx.com>
References: <CA+FsSoLqNX9j8SqjWa6-1tY-_-nCak1CfoMzkZDPd8RR-QsnBA@mail.gmail.com>
	<20110928113554.GF80782@nginx.com>
Message-ID: <AC85F1E8-D67B-4C51-ADE5-847D01C0C89F@alketech.com>



On Sep 28, 2011, at 7:35 AM, Igor Sysoev <igor at sysoev.ru> wrote:

> On Wed, Sep 28, 2011 at 12:20:21PM +0200, Antoine Bonavita wrote:
>> Given the amount of times Sep 28th 1970 is mentioned in the nginx code
>> (see below), I am putting a rough guess that this is your birthday.
>> So, in this very special day, I wanted to thank you for your amazing
>> little kid and I would have loved to buy you a beer but you're a bit
>> far away. If you ever get near the south of France, I'll buy you one.
>> 
>> Of course, I could be completely wrong.
> 
> You are right, this is my birthday. Thank you.
> 
> 
> -- 
> Igor Sysoev

A little late, but my best wishes and thanks from Bolivia!
You've done a great contribution. 
> 
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx


From martronic at martronic.ch  Thu Sep 29 12:11:17 2011
From: martronic at martronic.ch (Martronic SA)
Date: Thu, 29 Sep 2011 14:11:17 +0200 (CEST)
Subject: Nginx Substitutions Filter problem
In-Reply-To: <CAGxuLK2+3Z8M=UTfXvhEuyRgpJskXEX4EncQckNNRsRzbrJ-9A@mail.gmail.com>
References: <4E80E187.6030807@ntlworld.com>
	<CAJCGDi8ksNcSHyf-iLm5vZevAKBMew_M2NvA5s62_wyGRLGtGw@mail.gmail.com>
	<4E80EB49.9000108@ianhobson.co.uk>
	<CAJCGDi_jWABeVRNZUEeKmZ9q3cpdFOsRWgH9ASiU_=GNSycdBQ@mail.gmail.com>
	<20110927045203.20830@gmx.net> <4E81701D.408@martronic.ch>
	<CAGxuLK3+sVfD8WC=9gL6z0daokOB=qc8HaE3drqsxSgbocS6=A@mail.gmail.com>
	<2a4c6102eba7e97f63ad316476db06b8.squirrel@webmail.martronic.ch>
	<CAGxuLK1VRVX_6rsXDSomzcnpndg5L0Rd4CqCTfnU3PfJrJyvkw@mail.gmail.com>
	<744053bb97f08749b6230bfa2202e47e.squirrel@webmail.martronic.ch>
	<CAGxuLK2+3Z8M=UTfXvhEuyRgpJskXEX4EncQckNNRsRzbrJ-9A@mail.gmail.com>
Message-ID: <b8708a864833fdbf2c154127368f2d93.squirrel@webmail.martronic.ch>

hi

thank you very much for your help.
yes it was the gzip problematic.
maitenent it works great

Thanks
Martial

> Sorry. I can't find any log from my substitutions module in your debug log.
>
> And I find the possible problem. The response from backend  server is
> gzipped. You can compress the response with Nginx's gzip module.
>
> You should disable the gzipped response from backend server, like this:
>
> proxy_set_header Accept-Encoding "";
>
>
> 2011/9/29 Martronic SA <martronic at martronic.ch>
>
>> hi
>>
>> log file and configuration file for nginx hopla.net are attached
>>
>> Thanks
>> Martial
>>
>>
>> > It works in my box. Can you show me your debug log?
>> >
>> > http://wiki.nginx.org/Debugging
>> >
>> > Do't use this in the production server. The debug log will be very large.
>> >
>> >
>> > 2011/9/28 Martronic SA <martronic at martronic.ch>
>> >
>> >> full source attachment
>> >> Thanks.
>> >>
>> >>
>> >> <a id="portal-logo" href="http://bale.hopla.net" accesskey="1"
>> >> class="visualNoPrint"><img
>> >> src="http://bale.hopla.net/portal_logo/image" alt="Logo" title="Logo
>> >> Hopla-Net" height="175"
>> >> width="230" /></a>
>> >>
>> >> <div id="solgemabandeau" class="visualNoPrint">
>> >>
>> >>  <div id="portal-bandeau">
>> >>    <div id="inner-portal-bandeau" class="bannersContainer">
>> >>
>> >>          <div style="position:relative; display: block"
>> >> id="carousel-banner-0"
>> >> class="bando.jpg carousel-banner selected">
>> >>            <div style="height:200px; width:100%; background:transparent
>> >> url(http://bale.hopla.net/bando.jpg) no-repeat scroll left top;"
>> >> title="Une maison
>> >> dans un jardin, nouveau bandeau Hopla pour le printemps."
>> >> class="bandeau_image"></div>
>> >>          </div>
>> >>
>> >>
>> >>          <div style="display: none" id="carousel-banner-1"
>> >> class="bandeauhoplabale.jpg
>> >> carousel-banner">
>> >>            <div style="height:200px; width:100%; background:transparent
>> >> url(http://bale.hopla.net/bandeauhoplabale.jpg) no-repeat scroll left
>> >> top;"
>> >> title="Le printemps est de retour sur St Louis et B??le"
>> >> class="bandeau_image"></div>
>> >>          </div>
>> >>
>> >>        <span class="bandeauTopLeft"></span>
>> >>
>> >>        <span class="bandeauTopRight"></span>
>> >>        <span class="bandeauBottomLeft"></span>
>> >>        <span class="bandeauBottomRight"></span>
>> >>    </div>
>> >> </div>
>> >>
>> >> </div>
>> >>
>> >> > Can you show me your source page?
>> >> >
>> >> > Thanks.
>> >> >
>> >> > 2011/9/27 Martronic SA <martronic at martronic.ch>
>> >> >
>> >> >> Hello,
>> >> >>
>> >> >> We are using the "Substitutions Filter module"
>> >> >> to replace html content in our pages.
>> >> >>
>> >> >> Our configuration is :
>> >> >>
>> >> >> location /{
>> >> >> subs_filter_types text/html text/xml;
>> >> >> subs_filter 'http://bale.hopla.net/bons-**plans<
>> >> http://bale.hopla.net/bons-plans>'
>> >> >> 'http://bons-plans.hopla.net';
>> >> >> subs_filter 'id="portal-logo" href="http://bale.hopla.net"'
>> >> >> 'id="portal-logo" href="http://www.hopla.net"';
>> >> >> }
>> >> >>
>> >> >> The first substitution is working, but the second not.
>> >> >> Are we missing something? Is the "space" supported in substitution
>> >> module?
>> >> >>
>> >> >> Thank you for your help!
>> >> >>
>> >> >> Martronic SA
>> >> >>
>> >> >> nginx -V
>> >> >> nginx: nginx version: nginx/1.1.4
>> >> >> nginx: built by gcc 4.3.2 (Debian 4.3.2-1.1)
>> >> >> nginx: TLS SNI support enabled
>> >> >> nginx: configure arguments: --conf-path=/etc/nginx/nginx.**conf
>> >> >> --error-log-path=/var/log/**nginx/error.log
>> >> --pid-path=/var/run/nginx.pid
>> >> >> --lock-path=/var/lock/nginx.**lock --http-log-path=/var/log/**
>> >> >> nginx/access.log
>> >> >> --with-http_dav_module
>> >> --http-client-body-temp-path=/**var/lib/nginx/body
>> >> >> --with-http_ssl_module
>> >> >> --http-proxy-temp-path=/var/**lib/nginx/proxy
>> >> >> --with-http_stub_status_module
>> >> >> --http-fastcgi-temp-path=/var/**lib/nginx/fastcgi
>> >> >> --add-module=/var/lib/**substitutions4nginx-read-only
>> >> >>
>> >> >>
>> >> >>
>> >> >> ______________________________**_________________
>> >> >> nginx mailing list
>> >> >> nginx at nginx.org
>> >> >> http://mailman.nginx.org/**mailman/listinfo/nginx<
>> >> http://mailman.nginx.org/mailman/listinfo/nginx>
>> >> >>
>> >> > _______________________________________________
>> >> > nginx mailing list
>> >> > nginx at nginx.org
>> >> > http://mailman.nginx.org/mailman/listinfo/nginx
>> >>
>> >>
>> >>
>> >> _______________________________________________
>> >> nginx mailing list
>> >> nginx at nginx.org
>> >> http://mailman.nginx.org/mailman/listinfo/nginx
>> >>
>> > _______________________________________________
>> > nginx mailing list
>> > nginx at nginx.org
>> > http://mailman.nginx.org/mailman/listinfo/nginx
>>
>> _______________________________________________
>> nginx mailing list
>> nginx at nginx.org
>> http://mailman.nginx.org/mailman/listinfo/nginx
>>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx



From sb at waeme.net  Thu Sep 29 13:33:29 2011
From: sb at waeme.net (Sergey Budnevitch)
Date: Thu, 29 Sep 2011 17:33:29 +0400
Subject: NGINX On REDHAT LINUX 5
In-Reply-To: <OF46BDA6DE.B2B8F259-ON85257919.0059E70A-85257919.005A3D64@jlg.com>
References: <OF46BDA6DE.B2B8F259-ON85257919.0059E70A-85257919.005A3D64@jlg.com>
Message-ID: <3DE0EF14-8A9C-4576-ACFC-756922577146@waeme.net>


On 28.09.2011, at 20:25, RFGuerengomba at JLG.com wrote:

> 
> Hi,
> 
> I installed NGINX on Ubuntu before and it was working fine, but recently
> our company decided to move to RedHat Linux. I am having little difficulty.
> In Ubuntu under the NGINX system folder I have the folders called
> "sites-available" and "sites-enabled"
> I seem not to have that in REDHAT Linux. Can you please help me to
> understand better?

I am answering to nginx at nginx.org instead of nginx-devel, cause nginx-devel
is incorrect maillist for questions of this kind

In redhat linux you could place virtual sites configs in /etc/nginx/conf.d/ in
files with .conf extension.
ubuntu nginx package has "include /etc/nginx/sites-enabled;" directive
in default nginx.conf, redhat one has "include /etc/nginx/conf.d/*.conf;".


> 
> Roger F. Guerengomba
> Senior System Administrator
> JLG Industires, Inc.
> 1 JLG Drive
> McConnellsburg, PA 17233
> 717-485-6906
> rfguerengomba at jlg.com
> http://www.jlg.com
> 
> ***********************************************************************
> The information contained in this transmission is confidential.  It is
> intended solely for the use of the individual(s) or organization(s) to
> whom it is addressed.  Any disclosure, copying or further distribution
> is not permitted unless such privilege is explicitly granted in writing
> by JLG Industries, Inc.
> Further, JLG Industries, Inc. is not responsible for the proper and
> complete transmission of the substance of this communication nor for
> any delay in its receipt.
> 
> _______________________________________________
> nginx-devel mailing list
> nginx-devel at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx-devel


From nginx-forum at nginx.us  Thu Sep 29 14:16:21 2011
From: nginx-forum at nginx.us (firestorm)
Date: Thu, 29 Sep 2011 10:16:21 -0400
Subject: Problem with GZIP
Message-ID: <21793226c760e08d8926eb9b18c5d323.NginxMailingListEnglish@forum.nginx.org>

I have ubuntu server 10.04, nginx 1.1.4(compiled) and my gzip support is
not working. This is my configuration:

nginx: nginx version: nginx/1.1.4
nginx: built by gcc 4.4.3 (Ubuntu 4.4.3-4ubuntu5)
nginx: TLS SNI support enabled
nginx: configure arguments: --user=www-data --group=www-data
--with-http_ssl_module --with-http_realip_module
--with-http_addition_module --with-http_sub_module
--with-http_dav_module --with-http_flv_module
--with-http_gzip_static_module --with-http_random_index_module
--with-http_secure_link_module --with-http_stub_status_module
--with-debug

my nginx configuration is:

http {
    include       mime.types;
    default_type  application/octet-stream;

    #log_format  main  '$remote_addr - $remote_user [$time_local]
"$request" '
    #                  '$status $body_bytes_sent "$http_referer" '
    #                  '"$http_user_agent" "$http_x_forwarded_for"';

    #access_log  logs/access.log  main;

    sendfile        on;
    tcp_nodelay     on;
    tcp_nopush      off;

    #keepalive_timeout  0;
    keepalive_timeout  65;

    gzip on;
    gzip_min_length 1000;
    gzip_comp_level 6;
    gzip_proxied expired no-cache no-store private auth;
    gzip_types text/plain application/xml;
    #gzip_disable "MSIE [1-6]\.";


    server {
	listen 80;
	server_name localhost;

	#access_log /var/log/nginx/alfaomega.access_log main;
	#error_log /var/log/nginx/alfaomega.error_log info;

	root /var/www/alfaomega/web;
	charset utf-8;

       location = / {
	   index index.php;
       }
   }

#Other locations

}

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,216030,216030#msg-216030


From mdounin at mdounin.ru  Thu Sep 29 14:27:43 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Thu, 29 Sep 2011 18:27:43 +0400
Subject: Problem with GZIP
In-Reply-To: <21793226c760e08d8926eb9b18c5d323.NginxMailingListEnglish@forum.nginx.org>
References: <21793226c760e08d8926eb9b18c5d323.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20110929142742.GF1137@mdounin.ru>

Hello!

On Thu, Sep 29, 2011 at 10:16:21AM -0400, firestorm wrote:

> I have ubuntu server 10.04, nginx 1.1.4(compiled) and my gzip support is
> not working. This is my configuration:

Please show how do you test it's "not working".  If in doubt, 
please provide debug log, see http://wiki.nginx.org/Debugging.

Maxim Dounin


From nginx-forum at nginx.us  Thu Sep 29 15:06:22 2011
From: nginx-forum at nginx.us (tcbarrett)
Date: Thu, 29 Sep 2011 11:06:22 -0400
Subject: Best way to burst static file cache
In-Reply-To: <0668d2360623a0b4f9f9c84c0ab39f30.NginxMailingListEnglish@forum.nginx.org>
References: <0668d2360623a0b4f9f9c84c0ab39f30.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <da5c45696606c5553b7cdab562cd38b6.NginxMailingListEnglish@forum.nginx.org>

Nobody has any ideas?

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215835,216034#msg-216034


From RFGuerengomba at JLG.com  Thu Sep 29 15:44:51 2011
From: RFGuerengomba at JLG.com (RFGuerengomba at JLG.com)
Date: Thu, 29 Sep 2011 11:44:51 -0400
Subject: NGINX On REDHAT LINUX 5
In-Reply-To: <3DE0EF14-8A9C-4576-ACFC-756922577146@waeme.net>
References: <OF46BDA6DE.B2B8F259-ON85257919.0059E70A-85257919.005A3D64@jlg.com>
	<3DE0EF14-8A9C-4576-ACFC-756922577146@waeme.net>
Message-ID: <OF18EAFF31.FE20F463-ON8525791A.00561D2B-8525791A.00568133@jlg.com>

Thank you so much for your help; I have some issues when I issue:

/etc/init.d/nginx start or restart; it return not output.

when I issued the command

/usr/local/nginx/sbin/nginx -t  the test is successful

but

/usr/local/nginx/sbin/nginx-s reload  It gives me: invalid PID number "" in
"/usr/local/nginx/logs/nginx.pid"

when I look at the the nginx.pid number, there is no number in there.

Can you please help me on how to solve this problem?

Thanks,

Roger F. Guerengomba
Senior System Administrator
JLG Industires, Inc.
1 JLG Drive
McConnellsburg, PA 17233
717-485-6906
rfguerengomba at jlg.com
http://www.jlg.com



From:	sb at waeme.net
To:	nginx at nginx.org
Cc:	rfguerengomba at jlg.com
Date:	09/29/2011 09:33 AM
Subject:	Re: NGINX On REDHAT LINUX 5




On 28.09.2011, at 20:25, RFGuerengomba at JLG.com wrote:

>
> Hi,
>
> I installed NGINX on Ubuntu before and it was working fine, but recently
> our company decided to move to RedHat Linux. I am having little
difficulty.
> In Ubuntu under the NGINX system folder I have the folders called
> "sites-available" and "sites-enabled"
> I seem not to have that in REDHAT Linux. Can you please help me to
> understand better?

I am answering to nginx at nginx.org instead of nginx-devel, cause nginx-devel
is incorrect maillist for questions of this kind

In redhat linux you could place virtual sites configs in /etc/nginx/conf.d/
in
files with .conf extension.
ubuntu nginx package has "include /etc/nginx/sites-enabled;" directive
in default nginx.conf, redhat one has "include /etc/nginx/conf.d/*.conf;".


>
> Roger F. Guerengomba
> Senior System Administrator
> JLG Industires, Inc.
> 1 JLG Drive
> McConnellsburg, PA 17233
> 717-485-6906
> rfguerengomba at jlg.com
> http://www.jlg.com
>
> ***********************************************************************
> The information contained in this transmission is confidential.  It is
> intended solely for the use of the individual(s) or organization(s) to
> whom it is addressed.  Any disclosure, copying or further distribution
> is not permitted unless such privilege is explicitly granted in writing
> by JLG Industries, Inc.
> Further, JLG Industries, Inc. is not responsible for the proper and
> complete transmission of the substance of this communication nor for
> any delay in its receipt.
>
> _______________________________________________
> nginx-devel mailing list
> nginx-devel at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx-devel



***********************************************************************
The information contained in this transmission is confidential.  It is
intended solely for the use of the individual(s) or organization(s) to
whom it is addressed.  Any disclosure, copying or further distribution
is not permitted unless such privilege is explicitly granted in writing
by JLG Industries, Inc.
Further, JLG Industries, Inc. is not responsible for the proper and
complete transmission of the substance of this communication nor for
any delay in its receipt.


From agentzh at gmail.com  Thu Sep 29 16:08:10 2011
From: agentzh at gmail.com (agentzh)
Date: Thu, 29 Sep 2011 09:08:10 -0700
Subject: Nginx Substitutions Filter problem
In-Reply-To: <CAGxuLK2+3Z8M=UTfXvhEuyRgpJskXEX4EncQckNNRsRzbrJ-9A@mail.gmail.com>
References: <4E80E187.6030807@ntlworld.com>
	<CAJCGDi8ksNcSHyf-iLm5vZevAKBMew_M2NvA5s62_wyGRLGtGw@mail.gmail.com>
	<4E80EB49.9000108@ianhobson.co.uk>
	<CAJCGDi_jWABeVRNZUEeKmZ9q3cpdFOsRWgH9ASiU_=GNSycdBQ@mail.gmail.com>
	<20110927045203.20830@gmx.net> <4E81701D.408@martronic.ch>
	<CAGxuLK3+sVfD8WC=9gL6z0daokOB=qc8HaE3drqsxSgbocS6=A@mail.gmail.com>
	<2a4c6102eba7e97f63ad316476db06b8.squirrel@webmail.martronic.ch>
	<CAGxuLK1VRVX_6rsXDSomzcnpndg5L0Rd4CqCTfnU3PfJrJyvkw@mail.gmail.com>
	<744053bb97f08749b6230bfa2202e47e.squirrel@webmail.martronic.ch>
	<CAGxuLK2+3Z8M=UTfXvhEuyRgpJskXEX4EncQckNNRsRzbrJ-9A@mail.gmail.com>
Message-ID: <CAB4Tn6NTTkAVO=p8dVt9wzrJAszww=7Ddm7oEM6XU0Zxf6T3BQ@mail.gmail.com>

2011/9/29 ??? <nbubingo at gmail.com>:
> Sorry. I can't find any log from my substitutions module in your debug log.
> And I find the possible problem. The response from backend  server is
> gzipped.

Maybe we could check the Content-Encoding response header
automatically in ngx_subs and if it's gzip/deflate, just issue a
warning to error.log by default? ;)

Regards,
-agentzh


From sb at waeme.net  Thu Sep 29 20:02:10 2011
From: sb at waeme.net (Sergey Budnevitch)
Date: Fri, 30 Sep 2011 00:02:10 +0400
Subject: NGINX On REDHAT LINUX 5
In-Reply-To: <OF18EAFF31.FE20F463-ON8525791A.00561D2B-8525791A.00568133@jlg.com>
References: <OF46BDA6DE.B2B8F259-ON85257919.0059E70A-85257919.005A3D64@jlg.com>
	<3DE0EF14-8A9C-4576-ACFC-756922577146@waeme.net>
	<OF18EAFF31.FE20F463-ON8525791A.00561D2B-8525791A.00568133@jlg.com>
Message-ID: <1778F702-B039-4482-8AFC-3F0E1B8641BC@waeme.net>


On Sep 29, 2011, at 7:44 PM, RFGuerengomba at JLG.com wrote:

> Thank you so much for your help; I have some issues when I issue:
> 
> /etc/init.d/nginx start or restart; it return not output.
> 
> when I issued the command
> 
> /usr/local/nginx/sbin/nginx -t  the test is successful
> 
> but
> 
> /usr/local/nginx/sbin/nginx-s reload  It gives me: invalid PID number "" in
> "/usr/local/nginx/logs/nginx.pid"

It seems you install nginx from source. If you do not need third party modules
try to install prebuild package (http://wiki.nginx.org/Install), it should work
out of box

From RFGuerengomba at JLG.com  Thu Sep 29 20:26:14 2011
From: RFGuerengomba at JLG.com (RFGuerengomba at JLG.com)
Date: Thu, 29 Sep 2011 16:26:14 -0400
Subject: Having Serious trouble and need help with NGINX
Message-ID: <OF77FADB08.D48924FF-ON8525791A.0070045A-8525791A.0070440D@jlg.com>


 I have some issues when I issue: Can someone please help me. (Running
NGINX-1.0.4 On RedHat 5)

/etc/init.d/nginx start or restart; it return not output.

when I issued the command

/usr/local/nginx/sbin/nginx -t  the test is successful

but

/usr/local/nginx/sbin/nginx-s reload  It gives me: invalid PID number "" in
"/usr/local/nginx/logs/nginx.pid"

when I look at the the nginx.pid number, there is no number in there.

Can you please help me on how to solve this problem?

Thanks,

Rogergue

***********************************************************************
The information contained in this transmission is confidential.  It is
intended solely for the use of the individual(s) or organization(s) to
whom it is addressed.  Any disclosure, copying or further distribution
is not permitted unless such privilege is explicitly granted in writing
by JLG Industries, Inc.
Further, JLG Industries, Inc. is not responsible for the proper and
complete transmission of the substance of this communication nor for
any delay in its receipt.


From nginx-forum at nginx.us  Thu Sep 29 20:38:14 2011
From: nginx-forum at nginx.us (nailxx)
Date: Thu, 29 Sep 2011 16:38:14 -0400
Subject: Odd reverse proxy behavior
Message-ID: <63716caadbff839f3eced00f6ebce654.NginxMailingListEnglish@forum.nginx.org>

Hi, all!

I've stumbled upon a problem I can't find a solution for many hours in a
row. I use nginx as a reverse proxy to python/spawning web-server. I've
done it for 1000 times already and all worked fine before this day.

Nginx on a server I'm talking about serves 3 separate projects
(hostnames) as a reverse proxy. I've added fourth. Three existing
projects continued to work perfectly. But the new one sometimes works
and sometimes returns 502.

The observation is: once it began to return 502 it would return it on
every browser refresh. However if I'll throttle F5 fast, after 5-10
refreshes it starts to work properly. And vice versa: once it works like
it should, slow refreshes keep it working, but if I'd throttle F5 again
or if I'd wait for a minute or so nginx declines back to 502-state.

There is nothing recorded in neither log (access/error nginx/backend) if
it is in 502-state. The deal is not in a backend. I've tried to set
proxy_pass to the same value as in a sibling well-working project conf
and the same story repeats. Problem exists for static resources as well.
Is this conf cursed?

++++
user www-data;
worker_processes  1;

error_log  /var/log/nginx/error.log debug;
pid        /var/run/nginx.pid;

events {
    worker_connections  1024;
    # multi_accept on;
}

http {
    include       /etc/nginx/mime.types;

    access_log  /var/log/nginx/access.log;

    sendfile        on;
    #tcp_nopush     on;

    #keepalive_timeout  0;
    keepalive_timeout  65;
    tcp_nodelay        on;

    gzip  on;
    gzip_disable "MSIE [1-6]\.(?!.*SV1)";

    include /etc/nginx/conf.d/*.conf;
    include /etc/nginx/sites-enabled/*;
}

#### /etc/nginx/sites-enabled/my-cursed-conf ####

server {
    listen       80;
    server_name  hostname.iam.listening.to;

    # Main location
    location / {
        proxy_pass         http://127.0.0.1:8084;
        proxy_redirect     off;

        proxy_set_header   Host             $host;
        proxy_set_header   X-Real-IP        $remote_addr;
        proxy_set_header   X-Forwarded-For  $proxy_add_x_forwarded_for;
    }

    location /static/ { alias /srv/some/path/static/; }
}


++++

I feel I've tried all combinations of buffer-tweaking and none of them
helped. The system has > 700 Mb or RAM free and mostly idle CPU. Could
someone guess what the hell is going on?

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,216073,216073#msg-216073


From nginx-forum at nginx.us  Thu Sep 29 21:56:39 2011
From: nginx-forum at nginx.us (nailxx)
Date: Thu, 29 Sep 2011 17:56:39 -0400
Subject: Odd reverse proxy behavior
In-Reply-To: <63716caadbff839f3eced00f6ebce654.NginxMailingListEnglish@forum.nginx.org>
References: <63716caadbff839f3eced00f6ebce654.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <f48268b925107c30d41c4eb25f490645.NginxMailingListEnglish@forum.nginx.org>

Got it!

The problem was absolutely not related to nginx configuration. I found
it when I continued to get 502s even with nginx process down O_o

I've accidentally added two A-records to the DNS-zone with same name but
different IPs. Looks like the resulting IP was chosen in
semi-round-robin fashion. And when a not configured server was chosen,
I've got those 502s.

Whew!? Sorry for disturbing.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,216073,216075#msg-216075


From nginx-forum at nginx.us  Thu Sep 29 23:09:21 2011
From: nginx-forum at nginx.us (kaspars)
Date: Thu, 29 Sep 2011 19:09:21 -0400
Subject: Best way to burst static file cache
In-Reply-To: <da5c45696606c5553b7cdab562cd38b6.NginxMailingListEnglish@forum.nginx.org>
References: <0668d2360623a0b4f9f9c84c0ab39f30.NginxMailingListEnglish@forum.nginx.org>
	<da5c45696606c5553b7cdab562cd38b6.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <2ca1c0cfbd552de1033411ef7ba73187.NginxMailingListEnglish@forum.nginx.org>

Well, if you set cache to expire 30 years from now the browser should
honor that, that is if no Last-Modified is set.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215835,216076#msg-216076


From alex.thegreat at ambix.net  Thu Sep 29 23:48:41 2011
From: alex.thegreat at ambix.net (Alex)
Date: Thu, 29 Sep 2011 17:48:41 -0600
Subject: Nginx reverse proxy with cache - dynamic images do not display
Message-ID: <F3DA5B84-4F22-4984-B026-5F47933AD43E@ambix.net>

Hi,


I set up this configuration:  http://pastebin.com/R65RiX5n

The images that are served by a servelet do not appear.   Do you know  
how I can fix this so that images served by servelets (dotContent,  
contentAssets, or /resize_image) work?  So the image tags on the HTML  
are:



<img src="/resize_image? 
id=16a07d2b-126c-4176-8b43-369852683e3a&w=879&h=271">

<img src="/resize_image?path=/dotAsset/fe50b8c6-56ff-36985-aa6a- 
b8bee3f4ca6c.jpg&amp;w=498&amp;h=377">

<img src="/contentAsset/resize-image/7ba770c4-2903-4d5e-8952141- 
c6ada9ecba69/?w=175&h=127">

I appreciate any suggestions.

Alex

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110929/f14531d0/attachment.html>

From nginx-forum at nginx.us  Fri Sep 30 00:01:56 2011
From: nginx-forum at nginx.us (naytro)
Date: Thu, 29 Sep 2011 20:01:56 -0400
Subject: range request
Message-ID: <b8701f9d0cd129f19b7cf7c5a3297244.NginxMailingListEnglish@forum.nginx.org>

hello

how to disable range request ?

ciao

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,216078,216078#msg-216078


From nginx-forum at nginx.us  Fri Sep 30 00:32:59 2011
From: nginx-forum at nginx.us (naytro)
Date: Thu, 29 Sep 2011 20:32:59 -0400
Subject: range request
In-Reply-To: <b8701f9d0cd129f19b7cf7c5a3297244.NginxMailingListEnglish@forum.nginx.org>
References: <b8701f9d0cd129f19b7cf7c5a3297244.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <7dd4fdaf69e96cdb225040689263e52e.NginxMailingListEnglish@forum.nginx.org>

hello

a few minutes later I found a solution, module ngx_headers_more is
capable to remove input headers, so I removed a range and accept-rangers
headers

ciao

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,216078,216080#msg-216080


From yaniv at aknin.name  Fri Sep 30 08:11:53 2011
From: yaniv at aknin.name (Yaniv Aknin)
Date: Fri, 30 Sep 2011 11:11:53 +0300
Subject: nginx and upstream Content-Length
Message-ID: <CAPzVZKmd=uTkxvyLykRJ2pqJW_ZyR3Efsm33bgnQbjxqu58ACQ@mail.gmail.com>

Hi,

In a recent thread on the uwsgi mailing list[1], I began suspecting that
nginx will not honor an upstream's Content-Length header. i.e., if an
upstream mentions a Content-Length of 1,000 bytes, but the connection is
broken after 500 bytes, nginx will still happily serve this entity with a
200 OK status.

This may be a known bug in nginx, I wanted to be certain I indeed understand
it correctly and raise the attention to it on the nginx mailing list -
because I think this is a very serious bug with potentially disastrous
consequences, as I describe below.

I was able to confirm this both for uwsgi_pass and proxy_pass; if the
upstream sets a Content-Length and then breaks the connection before that
length was achieved, nginx will pass this onwards to the client.
Furthermore, since the upstream protocol is HTTP 1.0 but the nginx-client
protocl is HTTP 1.1 (with keepalive), the request will simply not terminate,
because the client can't tell that the server has nothing more to send and
nginx will not break the connection, despite the fact its connection with
the upstream was broken and there's no chance this request will ever be
fulfilled.

Things get far worse with gzip compression on - nginx will remove the
Content-Length header sent by the client and replace it with chunked
encoding - /incorrect chunked encoding/, that will make the client believe
it has the full entity, even though it has only a part of this.

Think about this with regard to caching, ETags and transparent proxy caching
- if something like this happens to a cachable entity, especially one with
an ETag, especially if a large ISP's transparent proxy intercepts the
request - you might end up serving an incorrect representation of the entity
for a very long time and for many thousands of requests (!).

Anyhow, I think the only sane resolution is that nginx will honor upstream
Content-Length (and chunked encoding, if and when nginx will support it),
and intentionally close the downstream connection prematurely in case the
upstream connection is closed before the end of the Content-Length or the
last chunk is received.

I suspect the recent work in nginx 1.1.4 for
ngx_http_upstream_keepalivecould be relevant here, but not sure,
didn't read the code.

I'll be happy to hear your thoughts or provide further data.

- Yaniv

1: http://comments.gmane.org/gmane.comp.python.wsgi.uwsgi.general/2061
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110930/9e291201/attachment.html>

From zjay1987 at gmail.com  Fri Sep 30 08:22:52 2011
From: zjay1987 at gmail.com (li zJay)
Date: Fri, 30 Sep 2011 16:22:52 +0800
Subject: nginx won't honor limit_rate
In-Reply-To: <5ac045ab558aa2e1c5c87c2eadb4a37a.NginxMailingListEnglish@forum.nginx.org>
References: <20110928232327.GA1137@mdounin.ru>
	<5ac045ab558aa2e1c5c87c2eadb4a37a.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <CABPr_TZuaQVBaTjRJm7VeWEYVAK_5-b1eBrHPo95J42amd3O6w@mail.gmail.com>

set "proxy_buffering on", the upstream response may touch the disk
this suffers a lot when nginx is used as a proxy to file download servers.

On Thu, Sep 29, 2011 at 7:31 AM, mojiz <nginx-forum at nginx.us> wrote:

> Thanks Max.
> It worked!
>
> Posted at Nginx Forum:
> http://forum.nginx.org/read.php?2,215989,215991#msg-215991
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx
>



--
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110930/c1c72009/attachment-0001.html>

From martronic at martronic.ch  Fri Sep 30 08:52:29 2011
From: martronic at martronic.ch (Martronic SA)
Date: Fri, 30 Sep 2011 10:52:29 +0200
Subject: Nginx Substitutions Filter problem
In-Reply-To: <CAB4Tn6NTTkAVO=p8dVt9wzrJAszww=7Ddm7oEM6XU0Zxf6T3BQ@mail.gmail.com>
References: <4E80E187.6030807@ntlworld.com>
	<CAJCGDi8ksNcSHyf-iLm5vZevAKBMew_M2NvA5s62_wyGRLGtGw@mail.gmail.com>
	<4E80EB49.9000108@ianhobson.co.uk>
	<CAJCGDi_jWABeVRNZUEeKmZ9q3cpdFOsRWgH9ASiU_=GNSycdBQ@mail.gmail.com>
	<20110927045203.20830@gmx.net> <4E81701D.408@martronic.ch>
	<CAGxuLK3+sVfD8WC=9gL6z0daokOB=qc8HaE3drqsxSgbocS6=A@mail.gmail.com>
	<2a4c6102eba7e97f63ad316476db06b8.squirrel@webmail.martronic.ch>
	<CAGxuLK1VRVX_6rsXDSomzcnpndg5L0Rd4CqCTfnU3PfJrJyvkw@mail.gmail.com>
	<744053bb97f08749b6230bfa2202e47e.squirrel@webmail.martronic.ch>
	<CAGxuLK2+3Z8M=UTfXvhEuyRgpJskXEX4EncQckNNRsRzbrJ-9A@mail.gmail.com>
	<CAB4Tn6NTTkAVO=p8dVt9wzrJAszww=7Ddm7oEM6XU0Zxf6T3BQ@mail.gmail.com>
Message-ID: <4E85834D.3060105@martronic.ch>

Maybe the documentation of the module should be completed to notify about
this problem.

Thanks!

Martronic SA

Le 29.09.2011 18:08, agentzh a ?crit :
> 2011/9/29 ??? <nbubingo at gmail.com>:
>> Sorry. I can't find any log from my substitutions module in your debug log.
>> And I find the possible problem. The response from backend  server is
>> gzipped.
> Maybe we could check the Content-Encoding response header
> automatically in ngx_subs and if it's gzip/deflate, just issue a
> warning to error.log by default? ;)
>
> Regards,
> -agentzh
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx


From j.vanarragon at lukkien.com  Fri Sep 30 09:06:27 2011
From: j.vanarragon at lukkien.com (Jaap van Arragon)
Date: Fri, 30 Sep 2011 11:06:27 +0200
Subject: HttpHealthcheckModule server not marked down
In-Reply-To: <CAP0FwPR78VwaXg4qbftEZbsPFeen-xwrnGQU3wFgdd8rPu9XOg@mail.gmail.com>
Message-ID: <CAAB5333.52F8B%j.vanarragon@lukkien.com>

Hi,

I believe that the healthcheck module is only used for checking if the
backend(node) is down.

As I read it, it doesn't do anything with bringing the node "Down" in the
nginx config but maybe I misunderstand it.

Regards,
Jaap


On 9/22/11 10:34 PM, "Sjaak Pieterse" <sjaak23 at gmail.com> wrote:

> Hi there,
> 
> i'm trying to use the HttpHealthcheckModule for nginx, but i have some
> troubles with it.
> 
> i have two servers in my upstream, when sabotaging the health for one
> server i see in the status view of healthcheck that the server is
> down(1), but if i go to the website i'm checking i still come out on
> it and see a broken page.
> 
> how can i arrange that the server automatically is marked as down when
> the check fails?
> 
> sorry for my bad english and maybe noob questions.
> 
> config:
> ?upstream www-health{
> ??????? server x.x.x.1 ;
> ??????? server x.x.x.2 ;
> ??? healthcheck_enabled;
> ??? healthcheck_delay 10000 ;
> ??? healthcheck_timeout 1000;
> ??? healthcheck_failcount 2;
> ??? #healthcheck_expected 'I_AM_ALIVE';
> ??? #Important: HTTP/1.0
> ??? healthcheck_send "GET / HTTP/1.0" 'Host: health.test.x.com'
> 'Conection: close' ;
> }
> 
> nginx: nginx version: nginx/1.0.6
> nginx: built by gcc 4.4.3 (Ubuntu 4.4.3-4ubuntu5)
> nginx: TLS SNI support enabled
> nginx: configure arguments: --with-http_ssl_module
> --add-module=/gnosek-nginx-upstream-fair-2131c73
> --with-http_stub_status_module
> --add-module=/cep21-healthcheck_nginx_upstreams-b33a846
> --prefix=/usr/local/nginx-1.0.6 --with-debug
> 
> used:
> peckhardt at test-nginx:~/nginx-1.0.6$patch -p1 <
> /cep21-healthcheck_nginx_upstreams-5fa4bff/nginx.patch
> 
> hope someone would help me.
> 
> greetings
> 
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx



From mdounin at mdounin.ru  Fri Sep 30 09:18:54 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Fri, 30 Sep 2011 13:18:54 +0400
Subject: nginx won't honor limit_rate
In-Reply-To: <CABPr_TZuaQVBaTjRJm7VeWEYVAK_5-b1eBrHPo95J42amd3O6w@mail.gmail.com>
References: <20110928232327.GA1137@mdounin.ru>
	<5ac045ab558aa2e1c5c87c2eadb4a37a.NginxMailingListEnglish@forum.nginx.org>
	<CABPr_TZuaQVBaTjRJm7VeWEYVAK_5-b1eBrHPo95J42amd3O6w@mail.gmail.com>
Message-ID: <20110930091854.GI1137@mdounin.ru>

Hello!

On Fri, Sep 30, 2011 at 04:22:52PM +0800, li zJay wrote:

> set "proxy_buffering on", the upstream response may touch the disk
> this suffers a lot when nginx is used as a proxy to file download servers.

If you want nginx to don't touch disk, use 

    proxy_max_temp_file_size 0;

instead.  This will still allow in-memory buffering and wouldn't 
touch disk.

Maxim Dounin

> 
> On Thu, Sep 29, 2011 at 7:31 AM, mojiz <nginx-forum at nginx.us> wrote:
> 
> > Thanks Max.
> > It worked!
> >
> > Posted at Nginx Forum:
> > http://forum.nginx.org/read.php?2,215989,215991#msg-215991
> >
> > _______________________________________________
> > nginx mailing list
> > nginx at nginx.org
> > http://mailman.nginx.org/mailman/listinfo/nginx
> >
> 
> 
> 
> --

> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx


From mdounin at mdounin.ru  Fri Sep 30 09:59:07 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Fri, 30 Sep 2011 13:59:07 +0400
Subject: nginx and upstream Content-Length
In-Reply-To: <CAPzVZKmd=uTkxvyLykRJ2pqJW_ZyR3Efsm33bgnQbjxqu58ACQ@mail.gmail.com>
References: <CAPzVZKmd=uTkxvyLykRJ2pqJW_ZyR3Efsm33bgnQbjxqu58ACQ@mail.gmail.com>
Message-ID: <20110930095907.GJ1137@mdounin.ru>

Hello!

On Fri, Sep 30, 2011 at 11:11:53AM +0300, Yaniv Aknin wrote:

> In a recent thread on the uwsgi mailing list[1], I began suspecting that
> nginx will not honor an upstream's Content-Length header. i.e., if an
> upstream mentions a Content-Length of 1,000 bytes, but the connection is
> broken after 500 bytes, nginx will still happily serve this entity with a
> 200 OK status.

Status code 200 is irrelevant - as it's generally not possible to 
know if connection will be broken in advance (i.e. before sending 
status).

> This may be a known bug in nginx, I wanted to be certain I indeed understand
> it correctly and raise the attention to it on the nginx mailing list -
> because I think this is a very serious bug with potentially disastrous
> consequences, as I describe below.
> 
> I was able to confirm this both for uwsgi_pass and proxy_pass; if the
> upstream sets a Content-Length and then breaks the connection before that
> length was achieved, nginx will pass this onwards to the client.
> Furthermore, since the upstream protocol is HTTP 1.0 but the nginx-client
> protocl is HTTP 1.1 (with keepalive), the request will simply not terminate,
> because the client can't tell that the server has nothing more to send and
> nginx will not break the connection, despite the fact its connection with
> the upstream was broken and there's no chance this request will ever be
> fulfilled.
> 
> Things get far worse with gzip compression on - nginx will remove the
> Content-Length header sent by the client and replace it with chunked
> encoding - /incorrect chunked encoding/, that will make the client believe
> it has the full entity, even though it has only a part of this.

Yes, this is a known problem.  Upstream module expects backend to 
behave properly, and if it misbehaves (or, more importantly, 
connection is broken for some reason) bad things may happen.

Upstream's module code needs carefull auditing to fix this.  It's 
somewhere in my TODO (though not very high).

Maxim Dounin


From yaniv at aknin.name  Fri Sep 30 14:35:59 2011
From: yaniv at aknin.name (Yaniv Aknin)
Date: Fri, 30 Sep 2011 17:35:59 +0300
Subject: nginx and upstream Content-Length
In-Reply-To: <20110930095907.GJ1137@mdounin.ru>
References: <CAPzVZKmd=uTkxvyLykRJ2pqJW_ZyR3Efsm33bgnQbjxqu58ACQ@mail.gmail.com>
	<20110930095907.GJ1137@mdounin.ru>
Message-ID: <CAPzVZKkbOFCLkkf5v1Rau1YCwVwPgP0Us0TJmW=JS3D2GL4agw@mail.gmail.com>

Well, I appreciate the time every contributor to nginx put into the project
and especially your and Igor's work, and I know I don't dictate the priority
of your TODO, but I humbly think this is a far more serious issue than
you're portraying.

Simply put, this bug can cause nginx to cause an otherwise *noticeable data
corruption* to become a *silent data corruption* (including the serious
cache poisoning ramifications), and I think there are few bugs less urgent
than that.

Anyhow, I've said all I have to say on the matter, so unless further
information comes up on this, I'll leave this topic alone (at least until
I'll get off my bum and come back here with a suggested patch to fix
this...).

Thank you for your concise, timely and complete reply,
 - Yaniv

On Fri, Sep 30, 2011 at 12:59 PM, Maxim Dounin <mdounin at mdounin.ru> wrote:

> Hello!
>
> On Fri, Sep 30, 2011 at 11:11:53AM +0300, Yaniv Aknin wrote:
>
> > In a recent thread on the uwsgi mailing list[1], I began suspecting that
> > nginx will not honor an upstream's Content-Length header. i.e., if an
> > upstream mentions a Content-Length of 1,000 bytes, but the connection is
> > broken after 500 bytes, nginx will still happily serve this entity with a
> > 200 OK status.
>
> Status code 200 is irrelevant - as it's generally not possible to
> know if connection will be broken in advance (i.e. before sending
> status).
>
> > This may be a known bug in nginx, I wanted to be certain I indeed
> understand
> > it correctly and raise the attention to it on the nginx mailing list -
> > because I think this is a very serious bug with potentially disastrous
> > consequences, as I describe below.
> >
> > I was able to confirm this both for uwsgi_pass and proxy_pass; if the
> > upstream sets a Content-Length and then breaks the connection before that
> > length was achieved, nginx will pass this onwards to the client.
> > Furthermore, since the upstream protocol is HTTP 1.0 but the nginx-client
> > protocl is HTTP 1.1 (with keepalive), the request will simply not
> terminate,
> > because the client can't tell that the server has nothing more to send
> and
> > nginx will not break the connection, despite the fact its connection with
> > the upstream was broken and there's no chance this request will ever be
> > fulfilled.
> >
> > Things get far worse with gzip compression on - nginx will remove the
> > Content-Length header sent by the client and replace it with chunked
> > encoding - /incorrect chunked encoding/, that will make the client
> believe
> > it has the full entity, even though it has only a part of this.
>
> Yes, this is a known problem.  Upstream module expects backend to
> behave properly, and if it misbehaves (or, more importantly,
> connection is broken for some reason) bad things may happen.
>
> Upstream's module code needs carefull auditing to fix this.  It's
> somewhere in my TODO (though not very high).
>
> Maxim Dounin
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110930/daf88372/attachment.html>

From zjay1987 at gmail.com  Fri Sep 30 14:56:41 2011
From: zjay1987 at gmail.com (li zJay)
Date: Fri, 30 Sep 2011 22:56:41 +0800
Subject: nginx won't honor limit_rate
In-Reply-To: <20110930091854.GI1137@mdounin.ru>
References: <20110928232327.GA1137@mdounin.ru>
	<5ac045ab558aa2e1c5c87c2eadb4a37a.NginxMailingListEnglish@forum.nginx.org>
	<CABPr_TZuaQVBaTjRJm7VeWEYVAK_5-b1eBrHPo95J42amd3O6w@mail.gmail.com>
	<20110930091854.GI1137@mdounin.ru>
Message-ID: <CABPr_TbHWcQnQ+g7hKCBZU9P=8OJr6DSrkT2n5PXKzy1Z5FBzg@mail.gmail.com>

hello!

If you want nginx to don't touch disk, use
>
>    proxy_max_temp_file_size 0;
>
> instead.  This will still allow in-memory buffering and wouldn't
> touch disk.
>

Because the client bandwidth is usually slow, in this case, tons of upstream
content will be buffer in memory before sent out to client, and then the
swap will be used and the server is going down soon.

On Fri, Sep 30, 2011 at 5:18 PM, Maxim Dounin <mdounin at mdounin.ru> wrote:

> Hello!
>
> On Fri, Sep 30, 2011 at 04:22:52PM +0800, li zJay wrote:
>
> > set "proxy_buffering on", the upstream response may touch the disk
> > this suffers a lot when nginx is used as a proxy to file download
> servers.
>
> If you want nginx to don't touch disk, use
>
>    proxy_max_temp_file_size 0;
>
> instead.  This will still allow in-memory buffering and wouldn't
> touch disk.
>
> Maxim Dounin
>
> >
> > On Thu, Sep 29, 2011 at 7:31 AM, mojiz <nginx-forum at nginx.us> wrote:
> >
> > > Thanks Max.
> > > It worked!
> > >
> > > Posted at Nginx Forum:
> > > http://forum.nginx.org/read.php?2,215989,215991#msg-215991
> > >
> > > _______________________________________________
> > > nginx mailing list
> > > nginx at nginx.org
> > > http://mailman.nginx.org/mailman/listinfo/nginx
> > >
> >
> >
> >
> > --
>
> > _______________________________________________
> > nginx mailing list
> > nginx at nginx.org
> > http://mailman.nginx.org/mailman/listinfo/nginx
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx
>



--
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110930/9781426e/attachment.html>

From tomlove at gmail.com  Fri Sep 30 15:08:32 2011
From: tomlove at gmail.com (Thomas Love)
Date: Fri, 30 Sep 2011 17:08:32 +0200
Subject: nginx and upstream Content-Length
In-Reply-To: <CAPzVZKkbOFCLkkf5v1Rau1YCwVwPgP0Us0TJmW=JS3D2GL4agw@mail.gmail.com>
References: <CAPzVZKmd=uTkxvyLykRJ2pqJW_ZyR3Efsm33bgnQbjxqu58ACQ@mail.gmail.com>
	<20110930095907.GJ1137@mdounin.ru>
	<CAPzVZKkbOFCLkkf5v1Rau1YCwVwPgP0Us0TJmW=JS3D2GL4agw@mail.gmail.com>
Message-ID: <CAJCGDi9WitbQ=Gt=z0+WkuNXwHC6io9ZoRUz+0UPn0TbuG+ekg@mail.gmail.com>

Hi Yaniv

Thanks for investigating and reporting on this. Do you know whether
the gzip chunked-encoding problem exists for fastcgi_pass as well?

Cheers
Thomas


From nginx-forum at nginx.us  Fri Sep 30 15:13:34 2011
From: nginx-forum at nginx.us (mojiz)
Date: Fri, 30 Sep 2011 11:13:34 -0400
Subject: nginx won't honor limit_rate
In-Reply-To: <20110930091854.GI1137@mdounin.ru>
References: <20110930091854.GI1137@mdounin.ru>
Message-ID: <f77f026064b51e7331bef2245e909603.NginxMailingListEnglish@forum.nginx.org>

What exactly do you mean by 'suffer'?
I'm using limit_rate to slow the user downloads.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215989,216112#msg-216112


From mdounin at mdounin.ru  Fri Sep 30 15:18:14 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Fri, 30 Sep 2011 19:18:14 +0400
Subject: nginx won't honor limit_rate
In-Reply-To: <CABPr_TbHWcQnQ+g7hKCBZU9P=8OJr6DSrkT2n5PXKzy1Z5FBzg@mail.gmail.com>
References: <20110928232327.GA1137@mdounin.ru>
	<5ac045ab558aa2e1c5c87c2eadb4a37a.NginxMailingListEnglish@forum.nginx.org>
	<CABPr_TZuaQVBaTjRJm7VeWEYVAK_5-b1eBrHPo95J42amd3O6w@mail.gmail.com>
	<20110930091854.GI1137@mdounin.ru>
	<CABPr_TbHWcQnQ+g7hKCBZU9P=8OJr6DSrkT2n5PXKzy1Z5FBzg@mail.gmail.com>
Message-ID: <20110930151814.GW1137@mdounin.ru>

Hello!

On Fri, Sep 30, 2011 at 10:56:41PM +0800, li zJay wrote:

> hello!
> 
> If you want nginx to don't touch disk, use
> >
> >    proxy_max_temp_file_size 0;
> >
> > instead.  This will still allow in-memory buffering and wouldn't
> > touch disk.
> >
> 
> Because the client bandwidth is usually slow, in this case, tons of upstream
> content will be buffer in memory before sent out to client, and then the
> swap will be used and the server is going down soon.

The limit on memory usage for proxy buffering is set with 
proxy_buffers directive, 8 * pagesize by default.  Nobody stops 
you from setting it even lower if you are memory-constrained.

The "proxy_buffering off;" is needed when you want response to be 
immediately passed to clients once it arrives.  It's critical for 
streaming and similar workloads where you can't afford delays.  
But it's not a switch to limit resource consumption.  Instead, it 
actually implies less effective resource usage: processing of 
requests without buffering takes more CPU time and may cause 
suboptimal network utilization.

Maxim Dounin

> 
> On Fri, Sep 30, 2011 at 5:18 PM, Maxim Dounin <mdounin at mdounin.ru> wrote:
> 
> > Hello!
> >
> > On Fri, Sep 30, 2011 at 04:22:52PM +0800, li zJay wrote:
> >
> > > set "proxy_buffering on", the upstream response may touch the disk
> > > this suffers a lot when nginx is used as a proxy to file download
> > servers.
> >
> > If you want nginx to don't touch disk, use
> >
> >    proxy_max_temp_file_size 0;
> >
> > instead.  This will still allow in-memory buffering and wouldn't
> > touch disk.
> >
> > Maxim Dounin
> >
> > >
> > > On Thu, Sep 29, 2011 at 7:31 AM, mojiz <nginx-forum at nginx.us> wrote:
> > >
> > > > Thanks Max.
> > > > It worked!
> > > >
> > > > Posted at Nginx Forum:
> > > > http://forum.nginx.org/read.php?2,215989,215991#msg-215991
> > > >
> > > > _______________________________________________
> > > > nginx mailing list
> > > > nginx at nginx.org
> > > > http://mailman.nginx.org/mailman/listinfo/nginx
> > > >
> > >
> > >
> > >
> > > --
> >
> > > _______________________________________________
> > > nginx mailing list
> > > nginx at nginx.org
> > > http://mailman.nginx.org/mailman/listinfo/nginx
> >
> > _______________________________________________
> > nginx mailing list
> > nginx at nginx.org
> > http://mailman.nginx.org/mailman/listinfo/nginx
> >
> 
> 
> 
> --

> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx


From igor at sysoev.ru  Fri Sep 30 15:41:00 2011
From: igor at sysoev.ru (Igor Sysoev)
Date: Fri, 30 Sep 2011 19:41:00 +0400
Subject: nginx-1.0.7
Message-ID: <20110930154100.GI72272@nginx.com>

Changes with nginx 1.0.7                                         30 Sep 2011

    *) Change: now if total size of all ranges is greater than source
       response size, then nginx disables ranges and returns just the source
       response.

    *) Feature: the "max_ranges" directive.

    *) Feature: the module ngx_http_mp4_module.

    *) Feature: the "worker_aio_requests" directive.

    *) Bugfix: if nginx was built --with-file-aio it could not be run on
       Linux kernel which did not support AIO.

    *) Bugfix: in Linux AIO error processing.
       Thanks to Hagai Avrahami.

    *) Bugfix: in Linux AIO combined with open_file_cache.

    *) Bugfix: open_file_cache did not update file info on retest if file
       was not atomically changed.

    *) Bugfix: reduced memory consumption for long-lived requests.

    *) Bugfix: in the "proxy/fastcgi/scgi/uwsgi_ignore_client_abort"
       directives.

    *) Bugfix: nginx could not be built on MacOSX 10.7.

    *) Bugfix: in the "proxy/fastcgi/scgi/uwsgi_ignore_client_abort"
       directives.

    *) Bugfix: request body might be processed incorrectly if client used
       pipelining.

    *) Bugfix: in the "request_body_in_single_buf" directive.

    *) Bugfix: in "proxy_set_body" and "proxy_pass_request_body" directives
       if SSL connection to backend was used.

    *) Bugfix: nginx hogged CPU if all servers in an upstream were marked as
       "down".

    *) Bugfix: a segmentation fault might occur during reconfiguration if
       ssl_session_cache was defined but not used in previous configuration.

    *) Bugfix: a segmentation fault might occur in a worker process if many
       backup servers were used in an upstream.


-- 
Igor Sysoev


From yaniv at aknin.name  Fri Sep 30 16:07:43 2011
From: yaniv at aknin.name (Yaniv Aknin)
Date: Fri, 30 Sep 2011 19:07:43 +0300
Subject: nginx and upstream Content-Length
In-Reply-To: <CAJCGDi9WitbQ=Gt=z0+WkuNXwHC6io9ZoRUz+0UPn0TbuG+ekg@mail.gmail.com>
References: <CAPzVZKmd=uTkxvyLykRJ2pqJW_ZyR3Efsm33bgnQbjxqu58ACQ@mail.gmail.com>
	<20110930095907.GJ1137@mdounin.ru>
	<CAPzVZKkbOFCLkkf5v1Rau1YCwVwPgP0Us0TJmW=JS3D2GL4agw@mail.gmail.com>
	<CAJCGDi9WitbQ=Gt=z0+WkuNXwHC6io9ZoRUz+0UPn0TbuG+ekg@mail.gmail.com>
Message-ID: <CAPzVZKmFraZ=DYK7_ifQik6VmamntTKzrmfAdfOEHcgtgJ9MnQ@mail.gmail.com>

Sorry, I didn't test FastCGI so can't say for sure. I suspect it will happen
there as well, but it's just an assumption.

 - Yaniv

On Fri, Sep 30, 2011 at 6:08 PM, Thomas Love <tomlove at gmail.com> wrote:

> Hi Yaniv
>
> Thanks for investigating and reporting on this. Do you know whether
> the gzip chunked-encoding problem exists for fastcgi_pass as well?
>
> Cheers
> Thomas
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20110930/0a3187e8/attachment-0001.html>

From zjay1987 at gmail.com  Fri Sep 30 16:14:35 2011
From: zjay1987 at gmail.com (li zJay)
Date: Sat, 1 Oct 2011 00:14:35 +0800
Subject: nginx won't honor limit_rate
In-Reply-To: <20110930151814.GW1137@mdounin.ru>
References: <20110928232327.GA1137@mdounin.ru>
	<5ac045ab558aa2e1c5c87c2eadb4a37a.NginxMailingListEnglish@forum.nginx.org>
	<CABPr_TZuaQVBaTjRJm7VeWEYVAK_5-b1eBrHPo95J42amd3O6w@mail.gmail.com>
	<20110930091854.GI1137@mdounin.ru>
	<CABPr_TbHWcQnQ+g7hKCBZU9P=8OJr6DSrkT2n5PXKzy1Z5FBzg@mail.gmail.com>
	<20110930151814.GW1137@mdounin.ru>
Message-ID: <CABPr_TafN4OVSez4CVQumZPn1=Y=v31OC33tF31h-YuRb6vQ+g@mail.gmail.com>

Thanks Maxim Dounin!

I thought that if proxy_max_temp_file_size is set to 0, all the upstream
content have to be buffered in memory, and I make a test and find I'm wrong.

On Fri, Sep 30, 2011 at 11:18 PM, Maxim Dounin <mdounin at mdounin.ru> wrote:

> Hello!
>
> On Fri, Sep 30, 2011 at 10:56:41PM +0800, li zJay wrote:
>
> > hello!
> >
> > If you want nginx to don't touch disk, use
> > >
> > >    proxy_max_temp_file_size 0;
> > >
> > > instead.  This will still allow in-memory buffering and wouldn't
> > > touch disk.
> > >
> >
> > Because the client bandwidth is usually slow, in this case, tons of
> upstream
> > content will be buffer in memory before sent out to client, and then the
> > swap will be used and the server is going down soon.
>
> The limit on memory usage for proxy buffering is set with
> proxy_buffers directive, 8 * pagesize by default.  Nobody stops
> you from setting it even lower if you are memory-constrained.
>
> The "proxy_buffering off;" is needed when you want response to be
> immediately passed to clients once it arrives.  It's critical for
> streaming and similar workloads where you can't afford delays.
> But it's not a switch to limit resource consumption.  Instead, it
> actually implies less effective resource usage: processing of
> requests without buffering takes more CPU time and may cause
> suboptimal network utilization.
>
> Maxim Dounin
>
> >
> > On Fri, Sep 30, 2011 at 5:18 PM, Maxim Dounin <mdounin at mdounin.ru>
> wrote:
> >
> > > Hello!
> > >
> > > On Fri, Sep 30, 2011 at 04:22:52PM +0800, li zJay wrote:
> > >
> > > > set "proxy_buffering on", the upstream response may touch the disk
> > > > this suffers a lot when nginx is used as a proxy to file download
> > > servers.
> > >
> > > If you want nginx to don't touch disk, use
> > >
> > >    proxy_max_temp_file_size 0;
> > >
> > > instead.  This will still allow in-memory buffering and wouldn't
> > > touch disk.
> > >
> > > Maxim Dounin
> > >
> > > >
> > > > On Thu, Sep 29, 2011 at 7:31 AM, mojiz <nginx-forum at nginx.us> wrote:
> > > >
> > > > > Thanks Max.
> > > > > It worked!
> > > > >
> > > > > Posted at Nginx Forum:
> > > > > http://forum.nginx.org/read.php?2,215989,215991#msg-215991
> > > > >
> > > > > _______________________________________________
> > > > > nginx mailing list
> > > > > nginx at nginx.org
> > > > > http://mailman.nginx.org/mailman/listinfo/nginx
> > > > >
> > > >
> > > >
> > > >
> > > > --
> > >
> > > > _______________________________________________
> > > > nginx mailing list
> > > > nginx at nginx.org
> > > > http://mailman.nginx.org/mailman/listinfo/nginx
> > >
> > > _______________________________________________
> > > nginx mailing list
> > > nginx at nginx.org
> > > http://mailman.nginx.org/mailman/listinfo/nginx
> > >
> >
> >
> >
> > --
>
> > _______________________________________________
> > nginx mailing list
> > nginx at nginx.org
> > http://mailman.nginx.org/mailman/listinfo/nginx
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx
>



-- 
??
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20111001/72658aea/attachment.html>

From nginx-forum at nginx.us  Fri Sep 30 18:37:10 2011
From: nginx-forum at nginx.us (dougconran)
Date: Fri, 30 Sep 2011 14:37:10 -0400
Subject: Question regarding Nginx Configuration
Message-ID: <2e6ad2d68b975059a0860f81a2f19b41.NginxMailingListEnglish@forum.nginx.org>

I have Nginx up and running but I think that the conf file could be
improved upon.

I have a single server hosting 2 domains (doma and domb).

doma is the default virtual host and supports a number of applications
all of which use Perl scripts.

domb has a separate domain name (obviously!) and supports one
application that has a Wordpress/PHP frontend plus a secure Perl
application

All doma static web pages are held in sub-directories under a root of
.../htdocs

The domb 'static' Wordpress pages (including all PHP pages) are in
.../htdocs/domb & sub-directories

All Perl scripts for both doma and domb are in .../cgi-bin
sub-directories including .../cgi-bin/domb for the secure domb apps.

As I understand it I have to have 3 server blocks:-

Block 1 - default port 80
Block 2 - domb port 80
Block 3 - domb port 443

Unfortunately it seems that I have to replicate the location ...htdocs
and cgi-bin*.pl$ blocks3 times (once for each server block) and the
/*.php$ block twice (for domb 80 and domb 443).  This, I think, is
because location blocks can only exist within server blocks.

Is it possible for location blocks to exist within the main http block
at a higher level than the server block (I've tried it and it didn't
work for me but that might have been because I had got some other
configuration statements wrong)?

TIA

Doug

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,216128,216128#msg-216128


From lists at ruby-forum.com  Fri Sep 30 20:13:02 2011
From: lists at ruby-forum.com (Jamie W.)
Date: Fri, 30 Sep 2011 22:13:02 +0200
Subject: intercepted 400 errors not displaying custom error page
Message-ID: <522f3010f47fced94ec3b41c0dc7646a@ruby-forum.com>

I am trying to intercept errors from a Unicorn proxy_pass and display a
custom 40x error page that is located on disk. This is what a subscript
of my config:

proxy_intercept_errors on;
error_page 400 403 =404 /404.html;
location /404.html {
  allow all;
  internal;
}

When I receive a 403 error from NGINX it is passing to the location and
rendering the 404.html error page in my root. When the upstream returns
a 400 response NGINX is intercepting it but not rendering the 404.html
file, instead it returns the generic NGINX 400 error response page.

Thank you in advance!

-- 
Posted via http://www.ruby-forum.com/.


From lists at ruby-forum.com  Fri Sep 30 20:15:41 2011
From: lists at ruby-forum.com (Jamie W.)
Date: Fri, 30 Sep 2011 22:15:41 +0200
Subject: intercepted 400 errors not displaying custom error page
In-Reply-To: <522f3010f47fced94ec3b41c0dc7646a@ruby-forum.com>
References: <522f3010f47fced94ec3b41c0dc7646a@ruby-forum.com>
Message-ID: <e27effbbd6e91bad2474a878241c4613@ruby-forum.com>

Update: It seems that all response codes I specify in error_page are 
intercepted by NGINX but do not pass to the /404.html location to render 
the custom error page.

-- 
Posted via http://www.ruby-forum.com/.


From lists at ruby-forum.com  Fri Sep 30 20:31:06 2011
From: lists at ruby-forum.com (Jamie W.)
Date: Fri, 30 Sep 2011 22:31:06 +0200
Subject: intercepted 400 errors not displaying custom error page
In-Reply-To: <522f3010f47fced94ec3b41c0dc7646a@ruby-forum.com>
References: <522f3010f47fced94ec3b41c0dc7646a@ruby-forum.com>
Message-ID: <d7d2a63dd73202984372b4e3306cdd96@ruby-forum.com>

My solution was to enable recursive_error_pages

-- 
Posted via http://www.ruby-forum.com/.


From liseen.wan at gmail.com  Fri Sep 30 22:06:28 2011
From: liseen.wan at gmail.com (liseen)
Date: Sat, 1 Oct 2011 06:06:28 +0800
Subject: HttpHealthcheckModule server not marked down
In-Reply-To: <CAP0FwPR78VwaXg4qbftEZbsPFeen-xwrnGQU3wFgdd8rPu9XOg@mail.gmail.com>
References: <CAP0FwPR78VwaXg4qbftEZbsPFeen-xwrnGQU3wFgdd8rPu9XOg@mail.gmail.com>
Message-ID: <CAB2wvz0rGmZYuO404hCF3GXb6QvwurCt=-Qd351FQsVrvtpm_g@mail.gmail.com>

Hi,

It is a bug.

the ngx_upstream_get_peer only check the index greater than i;  forgot to
check i itself.

I used my nginx patch for healthcheck,  I have used it in production more
than half a year. I will upload it to my github in some hours.

liseen

On Fri, Sep 23, 2011 at 4:34 AM, Sjaak Pieterse <sjaak23 at gmail.com> wrote:

> Hi there,
>
> i'm trying to use the HttpHealthcheckModule for nginx, but i have some
> troubles with it.
>
> i have two servers in my upstream, when sabotaging the health for one
> server i see in the status view of healthcheck that the server is
> down(1), but if i go to the website i'm checking i still come out on
> it and see a broken page.
>
> how can i arrange that the server automatically is marked as down when
> the check fails?
>
> sorry for my bad english and maybe noob questions.
>
> config:
>  upstream www-health{
>         server x.x.x.1 ;
>         server x.x.x.2 ;
>     healthcheck_enabled;
>     healthcheck_delay 10000 ;
>     healthcheck_timeout 1000;
>     healthcheck_failcount 2;
>     #healthcheck_expected 'I_AM_ALIVE';
>     #Important: HTTP/1.0
>     healthcheck_send "GET / HTTP/1.0" 'Host: health.test.x.com'
> 'Conection: close' ;
> }
>
> nginx: nginx version: nginx/1.0.6
> nginx: built by gcc 4.4.3 (Ubuntu 4.4.3-4ubuntu5)
> nginx: TLS SNI support enabled
> nginx: configure arguments: --with-http_ssl_module
> --add-module=/gnosek-nginx-upstream-fair-2131c73
> --with-http_stub_status_module
> --add-module=/cep21-healthcheck_nginx_upstreams-b33a846
> --prefix=/usr/local/nginx-1.0.6 --with-debug
>
> used:
> peckhardt at test-nginx:~/nginx-1.0.6$patch -p1 <
> /cep21-healthcheck_nginx_upstreams-5fa4bff/nginx.patch
>
> hope someone would help me.
>
> greetings
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20111001/1852d87e/attachment.html>

From liseen.wan at gmail.com  Fri Sep 30 23:16:02 2011
From: liseen.wan at gmail.com (liseen)
Date: Sat, 1 Oct 2011 07:16:02 +0800
Subject: HttpHealthcheckModule server not marked down
In-Reply-To: <CAB2wvz0rGmZYuO404hCF3GXb6QvwurCt=-Qd351FQsVrvtpm_g@mail.gmail.com>
References: <CAP0FwPR78VwaXg4qbftEZbsPFeen-xwrnGQU3wFgdd8rPu9XOg@mail.gmail.com>
	<CAB2wvz0rGmZYuO404hCF3GXb6QvwurCt=-Qd351FQsVrvtpm_g@mail.gmail.com>
Message-ID: <CAB2wvz3k8J5FFBNUX=t1UsAaXcc-mavegJBTQgdVwttf34XUeA@mail.gmail.com>

Hi,

Please try:

https://github.com/liseen/healthcheck_nginx_upstreams/blob/master/healthcheck.patch

patch -p1 < healthcheck.patch
./configure ....

if you use healthcheck with upstream hash, please compile with branch
support_http_healthchecks of cep21's fork

https://github.com/cep21/nginx_upstream_hash/tree/support_http_healthchecks

liseen


On Sat, Oct 1, 2011 at 6:06 AM, liseen <liseen.wan at gmail.com> wrote:

> Hi,
>
> It is a bug.
>
> the ngx_upstream_get_peer only check the index greater than i;  forgot to
> check i itself.
>
> I used my nginx patch for healthcheck,  I have used it in production more
> than half a year. I will upload it to my github in some hours.
>
> liseen
>
> On Fri, Sep 23, 2011 at 4:34 AM, Sjaak Pieterse <sjaak23 at gmail.com> wrote:
>
>> Hi there,
>>
>> i'm trying to use the HttpHealthcheckModule for nginx, but i have some
>> troubles with it.
>>
>> i have two servers in my upstream, when sabotaging the health for one
>> server i see in the status view of healthcheck that the server is
>> down(1), but if i go to the website i'm checking i still come out on
>> it and see a broken page.
>>
>> how can i arrange that the server automatically is marked as down when
>> the check fails?
>>
>> sorry for my bad english and maybe noob questions.
>>
>> config:
>>  upstream www-health{
>>         server x.x.x.1 ;
>>         server x.x.x.2 ;
>>     healthcheck_enabled;
>>     healthcheck_delay 10000 ;
>>     healthcheck_timeout 1000;
>>     healthcheck_failcount 2;
>>     #healthcheck_expected 'I_AM_ALIVE';
>>     #Important: HTTP/1.0
>>     healthcheck_send "GET / HTTP/1.0" 'Host: health.test.x.com'
>> 'Conection: close' ;
>> }
>>
>> nginx: nginx version: nginx/1.0.6
>> nginx: built by gcc 4.4.3 (Ubuntu 4.4.3-4ubuntu5)
>> nginx: TLS SNI support enabled
>> nginx: configure arguments: --with-http_ssl_module
>> --add-module=/gnosek-nginx-upstream-fair-2131c73
>> --with-http_stub_status_module
>> --add-module=/cep21-healthcheck_nginx_upstreams-b33a846
>> --prefix=/usr/local/nginx-1.0.6 --with-debug
>>
>> used:
>> peckhardt at test-nginx:~/nginx-1.0.6$patch -p1 <
>> /cep21-healthcheck_nginx_upstreams-5fa4bff/nginx.patch
>>
>> hope someone would help me.
>>
>> greetings
>>
>> _______________________________________________
>> nginx mailing list
>> nginx at nginx.org
>> http://mailman.nginx.org/mailman/listinfo/nginx
>>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20111001/ec0d3e12/attachment-0001.html>