From phil at pricom.com.au  Tue Nov  1 03:05:41 2011
From: phil at pricom.com.au (Philip Rhoades)
Date: Tue, 01 Nov 2011 14:05:41 +1100
Subject: First Question
In-Reply-To: <mailman.72.1320108576.3031.nginx@nginx.org>
References: <mailman.72.1320108576.3031.nginx@nginx.org>
Message-ID: <8f5f460e1ade4806f4427aafb4484401@www.pricom.com.au>

People,

I want to switch my Apache SSL to Nginx - that might help me resolve a 
Ruby on
Rails issue and it is too much work to change all the non-SSL stuff 
over as
well - I will do that later when I have time.  So for the time being, 
http will
be handled by Apache and https will be handled by Nginx.

I commented out the default server in:

   /etc/nginx/nginx.conf

and changed:

   /etc/nginx/conf.d/ssl.conf

thus:

#
# HTTPS server configuration
#

server {
     listen       443;
     server_name  www.pricom.com.au;
     access_log /var/log/nginx/nginx.vhost.access.log;
     error_log /var/log/nginx/nginx.vhost.error.log;

     ssl                  on;
     ssl_certificate      /etc/httpd/conf/ssl.crt/RapidSSL.crt;
     ssl_certificate_key  /etc/httpd/conf/ssl.key/mars-server.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   /var/www/ssl/wac;
     }
}

but in:

   /var/log/nginx/nginx.vhost.error.log

I get:

2011/11/01 13:50:28 [error] 9411#0: *7 open() 
"/usr/share/nginx/html/wac"
failed (2: No such file or directory), client: 203.206.181.78, server:
www.pricom.com.au, request: "GET /wac HTTP/1.1", host: "pricom.com.au"

Why is nginx reverting to the default path and how is it getting the 
right
dir without the correct path?

Thanks,

Phil.

-- 
Philip Rhoades

GPO Box 3411
Sydney NSW	2001
Australia
E-mail:  phil at pricom.com.au


From ilan at time4learning.com  Tue Nov  1 03:09:18 2011
From: ilan at time4learning.com (Ilan Berkner)
Date: Mon, 31 Oct 2011 23:09:18 -0400
Subject: First Question
In-Reply-To: <8f5f460e1ade4806f4427aafb4484401@www.pricom.com.au>
References: <mailman.72.1320108576.3031.nginx@nginx.org>
	<8f5f460e1ade4806f4427aafb4484401@www.pricom.com.au>
Message-ID: <CAOoWpPpFA6=zuesCz8LYvr1zMmDtqgFRHHRBMMuZFJbwbOMgyA@mail.gmail.com>

A beginner's guess...

Your server is listening on port 443 (HTTPS).  Your log file shows that the
file not found is being generated by the HTTP protocol for which your
configuration is not listening on.  Are you should that you're accessing
your site using "https://" and not "http://"?

...GET /wac HTTP/1.1", ...




On Mon, Oct 31, 2011 at 11:05 PM, Philip Rhoades <phil at pricom.com.au> wrote:

> People,
>
> I want to switch my Apache SSL to Nginx - that might help me resolve a
> Ruby on
> Rails issue and it is too much work to change all the non-SSL stuff over as
> well - I will do that later when I have time.  So for the time being, http
> will
> be handled by Apache and https will be handled by Nginx.
>
> I commented out the default server in:
>
>  /etc/nginx/nginx.conf
>
> and changed:
>
>  /etc/nginx/conf.d/ssl.conf
>
> thus:
>
> #
> # HTTPS server configuration
> #
>
> server {
>    listen       443;
>    server_name  www.pricom.com.au;
>    access_log /var/log/nginx/nginx.vhost.**access.log;
>    error_log /var/log/nginx/nginx.vhost.**error.log;
>
>    ssl                  on;
>    ssl_certificate      /etc/httpd/conf/ssl.crt/**RapidSSL.crt;
>    ssl_certificate_key  /etc/httpd/conf/ssl.key/mars-**server.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   /var/www/ssl/wac;
>    }
> }
>
> but in:
>
>  /var/log/nginx/nginx.vhost.**error.log
>
> I get:
>
> 2011/11/01 13:50:28 [error] 9411#0: *7 open() "/usr/share/nginx/html/wac"
> failed (2: No such file or directory), client: 203.206.181.78, server:
> www.pricom.com.au, request: "GET /wac HTTP/1.1", host: "pricom.com.au"
>
> Why is nginx reverting to the default path and how is it getting the right
> dir without the correct path?
>
> Thanks,
>
> Phil.
>
> --
> Philip Rhoades
>
> GPO Box 3411
> Sydney NSW      2001
> Australia
> E-mail:  phil at pricom.com.au
>
> ______________________________**_________________
> 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/20111031/0e9b1d56/attachment.html>

From btm at loftninjas.org  Tue Nov  1 05:00:25 2011
From: btm at loftninjas.org (Bryan McLellan)
Date: Tue, 1 Nov 2011 01:00:25 -0400
Subject: Timeout when sending over 16k of data with UTF-8 characters
Message-ID: <CAMojFHADVx9uB+KGixDnZL-h0B5xPot55H9DK1j-hD7NTNag0Q@mail.gmail.com>

I'm experiencing an issue where nginx and having a difficult time
determining the cause. It is timing out on an HTTPS PUT (to an HTTP
backend) when content length is greater than 16384 (per
$content_length) and it contains UTF-8 characters. The connection
hangs and I get a 408 returned by nginx after 60 seconds. When I make
the same call directly to the backend, it succeeds. Removing the UTF-8
characters or reducing the seize to 16384 or below will both allow the
request to succeed through nginx.

client_body_buffer_size is defaulting to 8k, and using
$request_body_file I've determined that it isn't the use of a
temporary file triggering the error. Are there any other configuration
values that default to 16k?

My current nginx configuration:
nginx: nginx version: nginx/1.0.5
nginx: built by gcc 4.4.3 (Ubuntu 4.4.3-4ubuntu5)
nginx: TLS SNI support enabled
nginx: configure arguments: --conf-path=/etc/nginx
--prefix=/srv/nginx/1.0.5 --with-http_ssl_module
--with-http_stub_status_module --add-module=../nginx-x-rid-header
--with-ld-opt=-luuid

I also tried 1.1.6 to see if its UTF-8 patch would help, to no avail:
nginx: nginx version: nginx/1.1.6
nginx: built by gcc 4.4.3 (Ubuntu 4.4.3-4ubuntu5)
nginx: TLS SNI support enabled
nginx: configure arguments: --conf-path=/etc/nginx
--prefix=/tmp/nginx-1.1.6 --with-http_ssl_module
--with-http_stub_status_module --add-module=../nginx-x-rid-header
--with-ld-opt=-luuid

Successful PUT:

$status 200
$request_time 0.943
$body_bytes_sent 16384
$upstream_status 200
$upstream_response_time 0.216
$request_length 17330
$content_length 16384
$request_body_file /srv/nginx/1.0.5/client_body_temp/0000000249

Failing PUT:
$status 408
$request_time 60.911
$body_bytes_sent -
$upstream_status -
$upstream_response_time -
$request_length 17330
$content_length 16385
$request_body_file /srv/nginx/1.0.5/client_body_temp/0000000250

Ideas?
Bryan


From igor at sysoev.ru  Tue Nov  1 06:21:12 2011
From: igor at sysoev.ru (Igor Sysoev)
Date: Tue, 1 Nov 2011 10:21:12 +0400
Subject: Timeout when sending over 16k of data with UTF-8 characters
In-Reply-To: <CAMojFHADVx9uB+KGixDnZL-h0B5xPot55H9DK1j-hD7NTNag0Q@mail.gmail.com>
References: <CAMojFHADVx9uB+KGixDnZL-h0B5xPot55H9DK1j-hD7NTNag0Q@mail.gmail.com>
Message-ID: <20111101062112.GA73819@nginx.com>

On Tue, Nov 01, 2011 at 01:00:25AM -0400, Bryan McLellan wrote:
> I'm experiencing an issue where nginx and having a difficult time
> determining the cause. It is timing out on an HTTPS PUT (to an HTTP
> backend) when content length is greater than 16384 (per
> $content_length) and it contains UTF-8 characters. The connection
> hangs and I get a 408 returned by nginx after 60 seconds. When I make
> the same call directly to the backend, it succeeds. Removing the UTF-8
> characters or reducing the seize to 16384 or below will both allow the
> request to succeed through nginx.
> 
> client_body_buffer_size is defaulting to 8k, and using
> $request_body_file I've determined that it isn't the use of a
> temporary file triggering the error. Are there any other configuration
> values that default to 16k?
> 
> My current nginx configuration:
> nginx: nginx version: nginx/1.0.5
> nginx: built by gcc 4.4.3 (Ubuntu 4.4.3-4ubuntu5)
> nginx: TLS SNI support enabled
> nginx: configure arguments: --conf-path=/etc/nginx
> --prefix=/srv/nginx/1.0.5 --with-http_ssl_module
> --with-http_stub_status_module --add-module=../nginx-x-rid-header
> --with-ld-opt=-luuid
> 
> I also tried 1.1.6 to see if its UTF-8 patch would help, to no avail:
> nginx: nginx version: nginx/1.1.6
> nginx: built by gcc 4.4.3 (Ubuntu 4.4.3-4ubuntu5)
> nginx: TLS SNI support enabled
> nginx: configure arguments: --conf-path=/etc/nginx
> --prefix=/tmp/nginx-1.1.6 --with-http_ssl_module
> --with-http_stub_status_module --add-module=../nginx-x-rid-header
> --with-ld-opt=-luuid
> 
> Successful PUT:
> 
> $status 200
> $request_time 0.943
> $body_bytes_sent 16384
> $upstream_status 200
> $upstream_response_time 0.216
> $request_length 17330
> $content_length 16384
> $request_body_file /srv/nginx/1.0.5/client_body_temp/0000000249
> 
> Failing PUT:
> $status 408
> $request_time 60.911
> $body_bytes_sent -
> $upstream_status -
> $upstream_response_time -
> $request_length 17330
> $content_length 16385
> $request_body_file /srv/nginx/1.0.5/client_body_temp/0000000250
> 
> Ideas?
> Bryan

Could you create debug log:
http://nginx.org/en/docs/debugging_log.html

What does this module do ?
--add-module=../nginx-x-rid-header


-- 
Igor Sysoev


From stef at caunter.ca  Tue Nov  1 07:41:55 2011
From: stef at caunter.ca (Stefan Caunter)
Date: Tue, 1 Nov 2011 03:41:55 -0400
Subject: First Question
In-Reply-To: <8f5f460e1ade4806f4427aafb4484401@www.pricom.com.au>
References: <mailman.72.1320108576.3031.nginx@nginx.org>
	<8f5f460e1ade4806f4427aafb4484401@www.pricom.com.au>
Message-ID: <CALdpnrwBGHCN-a9703LYAzPu5ZF2MNzwEGQWVq-XVmwu7MGGtQ@mail.gmail.com>

On Mon, Oct 31, 2011 at 11:05 PM, Philip Rhoades <phil at pricom.com.au> wrote:
> People,
>
> I want to switch my Apache SSL to Nginx - that might help me resolve a Ruby
> on
> Rails issue and it is too much work to change all the non-SSL stuff over as
> well - I will do that later when I have time. ?So for the time being, http
> will
> be handled by Apache and https will be handled by Nginx.
>
> I commented out the default server in:
>
> ?/etc/nginx/nginx.conf
>
> and changed:
>
> ?/etc/nginx/conf.d/ssl.conf
>
> thus:
>
> #
> # HTTPS server configuration
> #
>
> server {
> ? ?listen ? ? ? 443;
> ? ?server_name ?www.pricom.com.au;
> ? ?access_log /var/log/nginx/nginx.vhost.access.log;
> ? ?error_log /var/log/nginx/nginx.vhost.error.log;
>
> ? ?ssl ? ? ? ? ? ? ? ? ?on;
> ? ?ssl_certificate ? ? ?/etc/httpd/conf/ssl.crt/RapidSSL.crt;
> ? ?ssl_certificate_key ?/etc/httpd/conf/ssl.key/mars-server.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 ? /var/www/ssl/wac;
> ? ?}
> }

Did you restart nginx after these changes?

>
> but in:
>
> ?/var/log/nginx/nginx.vhost.error.log
>
> I get:
>
> 2011/11/01 13:50:28 [error] 9411#0: *7 open() "/usr/share/nginx/html/wac"
> failed (2: No such file or directory), client: 203.206.181.78, server:
> www.pricom.com.au, request: "GET /wac HTTP/1.1", host: "pricom.com.au"

You get 404, correct? Restart nginx and see if it still does this.

>
> Why is nginx reverting to the default path and how is it getting the right
> dir without the correct path?
>

Why not use nginx to proxy_pass ssl to http apache back end?

> Thanks,
>
> Phil.
>
> --
> Philip Rhoades
>
> GPO Box 3411
> Sydney NSW ? ? ?2001
> Australia
> E-mail: ?phil at pricom.com.au
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx
>


From kworthington at gmail.com  Tue Nov  1 08:20:35 2011
From: kworthington at gmail.com (Kevin Worthington)
Date: Tue, 1 Nov 2011 04:20:35 -0400
Subject: nginx-1.1.7
In-Reply-To: <20111031145842.GD45607@nginx.com>
References: <20111031145842.GD45607@nginx.com>
Message-ID: <CAGo79UVQN35bL5APFSdMiSP0EY_4+Jwf1miErEMJvy4ZWgrurA@mail.gmail.com>

Hello Nginx Users,

I just released Nginx 1.1.7 For Windows http://goo.gl/3UWH0 (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.
http://www.kevinworthington.com/




On Mon, Oct 31, 2011 at 10:58 AM, Igor Sysoev <igor at sysoev.ru> wrote:
> Changes with nginx 1.1.7 ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? 31 Oct 2011
>
> ? ?*) Feature: support of several resolvers in the "resolver" directive.
> ? ? ? Thanks to Kirill A. Korinskiy.
>
> ? ?*) Bugfix: a segmentation fault occurred on start or while
> ? ? ? reconfiguration if the "ssl" directive was used at http level and
> ? ? ? there was no "ssl_certificate" defined.
>
> ? ?*) Bugfix: reduced memory consumption while proxying of big files if
> ? ? ? they were buffered to disk.
>
> ? ?*) Bugfix: a segmentation fault might occur in a worker process if
> ? ? ? "proxy_http_version 1.1" directive was used.
>
> ? ?*) Bugfix: in the "expires @time" directive.
>
>
> --
> Igor Sysoev
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx
>


From sslavic at gmail.com  Tue Nov  1 10:54:27 2011
From: sslavic at gmail.com (=?UTF-8?Q?Stevo_Slavi=C4=87?=)
Date: Tue, 1 Nov 2011 11:54:27 +0100
Subject: nginx, jmeter and xml-rpc
Message-ID: <CAAUywg98Dje9t2LOHkQVhtiauOxcES6k-29RSF1bc-cDm9A_3A@mail.gmail.com>

Hello nginx users,

I'm experiencing issue testing java based xml-rpc service deployed on
tomcat 6 using jmeter 2.5.1. When the application server is fronted by
nginx 1.0.8 I'm getting http 408 error code when message being sent is
~30+ lines of indented XML, nginx doesn't seem to get whole content of
the message jmeter is sending and keeps waiting until timeout occurs.
When message is ~10 lines of XML, it gets passed through to the tomcat
and back to the client well. When I replace nginx with apache httpd it
works well for both bigger and smaller message.

Has anyone experienced anything similar? Any thoughts where to look
for the root cause are more than welcome.

IMO it's either a bug in nginx or configuration issue. At the moment I
can share just the output of nginx -V (see [1])

Regards,
Stevo.


[1] "nginx -V" output
[foo at bar ~]$ nginx -V
nginx: nginx version: nginx/1.0.8
nginx: built by gcc 4.1.2 20080704 (Red Hat 4.1.2-46)
nginx: TLS SNI support disabled
nginx: configure arguments: --user=nginx --group=nginx
--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
--http-uwsgi-temp-path=/var/lib/nginx/tmp/uwsgi
--http-scgi-temp-path=/var/lib/nginx/tmp/scgi
--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_xslt_module
--with-http_sub_module --with-http_gzip_static_module
--with-http_random_index_module --with-http_secure_link_module
--with-http_stub_status_module --with-cc-opt='-O2 -g -m64
-mtune=generic' --with-cc-opt='-O2 -g -m64 -mtune=generic'
--add-module=nginx_ajp_module --add-module=ngx_postgres-0.8
--add-module=agentzh-nginx-eval-module


From j.j.molenaar at gmail.com  Tue Nov  1 12:20:55 2011
From: j.j.molenaar at gmail.com (Joost Molenaar)
Date: Tue, 1 Nov 2011 13:20:55 +0100
Subject: nginx, jmeter and xml-rpc
In-Reply-To: <CAAUywg98Dje9t2LOHkQVhtiauOxcES6k-29RSF1bc-cDm9A_3A@mail.gmail.com>
References: <CAAUywg98Dje9t2LOHkQVhtiauOxcES6k-29RSF1bc-cDm9A_3A@mail.gmail.com>
Message-ID: <CAG07R1C7yGtLmE-MvQdR7jg=wK2To=Vc-uDEeB0cSiK1=Qmp_Q@mail.gmail.com>

Your issue is probably the buffer sizes:
http://wiki.nginx.org/HttpProxyModule#proxy_buffer_size

Greetings,
Joost
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20111101/302be3b2/attachment.html>

From btm at loftninjas.org  Tue Nov  1 13:13:44 2011
From: btm at loftninjas.org (Bryan McLellan)
Date: Tue, 1 Nov 2011 09:13:44 -0400
Subject: Timeout when sending over 16k of data with UTF-8 characters
In-Reply-To: <20111101062112.GA73819@nginx.com>
References: <CAMojFHADVx9uB+KGixDnZL-h0B5xPot55H9DK1j-hD7NTNag0Q@mail.gmail.com>
	<20111101062112.GA73819@nginx.com>
Message-ID: <CAMojFHD6QGtv8kvLijB3=iC-ZsrL_8Lseccm6CCNh4Hfz_xCyw@mail.gmail.com>

On Tue, Nov 1, 2011 at 2:21 AM, Igor Sysoev <igor at sysoev.ru> wrote:
> Could you create debug log:
> http://nginx.org/en/docs/debugging_log.html

https://gist.github.com/1330454

You

Ruby throws an EOFError when this occurs:

EOFError: end of file reached
	from /home/btm/.rvm/rubies/ruby-1.9.1-p431/lib/ruby/1.9.1/net/protocol.rb:135:in
`sysread'
	from /home/btm/.rvm/rubies/ruby-1.9.1-p431/lib/ruby/1.9.1/net/protocol.rb:135:in
`block in rbuf_fill'
	from /home/btm/.rvm/rubies/ruby-1.9.1-p431/lib/ruby/1.9.1/timeout.rb:52:in
`timeout'
	from /home/btm/.rvm/rubies/ruby-1.9.1-p431/lib/ruby/1.9.1/timeout.rb:82:in
`timeout'
	from /home/btm/.rvm/rubies/ruby-1.9.1-p431/lib/ruby/1.9.1/net/protocol.rb:134:in
`rbuf_fill'
	from /home/btm/.rvm/rubies/ruby-1.9.1-p431/lib/ruby/1.9.1/net/protocol.rb:116:in
`readuntil'
	from /home/btm/.rvm/rubies/ruby-1.9.1-p431/lib/ruby/1.9.1/net/protocol.rb:126:in
`readline'
	from /home/btm/.rvm/rubies/ruby-1.9.1-p431/lib/ruby/1.9.1/net/http.rb:2136:in
`read_status_line'
	from /home/btm/.rvm/rubies/ruby-1.9.1-p431/lib/ruby/1.9.1/net/http.rb:2125:in
`read_new'
	from /home/btm/.rvm/rubies/ruby-1.9.1-p431/lib/ruby/1.9.1/net/http.rb:1117:in
`transport_request'
	from /home/btm/.rvm/rubies/ruby-1.9.1-p431/lib/ruby/1.9.1/net/http.rb:1103:in
`request'
	from /home/btm/.rvm/rubies/ruby-1.9.1-p431/lib/ruby/1.9.1/net/http.rb:1096:in
`block in request'
	from /home/btm/.rvm/rubies/ruby-1.9.1-p431/lib/ruby/1.9.1/net/http.rb:564:in
`start'
	from /home/btm/.rvm/rubies/ruby-1.9.1-p431/lib/ruby/1.9.1/net/http.rb:1094:in
`request'
	from /home/btm/.rvm/gems/ruby-1.9.1-p431/gems/chef-0.10.4/lib/chef/rest/rest_request.rb:84:in
`block in call'
	from /home/btm/.rvm/gems/ruby-1.9.1-p431/gems/chef-0.10.4/lib/chef/rest/rest_request.rb:99:in
`hide_net_http_bug'
	from /home/btm/.rvm/gems/ruby-1.9.1-p431/gems/chef-0.10.4/lib/chef/rest/rest_request.rb:83:in
`call'
	from /home/btm/.rvm/gems/ruby-1.9.1-p431/gems/chef-0.10.4/lib/chef/rest.rb:219:in
`block in api_request'
	from /home/btm/.rvm/gems/ruby-1.9.1-p431/gems/chef-0.10.4/lib/chef/rest.rb:288:in
`retriable_rest_request'
	from /home/btm/.rvm/gems/ruby-1.9.1-p431/gems/chef-0.10.4/lib/chef/rest.rb:218:in
`api_request'
	from /home/btm/.rvm/gems/ruby-1.9.1-p431/gems/chef-0.10.4/lib/chef/rest.rb:130:in
`put_rest'
	from /home/btm/.rvm/gems/ruby-1.9.1-p431/gems/chef-0.10.4/lib/chef/node.rb:626:in
`save'
	from (irb):9
	from /home/btm/.rvm/gems/ruby-1.9.1-p431/gems/chef-0.10.4/lib/chef/shef.rb:73:in
`block in start'
	from /home/btm/.rvm/gems/ruby-1.9.1-p431/gems/chef-0.10.4/lib/chef/shef.rb:72:in
`catch'
	from /home/btm/.rvm/gems/ruby-1.9.1-p431/gems/chef-0.10.4/lib/chef/shef.rb:72:in
`start'
	from /home/btm/.rvm/gems/ruby-1.9.1-p431/gems/chef-0.10.4/bin/shef:34:in
`<top (required)>'
	from /home/btm/.rvm/gems/ruby-1.9.1-p431/bin/shef:19:in `load'
	from /home/btm/.rvm/gems/ruby-1.9.1-p431/bin/shef:19:in `<main>'chef > quit


> What does this module do ?
> --add-module=../nginx-x-rid-header

https://github.com/newobj/nginx-x-rid-header

It adds a header line with a uuid for associating frontend and backend requests.

It fails without that module as well with this configuration:

nginx: nginx version: nginx/1.1.6
nginx: built by gcc 4.4.3 (Ubuntu 4.4.3-4ubuntu5)
nginx: TLS SNI support enabled
nginx: configure arguments: --conf-path=/etc/nginx
--prefix=/tmp/nginx-1.1.6 --with-http_ssl_module
--with-http_stub_status_module --with-debug

Bryan


From mdounin at mdounin.ru  Tue Nov  1 13:56:22 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Tue, 1 Nov 2011 17:56:22 +0400
Subject: Timeout when sending over 16k of data with UTF-8 characters
In-Reply-To: <CAMojFHD6QGtv8kvLijB3=iC-ZsrL_8Lseccm6CCNh4Hfz_xCyw@mail.gmail.com>
References: <CAMojFHADVx9uB+KGixDnZL-h0B5xPot55H9DK1j-hD7NTNag0Q@mail.gmail.com>
	<20111101062112.GA73819@nginx.com>
	<CAMojFHD6QGtv8kvLijB3=iC-ZsrL_8Lseccm6CCNh4Hfz_xCyw@mail.gmail.com>
Message-ID: <20111101135622.GI95664@mdounin.ru>

Hello!

On Tue, Nov 01, 2011 at 09:13:44AM -0400, Bryan McLellan wrote:

> On Tue, Nov 1, 2011 at 2:21 AM, Igor Sysoev <igor at sysoev.ru> wrote:
> > Could you create debug log:
> > http://nginx.org/en/docs/debugging_log.html
> 
> https://gist.github.com/1330454

2011/11/01 12:44:23 [debug] 13689#0: *129 http header: "Content-Length: 16385"
...
2011/11/01 12:44:23 [debug] 13689#0: *129 http header: "User-Agent: Chef Client/0.10.4 (ruby-1.9.1-p431; ohai-0.6.10; x86_64-linux; +http://opscode.com)"
...

Note: content length is 16385.

...
2011/11/01 12:44:23 [debug] 13689#0: *129 http read client request body
2011/11/01 12:44:23 [debug] 13689#0: *129 SSL_read: -1
2011/11/01 12:44:23 [debug] 13689#0: *129 SSL_get_error: 2
2011/11/01 12:44:23 [debug] 13689#0: *129 http client request body recv -2
2011/11/01 12:44:23 [debug] 13689#0: *129 http client request body rest 16385
...
2011/11/01 12:44:24 [debug] 13689#0: *129 http read client request body
2011/11/01 12:44:24 [debug] 13689#0: *129 SSL_read: 8192
2011/11/01 12:44:24 [debug] 13689#0: *129 http client request body recv 8192
...
2011/11/01 12:44:24 [debug] 13689#0: *129 SSL_read: 8192
2011/11/01 12:44:24 [debug] 13689#0: *129 http client request body recv 8192
2011/11/01 12:44:24 [debug] 13689#0: *129 write: 13, 0870D8E0, 8192, 8192
2011/11/01 12:44:24 [debug] 13689#0: *129 SSL_read: -1
2011/11/01 12:44:24 [debug] 13689#0: *129 SSL_get_error: 2
2011/11/01 12:44:24 [debug] 13689#0: *129 http client request body recv -2
2011/11/01 12:44:24 [debug] 13689#0: *129 http client request body rest 1

Here 16384 bytes has been read from client, and one byte remains 
("rest 1").  OpenSSL doesn't provide any more bytes and claims it 
needs more network input (2, SSL_ERROR_WANT_READ).

2011/11/01 12:44:24 [debug] 13689#0: *129 event timer del: 12: 1596563798
2011/11/01 12:44:24 [debug] 13689#0: *129 event timer add: 12: 60000:1596564354
2011/11/01 12:45:24 [debug] 13689#0: *129 event timer del: 12: 1596564354
2011/11/01 12:45:24 [debug] 13689#0: *129 http run request: "/organizations/opscode-btm/nodes/broken?"
2011/11/01 12:45:24 [debug] 13689#0: *129 http finalize request: 408, "/organizations/opscode-btm/nodes/broken?" a:1, c:1

Though client fails to provide one more byte.

For me, it looks like problem in client.  It's either calculate 
content length incorrectly or fails to properly flush ssl buffers 
on it's side.

Maxim Dounin


From igor at sysoev.ru  Tue Nov  1 14:57:36 2011
From: igor at sysoev.ru (Igor Sysoev)
Date: Tue, 1 Nov 2011 18:57:36 +0400
Subject: nginx-1.0.9
Message-ID: <20111101145736.GA89884@nginx.com>

Changes with nginx 1.0.9                                         01 Nov 2011

    *) Change: now the 0x7F-0x1F characters are escaped as \xXX in an
       access_log.

    *) Change: now SIGWINCH signal works only in daemon mode.

    *) Feature: "proxy/fastcgi/scgi/uwsgi_ignore_headers" directives support
       the following additional values: X-Accel-Limit-Rate,
       X-Accel-Buffering, X-Accel-Charset.

    *) Feature: decrease of memory consumption if SSL is used.

    *) Feature: accept filters are now supported on NetBSD.

    *) Feature: the "uwsgi_buffering" and "scgi_buffering" directives.
       Thanks to Peter Smit.

    *) Bugfix: a segmentation fault occurred on start or while
       reconfiguration if the "ssl" directive was used at http level and
       there was no "ssl_certificate" defined.

    *) Bugfix: some UTF-8 characters were processed incorrectly.
       Thanks to Alexey Kuts.

    *) Bugfix: the ngx_http_rewrite_module directives specified at "server"
       level were executed twice if no matching locations were defined.

    *) Bugfix: a socket leak might occurred if "aio sendfile" was used.

    *) Bugfix: connections with fast clients might be closed after
       send_timeout if file AIO was used.

    *) Bugfix: in the ngx_http_autoindex_module.

    *) Bugfix: the module ngx_http_mp4_module did not support seeking on
       32-bit platforms.

    *) Bugfix: non-cacheable responses might be cached if
       "proxy_cache_bypass" directive was used.
       Thanks to John Ferlito.

    *) Bugfix: cached responses with an empty body were returned
       incorrectly; the bug had appeared in 0.8.31.

    *) Bugfix: 201 responses of the ngx_http_dav_module were incorrect; the
       bug had appeared in 0.8.32.

    *) Bugfix: in the "return" directive.

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


-- 
Igor Sysoev


From ilan at time4learning.com  Tue Nov  1 15:00:19 2011
From: ilan at time4learning.com (Ilan Berkner)
Date: Tue, 1 Nov 2011 11:00:19 -0400
Subject: nginx-1.0.9
In-Reply-To: <20111101145736.GA89884@nginx.com>
References: <20111101145736.GA89884@nginx.com>
Message-ID: <CAOoWpPpfvB-9N06zwwjF7KLMkQXf=-XfoNhB1Vb3fdiMxVsoeg@mail.gmail.com>

Thanks, question:

Do these bug fixes and changes make it into the current development branch
as well (1.1.17)?

On Tue, Nov 1, 2011 at 10:57 AM, Igor Sysoev <igor at sysoev.ru> wrote:

> Changes with nginx 1.0.9                                         01 Nov
> 2011
>
>    *) Change: now the 0x7F-0x1F characters are escaped as \xXX in an
>       access_log.
>
>    *) Change: now SIGWINCH signal works only in daemon mode.
>
>    *) Feature: "proxy/fastcgi/scgi/uwsgi_ignore_headers" directives support
>       the following additional values: X-Accel-Limit-Rate,
>       X-Accel-Buffering, X-Accel-Charset.
>
>    *) Feature: decrease of memory consumption if SSL is used.
>
>    *) Feature: accept filters are now supported on NetBSD.
>
>    *) Feature: the "uwsgi_buffering" and "scgi_buffering" directives.
>       Thanks to Peter Smit.
>
>    *) Bugfix: a segmentation fault occurred on start or while
>       reconfiguration if the "ssl" directive was used at http level and
>       there was no "ssl_certificate" defined.
>
>    *) Bugfix: some UTF-8 characters were processed incorrectly.
>       Thanks to Alexey Kuts.
>
>    *) Bugfix: the ngx_http_rewrite_module directives specified at "server"
>       level were executed twice if no matching locations were defined.
>
>    *) Bugfix: a socket leak might occurred if "aio sendfile" was used.
>
>    *) Bugfix: connections with fast clients might be closed after
>       send_timeout if file AIO was used.
>
>    *) Bugfix: in the ngx_http_autoindex_module.
>
>    *) Bugfix: the module ngx_http_mp4_module did not support seeking on
>       32-bit platforms.
>
>    *) Bugfix: non-cacheable responses might be cached if
>       "proxy_cache_bypass" directive was used.
>       Thanks to John Ferlito.
>
>    *) Bugfix: cached responses with an empty body were returned
>       incorrectly; the bug had appeared in 0.8.31.
>
>    *) Bugfix: 201 responses of the ngx_http_dav_module were incorrect; the
>       bug had appeared in 0.8.32.
>
>    *) Bugfix: in the "return" directive.
>
>    *) Bugfix: the "ssl_verify_client", "ssl_verify_depth", and
>       "ssl_prefer_server_ciphers" directives might work incorrectly if SNI
>       was used.
>
>
> --
> 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/20111101/ab506e49/attachment.html>

From igor at sysoev.ru  Tue Nov  1 15:04:48 2011
From: igor at sysoev.ru (Igor Sysoev)
Date: Tue, 1 Nov 2011 19:04:48 +0400
Subject: nginx-1.0.9
In-Reply-To: <CAOoWpPpfvB-9N06zwwjF7KLMkQXf=-XfoNhB1Vb3fdiMxVsoeg@mail.gmail.com>
References: <20111101145736.GA89884@nginx.com>
	<CAOoWpPpfvB-9N06zwwjF7KLMkQXf=-XfoNhB1Vb3fdiMxVsoeg@mail.gmail.com>
Message-ID: <20111101150448.GE89884@nginx.com>

On Tue, Nov 01, 2011 at 11:00:19AM -0400, Ilan Berkner wrote:
> Thanks, question:
> 
> Do these bug fixes and changes make it into the current development branch
> as well (1.1.17)?

Quite the opposite, these fixes and changes were merged from development
branch.


-- 
Igor Sysoev


From nginx-forum at nginx.us  Tue Nov  1 15:29:44 2011
From: nginx-forum at nginx.us (artemg)
Date: Tue, 01 Nov 2011 11:29:44 -0400
Subject: What is the right way if module doesnt provide data for upstream
	filters
Message-ID: <bb138c7c6fd2eb441177ffbfb1c8a03e.NginxMailingListEnglish@forum.nginx.org>

I have module, that matches data going via nginx. I am buffering data,
so there is situation that there is no data to provide to upstream
filters. What is the right way to do in such case? Now I am calling
next_body_filter(r, NULL); Is that ok?

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,217654,217654#msg-217654


From quintinpar at gmail.com  Tue Nov  1 15:44:29 2011
From: quintinpar at gmail.com (Quintin Par)
Date: Tue, 1 Nov 2011 21:14:29 +0530
Subject: =?UTF-8?Q?Cache_HIT=E2=80=99s_not_happening?=
Message-ID: <CACW9FdOM2P2Kv93pjOnaNruvByAPYCuo+R_qqR7j+COXQyLH+A@mail.gmail.com>

Hi all,

Can someone please answer this question on serverfault(Its difficult to
retain code formatting and images in the group mail, hence the post in
serverfault.com).

http://serverfault.com/questions/326545/nginx-cache-hits-not-happening

Thanks,
Quintin
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20111101/d8266358/attachment-0001.html>

From mdounin at mdounin.ru  Tue Nov  1 16:22:54 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Tue, 1 Nov 2011 20:22:54 +0400
Subject: =?UTF-8?Q?Re=3A_Cache_HIT=E2=80=99s_not_happening?=
In-Reply-To: <CACW9FdOM2P2Kv93pjOnaNruvByAPYCuo+R_qqR7j+COXQyLH+A@mail.gmail.com>
References: <CACW9FdOM2P2Kv93pjOnaNruvByAPYCuo+R_qqR7j+COXQyLH+A@mail.gmail.com>
Message-ID: <20111101162254.GK95664@mdounin.ru>

Hello!

On Tue, Nov 01, 2011 at 09:14:29PM +0530, Quintin Par wrote:

> Hi all,
> 
> Can someone please answer this question on serverfault(Its difficult to
> retain code formatting and images in the group mail, hence the post in
> serverfault.com).
> 
> http://serverfault.com/questions/326545/nginx-cache-hits-not-happening

Answer from kolbyjack suggesting to add

    proxy_ignore_headers Set-Cookie; 

is likely correct as long as your backend actually sends 
Set-Cookie header.

If it doesn't fix your problem, please provide full config and 
debug log.  See here http://wiki.nginx.org/Debugging for more 
details.

Maxim Dounin


From stef at caunter.ca  Tue Nov  1 16:25:12 2011
From: stef at caunter.ca (Stefan Caunter)
Date: Tue, 1 Nov 2011 12:25:12 -0400
Subject: nginx, jmeter and xml-rpc
In-Reply-To: <CAAUywg98Dje9t2LOHkQVhtiauOxcES6k-29RSF1bc-cDm9A_3A@mail.gmail.com>
References: <CAAUywg98Dje9t2LOHkQVhtiauOxcES6k-29RSF1bc-cDm9A_3A@mail.gmail.com>
Message-ID: <CALdpnrzf2fp2cYj0pxejfk4PJ=PwazNzY115jGs=phUL8DjXBg@mail.gmail.com>

On Tue, Nov 1, 2011 at 6:54 AM, Stevo Slavi? <sslavic at gmail.com> wrote:
> Hello nginx users,
>
> I'm experiencing issue testing java based xml-rpc service deployed on
> tomcat 6 using jmeter 2.5.1. When the application server is fronted by
> nginx 1.0.8 I'm getting http 408 error code when message being sent is
> ~30+ lines of indented XML, nginx doesn't seem to get whole content of
> the message jmeter is sending and keeps waiting until timeout occurs.
> When message is ~10 lines of XML, it gets passed through to the tomcat
> and back to the client well. When I replace nginx with apache httpd it
> works well for both bigger and smaller message.
>
> Has anyone experienced anything similar? Any thoughts where to look
> for the root cause are more than welcome.
>
> IMO it's either a bug in nginx or configuration issue. At the moment I
> can share just the output of nginx -V (see [1])
>

You need to provide nginx configuration, and log output at a minimum.
It is probably configuration, but no one can help you without you
providing data.

> Regards,
> Stevo.
>
>
> [1] "nginx -V" output
> [foo at bar ~]$ nginx -V
> nginx: nginx version: nginx/1.0.8
> nginx: built by gcc 4.1.2 20080704 (Red Hat 4.1.2-46)
> nginx: TLS SNI support disabled
> nginx: configure arguments: --user=nginx --group=nginx
> --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
> --http-uwsgi-temp-path=/var/lib/nginx/tmp/uwsgi
> --http-scgi-temp-path=/var/lib/nginx/tmp/scgi
> --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_xslt_module
> --with-http_sub_module --with-http_gzip_static_module
> --with-http_random_index_module --with-http_secure_link_module
> --with-http_stub_status_module --with-cc-opt='-O2 -g -m64
> -mtune=generic' --with-cc-opt='-O2 -g -m64 -mtune=generic'
> --add-module=nginx_ajp_module --add-module=ngx_postgres-0.8
> --add-module=agentzh-nginx-eval-module
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx
>


From nginx-forum at nginx.us  Tue Nov  1 17:55:45 2011
From: nginx-forum at nginx.us (firestorm)
Date: Tue, 01 Nov 2011 13:55:45 -0400
Subject: Problem with fastcgi cache
In-Reply-To: <4974ddde02c516685929d60a0ee705f9.NginxMailingListEnglish@forum.nginx.org>
References: <4974ddde02c516685929d60a0ee705f9.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <28963de8765052c595f6afb7a55e4e44.NginxMailingListEnglish@forum.nginx.org>

For more information my nginx server is installed in Ubuntu 10.04,
compiled from sources the following compilation options:

./configure --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

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,217538,217668#msg-217668


From diniz at freeddom.com  Tue Nov  1 19:22:47 2011
From: diniz at freeddom.com (=?utf-8?Q?Felipe_Jos=C3=A9_Diniz?=)
Date: Tue, 1 Nov 2011 17:22:47 -0200 (BRST)
Subject: Proxy_upstrem error
In-Reply-To: <344842845.212894.1320174426177.JavaMail.root@webmail>
Message-ID: <1511750432.213022.1320175367217.JavaMail.root@webmail>

Hi, 

We are using nginx 0.7.67, and we need to filter a specific error, and match it with proxy_next_upstream.
The error we need is when a client send a Connection Reset. I believe nginx filter it with "proxy_next_upstream error" along with all the others connection errors. For ours application the reset error must be the only one filtered and then the nginx would use proxy_next_upstream to send the request to another machine.

Is there a way to do such filter in nginx configuration?

Thanx

Felipe Diniz 



From ehabkost at raisama.net  Wed Nov  2 00:27:46 2011
From: ehabkost at raisama.net (Eduardo Habkost)
Date: Tue, 1 Nov 2011 22:27:46 -0200
Subject: nginx-1.0.9
In-Reply-To: <20111101145736.GA89884@nginx.com>
References: <20111101145736.GA89884@nginx.com>
Message-ID: <CAFszQJjArw62D0NKhO2N8D5pzPqePeRjsqACmdHYNkUwNW3a0g@mail.gmail.com>

Hi,

I don't know if others reported this, already: I don't see a PGP
signature file available for 1.0.9; the pgp link on nginx.org
(http://nginx.org/download/nginx-1.0.9.tar.gz.asc) returns a 404
error.

-- 
Eduardo

On Tue, Nov 1, 2011 at 12:57 PM, Igor Sysoev <igor at sysoev.ru> wrote:
> Changes with nginx 1.0.9 ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? 01 Nov 2011
>
[...]
>
> --
> Igor Sysoev
>


From igor at sysoev.ru  Wed Nov  2 05:27:09 2011
From: igor at sysoev.ru (Igor Sysoev)
Date: Wed, 2 Nov 2011 09:27:09 +0400
Subject: nginx-1.0.9
In-Reply-To: <CAFszQJjArw62D0NKhO2N8D5pzPqePeRjsqACmdHYNkUwNW3a0g@mail.gmail.com>
References: <20111101145736.GA89884@nginx.com>
	<CAFszQJjArw62D0NKhO2N8D5pzPqePeRjsqACmdHYNkUwNW3a0g@mail.gmail.com>
Message-ID: <20111102052708.GA8238@nginx.com>

On Tue, Nov 01, 2011 at 10:27:46PM -0200, Eduardo Habkost wrote:
> Hi,
> 
> I don't know if others reported this, already: I don't see a PGP
> signature file available for 1.0.9; the pgp link on nginx.org
> (http://nginx.org/download/nginx-1.0.9.tar.gz.asc) returns a 404
> error.

Fixed, thank you.


-- 
Igor Sysoev


From nginx-forum at nginx.us  Wed Nov  2 06:21:31 2011
From: nginx-forum at nginx.us (wangbin579)
Date: Wed, 02 Nov 2011 02:21:31 -0400
Subject: Tcpcopy,an online request replication tool fit for nginx
Message-ID: <141f5701aa1bdc50e4d7a29c237508a3.NginxMailingListEnglish@forum.nginx.org>

Tcpcopy  is an online request replication tool .
It may be useful for migrating to nginx from apache

https://github.com/wangbin579/tcpcopy or
http://code.google.com/p/tcpcopy/

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,217680,217680#msg-217680


From nginx-forum at nginx.us  Wed Nov  2 06:32:21 2011
From: nginx-forum at nginx.us (wangbin579)
Date: Wed, 02 Nov 2011 02:32:21 -0400
Subject: nginx_hmux_module - support hmux protocol proxy with Nginx
Message-ID: <2d7909af72c23eb07f1385d88d6ae095.NginxMailingListEnglish@forum.nginx.org>

With this module, Nginx can connect to Resin through hmux protocol
directly.
You also can use tcpcopy to test this module.

nginx_hmux_module:
https://github.com/wangbin579/nginx-hmux-module or
http://code.google.com/p/nginx-hmux-module/
tcpcopy:
https://github.com/wangbin579/tcpcopy or
http://code.google.com/p/tcpcopy/

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,217681,217681#msg-217681


From nginx-forum at nginx.us  Wed Nov  2 06:34:46 2011
From: nginx-forum at nginx.us (wangbin579)
Date: Wed, 02 Nov 2011 02:34:46 -0400
Subject: Tcpcopy,an online request replication tool fit for nginx
In-Reply-To: <141f5701aa1bdc50e4d7a29c237508a3.NginxMailingListEnglish@forum.nginx.org>
References: <141f5701aa1bdc50e4d7a29c237508a3.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <2845639b8a3be7293ab599241e0689ed.NginxMailingListEnglish@forum.nginx.org>

you can use tcpcopy to compare the performances of apache and nginx.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,217680,217682#msg-217682


From nginx-forum at nginx.us  Wed Nov  2 06:56:38 2011
From: nginx-forum at nginx.us (wangbin579)
Date: Wed, 02 Nov 2011 02:56:38 -0400
Subject: Tcpcopy,an online request replication tool fit for nginx
In-Reply-To: <141f5701aa1bdc50e4d7a29c237508a3.NginxMailingListEnglish@forum.nginx.org>
References: <141f5701aa1bdc50e4d7a29c237508a3.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <913df437a76e6c067628e67705adcdfd.NginxMailingListEnglish@forum.nginx.org>

tcpcopy helps me a lot when deveploping the netease ad system and other
systems

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,217680,217684#msg-217684


From kworthington at gmail.com  Wed Nov  2 09:45:05 2011
From: kworthington at gmail.com (Kevin Worthington)
Date: Wed, 2 Nov 2011 05:45:05 -0400
Subject: nginx-1.0.9
In-Reply-To: <20111101145736.GA89884@nginx.com>
References: <20111101145736.GA89884@nginx.com>
Message-ID: <CAGo79UWH8kDPvW_czf23848dnDHFMgL+a6z-rqcq7DGnB3i5bw@mail.gmail.com>

Hello Nginx Users,
Just released: Nginx 1.0.9 For Windows http://goo.gl/pZnNA (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
Thank you,Kevin--Kevin Worthingtonkworthington (at] gmail [dot}
.com.http://www.kevinworthington.com/


On Tue, Nov 1, 2011 at 10:57 AM, Igor Sysoev <igor at sysoev.ru> wrote:
> Changes with nginx 1.0.9 ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? 01 Nov 2011
>
> ? ?*) Change: now the 0x7F-0x1F characters are escaped as \xXX in an
> ? ? ? access_log.
>
> ? ?*) Change: now SIGWINCH signal works only in daemon mode.
>
> ? ?*) Feature: "proxy/fastcgi/scgi/uwsgi_ignore_headers" directives support
> ? ? ? the following additional values: X-Accel-Limit-Rate,
> ? ? ? X-Accel-Buffering, X-Accel-Charset.
>
> ? ?*) Feature: decrease of memory consumption if SSL is used.
>
> ? ?*) Feature: accept filters are now supported on NetBSD.
>
> ? ?*) Feature: the "uwsgi_buffering" and "scgi_buffering" directives.
> ? ? ? Thanks to Peter Smit.
>
> ? ?*) Bugfix: a segmentation fault occurred on start or while
> ? ? ? reconfiguration if the "ssl" directive was used at http level and
> ? ? ? there was no "ssl_certificate" defined.
>
> ? ?*) Bugfix: some UTF-8 characters were processed incorrectly.
> ? ? ? Thanks to Alexey Kuts.
>
> ? ?*) Bugfix: the ngx_http_rewrite_module directives specified at "server"
> ? ? ? level were executed twice if no matching locations were defined.
>
> ? ?*) Bugfix: a socket leak might occurred if "aio sendfile" was used.
>
> ? ?*) Bugfix: connections with fast clients might be closed after
> ? ? ? send_timeout if file AIO was used.
>
> ? ?*) Bugfix: in the ngx_http_autoindex_module.
>
> ? ?*) Bugfix: the module ngx_http_mp4_module did not support seeking on
> ? ? ? 32-bit platforms.
>
> ? ?*) Bugfix: non-cacheable responses might be cached if
> ? ? ? "proxy_cache_bypass" directive was used.
> ? ? ? Thanks to John Ferlito.
>
> ? ?*) Bugfix: cached responses with an empty body were returned
> ? ? ? incorrectly; the bug had appeared in 0.8.31.
>
> ? ?*) Bugfix: 201 responses of the ngx_http_dav_module were incorrect; the
> ? ? ? bug had appeared in 0.8.32.
>
> ? ?*) Bugfix: in the "return" directive.
>
> ? ?*) Bugfix: the "ssl_verify_client", "ssl_verify_depth", and
> ? ? ? "ssl_prefer_server_ciphers" directives might work incorrectly if SNI
> ? ? ? was used.
>
>
> --
> Igor Sysoev
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx
>


From nginx-forum at nginx.us  Wed Nov  2 10:09:59 2011
From: nginx-forum at nginx.us (est)
Date: Wed, 02 Nov 2011 06:09:59 -0400
Subject: nginx parse var in if file exist statement error?
Message-ID: <c33a9d760575d49a6861fc2dc02bc123.NginxMailingListEnglish@forum.nginx.org>

Hi guys,

I found a stange bug in nginx 0.7.65, 0.8.54, and 1.0.6

Here is my setup of three machines


$ nginx -V
nginx version: nginx/0.7.65
TLS SNI support enabled
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
--http-client-body-temp-path=/var/lib/nginx/body
--http-proxy-temp-path=/var/lib/nginx/proxy
--http-fastcgi-temp-path=/var/lib/nginx/fastcgi --with-debug
--with-http_stub_status_module --with-http_flv_module
--with-http_ssl_module --with-http_dav_module
--with-http_gzip_static_module --with-http_realip_module --with-mail
--with-mail_ssl_module --with-ipv6
--add-module=/build/buildd/nginx-0.7.65/modules/nginx-upstream-fair


$nginx -V
nginx version: nginx/0.8.54
built by gcc 4.2.4 (Ubuntu 4.2.4-1ubuntu4)
configure arguments: --user=www-data --group=www-data
--prefix=/usr/local/nginx --with-http_stub_status_module

$ nginx -V
nginx: nginx version: nginx/1.0.6
nginx: TLS SNI support enabled
nginx: configure arguments: --prefix=/etc/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 --pid-path=/var/run/nginx.pid
--lock-path=/var/run/nginx.lock
--http-client-body-temp-path=/var/cache/nginx/client_temp
--http-proxy-temp-path=/var/cache/nginx/proxy_temp
--http-fastcgi-temp-path=/var/cache/nginx/fastcgi_temp
--http-uwsgi-temp-path=/var/cache/nginx/uwcgi_temp
--http-scgi-temp-path=/var/cache/nginx/scgi_temp --user=nginx
--group=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_random_index_module
--with-http_secure_link_module --with-http_stub_status_module
--with-mail --with-mail_ssl_module --with-file-aio --with-ipv6


So I have dir to store images, either jpg or png, the file name (image
id) is unique. I want the URL to provide image id only, nginx to serve
static jpg or png.

the config snippet is something like this:


server {
  listen 80;
  error_log /var/log/nginx/error_imglib.log debug;
  root /home/develop/image_library;

  location ~* ^/img/small/(\d+)/?$ {
    set $fext 'jpg';
    if (!-f '/home/develop/image_library/dump/$1-s.jpg'){
      set $fext png;
    }
    alias '/home/develop/image_library/dump/$1-s.$fext' ;
  }
}

When running curl it returls something like this


$ curl  "http://127.0.0.1/img/small/1"
<html>
<head><title>404 Not Found</title></head>
<body bgcolor="white">
<center><h1>404 Not Found</h1></center>
<hr><center>nginx/1.0.6</center>
</body>
</html>

Now the bug:

in three of my machines I set the error_log level to "debug", here is
the output:

2011/11/02 14:01:09 [debug] 4194#0: *24 http script capture: "1"
2011/11/02 14:01:09 [debug] 4194#0: *24 http script copy: "-s."
2011/11/02 14:01:09 [debug] 4194#0: *24 http script var: "png"
2011/11/02 14:01:09 [debug] 4194#0: *24 http filename:
"/home/steve/image_library/dump/1-s.png1
User-Agent"
2011/11/02 14:01:09 [debug] 4194#0: *24 add cleanup: 0833B2F0
2011/11/02 14:01:09 [error] 4194#0: *24 open()
"/home/steve/image_library/dump/1" failed (2: No such file or
directory), client: 127.0.0.1, request: "GET /img/small/1 HTTP/1.1"



2011/11/01 23:00:53 [alert] 7793#0: *38
"/home/develop/image_library/dump/1-s.jpgindex.html" is not a directory,
request: "GET /img/small/1 HTTP/1.1"

2011/11/02 02:30:18 [error] 7666#0: *2491 open()
"/home/develop/image_library/dump/1-s.pn" failed (2: No such file or
directory), client: 127.0.0.1,  request: "GET /img/small/1 HTTP/1.1"

2011/11/01 22:47:45 [error] 6740#0: *24
"/home/develop/image_library/dump/1-s.jpgTP/1.1
Hostindex.html" is not found (2: No such file or directory), request:
"GET /img/small/1 HTTP/1.1"

So clearly somehow, nginx managed to mess HTTP headers into filename
parser.

Can anyone help me? This is the weirdest bug I have ever encountered
with nginx.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,217683,217683#msg-217683


From david.yu.ftw at gmail.com  Wed Nov  2 13:14:40 2011
From: david.yu.ftw at gmail.com (David Yu)
Date: Wed, 2 Nov 2011 21:14:40 +0800
Subject: nginx_hmux_module - support hmux protocol proxy with Nginx
In-Reply-To: <2d7909af72c23eb07f1385d88d6ae095.NginxMailingListEnglish@forum.nginx.org>
References: <2d7909af72c23eb07f1385d88d6ae095.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <CAOkdov+oaG7HfzOPmmpdM7u+d98oRFsT7px6EZ0JP=z__3ju6Q@mail.gmail.com>

On Wed, Nov 2, 2011 at 2:32 PM, wangbin579 <nginx-forum at nginx.us> wrote:

> With this module, Nginx can connect to Resin through hmux protocol
> directly.
>
Cool.  Do you know where I can some documentation on the hmux protocol
(already tried googling)?
Judging by the name "mux", I couldn't help but think this is a binary
protocol with multiplexing+keep-alive (w/c is awesome if it is).

> You also can use tcpcopy to test this module.
>
> nginx_hmux_module:
> https://github.com/wangbin579/nginx-hmux-module or
> http://code.google.com/p/nginx-hmux-module/
> tcpcopy:
> https://github.com/wangbin579/tcpcopy or
> http://code.google.com/p/tcpcopy/
>
> Posted at Nginx Forum:
> http://forum.nginx.org/read.php?2,217681,217681#msg-217681
>
> _______________________________________________
> 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/20111102/db839388/attachment.html>

From i at vbart.ru  Wed Nov  2 13:34:44 2011
From: i at vbart.ru (Valentin V. Bartenev)
Date: Wed, 2 Nov 2011 17:34:44 +0400
Subject: nginx parse var in if file exist statement error?
In-Reply-To: <c33a9d760575d49a6861fc2dc02bc123.NginxMailingListEnglish@forum.nginx.org>
References: <c33a9d760575d49a6861fc2dc02bc123.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <201111021734.44906.i@vbart.ru>

On Wednesday 02 November 2011 14:09:59 est wrote:
[...]
> So I have dir to store images, either jpg or png, the file name (image
> id) is unique. I want the URL to provide image id only, nginx to serve
> static jpg or png.
> 
> the config snippet is something like this:
> 
> 
> server {
>   listen 80;
>   error_log /var/log/nginx/error_imglib.log debug;
>   root /home/develop/image_library;
> 
>   location ~* ^/img/small/(\d+)/?$ {
>     set $fext 'jpg';
>     if (!-f '/home/develop/image_library/dump/$1-s.jpg'){
>       set $fext png;
>     }
>     alias '/home/develop/image_library/dump/$1-s.$fext' ;
>   }
> }
> 
[...]

server {
  listen 80;
  error_log /var/log/nginx/error_imglib.log debug;
  root /home/develop/image_library;

  location ~* ^/img/small/(\d+)/?$ {
    try_files /dump/$1-s.jpg /dump/$1-s.png =404
  }
}

wbr, Valentin V. Bartenev


From mdounin at mdounin.ru  Wed Nov  2 13:51:06 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Wed, 2 Nov 2011 17:51:06 +0400
Subject: nginx parse var in if file exist statement error?
In-Reply-To: <c33a9d760575d49a6861fc2dc02bc123.NginxMailingListEnglish@forum.nginx.org>
References: <c33a9d760575d49a6861fc2dc02bc123.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20111102135106.GR95664@mdounin.ru>

Hello!

On Wed, Nov 02, 2011 at 06:09:59AM -0400, est wrote:

[...]

> So clearly somehow, nginx managed to mess HTTP headers into filename
> parser.
> 
> Can anyone help me? This is the weirdest bug I have ever encountered
> with nginx.

Valentin already replied how to do this properly.  As for the bug 
itself, it's documented here:

http://wiki.nginx.org/IfIsEvil

Maxim Dounin


From phil at pricom.com.au  Wed Nov  2 13:59:24 2011
From: phil at pricom.com.au (Philip Rhoades)
Date: Thu, 03 Nov 2011 00:59:24 +1100
Subject: Nginx + RoundCubeMail + SSL ?
Message-ID: <40a725e8b5d18b56d51c9bb938b71792@pricom.com.au>

People,

Does anyone have a working setup with this combination?  If so, could I 
see the nginx.conf (ssl.conf) file? (I have tried all sorts of Google 
solutions with no success).

Thanks,

Phil.
-- 
Philip Rhoades

GPO Box 3411
Sydney NSW	2001
Australia
E-mail:  phil at pricom.com.au


From ml-nginx at zu-con.org  Wed Nov  2 16:54:24 2011
From: ml-nginx at zu-con.org (Matthias Rieber)
Date: Wed, 02 Nov 2011 17:54:24 +0100
Subject: large_client_header_buffers - meaning of number
Message-ID: <4EB175C0.8060504@zu-con.org>

Hi,

the doc says: large_client_header_buffers number size

What's the meaning of number? At most number of large headers can be
'processed' at once? Per request or globally? If yes, what happens to
the request that exceeds that limit?

Matthias


From mdounin at mdounin.ru  Wed Nov  2 17:57:37 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Wed, 2 Nov 2011 21:57:37 +0400
Subject: large_client_header_buffers - meaning of number
In-Reply-To: <4EB175C0.8060504@zu-con.org>
References: <4EB175C0.8060504@zu-con.org>
Message-ID: <20111102175737.GV95664@mdounin.ru>

Hello!

On Wed, Nov 02, 2011 at 05:54:24PM +0100, Matthias Rieber wrote:

> Hi,
> 
> the doc says: large_client_header_buffers number size

You probably mean wiki.

> What's the meaning of number? At most number of large headers can be
> 'processed' at once? Per request or globally? If yes, what happens to
> the request that exceeds that limit?

Docs may be found here:
http://nginx.org/en/docs/http/ngx_http_core_module.html#large_client_header_buffers

I believe the description is quite complete and answers all of 
your questions.

Maxim Dounin


From al-nginx at none.at  Wed Nov  2 19:10:12 2011
From: al-nginx at none.at (Aleksandar Lazic)
Date: Wed, 02 Nov 2011 20:10:12 +0100
Subject: Nginx + RoundCubeMail + SSL ?
In-Reply-To: <40a725e8b5d18b56d51c9bb938b71792@pricom.com.au>
References: <40a725e8b5d18b56d51c9bb938b71792@pricom.com.au>
Message-ID: <2b62d3ebdbe92614df34890fd7718e7b@none.at>

Hi Philip,

On 02.11.2011 14:59, Philip Rhoades wrote:
> People,
>
> Does anyone have a working setup with this combination?  If so, could
> I see the nginx.conf (ssl.conf) file? (I have tried all sorts of
> Google solutions with no success).

What's the problem.

it's pretty straight forward.

1.) php setup

###
         location ~ \.php {
             include        fastcgi_params;
             fastcgi_split_path_info ^(.+\.php)(.*)$;
             fastcgi_pass   php; # => upstream php { ... }
             fastcgi_index  index.php;
             fastcgi_param  SCRIPT_FILENAME  
/installed$fastcgi_script_name;
             fastcgi_param PATH_INFO $fastcgi_path_info;
             fastcgi_param PATH_TRANSLATED 
$document_root$fastcgi_path_info;
         }

###

2.) location setup.

###
location /roundcube{
             alias /installed/roundcube;
             index index.php;
             try_files $uri $uri/ index.php;
         }
###

BR
Aleks


From nginx-forum at nginx.us  Wed Nov  2 21:43:22 2011
From: nginx-forum at nginx.us (firestorm)
Date: Wed, 02 Nov 2011 17:43:22 -0400
Subject: Problem with fastcgi cache
In-Reply-To: <4974ddde02c516685929d60a0ee705f9.NginxMailingListEnglish@forum.nginx.org>
References: <4974ddde02c516685929d60a0ee705f9.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <997d175cf15b53661ade07b51a4be200.NginxMailingListEnglish@forum.nginx.org>

Problem solved. The solution was include the statements:

fastcgi_pass_header Cookie;
fastcgi_ignore_headers Cache-Control Expires Set-Cookie;

I added the configuration to log the cache request result (HIT or
MISS):

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

    access_log  logs/access.log  cache;


And now you can see in the log:

10.35.9.129 - - [02/Nov/2011:12:41:00 -0400] "GET
/administration.php/school HTTP/1.1" 200 HIT 2712
"http://10.128.50.101/administration.php/school" "Mozilla/5.0 (Windows
NT 5.1; rv:7.0) Gecko/20100101 Firefox/7.0" "-"

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,217538,217725#msg-217725


From nginx-forum at nginx.us  Thu Nov  3 01:24:08 2011
From: nginx-forum at nginx.us (est)
Date: Wed, 02 Nov 2011 21:24:08 -0400
Subject: nginx parse var in if file exist statement error?
In-Reply-To: <201111021734.44906.i@vbart.ru>
References: <201111021734.44906.i@vbart.ru>
Message-ID: <8ba8f7a6b724aefc68b0b7d469c45ef4.NginxMailingListEnglish@forum.nginx.org>

Oh thanks very much guys. try_files worked well. I tried using try_files
before, but mistakenly wrote it as something like this

try_files dump/$1-s.jpg dump/$1-s.png 404;

The correct working line should be like this

try_files /dump/$1-s.jpg /dump/$1-s.png =404;

And I didn't expect IF is evil. Suprise!

Thanks again.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,217683,217730#msg-217730


From delta.yeh at gmail.com  Thu Nov  3 02:49:52 2011
From: delta.yeh at gmail.com (Delta Yeh)
Date: Thu, 3 Nov 2011 10:49:52 +0800
Subject: how to set the proxy buffer if web server return large cookie via
	Set-Cookie
Message-ID: <CADKMG51Cvp8Nd_xwvhmSA_=FydtzM=Qiyc=Q8+AHq3-4JVmzAQ@mail.gmail.com>

Hi,

  According to http://nginx.org/en/docs/http/ngx_http_core_module.html#large_client_header_buffers,
  for long uri or large cookie from browser, I can use
large_client_header_buffers  .
  For the large cookie from web server via Set-Cookie, how to set proxy buffer?

  Should I set proxy_buffer_size  the same size  as large_client_header_buffers?
  Or set the proxy_buffers  the same as   large_client_header_buffers?

  For example, there is a 32k cookie
    location / {
    large_client_header_buffers  4 32k;
    proxy_buffer_size  32k;
   proxy_pass ....
}

  or

 location / {
    large_client_header_buffers  4 32k;
    proxy_buffers  4 32k;

   proxy_pass ....
}


According to the WIKI, I prefer proxy_buffers directive.

BR,

DeltaY


From nginx-forum at nginx.us  Thu Nov  3 02:59:45 2011
From: nginx-forum at nginx.us (wangbin579)
Date: Wed, 02 Nov 2011 22:59:45 -0400
Subject: nginx_hmux_module - support hmux protocol proxy with Nginx
In-Reply-To: <2d7909af72c23eb07f1385d88d6ae095.NginxMailingListEnglish@forum.nginx.org>
References: <2d7909af72c23eb07f1385d88d6ae095.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <2ac5444cde8a1d83cb20abb48b3593e3.NginxMailingListEnglish@forum.nginx.org>

source code from mod_caucho.c (apache module) and resin source code

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,217681,217733#msg-217733


From nginx-forum at nginx.us  Thu Nov  3 03:03:02 2011
From: nginx-forum at nginx.us (wangbin579)
Date: Wed, 02 Nov 2011 23:03:02 -0400
Subject: Tcpcopy,an online request replication tool fit for nginx
In-Reply-To: <141f5701aa1bdc50e4d7a29c237508a3.NginxMailingListEnglish@forum.nginx.org>
References: <141f5701aa1bdc50e4d7a29c237508a3.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <6f2273e5254cfc91047a0e409d84b19c.NginxMailingListEnglish@forum.nginx.org>

An example:
    Suppose 13 and 14 are online machines, 148 is a target machine which
is similar to 
    the online machines, and 12321 is used both as local port and remote
port. We 
    use tcpcopy to test if 148 can endure two times of current online
stress.
    
    Using tcpcopy to perform the above test task.
    the target machine(148)
       # modprobe ip_queue (if not run up)
       # iptables -I OUTPUT -p tcp --sport 12321 -j QUEUE (if not set)
       # ./interception
    online machine(13):
       # ./tcpcopy xx.xx.xx.13 12321 xx.xx.xx.148 12321
    online machine(14):
       # ./tcpcopy xx.xx.xx.14 12321 xx.xx.xx.148 12321

    Cpu load and memory usage are as follows:
        13 cpu:
           11124 adrun 15 0 193m 146m 744 S 18.6 7.3 495:31.56
asyn_server
           11281 root 15 0 65144 40m 1076 S 12.3 2.0 0:47.89 tcpcopy
        14 cpu:
           16855 adrun 15 0 98.7m 55m 744 S 21.6 2.7 487:49.51
asyn_server
           16429 root 15 0 41156 17m 1076 S 14.0 0.9 0:33.63 tcpcopy
        148 cpu :
           25609 root 15 0 76892 59m 764 S 49.6 2.9 63:03.14
asyn_server
           20184 root 15 0 5624 4232 292 S 17.0 0.2 0:52.82
interception

    Access log analysis:
        13 online machine: grep 'Tue 11:08' access_0913_11.log |wc -l
:89316,  1489 reqs/sec
        14 online machine: grep 'Tue 11:08' access_0913_11.log |wc -l
:89309,  1488 reqs/sec
        148 test  machine: grep 'Tue 11:08' access_0913_11.log |wc -l
:178175, 2969 reqs/sec
        request loss rate:
           (89316+89309-178175)/(89316+89309)=0.25%

    From the above, we can see that the target machine can endure two
times of current 
    online stress.
    
    What about the cpu load ?
    tcpcopy on online machine 13 occupies 12.3% of cpu load, tcpcopy on
online 14 
    occupies 14% and interception on target machine 148 occupies 17%. We
can see 
    that the cpu load is very low here, and so is the memory usage.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,217680,217734#msg-217734


From nginx-forum at nginx.us  Thu Nov  3 03:07:32 2011
From: nginx-forum at nginx.us (wangbin579)
Date: Wed, 02 Nov 2011 23:07:32 -0400
Subject: Tcpcopy,an online request replication tool fit for nginx
In-Reply-To: <141f5701aa1bdc50e4d7a29c237508a3.NginxMailingListEnglish@forum.nginx.org>
References: <141f5701aa1bdc50e4d7a29c237508a3.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <f54568cc41662288ce2f17ebca291a41.NginxMailingListEnglish@forum.nginx.org>

a stress test for nginx  written in chinese.

http://blog.csdn.net/wangbin579/article/details/6929495

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,217680,217735#msg-217735


From phil at pricom.com.au  Thu Nov  3 05:17:52 2011
From: phil at pricom.com.au (Philip Rhoades)
Date: Thu, 03 Nov 2011 16:17:52 +1100
Subject: Getting my own posts but NOT replies?
Message-ID: <7990b6f18be850ab94f4d37ee4f0a868@pricom.com.au>

People,

I am getting my own posts to this list but NOT any replies (I have to 
look them up on the nginx email archive pages).  I have checked out my 
preferences and everything looks OK - anyone got any ideas why this 
might be happening?

Thanks,

Phil.
-- 
Philip Rhoades

GPO Box 3411
Sydney NSW	2001
Australia
E-mail:  phil at pricom.com.au


From nginx-forum at nginx.us  Thu Nov  3 07:06:35 2011
From: nginx-forum at nginx.us (forum_id)
Date: Thu, 03 Nov 2011 03:06:35 -0400
Subject: HTTP Request filter module
Message-ID: <d827925076d1841e37f46b3add56084f.NginxMailingListEnglish@forum.nginx.org>

I have a requirement for HTTP request filter. Whenever I receive a HTTP
request this filter module should read the request and does some
internal statistical work, it doesn't manipulate the request page nor
creates response page. It just need to peep into request that's it.
After this filter module, nginx HTTP server should continue to process
the request and serve the response to user. Can we do it without a proxy
handler, because nginx serves the responses directly here.

Please help me.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,217739,217739#msg-217739


From matt.starcrest at yahoo.com  Thu Nov  3 07:08:49 2011
From: matt.starcrest at yahoo.com (Matt Starcrest)
Date: Thu, 3 Nov 2011 00:08:49 -0700 (PDT)
Subject: python wsgi behind nginx:  yield not behaving as expected
Message-ID: <1320304129.63380.YahooMailNeo@web46009.mail.sp1.yahoo.com>

Hi,
I want to run wsgi python code on a web server behind nginx. ?The server needs to respond quickly to an http request, then continue to do some (slow) work after responding. ?Python's yield statement seems to fit the bill, as follows:

def application(environ, start_response):
? ? output = get_response_quickly(environ)

? ? start_response('200 OK',?[('Content-type', 'text/plain'),?('Content-Length', str(len(output)))])

? ? yield output
? ? do_slow_work()

If I run this wsgi in a standalone python server (I tried wsgiref, fapws, uwsgi), the caller receives the output immediately (after get_response_quickly()), as desired. ?Great. ?However, if I run any of these servers behind nginx, the caller doesn't receive a response until *after* do_slow_work() -- thus defeating the purpose.

Is there a way to make this pattern work with nginx? ?Or is there a better way in general to respond quickly but continue work, without manually creating python threads / other clumsiness?

Thanks,
Matt
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20111103/80b81e19/attachment.html>

From igor at sysoev.ru  Thu Nov  3 07:12:47 2011
From: igor at sysoev.ru (Igor Sysoev)
Date: Thu, 3 Nov 2011 11:12:47 +0400
Subject: python wsgi behind nginx:  yield not behaving as expected
In-Reply-To: <1320304129.63380.YahooMailNeo@web46009.mail.sp1.yahoo.com>
References: <1320304129.63380.YahooMailNeo@web46009.mail.sp1.yahoo.com>
Message-ID: <20111103071247.GB37071@nginx.com>

On Thu, Nov 03, 2011 at 12:08:49AM -0700, Matt Starcrest wrote:
> Hi,
> I want to run wsgi python code on a web server behind nginx. ?The server needs to respond quickly to an http request, then continue to do some (slow) work after responding. ?Python's yield statement seems to fit the bill, as follows:
> 
> def application(environ, start_response):
> ? ? output = get_response_quickly(environ)
> 
> ? ? start_response('200 OK',?[('Content-type', 'text/plain'),?('Content-Length', str(len(output)))])
> 
> ? ? yield output
> ? ? do_slow_work()
> 
> If I run this wsgi in a standalone python server (I tried wsgiref, fapws, uwsgi), the caller receives the output immediately (after get_response_quickly()), as desired. ?Great. ?However, if I run any of these servers behind nginx, the caller doesn't receive a response until *after* do_slow_work() -- thus defeating the purpose.
> 
> Is there a way to make this pattern work with nginx? ?Or is there a better way in general to respond quickly but continue work, without manually creating python threads / other clumsiness?

In nginx-1.0.9 or nginx-1.1.5 you can try
   uwsgi_buffering  off;
or 
   sgi_buffering  off;
depending on protocol. In any modern enough nginx version you can use
    proxy_buffering off;

There is no way currently to disable buffering of FastCGI servers.


-- 
Igor Sysoev


From mdounin at mdounin.ru  Thu Nov  3 07:44:43 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Thu, 3 Nov 2011 11:44:43 +0400
Subject: how to set the proxy buffer if web server return large cookie via
	Set-Cookie
In-Reply-To: <CADKMG51Cvp8Nd_xwvhmSA_=FydtzM=Qiyc=Q8+AHq3-4JVmzAQ@mail.gmail.com>
References: <CADKMG51Cvp8Nd_xwvhmSA_=FydtzM=Qiyc=Q8+AHq3-4JVmzAQ@mail.gmail.com>
Message-ID: <20111103074443.GA95664@mdounin.ru>

Hello!

On Thu, Nov 03, 2011 at 10:49:52AM +0800, Delta Yeh wrote:

> Hi,
> 
>   According to http://nginx.org/en/docs/http/ngx_http_core_module.html#large_client_header_buffers,
>   for long uri or large cookie from browser, I can use
> large_client_header_buffers  .
>   For the large cookie from web server via Set-Cookie, how to set proxy buffer?
> 
>   Should I set proxy_buffer_size  the same size  as large_client_header_buffers?
>   Or set the proxy_buffers  the same as   large_client_header_buffers?
> 
>   For example, there is a 32k cookie
>     location / {
>     large_client_header_buffers  4 32k;
>     proxy_buffer_size  32k;
>    proxy_pass ....
> }
> 
>   or
> 
>  location / {
>     large_client_header_buffers  4 32k;
>     proxy_buffers  4 32k;
> 
>    proxy_pass ....
> }
> 
> 
> According to the WIKI, I prefer proxy_buffers directive.

Response headers from upstream have to fit into proxy_buffer_size 
buffer.  The proxy_buffers directive is used when reading response 
body from upstream, not headers.

Maxim Dounin


From matt.starcrest at yahoo.com  Thu Nov  3 07:48:30 2011
From: matt.starcrest at yahoo.com (Matt Starcrest)
Date: Thu, 3 Nov 2011 00:48:30 -0700 (PDT)
Subject: python wsgi behind nginx:  yield not behaving as expected
In-Reply-To: <20111103071247.GB37071@nginx.com>
References: <1320304129.63380.YahooMailNeo@web46009.mail.sp1.yahoo.com>
	<20111103071247.GB37071@nginx.com>
Message-ID: <1320306510.61261.YahooMailNeo@web46005.mail.sp1.yahoo.com>

Looks like that solved it! ?Thanks Igor!


________________________________
From: Igor Sysoev <igor at sysoev.ru>
To: nginx at nginx.org
Sent: Thursday, November 3, 2011 12:12 AM
Subject: Re: python wsgi behind nginx:  yield not behaving as expected

On Thu, Nov 03, 2011 at 12:08:49AM -0700, Matt Starcrest wrote:
> Hi,
> I want to run wsgi python code on a web server behind nginx. ?The server needs to respond quickly to an http request, then continue to do some (slow) work after responding. ?Python's yield statement seems to fit the bill, as follows:
> 
> def application(environ, start_response):
> ? ? output = get_response_quickly(environ)
> 
> ? ? start_response('200 OK',?[('Content-type', 'text/plain'),?('Content-Length', str(len(output)))])
> 
> ? ? yield output
> ? ? do_slow_work()
> 
> If I run this wsgi in a standalone python server (I tried wsgiref, fapws, uwsgi), the caller receives the output immediately (after get_response_quickly()), as desired. ?Great. ?However, if I run any of these servers behind nginx, the caller doesn't receive a response until *after* do_slow_work() -- thus defeating the purpose.
> 
> Is there a way to make this pattern work with nginx? ?Or is there a better way in general to respond quickly but continue work, without manually creating python threads / other clumsiness?

In nginx-1.0.9 or nginx-1.1.5 you can try
?  uwsgi_buffering? off;
or 
?  sgi_buffering? off;
depending on protocol. In any modern enough nginx version you can use
? ? proxy_buffering off;

There is no way currently to disable buffering of FastCGI servers.


-- 
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/20111103/3c3551ce/attachment.html>

From delta.yeh at gmail.com  Thu Nov  3 07:51:07 2011
From: delta.yeh at gmail.com (Delta Yeh)
Date: Thu, 3 Nov 2011 15:51:07 +0800
Subject: how to set the proxy buffer if web server return large cookie via
	Set-Cookie
In-Reply-To: <20111103074443.GA95664@mdounin.ru>
References: <CADKMG51Cvp8Nd_xwvhmSA_=FydtzM=Qiyc=Q8+AHq3-4JVmzAQ@mail.gmail.com>
	<20111103074443.GA95664@mdounin.ru>
Message-ID: <CADKMG51yDrj-WVqTP8mAdqOZjOjC++pDO8+2nvkW2gKLEe5=gg@mail.gmail.com>

So to resolve large cookie issue, the config should be:

    location / {
     large_client_header_buffers  4 32k;
     proxy_buffer_size  32k;
    proxy_pass ....
 }

For  response, default proxy_buffers should be OK for most web application.


2011/11/3 Maxim Dounin <mdounin at mdounin.ru>:
> Hello!
>
> On Thu, Nov 03, 2011 at 10:49:52AM +0800, Delta Yeh wrote:
>
>> Hi,
>>
>> ? According to http://nginx.org/en/docs/http/ngx_http_core_module.html#large_client_header_buffers,
>> ? for long uri or large cookie from browser, I can use
>> large_client_header_buffers ?.
>> ? For the large cookie from web server via Set-Cookie, how to set proxy buffer?
>>
>> ? Should I set proxy_buffer_size ?the same size ?as large_client_header_buffers?
>> ? Or set the proxy_buffers ?the same as ? large_client_header_buffers?
>>
>> ? For example, there is a 32k cookie
>> ? ? location / {
>> ? ? large_client_header_buffers ?4 32k;
>> ? ? proxy_buffer_size ?32k;
>> ? ?proxy_pass ....
>> }
>>
>> ? or
>>
>> ?location / {
>> ? ? large_client_header_buffers ?4 32k;
>> ? ? proxy_buffers ?4 32k;
>>
>> ? ?proxy_pass ....
>> }
>>
>>
>> According to the WIKI, I prefer proxy_buffers directive.
>
> Response headers from upstream have to fit into proxy_buffer_size
> buffer. ?The proxy_buffers directive is used when reading response
> body from upstream, not headers.
>
> Maxim Dounin
>


From igor at sysoev.ru  Thu Nov  3 07:51:55 2011
From: igor at sysoev.ru (Igor Sysoev)
Date: Thu, 3 Nov 2011 11:51:55 +0400
Subject: python wsgi behind nginx:  yield not behaving as expected
In-Reply-To: <1320306510.61261.YahooMailNeo@web46005.mail.sp1.yahoo.com>
References: <1320304129.63380.YahooMailNeo@web46009.mail.sp1.yahoo.com>
	<20111103071247.GB37071@nginx.com>
	<1320306510.61261.YahooMailNeo@web46005.mail.sp1.yahoo.com>
Message-ID: <20111103075154.GB37679@nginx.com>

On Thu, Nov 03, 2011 at 12:48:30AM -0700, Matt Starcrest wrote:
> Looks like that solved it! ?Thanks Igor!

Please note, that any "..._buffering off" disables caching,
so you can set it exactly on locations which should be cached:

   location /some/page {
       ..._buffering off;
   }

> From: Igor Sysoev <igor at sysoev.ru>
> To: nginx at nginx.org
> Sent: Thursday, November 3, 2011 12:12 AM
> Subject: Re: python wsgi behind nginx:  yield not behaving as expected
> 
> On Thu, Nov 03, 2011 at 12:08:49AM -0700, Matt Starcrest wrote:
> > Hi,
> > I want to run wsgi python code on a web server behind nginx. ?The server needs to respond quickly to an http request, then continue to do some (slow) work after responding. ?Python's yield statement seems to fit the bill, as follows:
> > 
> > def application(environ, start_response):
> > ? ? output = get_response_quickly(environ)
> > 
> > ? ? start_response('200 OK',?[('Content-type', 'text/plain'),?('Content-Length', str(len(output)))])
> > 
> > ? ? yield output
> > ? ? do_slow_work()
> > 
> > If I run this wsgi in a standalone python server (I tried wsgiref, fapws, uwsgi), the caller receives the output immediately (after get_response_quickly()), as desired. ?Great. ?However, if I run any of these servers behind nginx, the caller doesn't receive a response until *after* do_slow_work() -- thus defeating the purpose.
> > 
> > Is there a way to make this pattern work with nginx? ?Or is there a better way in general to respond quickly but continue work, without manually creating python threads / other clumsiness?
> 
> In nginx-1.0.9 or nginx-1.1.5 you can try
> ?  uwsgi_buffering? off;
> or 
> ?  sgi_buffering? off;
> depending on protocol. In any modern enough nginx version you can use
> ? ? proxy_buffering off;
> 
> There is no way currently to disable buffering of FastCGI servers.
> 
> 
> -- 
> 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


-- 
Igor Sysoev


From quintinpar at gmail.com  Thu Nov  3 09:17:34 2011
From: quintinpar at gmail.com (Quintin Par)
Date: Thu, 3 Nov 2011 14:47:34 +0530
Subject: Nginx & long poll: Best practices to reduce memory and bandwidth
	footprint
Message-ID: <CACW9FdPYnJ_TFGJkv202aqaWFpfA-L18-w52futK7Bb0XMi97g@mail.gmail.com>

Can someone please answer this question on serverfault.com?

http://serverfault.com/questions/327301/nginx-long-poll-best-practices-to-reduce-memory-and-bandwidth-footprint


-Quintin
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20111103/736ef33a/attachment.html>

From agentzh at gmail.com  Thu Nov  3 09:34:09 2011
From: agentzh at gmail.com (agentzh)
Date: Thu, 3 Nov 2011 17:34:09 +0800
Subject: [ANN] ngx_openresty 1.0.8.26 (stable) released
In-Reply-To: <CAB4Tn6M0C8BAXRJwoWVBFyf=ybXVLusCFA9k0WwXDBADUgGMWw@mail.gmail.com>
References: <CAB4Tn6M0C8BAXRJwoWVBFyf=ybXVLusCFA9k0WwXDBADUgGMWw@mail.gmail.com>
Message-ID: <CAB4Tn6PqPgH6aeu9A-vDq2T2b30HXrRDePe3j8jKSuCg=bydBA@mail.gmail.com>

Hi, folks!

I'm happy to announce that the new stable release of ngx_openresty,
1.0.8.26, has just been kicked out of door:

  http://openresty.org/#Download

This is the first stable release of ngx_openresty that is based on the
Nginx core 1.0.8. And this is a big release with a *lot* of bug fixes and
new features.

Special thanks go to all our contributors and users to help make this
happen over the last month :)

Here goes the complete change log for this release, as compared to the last
stable release, 1.0.6.22, released nearly a month ago:

   - upgraded the Nginx core to 1.0.8.
   - upgraded LuaNginxModule to 0.3.1rc23.
      - feature: added new directive lua_shared_dict:
      http://wiki.nginx.org/HttpLuaModule#lua_shared_dict
      - feature: added Lua API for the shm-based dictionary:
      http://wiki.nginx.org/HttpLuaModule#ngx.shared.DICT
      - feature: now we apply the patch to the nginx core so as to allow
      main request body modifications.
      - feature: added new Lua API ngx.req.set_body_file():
      http://wiki.nginx.org/HttpLuaModule#ngx.req.set_body_file
      - feature: added new Lua API ngx.req.set_body_data():
      http://wiki.nginx.org/HttpLuaModule#ngx.req.set_body_data
      - feature: added new Lua functions ngx.req.read_body(),
      ngx.req.discard_body(), ngx.req.get_body_data(), and
      ngx.req.get_body_file(). see the docs here:
      http://wiki.nginx.org/HttpLuaModule#ngx.req.read_body
      - feature: now we implemented ngx.req.set_uri() and
      ngx.req.set_uri_args() to emulate ngx_rewrite's rewrite directive
      (without redirect or permanent modifiers). thanks Vladimir Protasov
      (utros) and Nginx User.
      - feature: added constant ngx.HTTP_METHOD_NOT_IMPLEMENTED (501).
      thanks Nginx User.
      - feature: now for HTTP 1.0 requests, we disable the automatic full
      buffering mode if the user sets the Content-Length response header
      before sending out the headers. this allows streaming output for HTTP 1.0
      requests if the content length can be calculated beforehand.
thanks Li Ziyi.
      - bugfix: now we properly support setting the Cache-Control response
      header via the ngx.header.HEADER interface.
      - bugfix: no longer set header hash to 1. use the ngx_hash_key_lcinstead.
      - bugfix: now we skip rewrite phase Lua handlers altogether if
      ngx_rewrite's rewrite directive issue a location re-lookup by
      changing URIs (but not including rewrite ... break). thanks Nginx User.
      - bugfix: fixed hanging issues when using ngx.exec() within
      rewrite_by_lua and access_by_lua. thanks Nginx User for reporting it.
      - bugfix: lua_need_request_body should not skip requests with methods
      other than POST and PUT. thanks Nginx User.
      - bugfix: ndk.set_var.DIRECTIVE had a memory issue and might pass
      empty argument values to the directive being called. thanks dannynoonan.
      - bugfix: no longer free request body buffers that are not allocated
      by ourselves.
      - bugfix: now we allow setting ngx.var.VARIABLE to nil.
      - bugfix: now we explicitly clear all the modules' contexts before
      dump to named location with ngx.exec. thanks Nginx User.
   - upgraded EchoNginxModule to 0.37rc7.
      - bugfix: fixed a memory issue in both echo_sleep and
      echo_blocking_sleep: we should not pass ngx_str_t strings to
atof()which expects C strings.
      - bugfix: now we explicitly clear all the modules' contexts before
      dump to named location with echo_exec.
      - bugfix: bugfix: echo_exec may hang when running after
echo_sleep(or other I/O interruption calls): we should have called
      ngx_http_finalize_request on NGX_DONE to decrement r->main->countanyway.
      - bugfix: now we properly set the Content-Length request header for
      subrequests.
   - upgraded SrcacheNginxModule to 0.13rc2.
      - feature: implemented response status line and general response
      header cachin and added new directives srcache_store_hide_header and
      srcache_store_pass_header to control which headers to cache and which
      not.
      - feature: added new directive srcache_response_cache_control to
      control whether honor response headers Cache-Control and Expires,
      default on.
      - feature: we disable srcache_store automatically by default
when Cache-Control:
      max-age=0 and Expires: <date no more recently than now> are seen.
      - feature: implemented builtin nginx variable $srcache_expire for
      automatic expiration time calculation based on response headers
      Cache-Control (max-age) and Expires; also added new directives
      srcache_max_expire and srcache_default_expire.
      - feature: implemented the srcache_store_no_cache directive; now by
      default, we do not store responses with the header Cache-Control:
      no-cache into the cache.
      - feature: implemented the srcache_store_no_store directive (default
      off). Now by default, responses with the header Cache-Control:
      no-store will not be stored into the cache.
      - feature: implemented the srcache_store_private directive to control
      whether to store responses with the header Cache-Control: private.
      - feature: implemented the srcache_request_cache_control directive to
      allow request headers Cache-Control: no-cache or Pragma: no-cache to
      force bypassing cache lookup. it also honors the request header
Cache-Control:
      no-store. this directive is turned off by default.
      - feature: now we check response header Content-Encoding by default
      and a non-empty header value will skip srcache_store; also introduced
      a new directive named srcache_ignore_content_encoding to ignore this
      response header.
      - feature: implemented the srcache_methods directive to specify
      request methods that are cacheable, by default, only GET and HEAD are
      cacheable.
      - bugfix: we no longer set header hash to 1; we use
ngx_hash_key_lcinstead.
      - bugfix: when we skip srcache_fetch by means of srcache_fetch_skip,
      we should not automatically skip srcache_store.
      - bugfix: now we ignore the Content-Length header (if any) of the
      main request for the subrequests.
      - bugfix: there might be a segfault when failing to allocate memory
      in ngx_http_srcache_add_copy_chain. thanks Shaun savage.
      - feature: implemented new directive srcache_store_statuses to allow
      the user to specify the response status code list that is to be
stored into
      the cache.
      - bugfix: we now only cache 200, 301, and 302 responses by default.
   - upgraded IconvNginxModule to 0.10rc5.
      - bugfix: fixed -Wset-but-not-used warnings issued by gcc 4.6.0.
      thanks Zhi Jiale (Calio).
   - upgraded HeadersMoreNginxModule to 0.16rc3.
      - bugfix: we should set header hash using ngx_hash_key_lc, not simply
      to 1.
      - bugfix: fixed setting Cache-Control response headers. we should
      properly prepare the r->cache_control array as well.
   - upgraded RdsJsonNginxModule to 0.12rc6.
      - bugfix: fixed compatibility with nginx 1.1.4+.
   - upgraded RdsCsvNginxModule to 0.04.
      - bugfix: fixed compatibility issues with nginx 1.1.4+.
      - optimization: now we only register our filters when rds_csv is
      actually used in nginx.conf.
   - upgraded Redis2NginxModule to 0.08rc1.
      - bugfix: fixed compatibility with nginx 1.1.4+.
   - upgraded DrizzleNginxModule to 0.1.2rc2.
      - bugfix: fixed compatibility with nginx 1.1.4+
   - upgraded MemcNginxModule to 0.13rc1.
      - bugfix: fixed compatibility with nginx 1.1.4+.
   - upgraded SetMiscNginxModule to v0.22rc3.
      - minor code cleanup.
   - applied the patch to the Nginx core that always clears all modules'
   contexts in ngx_http_named_location.
   - applied the patch for the variable-header-ignore-no-hash issue. see
   http://forum.nginx.org/read.php?29,216062 for details.

As always, you're welcome to report bugs and feature requests either here
or directly to me :)

It'll also be highly appreciated to try out the devel releases (based on
the Nginx core 1.0.9+) that are coming out later ;)

OpenResty (aka. ngx_openresty) is a full-fledged web application server by
bundling the standard Nginx core, lots of 3rd-party Nginx
modules<http://wiki.nginx.org/3rdPartyModules>,
as well as most of their external dependencies.

By taking adantage of various well-designed Nginx modules, OpenResty
effectively turns the nginx server into a powerful web app server, in which
the web developers can use the Lua programming language to script various
existing nginx C modules and Lua modules and construct extremely
high-performance web applications that is capable to handle 10K+
connections.

OpenResty aims to run your server-side web app completely in the Nginx
server, leveraging Nginx's event model to do non-blocking I/O not only with
the HTTP clients, but also with remote backends like MySQL, PostgreSQL,
Memcached, and Redis.

You can find more details on the homepage of ngx_openresty here:

    http://openresty.org

Have fun!
-agentzh
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20111103/8d74cc64/attachment-0001.html>

From ft at falkotimme.com  Thu Nov  3 09:43:06 2011
From: ft at falkotimme.com (Falko Timme)
Date: Thu, 3 Nov 2011 10:43:06 +0100
Subject: ISPConfig 3 now with full nginx support
Message-ID: <AC318A041118483EAD34D4EE1ED22991@notebook>

Hi,

ISPConfig 3 is an open source hosting control panel for Linux which is
capable of managing multiple servers from one control panel. ISPConfig is
licensed under the BSD license. Its feature list is here:
http://www.ispconfig.org/ispconfig-3/

Today we have released ISPConfig 3.0.4, and one of the new features is full
support for nginx, i.e., ISPConfig can now create and manage websites on an
nginx server. It has support for SSL (SNI is possible as well),
rewrites/redirects, CGI (through fcgiwrap), PHP-FPM (both TCP connections
and sockets are supported), custom php.ini settings, custom nginx
directives, basic http authentication, subdomains, alias domains, IPv6, etc.

You can find the full announcement on
http://www.ispconfig.org/releases/ispconfig-3-0-4-released/

ISPConfig can be downloaded from
http://www.ispconfig.org/ispconfig-3/download/

Best Regards,

Falko Timme
ISPConfig Team


From lists at ruby-forum.com  Thu Nov  3 09:46:14 2011
From: lists at ruby-forum.com (Noah C.)
Date: Thu, 03 Nov 2011 10:46:14 +0100
Subject: DNS TTLs being ignored
In-Reply-To: <CAAqHd6v_LxnOpfoE6Fq+sKebtwHiguse4d6GmmvpAUayR5eqaA@mail.gmail.com>
References: <CAAqHd6v_LxnOpfoE6Fq+sKebtwHiguse4d6GmmvpAUayR5eqaA@mail.gmail.com>
Message-ID: <d7ffe70b7f12b278cb9ffeac1a02c358@ruby-forum.com>

Thanks for the reply Andrew. Do you have any idea when it's likely to be 
generally available? This is a pretty big nuisance for us, and I'd like 
to be able to figure out if I need to look at using a new reverse proxy, 
at least for the time being.

--Noah

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


From andrew at nginx.com  Thu Nov  3 09:50:58 2011
From: andrew at nginx.com (Andrew Alexeev)
Date: Thu, 3 Nov 2011 13:50:58 +0400
Subject: DNS TTLs being ignored
In-Reply-To: <d7ffe70b7f12b278cb9ffeac1a02c358@ruby-forum.com>
References: <CAAqHd6v_LxnOpfoE6Fq+sKebtwHiguse4d6GmmvpAUayR5eqaA@mail.gmail.com>
	<d7ffe70b7f12b278cb9ffeac1a02c358@ruby-forum.com>
Message-ID: <AB401AA3-211C-4401-B050-712D7CBBB054@nginx.com>

Noah,

This fix/improvement be introduced in 1.1.8 which will come out around Nov 14.

Hope this helps

On Nov 3, 2011, at 1:46 PM, Noah C. wrote:

> Thanks for the reply Andrew. Do you have any idea when it's likely to be 
> generally available? This is a pretty big nuisance for us, and I'd like 
> to be able to figure out if I need to look at using a new reverse proxy, 
> at least for the time being.
> 
> --Noah
> 
> -- 
> Posted via http://www.ruby-forum.com/.
> 
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx



From calin.don at gmail.com  Thu Nov  3 10:41:56 2011
From: calin.don at gmail.com (Calin Don)
Date: Thu, 3 Nov 2011 12:41:56 +0200
Subject: HTTP Request filter module
In-Reply-To: <d827925076d1841e37f46b3add56084f.NginxMailingListEnglish@forum.nginx.org>
References: <d827925076d1841e37f46b3add56084f.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <CAEOe2JwU42j0_cmuZC=v5=QJ8spVvqv_4hEBnGTeeUc7H7PNWA@mail.gmail.com>

Doesn't post_action do what you need?
http://wiki.nginx.org/HttpCoreModule#post_action

On Thu, Nov 3, 2011 at 09:06, forum_id <nginx-forum at nginx.us> wrote:

> I have a requirement for HTTP request filter. Whenever I receive a HTTP
> request this filter module should read the request and does some
> internal statistical work, it doesn't manipulate the request page nor
> creates response page. It just need to peep into request that's it.
> After this filter module, nginx HTTP server should continue to process
> the request and serve the response to user. Can we do it without a proxy
> handler, because nginx serves the responses directly here.
>
> Please help me.
>
> Posted at Nginx Forum:
> http://forum.nginx.org/read.php?2,217739,217739#msg-217739
>
> _______________________________________________
> 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/20111103/72a3c223/attachment.html>

From agentzh at gmail.com  Thu Nov  3 11:00:37 2011
From: agentzh at gmail.com (agentzh)
Date: Thu, 3 Nov 2011 19:00:37 +0800
Subject: HTTP Request filter module
In-Reply-To: <d827925076d1841e37f46b3add56084f.NginxMailingListEnglish@forum.nginx.org>
References: <d827925076d1841e37f46b3add56084f.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <CAB4Tn6N1_JTXZGknZd9kMvuM4bjBCKb_nZ3f3UB2LudPj3CaDQ@mail.gmail.com>

On Thu, Nov 3, 2011 at 3:06 PM, forum_id <nginx-forum at nginx.us> wrote:
>
> I have a requirement for HTTP request filter. Whenever I receive a HTTP
> request this filter module should read the request and does some
> internal statistical work, it doesn't manipulate the request page nor
> creates response page. It just need to peep into request that's it.
> After this filter module, nginx HTTP server should continue to process
> the request and serve the response to user. Can we do it without a proxy
> handler, because nginx serves the responses directly here.
>

I think this is a perfect use case for the rewrite_by_lua or
access_by_lua directives provided by the ngx_lua module. See

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

You can read and modify the request URI, query args, or even request
bodies in your Lua code and save the statistical results in the
shm-based dictionary (shared_dict) provided by ngx_lua or issue a
subrequest from within Lua to internal locations configured by
ngx_memc or ngx_redis2 via the ngx.location.capture API.

Because LuaJIT 2.0 is so light and so fast, this setting can achieve
performance comparable with an Nginx C module and also provide
scripting flexibility :)

Regards,
-agentzh


From nginx-forum at nginx.us  Thu Nov  3 12:12:02 2011
From: nginx-forum at nginx.us (forum_id)
Date: Thu, 03 Nov 2011 08:12:02 -0400
Subject: HTTP Request filter module
In-Reply-To: <CAEOe2JwU42j0_cmuZC=v5=QJ8spVvqv_4hEBnGTeeUc7H7PNWA@mail.gmail.com>
References: <CAEOe2JwU42j0_cmuZC=v5=QJ8spVvqv_4hEBnGTeeUc7H7PNWA@mail.gmail.com>
Message-ID: <415d07b2a31d75210dfceb8ef4c979b1.NginxMailingListEnglish@forum.nginx.org>

Thank you. I am new to nginx, so spare me if I talk anything nonsense. I
want through link you provided, and then went through code as well. I
couldn't understand few things here
1) Can we override nginx core's directive?
2) Since that post action is been used in multiple place, how far it
safe to over write it?

Thank you once again.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,217739,217761#msg-217761


From nginx-forum at nginx.us  Thu Nov  3 12:13:24 2011
From: nginx-forum at nginx.us (forum_id)
Date: Thu, 03 Nov 2011 08:13:24 -0400
Subject: HTTP Request filter module
In-Reply-To: <CAB4Tn6N1_JTXZGknZd9kMvuM4bjBCKb_nZ3f3UB2LudPj3CaDQ@mail.gmail.com>
References: <CAB4Tn6N1_JTXZGknZd9kMvuM4bjBCKb_nZ3f3UB2LudPj3CaDQ@mail.gmail.com>
Message-ID: <917477f9cde52b53ea5f9cd28ea54872.NginxMailingListEnglish@forum.nginx.org>

Thank you. As per the project requirements need to develop C modules
only, so if @agentzh can suggest me in this way, that would be helpful.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,217739,217762#msg-217762


From agentzh at gmail.com  Thu Nov  3 12:54:57 2011
From: agentzh at gmail.com (agentzh)
Date: Thu, 3 Nov 2011 20:54:57 +0800
Subject: HTTP Request filter module
In-Reply-To: <917477f9cde52b53ea5f9cd28ea54872.NginxMailingListEnglish@forum.nginx.org>
References: <CAB4Tn6N1_JTXZGknZd9kMvuM4bjBCKb_nZ3f3UB2LudPj3CaDQ@mail.gmail.com>
	<917477f9cde52b53ea5f9cd28ea54872.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <CAB4Tn6MfVkLavz4Pa+FyqT1z1K6+feoXbAuFBVcsLUEaRweL9Q@mail.gmail.com>

On Thu, Nov 3, 2011 at 8:13 PM, forum_id <nginx-forum at nginx.us> wrote:
> Thank you. As per the project requirements need to develop C modules
> only, so if @agentzh can suggest me in this way, that would be helpful.
>

Just take a look at how ngx_lua does those things on the C level :)

Regards,
-agentzh


From wendal1985 at gmail.com  Thu Nov  3 13:39:45 2011
From: wendal1985 at gmail.com (Wendal Chen)
Date: Thu, 3 Nov 2011 21:39:45 +0800
Subject: [ANN] ngx_openresty 1.0.8.26 (stable) released
In-Reply-To: <CAB4Tn6PqPgH6aeu9A-vDq2T2b30HXrRDePe3j8jKSuCg=bydBA@mail.gmail.com>
References: <CAB4Tn6M0C8BAXRJwoWVBFyf=ybXVLusCFA9k0WwXDBADUgGMWw@mail.gmail.com>
	<CAB4Tn6PqPgH6aeu9A-vDq2T2b30HXrRDePe3j8jKSuCg=bydBA@mail.gmail.com>
Message-ID: <CAM79QPXmGng5_tkuLAmTPocFioEMzWUBwa6kHom=WgVSCPQf2w@mail.gmail.com>

Great!!

2011/11/3 agentzh <agentzh at gmail.com>

> Hi, folks!
>
> I'm happy to announce that the new stable release of ngx_openresty,
> 1.0.8.26, has just been kicked out of door:
>
>   http://openresty.org/#Download
>
> This is the first stable release of ngx_openresty that is based on the
> Nginx core 1.0.8. And this is a big release with a *lot* of bug fixes and
> new features.
>
> Special thanks go to all our contributors and users to help make this
> happen over the last month :)
>
> Here goes the complete change log for this release, as compared to the
> last stable release, 1.0.6.22, released nearly a month ago:
>
>    - upgraded the Nginx core to 1.0.8.
>    - upgraded LuaNginxModule to 0.3.1rc23.
>       - feature: added new directive lua_shared_dict:
>       http://wiki.nginx.org/HttpLuaModule#lua_shared_dict
>       - feature: added Lua API for the shm-based dictionary:
>       http://wiki.nginx.org/HttpLuaModule#ngx.shared.DICT
>       - feature: now we apply the patch to the nginx core so as to allow
>       main request body modifications.
>       - feature: added new Lua API ngx.req.set_body_file():
>       http://wiki.nginx.org/HttpLuaModule#ngx.req.set_body_file
>       - feature: added new Lua API ngx.req.set_body_data():
>       http://wiki.nginx.org/HttpLuaModule#ngx.req.set_body_data
>       - feature: added new Lua functions ngx.req.read_body(),
>       ngx.req.discard_body(), ngx.req.get_body_data(), and
>       ngx.req.get_body_file(). see the docs here:
>       http://wiki.nginx.org/HttpLuaModule#ngx.req.read_body
>       - feature: now we implemented ngx.req.set_uri() and
>       ngx.req.set_uri_args() to emulate ngx_rewrite's rewrite directive
>       (without redirect or permanent modifiers). thanks Vladimir Protasov
>       (utros) and Nginx User.
>       - feature: added constant ngx.HTTP_METHOD_NOT_IMPLEMENTED (501).
>       thanks Nginx User.
>       - feature: now for HTTP 1.0 requests, we disable the automatic full
>       buffering mode if the user sets the Content-Length response header
>       before sending out the headers. this allows streaming output for HTTP 1.0
>       requests if the content length can be calculated beforehand. thanks Li Ziyi.
>       - bugfix: now we properly support setting the Cache-Controlresponse header via the
>       ngx.header.HEADER interface.
>       - bugfix: no longer set header hash to 1. use the ngx_hash_key_lcinstead.
>       - bugfix: now we skip rewrite phase Lua handlers altogether if
>       ngx_rewrite's rewrite directive issue a location re-lookup by
>       changing URIs (but not including rewrite ... break). thanks Nginx User.
>       - bugfix: fixed hanging issues when using ngx.exec() within
>       rewrite_by_lua and access_by_lua. thanks Nginx User for reporting
>       it.
>       - bugfix: lua_need_request_body should not skip requests with
>       methods other than POST and PUT. thanks Nginx User.
>       - bugfix: ndk.set_var.DIRECTIVE had a memory issue and might pass
>       empty argument values to the directive being called. thanks dannynoonan.
>       - bugfix: no longer free request body buffers that are not
>       allocated by ourselves.
>       - bugfix: now we allow setting ngx.var.VARIABLE to nil.
>       - bugfix: now we explicitly clear all the modules' contexts before
>       dump to named location with ngx.exec. thanks Nginx User.
>    - upgraded EchoNginxModule to 0.37rc7.
>       - bugfix: fixed a memory issue in both echo_sleep and
>       echo_blocking_sleep: we should not pass ngx_str_t strings to atof()which expects C strings.
>       - bugfix: now we explicitly clear all the modules' contexts before
>       dump to named location with echo_exec.
>       - bugfix: bugfix: echo_exec may hang when running after echo_sleep(or other I/O interruption calls): we should have called
>       ngx_http_finalize_request on NGX_DONE to decrement r->main->countanyway.
>       - bugfix: now we properly set the Content-Length request header for
>       subrequests.
>    - upgraded SrcacheNginxModule to 0.13rc2.
>       - feature: implemented response status line and general response
>       header cachin and added new directives srcache_store_hide_headerand
>       srcache_store_pass_header to control which headers to cache and
>       which not.
>       - feature: added new directive srcache_response_cache_control to
>       control whether honor response headers Cache-Control and Expires,
>       default on.
>       - feature: we disable srcache_store automatically by default when Cache-Control:
>       max-age=0 and Expires: <date no more recently than now> are seen.
>       - feature: implemented builtin nginx variable $srcache_expire for
>       automatic expiration time calculation based on response headers
>       Cache-Control (max-age) and Expires; also added new directives
>       srcache_max_expire and srcache_default_expire.
>       - feature: implemented the srcache_store_no_cache directive; now by
>       default, we do not store responses with the header Cache-Control:
>       no-cache into the cache.
>       - feature: implemented the srcache_store_no_store directive(default
>       off). Now by default, responses with the header Cache-Control:
>       no-store will not be stored into the cache.
>       - feature: implemented the srcache_store_private directive to
>       control whether to store responses with the header Cache-Control:
>       private.
>       - feature: implemented the srcache_request_cache_control directive
>       to allow request headers Cache-Control: no-cache or Pragma: no-cacheto force bypassing cache lookup. it also honors the request header Cache-Control:
>       no-store. this directive is turned off by default.
>       - feature: now we check response header Content-Encoding by default
>       and a non-empty header value will skip srcache_store; also
>       introduced a new directive named srcache_ignore_content_encoding to
>       ignore this response header.
>       - feature: implemented the srcache_methods directive to specify
>       request methods that are cacheable, by default, only GET and HEADare cacheable.
>       - bugfix: we no longer set header hash to 1; we use ngx_hash_key_lcinstead.
>       - bugfix: when we skip srcache_fetch by means of srcache_fetch_skip,
>       we should not automatically skip srcache_store.
>       - bugfix: now we ignore the Content-Length header (if any) of the
>       main request for the subrequests.
>       - bugfix: there might be a segfault when failing to allocate memory
>       in ngx_http_srcache_add_copy_chain. thanks Shaun savage.
>       - feature: implemented new directive srcache_store_statuses to
>       allow the user to specify the response status code list that is to be
>       stored into the cache.
>       - bugfix: we now only cache 200, 301, and 302 responses by default.
>    - upgraded IconvNginxModule to 0.10rc5.
>       - bugfix: fixed -Wset-but-not-used warnings issued by gcc 4.6.0.
>       thanks Zhi Jiale (Calio).
>    - upgraded HeadersMoreNginxModule to 0.16rc3.
>       - bugfix: we should set header hash using ngx_hash_key_lc, not
>       simply to 1.
>       - bugfix: fixed setting Cache-Control response headers. we should
>       properly prepare the r->cache_control array as well.
>    - upgraded RdsJsonNginxModule to 0.12rc6.
>       - bugfix: fixed compatibility with nginx 1.1.4+.
>    - upgraded RdsCsvNginxModule to 0.04.
>       - bugfix: fixed compatibility issues with nginx 1.1.4+.
>       - optimization: now we only register our filters when rds_csv is
>       actually used in nginx.conf.
>    - upgraded Redis2NginxModule to 0.08rc1.
>       - bugfix: fixed compatibility with nginx 1.1.4+.
>    - upgraded DrizzleNginxModule to 0.1.2rc2.
>       - bugfix: fixed compatibility with nginx 1.1.4+
>    - upgraded MemcNginxModule to 0.13rc1.
>       - bugfix: fixed compatibility with nginx 1.1.4+.
>    - upgraded SetMiscNginxModule to v0.22rc3.
>       - minor code cleanup.
>    - applied the patch to the Nginx core that always clears all modules'
>    contexts in ngx_http_named_location.
>    - applied the patch for the variable-header-ignore-no-hash issue. see
>    http://forum.nginx.org/read.php?29,216062 for details.
>
> As always, you're welcome to report bugs and feature requests either here
> or directly to me :)
>
> It'll also be highly appreciated to try out the devel releases (based on
> the Nginx core 1.0.9+) that are coming out later ;)
>
> OpenResty (aka. ngx_openresty) is a full-fledged web application server
> by bundling the standard Nginx core, lots of 3rd-party Nginx modules<http://wiki.nginx.org/3rdPartyModules>,
> as well as most of their external dependencies.
>
> By taking adantage of various well-designed Nginx modules, OpenResty
> effectively turns the nginx server into a powerful web app server, in which
> the web developers can use the Lua programming language to script various
> existing nginx C modules and Lua modules and construct extremely
> high-performance web applications that is capable to handle 10K+
> connections.
>
> OpenResty aims to run your server-side web app completely in the Nginx
> server, leveraging Nginx's event model to do non-blocking I/O not only with
> the HTTP clients, but also with remote backends like MySQL, PostgreSQL,
> Memcached, and Redis.
>
> You can find more details on the homepage of ngx_openresty here:
>
>     http://openresty.org
>
> Have fun!
> -agentzh
>
>
> _______________________________________________
> 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/20111103/e2a6f9bf/attachment-0001.html>

From lists at ruby-forum.com  Thu Nov  3 15:29:45 2011
From: lists at ruby-forum.com (Roger Gue)
Date: Thu, 03 Nov 2011 16:29:45 +0100
Subject: Upstream time out
Message-ID: <ac4573fd6465e0b060bcdcfada0b5e61@ruby-forum.com>

I am using NGINX 1.0.4 on Redhat Linux 5 and I am having the error
message: 504 Gateway Time out NGINX 1.0.4

HERE IS THE DETAIL OF MY ERROR.LOG FILE;

2011/11/03 09:05:11 [error] 22348#0: *898 upstream timed out (110
Connection timed out) while reading response header from upstream,
client:172.16.19.180, server: mcb-http-t.jlg.com, request: "POST
/ole/findshipto.jsp HTTP/1.1", upstream:
"http://172.21.68:80/ole/findshipto.jsp", host: "mcb-htp-t.jlg.com",
referrer: :https://mcb-http-t.jlg.com/ole/findshipto.jsp"

Please some one help me as I am new to NGINX:

HERE IS MY PROXY CONFIGURATION FILE:

proxy_set_header X-Real-IP $remote_addr;
proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
client_max_body-size 50m;
client_body_buffer_size 256k;
proxy_connect_timeout   300;
proxy_send_timeout      300;
proxy_read_timeout      300;
proxy_buffers           32 4k;
client_body_timeout     300;
proxy_ignore_client_abort on;

I WILL APPRECIATE YOUR HELP IN SOLVING THIS PROBLEM. THANKS IN ADVANCE

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


From ilan at time4learning.com  Thu Nov  3 15:34:05 2011
From: ilan at time4learning.com (Ilan Berkner)
Date: Thu, 3 Nov 2011 11:34:05 -0400
Subject: Slow loading SWF files during high load
Message-ID: <CAOoWpPrvNECqU7+r8RvRfM=km1Ra+sY5CgQMt4p0h6MKuHmFQA@mail.gmail.com>

During our busier days and hours, pages that have SWF files loading on them
take a long time to load.  Granted, some of our SWF files are around half a
meg (.5 MB) so that's expected, but what specific settings in the Nginx
configuration file could assist with speeding up downloads of large files?

Thanks
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20111103/e9db62da/attachment.html>

From mdounin at mdounin.ru  Thu Nov  3 16:58:42 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Thu, 3 Nov 2011 20:58:42 +0400
Subject: Upstream time out
In-Reply-To: <ac4573fd6465e0b060bcdcfada0b5e61@ruby-forum.com>
References: <ac4573fd6465e0b060bcdcfada0b5e61@ruby-forum.com>
Message-ID: <20111103165842.GD95664@mdounin.ru>

Hello!

On Thu, Nov 03, 2011 at 04:29:45PM +0100, Roger Gue wrote:

> I am using NGINX 1.0.4 on Redhat Linux 5 and I am having the error
> message: 504 Gateway Time out NGINX 1.0.4
> 
> HERE IS THE DETAIL OF MY ERROR.LOG FILE;
> 
> 2011/11/03 09:05:11 [error] 22348#0: *898 upstream timed out (110
> Connection timed out) while reading response header from upstream,
> client:172.16.19.180, server: mcb-http-t.jlg.com, request: "POST
> /ole/findshipto.jsp HTTP/1.1", upstream:
> "http://172.21.68:80/ole/findshipto.jsp", host: "mcb-htp-t.jlg.com",
> referrer: :https://mcb-http-t.jlg.com/ole/findshipto.jsp"

[...]

> proxy_read_timeout      300;

The error suggests you backend doesn't handle requests properly/in 
time.  If it's expected that request may take longer than 300 
seconds to complete - you may want to tune proxy_read_timeout.  If 
not - your backend is probably overloaded or not working at all, 
and you may want to fix it or add more backends.

Maxim Dounin


From lists at ruby-forum.com  Thu Nov  3 17:47:51 2011
From: lists at ruby-forum.com (Roger Gue)
Date: Thu, 03 Nov 2011 18:47:51 +0100
Subject: 504 Gateway Time out Nginx 1.0.4: Upstream time out
In-Reply-To: <ac4573fd6465e0b060bcdcfada0b5e61@ruby-forum.com>
References: <ac4573fd6465e0b060bcdcfada0b5e61@ruby-forum.com>
Message-ID: <0c4c2869127d45aaf119cad0fb6df639@ruby-forum.com>

Thanks for your help.  I know my question will sound little ackward but 
I am new to all this: How do I tune proxy_read_tomeout?

How do I check the backend to see if it is overloaded? or not working at 
all?

How do I fix the backends or add more to it?
 Thanks a lot for taken time to help me on this.

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


From quintinpar at gmail.com  Thu Nov  3 18:30:00 2011
From: quintinpar at gmail.com (Quintin Par)
Date: Fri, 4 Nov 2011 00:00:00 +0530
Subject: Correct way to setup maintenance page in nginx
Message-ID: <CACW9FdNB2iERDjQ9QUZ=JMpDRiqhpK9J__MRq5k1joudrYj1bQ@mail.gmail.com>

Hi all,

I tried using this method with `try_files`. Didn?t work. Also the HTTP
status code(503) is not being set.

    location / {
        try_files /var/www/during_build.html @maintenance;
            proxy_pass  http://localhost:82;
    }

    location @maintenance {
          return 503;
    }

This method outside of `location` directive

 error_page 503 /var/www/during_build.html;

 ## System Maintenance (Service Unavailable)
 if (-f  /var/www/during_build.html) {
     return 503;
 }

Is also not working. Nginx just returns 503 without the custom page.

What is the correct way to show system down pages?


-Quintin
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20111104/171ecb6d/attachment.html>

From i at vbart.ru  Thu Nov  3 20:18:03 2011
From: i at vbart.ru (Valentin V. Bartenev)
Date: Fri, 4 Nov 2011 00:18:03 +0400
Subject: Correct way to setup maintenance page in nginx
In-Reply-To: <CACW9FdNB2iERDjQ9QUZ=JMpDRiqhpK9J__MRq5k1joudrYj1bQ@mail.gmail.com>
References: <CACW9FdNB2iERDjQ9QUZ=JMpDRiqhpK9J__MRq5k1joudrYj1bQ@mail.gmail.com>
Message-ID: <201111040018.04115.i@vbart.ru>

On Thursday 03 November 2011 22:30:00 Quintin Par wrote:
[...]
> This method outside of `location` directive
> 
>  error_page 503 /var/www/during_build.html;
> 
>  ## System Maintenance (Service Unavailable)
>  if (-f  /var/www/during_build.html) {
>      return 503;
>  }
> 
> Is also not working. Nginx just returns 503 without the custom page.
> 
> What is the correct way to show system down pages?
> 

Try this one:

    error_page 503 /during_build.html;

    location / {

        if (-f /var/www/during_build.html) {
            return 503;
        }

    }

    location = /during_build.html {
        root /var/www/;
        internal;
    }


wbr, Valentin V. Bartenev


From appa at perusio.net  Thu Nov  3 21:02:11 2011
From: appa at perusio.net (=?UTF-8?B?QW50w7NuaW8=?= P. P. Almeida)
Date: Thu, 03 Nov 2011 21:02:11 +0000
Subject: Correct way to setup maintenance page in nginx
In-Reply-To: <201111040018.04115.i@vbart.ru>
References: <CACW9FdNB2iERDjQ9QUZ=JMpDRiqhpK9J__MRq5k1joudrYj1bQ@mail.gmail.com>
	<201111040018.04115.i@vbart.ru>
Message-ID: <87bossrkbw.wl%appa@perusio.net>

On 3 Nov 2011 20h18 WET, i at vbart.ru wrote:

> On Thursday 03 November 2011 22:30:00 Quintin Par wrote:
> [...]
>> This method outside of `location` directive
>>
>> error_page 503 /var/www/during_build.html;
>>
>> ## System Maintenance (Service Unavailable)
>> if (-f  /var/www/during_build.html) {
>> return 503;
>> }
>>
>> Is also not working. Nginx just returns 503 without the custom
>> page.
>>
>> What is the correct way to show system down pages?
>>
>
> Try this one:
>
> error_page 503 /during_build.html;
>
> location / {
>
> if (-f /var/www/during_build.html) {
> return 503;
> }
>
> }
>
> location = /during_build.html {
> root /var/www/;
> internal;
> }

Why not this?

root /var/www;

location / {

   error_page 503 @unavailable;

   location @unavailable {
      try_files /during_building.html @503;
   }
}

location @503 {
   return 503;
}

No need to use the if and the internal is implicit on the try_files.

--- appa


From i at vbart.ru  Thu Nov  3 21:42:50 2011
From: i at vbart.ru (Valentin V. Bartenev)
Date: Fri, 4 Nov 2011 01:42:50 +0400
Subject: Correct way to setup maintenance page in nginx
In-Reply-To: <87bossrkbw.wl%appa@perusio.net>
References: <CACW9FdNB2iERDjQ9QUZ=JMpDRiqhpK9J__MRq5k1joudrYj1bQ@mail.gmail.com>
	<201111040018.04115.i@vbart.ru> <87bossrkbw.wl%appa@perusio.net>
Message-ID: <201111040142.50947.i@vbart.ru>

On Friday 04 November 2011 01:02:11 Ant?nio P. P. Almeida wrote:
> Why not this?
> 
> root /var/www;
> 
> location / {
> 
>    error_page 503 @unavailable;
> 
>    location @unavailable {
>       try_files /during_building.html @503;
>    }
> }
> 
> location @503 {
>    return 503;
> }
> 
> No need to use the if and the internal is implicit on the try_files.

First of all, named locations can be on the server level only.

And, how does a request get into @unavailable? Only after 503 has occurred.
But, what will cause it?

I think the main idea was: when we need to do some maintenance, we just create 
a specific file (probably with explaining message for users) and Nginx starts 
to return 503 on every request.

wbr, Valentin V. Bartenev


From zzz at zzz.org.ua  Thu Nov  3 21:54:02 2011
From: zzz at zzz.org.ua (Alexandr Gomoliako)
Date: Thu, 3 Nov 2011 23:54:02 +0200
Subject: Correct way to setup maintenance page in nginx
In-Reply-To: <201111040142.50947.i@vbart.ru>
References: <CACW9FdNB2iERDjQ9QUZ=JMpDRiqhpK9J__MRq5k1joudrYj1bQ@mail.gmail.com>
	<201111040018.04115.i@vbart.ru> <87bossrkbw.wl%appa@perusio.net>
	<201111040142.50947.i@vbart.ru>
Message-ID: <CAKb9G9h8bUWN=vXCCHHG4nUQ73pYd4ZG3X0QOj8N3Rii4r-3Pw@mail.gmail.com>

On 11/3/11, Valentin V. Bartenev <i at vbart.ru> wrote:
> I think the main idea was: when we need to do some maintenance, we just
> create
> a specific file (probably with explaining message for users) and Nginx
> starts
> to return 503 on every request.

If you are willing to create some file for this you can just as well swap
nginx.conf with special maintenance config, where you can specify a single
server { } block with return 503 or anything you like.


From i at vbart.ru  Thu Nov  3 22:06:24 2011
From: i at vbart.ru (Valentin V. Bartenev)
Date: Fri, 4 Nov 2011 02:06:24 +0400
Subject: Correct way to setup maintenance page in nginx
In-Reply-To: <CAKb9G9h8bUWN=vXCCHHG4nUQ73pYd4ZG3X0QOj8N3Rii4r-3Pw@mail.gmail.com>
References: <CACW9FdNB2iERDjQ9QUZ=JMpDRiqhpK9J__MRq5k1joudrYj1bQ@mail.gmail.com>
	<201111040142.50947.i@vbart.ru>
	<CAKb9G9h8bUWN=vXCCHHG4nUQ73pYd4ZG3X0QOj8N3Rii4r-3Pw@mail.gmail.com>
Message-ID: <201111040206.24783.i@vbart.ru>

On Friday 04 November 2011 01:54:02 Alexandr Gomoliako wrote:
> If you are willing to create some file for this you can just as well swap
> nginx.conf with special maintenance config, where you can specify a single
> server { } block with return 503 or anything you like.

Of course, you have many options to do this with different pros and cons.
And I just help to solve the problem the way that topic starter wants.

wbr, Valentin V. Bartenev


From coderight at gmail.com  Fri Nov  4 01:51:25 2011
From: coderight at gmail.com (Chris)
Date: Thu, 3 Nov 2011 21:51:25 -0400
Subject: "if" statement breaks try_files?
Message-ID: <CAPaFLLTmuxOOFLBP1gYQCL-DA8J3h0xz52rqtLHExQFcikuzuw@mail.gmail.com>

I'm seeing something really weird.  If I put an "if" statement
anywhere within the same block as a "try_files" then it stops working.
 It always returns that the file exists even when it doesn't.

I'm testing against nginx-1.0.9 on a Linux host.

This can be tested with this simple.conf:

worker_processes  1;
events { worker_connections  1024; }
http {
   server {
      listen 80;
      server_name localhost;
      root /var/www;
      index index.html index.htm index.php;

      location ~ \.php($|/) {
         try_files doesnotexist =405;
         fastcgi_index index.php;
         fastcgi_pass unix:/tmp/php.socket;

         if ($uri) {}

         fastcgi_param  SCRIPT_FILENAME    /var/www$fastcgi_script_name;
         fastcgi_param  HTTPS              off;
         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_param  SCRIPT_NAME        $fastcgi_script_name;
         fastcgi_param  REQUEST_URI        $request_uri;
         fastcgi_param  DOCUMENT_URI       $document_uri;
         fastcgi_param  DOCUMENT_ROOT      $document_root;
         fastcgi_param  SERVER_PROTOCOL    $server_protocol;
         fastcgi_param  GATEWAY_INTERFACE  CGI/1.1;
         fastcgi_param  SERVER_SOFTWARE    nginx;
         fastcgi_param  REMOTE_ADDR        $remote_addr;
         fastcgi_param  REMOTE_PORT        $remote_port;
         fastcgi_param  SERVER_ADDR        $server_addr;
         fastcgi_param  SERVER_PORT        $server_port;
         fastcgi_param  SERVER_NAME        $server_name;
         fastcgi_param  REDIRECT_STATUS    200;
      }
   }
}

If that "if ($uri)" statement is left in it never returns the 405
error when accessing whatever.php.  If you comment out that "if" then
it works as expected and always returns the 405 error.

Any idea what is going on here?

CR


From quintinpar at gmail.com  Fri Nov  4 03:13:40 2011
From: quintinpar at gmail.com (Quintin Par)
Date: Fri, 4 Nov 2011 08:43:40 +0530
Subject: Correct way to setup maintenance page in nginx
In-Reply-To: <201111040018.04115.i@vbart.ru>
References: <CACW9FdNB2iERDjQ9QUZ=JMpDRiqhpK9J__MRq5k1joudrYj1bQ@mail.gmail.com>
	<201111040018.04115.i@vbart.ru>
Message-ID: <CACW9FdOeZFmVLJFFC53xnq08fuEk+D9BJAMD3s_bEO9ZHkCy+w@mail.gmail.com>

Thanks Valentin. This worked like a charm.

I spend nearly 2-3 hours trying for a solution that evaded ?if? and just
?try_files?. Even the official wiki supported that.

There are a lot of solutions out in the wild and most for them don?t work.
Thanks again for the help.

 -Quintin
On Fri, Nov 4, 2011 at 1:48 AM, Valentin V. Bartenev <i at vbart.ru> wrote:

> location = /during_build.html {
>        root /var/www/;
>        internal;
>    }
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20111104/b60a9656/attachment.html>

From quintinpar at gmail.com  Fri Nov  4 03:15:20 2011
From: quintinpar at gmail.com (Quintin Par)
Date: Fri, 4 Nov 2011 08:45:20 +0530
Subject: Nginx & long poll: Best practices to reduce memory and bandwidth
	footprint
In-Reply-To: <CACW9FdPYnJ_TFGJkv202aqaWFpfA-L18-w52futK7Bb0XMi97g@mail.gmail.com>
References: <CACW9FdPYnJ_TFGJkv202aqaWFpfA-L18-w52futK7Bb0XMi97g@mail.gmail.com>
Message-ID: <CACW9FdO1oFdmcQkawc32O-YSFLyOyAdN2dDRPjYChguODMj4pQ@mail.gmail.com>

Posting the contents inline. Can someone review this?
-----------------------------------------------------------------------------------------------------
I use nginx in this mode for [BOSH][1] and chat clients along with gzip.

location ~* /http-bind/ {
     proxy_buffering off;
     keepalive_timeout 55;
     access_log off;
     tcp_nodelay on;
     proxy_pass http://x.x.x.x:1111;
 }

Is this the best approach to **managing long polling** in nginx.

I also use just one worker process for altogether for web & chat (single
CPU). Is that fine?


  [1]: http://xmpp.org/extensions/xep-0206.html
-Quintin

On Thu, Nov 3, 2011 at 2:47 PM, Quintin Par <quintinpar at gmail.com> wrote:

> Can someone please answer this question on serverfault.com?
>
>
> http://serverfault.com/questions/327301/nginx-long-poll-best-practices-to-reduce-memory-and-bandwidth-footprint
>
>
> -Quintin
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20111104/c1191e1b/attachment.html>

From nginx at nginxuser.net  Fri Nov  4 06:13:47 2011
From: nginx at nginxuser.net (Nginx User)
Date: Fri, 4 Nov 2011 09:13:47 +0300
Subject: "if" statement breaks try_files?
In-Reply-To: <CAPaFLLTmuxOOFLBP1gYQCL-DA8J3h0xz52rqtLHExQFcikuzuw@mail.gmail.com>
References: <CAPaFLLTmuxOOFLBP1gYQCL-DA8J3h0xz52rqtLHExQFcikuzuw@mail.gmail.com>
Message-ID: <CAN=WzV0wZL9o-M7dJXAbzmoafr-74=nCS3GTa1wC7hzG0RrKCA@mail.gmail.com>

See: http://wiki.nginx.org/IfIsEvil


From quintinpar at gmail.com  Fri Nov  4 08:55:45 2011
From: quintinpar at gmail.com (Quintin Par)
Date: Fri, 4 Nov 2011 14:25:45 +0530
Subject: Correct way to setup maintenance page in nginx
In-Reply-To: <CACW9FdOeZFmVLJFFC53xnq08fuEk+D9BJAMD3s_bEO9ZHkCy+w@mail.gmail.com>
References: <CACW9FdNB2iERDjQ9QUZ=JMpDRiqhpK9J__MRq5k1joudrYj1bQ@mail.gmail.com>
	<201111040018.04115.i@vbart.ru>
	<CACW9FdOeZFmVLJFFC53xnq08fuEk+D9BJAMD3s_bEO9ZHkCy+w@mail.gmail.com>
Message-ID: <CACW9FdM1e3mNg9s3Of4Ux_PvPNOk4dsVy00Ms0n-MEgYjkoHjA@mail.gmail.com>

On second thought don?t you think the file check in the ?if? condition is
expensive considering it will be executed on every hit?

-Quintin

On Fri, Nov 4, 2011 at 8:43 AM, Quintin Par <quintinpar at gmail.com> wrote:

> Thanks Valentin. This worked like a charm.
>
> I spend nearly 2-3 hours trying for a solution that evaded ?if? and just
> ?try_files?. Even the official wiki supported that.
>
> There are a lot of solutions out in the wild and most for them don?t work.
> Thanks again for the help.
>
>  -Quintin
> On Fri, Nov 4, 2011 at 1:48 AM, Valentin V. Bartenev <i at vbart.ru> wrote:
>
>> location = /during_build.html {
>>        root /var/www/;
>>        internal;
>>    }
>>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20111104/5a26987b/attachment.html>

From i at vbart.ru  Fri Nov  4 09:48:15 2011
From: i at vbart.ru (Valentin V. Bartenev)
Date: Fri, 4 Nov 2011 13:48:15 +0400
Subject: Correct way to setup maintenance page in nginx
In-Reply-To: <CACW9FdM1e3mNg9s3Of4Ux_PvPNOk4dsVy00Ms0n-MEgYjkoHjA@mail.gmail.com>
References: <CACW9FdNB2iERDjQ9QUZ=JMpDRiqhpK9J__MRq5k1joudrYj1bQ@mail.gmail.com>
	<CACW9FdOeZFmVLJFFC53xnq08fuEk+D9BJAMD3s_bEO9ZHkCy+w@mail.gmail.com>
	<CACW9FdM1e3mNg9s3Of4Ux_PvPNOk4dsVy00Ms0n-MEgYjkoHjA@mail.gmail.com>
Message-ID: <201111041348.15401.i@vbart.ru>

On Friday 04 November 2011 12:55:45 Quintin Par wrote:
> On second thought don?t you think the file check in the ?if? condition is
> expensive considering it will be executed on every hit?

If you don't deal with high-load or dedicated file storage with big access 
latency, then don't worry. IMHO.

Also, you can try to tune Nginx open file cache:
http://nginx.org/en/docs/http/ngx_http_core_module.html#open_file_cache

wbr, Valentin V. Bartenev


From quintinpar at gmail.com  Fri Nov  4 09:57:19 2011
From: quintinpar at gmail.com (Quintin Par)
Date: Fri, 4 Nov 2011 15:27:19 +0530
Subject: Correct way to setup maintenance page in nginx
In-Reply-To: <201111041348.15401.i@vbart.ru>
References: <CACW9FdNB2iERDjQ9QUZ=JMpDRiqhpK9J__MRq5k1joudrYj1bQ@mail.gmail.com>
	<CACW9FdOeZFmVLJFFC53xnq08fuEk+D9BJAMD3s_bEO9ZHkCy+w@mail.gmail.com>
	<CACW9FdM1e3mNg9s3Of4Ux_PvPNOk4dsVy00Ms0n-MEgYjkoHjA@mail.gmail.com>
	<201111041348.15401.i@vbart.ru>
Message-ID: <CACW9FdM0G=7H9hUZpz6nM4zpdQic=gd7FHfMwTnGW=e3R5_CXw@mail.gmail.com>

Thanks.

On Fri, Nov 4, 2011 at 3:18 PM, Valentin V. Bartenev <i at vbart.ru> wrote:

> On Friday 04 November 2011 12:55:45 Quintin Par wrote:
> > On second thought don?t you think the file check in the ?if? condition is
> > expensive considering it will be executed on every hit?
>
> If you don't deal with high-load or dedicated file storage with big access
> latency, then don't worry. IMHO.
>
> Also, you can try to tune Nginx open file cache:
> http://nginx.org/en/docs/http/ngx_http_core_module.html#open_file_cache
>
> wbr, Valentin V. Bartenev
>
> _______________________________________________
> 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/20111104/61d6ed85/attachment.html>

From nginx-forum at nginx.us  Fri Nov  4 10:50:58 2011
From: nginx-forum at nginx.us (forum_id)
Date: Fri, 04 Nov 2011 06:50:58 -0400
Subject: HTTP Request filter module
In-Reply-To: <CAB4Tn6MfVkLavz4Pa+FyqT1z1K6+feoXbAuFBVcsLUEaRweL9Q@mail.gmail.com>
References: <CAB4Tn6MfVkLavz4Pa+FyqT1z1K6+feoXbAuFBVcsLUEaRweL9Q@mail.gmail.com>
Message-ID: <16ecfbaa9bb2dbf62dda34a3415b2c44.NginxMailingListEnglish@forum.nginx.org>

@agentzh, thank you. As you suggested, I went through the ngx_lua code.
Since I am new to nginx, my understanding was not so good, but I want to
write it in few points, 
1) Lua handler receives request 
2) Forwards request to Lua interpreter 
3) Send Lua response back to user.

Let me know if my understanding itself is wrong. Give me some simple
pointers to understand, because I need to start with my module as soon
as possible. Thank you once again.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,217739,217810#msg-217810


From nginx-forum at nginx.us  Fri Nov  4 11:00:53 2011
From: nginx-forum at nginx.us (forum_id)
Date: Fri, 04 Nov 2011 07:00:53 -0400
Subject: Proxy handler to multiple servers for single request.
Message-ID: <8424e8ab817f240b5c6e51e54bd3e760.NginxMailingListEnglish@forum.nginx.org>

I have requirement where proxy handler needs to talk to multiple server
before sending request to actual server where it gets processed. For
example proxy handler receives user request, talks to virus scanner to
see user data contain any virus or not then only forward to actual
server. Here virus scanner may be another server or a thread part of
nginx. My doubts are

1) When I went through proxy module, ngx_http_proxy_create_request is
the callback responsible to prepare request to send to another server.
As per understanding, I need to call here virus scanner before sending
request. Is it correct? Or is there any other place I can do this work?
In this function if I call virus scanner API's, how will the request
forwarded the server which serves request?

2) How to make it asynchronous with out effecting nginx's performance.

--
Thanks

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,217811,217811#msg-217811


From nginx-forum at nginx.us  Fri Nov  4 11:58:10 2011
From: nginx-forum at nginx.us (forum_id)
Date: Fri, 04 Nov 2011 07:58:10 -0400
Subject: Proxy handler to multiple servers for single request.
In-Reply-To: <8424e8ab817f240b5c6e51e54bd3e760.NginxMailingListEnglish@forum.nginx.org>
References: <8424e8ab817f240b5c6e51e54bd3e760.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <9596b7b7d94f7d74660ee753aea0a05b.NginxMailingListEnglish@forum.nginx.org>

I am trying to develop some thing similar to mod_clamav for apache. This
module scans only responses, I need to scan even requests.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,217811,217814#msg-217814


From tomlove at gmail.com  Fri Nov  4 12:50:52 2011
From: tomlove at gmail.com (Thomas Love)
Date: Fri, 4 Nov 2011 14:50:52 +0200
Subject: Nginx & long poll: Best practices to reduce memory and bandwidth
	footprint
In-Reply-To: <CACW9FdO1oFdmcQkawc32O-YSFLyOyAdN2dDRPjYChguODMj4pQ@mail.gmail.com>
References: <CACW9FdPYnJ_TFGJkv202aqaWFpfA-L18-w52futK7Bb0XMi97g@mail.gmail.com>
	<CACW9FdO1oFdmcQkawc32O-YSFLyOyAdN2dDRPjYChguODMj4pQ@mail.gmail.com>
Message-ID: <CAJCGDi86PA8bJE0=M3EYJjqw5t+a_NpaZRW4PXErOMErdF59fA@mail.gmail.com>

On 4 November 2011 05:15, Quintin Par <quintinpar at gmail.com> wrote:

> Posting the contents inline. Can someone review this?
>
> -----------------------------------------------------------------------------------------------------
> I use nginx in this mode for [BOSH][1] and chat clients along with gzip.
>
> location ~* /http-bind/ {
>      proxy_buffering off;
>      keepalive_timeout 55;
>      access_log off;
>      tcp_nodelay on;
>      proxy_pass http://x.x.x.x:1111;
>  }
>
> Is this the best approach to **managing long polling** in nginx.
>
> I also use just one worker process for altogether for web & chat (single
> CPU). Is that fine?
>
>
It looks fine. Use that until you have a performance problem, and then
gather evidence identifying your bottleneck before changing anything.

You might as well start with n worker processes though, where n is the
number of cores on your CPU. Give yourself a few thousand
worker_connections, because you'll have a relatively large number of
relatively idle sockets.

Long-polling is very simple and nginx is good at it. I would advise not to
try to fix anything until you find a real problem.

Thomas
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20111104/2d8a7165/attachment.html>

From agentzh at gmail.com  Fri Nov  4 13:02:23 2011
From: agentzh at gmail.com (agentzh)
Date: Fri, 4 Nov 2011 21:02:23 +0800
Subject: HTTP Request filter module
In-Reply-To: <16ecfbaa9bb2dbf62dda34a3415b2c44.NginxMailingListEnglish@forum.nginx.org>
References: <CAB4Tn6MfVkLavz4Pa+FyqT1z1K6+feoXbAuFBVcsLUEaRweL9Q@mail.gmail.com>
	<16ecfbaa9bb2dbf62dda34a3415b2c44.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <CAB4Tn6PUGgrWwc3CaHzh2J8DjGNgtFWWvVCAqocQ+QB62Nk42A@mail.gmail.com>

On Fri, Nov 4, 2011 at 6:50 PM, forum_id <nginx-forum at nginx.us> wrote:
> @agentzh, thank you. As you suggested, I went through the ngx_lua code.
> Since I am new to nginx, my understanding was not so good, but I want to
> write it in few points,

If you're not familiar with Nginx internals, then I think use ngx_lua
and Lua scripting is the quickest and easiest way to get the job done.
I don't think I have the time to explain every thing here because it's
really complicated for freshmen :)

If you insist in Nginx C module development, I suggest you start with
the resources listed here:

    http://wiki.nginx.org/Resources

Regards,
-agentzh


From coderight at gmail.com  Fri Nov  4 14:02:04 2011
From: coderight at gmail.com (Chris)
Date: Fri, 4 Nov 2011 10:02:04 -0400
Subject: "if" statement breaks try_files?
In-Reply-To: <CAN=WzV0wZL9o-M7dJXAbzmoafr-74=nCS3GTa1wC7hzG0RrKCA@mail.gmail.com>
References: <CAPaFLLTmuxOOFLBP1gYQCL-DA8J3h0xz52rqtLHExQFcikuzuw@mail.gmail.com>
	<CAN=WzV0wZL9o-M7dJXAbzmoafr-74=nCS3GTa1wC7hzG0RrKCA@mail.gmail.com>
Message-ID: <CAPaFLLTRrD9u+D5EdHpBXXpoeUzXpa6YJkXp=ZBCuNvtn_WB-g@mail.gmail.com>

On Fri, Nov 4, 2011 at 2:13 AM, Nginx User <nginx at nginxuser.net> wrote:
> See: http://wiki.nginx.org/IfIsEvil

Thanks, I should have known.  I had read that but forgotten the most
important parts.

Shame so many projects design their own language when so many proven
works are out there, but I digress.

CR


From simone.fumagalli at contactlab.com  Fri Nov  4 14:04:01 2011
From: simone.fumagalli at contactlab.com (Simone Fumagalli)
Date: Fri, 4 Nov 2011 15:04:01 +0100
Subject: NGINX cache. Real meaning of zone_size
Message-ID: <4EB3F0D1.6040604@contactlab.com>

Hello, I read from the list

"zone_size" is a size of keys_zone, i.e. shared memory zone used to store cache keys (some minimal metadata about cached pages, about 64 bytes on 32-bit platforms).

but what does this practically mean ?

Let say a new request arrive and the request has to be cached. What does NGINX do if the "space" in the shared memory is over ? It delete the oldest file in the cache (and the key too) ?

Is there a rule to size this parameter ?

Thanks

--
Simone


From nginx-forum at nginx.us  Fri Nov  4 14:11:59 2011
From: nginx-forum at nginx.us (forum_id)
Date: Fri, 04 Nov 2011 10:11:59 -0400
Subject: HTTP Request filter module
In-Reply-To: <CAB4Tn6PUGgrWwc3CaHzh2J8DjGNgtFWWvVCAqocQ+QB62Nk42A@mail.gmail.com>
References: <CAB4Tn6PUGgrWwc3CaHzh2J8DjGNgtFWWvVCAqocQ+QB62Nk42A@mail.gmail.com>
Message-ID: <e046a3f40bc26ef6b043bd20e6f85746.NginxMailingListEnglish@forum.nginx.org>

@agentzh, thank you. Even though I am new to nginx I went through to
understand how nginx works and understood how nginx modules work. I just
need very few pointers, like lua modules receives request and forwards
to lua interpreter, after processing the request how it sends details. I
need to overall details not complete in depth details, so that I can
spend more time on code in productive way. It will make easy to
understand in better way. I need to use C language. Please give very
small notes with 5 points are also fine for me to start on it.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,217739,217820#msg-217820


From lists at ruby-forum.com  Fri Nov  4 19:38:37 2011
From: lists at ruby-forum.com (Roger Gue)
Date: Fri, 04 Nov 2011 20:38:37 +0100
Subject: 504 Gateway Time out Nginx 1.0.4: Upstream time out
In-Reply-To: <ac4573fd6465e0b060bcdcfada0b5e61@ruby-forum.com>
References: <ac4573fd6465e0b060bcdcfada0b5e61@ruby-forum.com>
Message-ID: <073867d2c99fa075d96095ce39dd647c@ruby-forum.com>

HERE IS MY NGINX.CONF FILE DETAIL:

user nginx;
worker_processes  1;

error_log  /var/log/nginx/error.log;
pid        /var/run/nginx/nginx.pid;

events {
    worker_connections  2048;
    # 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   3;
    tcp_nodelay        off;
    ignore_invalid_headers on;

    gzip  on;
    gzip_disable "MSIE [1-6]\.(?!.*SV1)";

    gzip_comp_level 6;
    gzip_static on;
    gzip_min_length 2200;
    gzip_buffers 16 8k;
    gzip_http_version 1.1;
    gzip_types text/plain text/css application/x-javascript image/x-icon 
text/xml application/xml application/xml+rss text/javascript;

    include /etc/nginx/conf.d/*.conf;
    include /etc/nginx/sites-enabled/*;
}


AND HERE IS MY PROXY.CONF FILE DETAIL

#proxy.conf
#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_body_buffer_size 256k;
     proxy_connect_timeout   10m;
     proxy_send_timeout      10m;
      proxy_read_timeout      8m;
     client_header_timeout   10m;
   proxy_buffers           32 4k;
     client_body_timeout     800;
    proxy_ignore_client_abort on;

client_header_buffer_size     1k;
large_client_header_buffers 8 8k;
client_max_body_size          2g;

            postpone_output 1460;

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


From djdarkbeat at gmail.com  Fri Nov  4 23:04:36 2011
From: djdarkbeat at gmail.com (Brian Loomis)
Date: Fri, 4 Nov 2011 17:04:36 -0600
Subject: beyond proxy_pass
Message-ID: <CAPES3LLTKqqH6WL+bAcBiLa68W74txncORHZp2BcMNf0mduvbw@mail.gmail.com>

I have a situation where I have a very high write intensive site in
nginx.  This site aggregates GPS positions for transit from a high
amount of devices.

The incoming traffic generates push notifications with the push_module
from slack.net (Leo)

I need to get nginx to send the same packet it receives to a
development server so that we can in essence `fork` the traffic so the
second server can make it's own set of push notifications to have live
http activity but completely sandboxed from the production site.

I've got the two packt books and have been going through them and am
using proxy_pass already o serve JSON across domains in productions
but can't seem to get proxy_pass to send to the development server
successfully.

What's the best way to fork or mirror https requests to two locations like this?

Brian Loomis


From nginx at nginxuser.net  Sat Nov  5 16:04:41 2011
From: nginx at nginxuser.net (Nginx User)
Date: Sat, 5 Nov 2011 19:04:41 +0300
Subject: ngx.req.get_post_args Issue
Message-ID: <CAN=WzV25TdtiEW99gBYRoGj2V_-j9YZKZ39b=G=f4Nh6637ORg@mail.gmail.com>

-************************
           START
-************************
if ngx.var.request_method == "POST" then
	ngx.req.read_body()
	local post_args = ngx.req.get_post_args()
	for my_key, v in pairs( post_args ) do
		if type(v) == "table" then
			my_arg = table.concat(v, " ")
		else
			my_arg = v
		end
		regex_rules.log_alert("Alert:- " .. my_key .. " ::: " .. my_arg)
	end
end
-************************
           END
-************************


The code in ngx_lua below which should log key/value pairs from a post
request results in the log output below:


-************************
           START
-************************
[alert] 17670#0: *1 Alert:- app_form[localServerPath] :::
/home/user/testsite.com/public_html/share/data/505
/home/user/testsite.com/public_html/share/data/505, client:
89.148.6.249, server: testsite.com, request: "POST // HTTP/1.1", host:
"testsite.com", referrer:
"http://testsite.com/?app_view=core.ItemAdmin&app_subView=core.ItemAdd&app_addPlugin=ItemAddFromServer&app_form%5BlocalServerPath%5D=/home/user/testsite.com/public_html/share/data/505&app_itemId=470&app_form%5Baction%5D%5BfindFilesFromLocalServer%5D=1&app_form%5BformName%5D=ItemAddFromServer"

[alert] 17670#0: *1 Alert:- app_form[formName] ::: ItemAddFromServer
ItemAddFromServer, client: 89.148.6.249, server: testsite.com,
request: "POST // HTTP/1.1", host: "testsite.com", referrer:
"http://testsite.com/?app_view=core.ItemAdmin&app_subView=core.ItemAdd&app_addPlugin=ItemAddFromServer&app_form%5BlocalServerPath%5D=/home/user/testsite.com/public_html/share/data/505&app_itemId=470&app_form%5Baction%5D%5BfindFilesFromLocalServer%5D=1&app_form%5BformName%5D=ItemAddFromServer"
[debug] 17670#0: *1 posix_memalign: 08A6A910:4096 @16

[alert] 17670#0: *1 Alert:- app_itemId ::: 470
------WebKitFormBoundary3c6CizwarDCoyCst
Content-Disposition: form-data; name="app_formUrl"

//?app_view=core.ItemAdmin 470
------WebKitFormBoundary3c6CizwarDCoyCst
Content-Disposition: form-data; name="app_authToken"

8845e98e9296
------WebKitFormBoundary3c6CizwarDCoyCst
Content-Disposition: form-data; name="app_controller"

core.ItemAdd
------WebKitFormBoundary3c6CizwarDCoyCst
Content-Disposition: form-data; name="app_form[formName]"

ItemAddFromServer
------WebKitFormBoundary3c6CizwarDCoyCst
Content-Disposition: form-data; name="app_itemId"

470
------WebKitFormBoundary3c6CizwarDCoyCst
Content-Disposition: form-data; name="app_addPlugin"

ItemAddFromServer
------WebKitFormBoundary3c6CizwarDCoyCst
Content-Disposition: form-data; name="app_form[localServerPath]"

/home/user/testsite.com/public_html/share/data/505
------WebKitFormBoundary3c6CizwarDCoyCst
Content-Disposition: form-data; name="app_form[set][title]"

on
------WebKitFormBoundary3c6CizwarDCoyCst
Content-Disposition: form-data;
name="app_form[CreateThumbnailOption][createThumbnail]"

on
------WebKitFormBoundary3c6CizwarDCoyCst
Content-Disposition: form-data; name="app_form[action][addFromLocalServer]"

Add Files
------WebKitFormBoundary3c6CizwarDCoyCst--
, client: 89.148.6.249, server: testsite.com, request: "POST //
HTTP/1.1", host: "testsite.com", referrer:
"http://testsite.com/?app_view=core.ItemAdmin&app_subView=core.ItemAdd&app_addPlugin=ItemAddFromServer&app_form%5BlocalServerPath%5D=/home/user/testsite.com/public_html/share/data/505&app_itemId=470&app_form%5Baction%5D%5BfindFilesFromLocalServer%5D=1&app_form%5BformName%5D=ItemAddFromServer"

[alert] 17670#0: *1 Alert:- app_form[action][findFilesFromLocalServer]
::: 1 1, client: 89.148.6.249, server: testsite.com, request: "POST //
HTTP/1.1", host: "testsite.com", referrer:
"http://testsite.com/?app_view=core.ItemAdmin&app_subView=core.ItemAdd&app_addPlugin=ItemAddFromServer&app_form%5BlocalServerPath%5D=/home/user/testsite.com/public_html/share/data/505&app_itemId=470&app_form%5Baction%5D%5BfindFilesFromLocalServer%5D=1&app_form%5BformName%5D=ItemAddFromServer"

[alert] 17670#0: *1 Alert:- ------WebKitFormBoundary3c6CizwarDCoyCst
Content-Disposition: form-data; name ::: "app_return"

//?app_view=core.ItemAdmin, client: 89.148.6.249, server:
testsite.com, request: "POST // HTTP/1.1", host: "testsite.com",
referrer: "http://testsite.com/?app_view=core.ItemAdmin&app_subView=core.ItemAdd&app_addPlugin=ItemAddFromServer&app_form%5BlocalServerPath%5D=/home/user/testsite.com/public_html/share/data/505&app_itemId=470&app_form%5Baction%5D%5BfindFilesFromLocalServer%5D=1&app_form%5BformName%5D=ItemAddFromServer"

[alert] 17670#0: *1 Alert:- app_subView ::: core.ItemAdd core.ItemAdd,
client: 89.148.6.249, server: testsite.com, request: "POST //
HTTP/1.1", host: "testsite.com", referrer:
"http://testsite.com/?app_view=core.ItemAdmin&app_subView=core.ItemAdd&app_addPlugin=ItemAddFromServer&app_form%5BlocalServerPath%5D=/home/user/testsite.com/public_html/share/data/505&app_itemId=470&app_form%5Baction%5D%5BfindFilesFromLocalServer%5D=1&app_form%5BformName%5D=ItemAddFromServer"

[alert] 17670#0: *1 Alert:- app_addPlugin ::: ItemAddFromServer
ItemAddFromServer, client: 89.148.6.249, server: testsite.com,
request: "POST // HTTP/1.1", host: "testsite.com", referrer:
"http://testsite.com/?app_view=core.ItemAdmin&app_subView=core.ItemAdd&app_addPlugin=ItemAddFromServer&app_form%5BlocalServerPath%5D=/home/user/testsite.com/public_html/share/data/505&app_itemId=470&app_form%5Baction%5D%5BfindFilesFromLocalServer%5D=1&app_form%5BformName%5D=ItemAddFromServer"
-************************
           END
-************************


1. The concat command suggests that when "v" is a table, the table
holds the same value twice. E.G. "[alert] 17670#0: *1 Alert:-
app_form[action][findFilesFromLocalServer] ::: 1 1," and "[alert]
17670#0: *1 Alert:- app_form[localServerPath] :::
/home/user/testsite.com/public_html/share/data/505
/home/user/testsite.com/public_html/share/data/505,"

2. The "------WebKitFormBoundary3c6CizwarDCoyCst" lines seem to
suggest the request body has been somehow attached to one of the
variables "app_itemId". This should just have a value of 140.


I have posted the form that generates this at:
http://pastebin.com/6scK6cWP. (Pls widen page view to see properly)

Thanks


From agentzh at gmail.com  Sun Nov  6 02:38:53 2011
From: agentzh at gmail.com (agentzh)
Date: Sun, 6 Nov 2011 10:38:53 +0800
Subject: ngx.req.get_post_args Issue
In-Reply-To: <CAN=WzV25TdtiEW99gBYRoGj2V_-j9YZKZ39b=G=f4Nh6637ORg@mail.gmail.com>
References: <CAN=WzV25TdtiEW99gBYRoGj2V_-j9YZKZ39b=G=f4Nh6637ORg@mail.gmail.com>
Message-ID: <CAB4Tn6NuR-5Cqqz27HE0jYj1KqYLvujkruR9xW6crV7v2HgXRA@mail.gmail.com>

On Sun, Nov 6, 2011 at 12:04 AM, Nginx User <nginx at nginxuser.net> wrote:
> [alert] 17670#0: *1 Alert:- app_itemId ::: 470
> ------WebKitFormBoundary3c6CizwarDCoyCst
> Content-Disposition: form-data; name="app_formUrl"
>
> //?app_view=core.ItemAdmin 470
> ------WebKitFormBoundary3c6CizwarDCoyCst
> Content-Disposition: form-data; name="app_authToken"
>

Multipart form format is not supported in ngx.req.get_post_args. Only
urlencoded format is supported for now. I'll make this clear in the
documentation.

Regards,
-agentzh


From nginx at nginxuser.net  Sun Nov  6 07:12:28 2011
From: nginx at nginxuser.net (Nginx User)
Date: Sun, 6 Nov 2011 10:12:28 +0300
Subject: ngx.req.get_post_args Issue
In-Reply-To: <CAB4Tn6NuR-5Cqqz27HE0jYj1KqYLvujkruR9xW6crV7v2HgXRA@mail.gmail.com>
References: <CAN=WzV25TdtiEW99gBYRoGj2V_-j9YZKZ39b=G=f4Nh6637ORg@mail.gmail.com>
	<CAB4Tn6NuR-5Cqqz27HE0jYj1KqYLvujkruR9xW6crV7v2HgXRA@mail.gmail.com>
Message-ID: <CAN=WzV0OdNjnqpVi2q0UTa0Hr_-M-=hzy5XewWCgG7JPqNs1Wg@mail.gmail.com>

On 6 November 2011 05:38, agentzh <agentzh at gmail.com> wrote:
> Multipart form format is not supported in ngx.req.get_post_args. Only
> urlencoded format is supported for now. I'll make this clear in the
> documentation.

Thanks for the clarification..

As I need to have a mix of both types, I have combined three files
from the cgilua package into one module (cgilua_module) and using this
instead.

Handles both multipart and urlencoded formats and outputs a table of
key/pair values. (Also has "application/xml", "text/xml", and
"text/plain" under the POST request handler. Not yet sure where these
fit in and just left them in)

My code (work in progress) is now
-************************
          START
-************************
local ngx_cgi = require "cgilua_module"
if ngx.var.request_method == "POST" then
       ngx.req.read_body()
       local post_args = ngx_cgi.get_post_args()
       for my_key, v in pairs( post_args ) do
               if type(v) == "table" then
                       my_arg = table.concat(v, " ")
               else
                       my_arg = v
               end
               regex_rules.log_alert("Alert:- " .. my_key .. " ::: " .. my_arg)
       end
end
-************************
          END
-************************

Since cgilua has a permissive license, perhaps this, although in lua
and not C, will be helpful in extending the coverage of
ngx.req.get_post_args to multipart encoded forms.

Cheers!


From nginx at nginxuser.net  Sun Nov  6 10:20:45 2011
From: nginx at nginxuser.net (Nginx User)
Date: Sun, 6 Nov 2011 13:20:45 +0300
Subject: ngx.req.get_post_args Issue
In-Reply-To: <CAN=WzV0OdNjnqpVi2q0UTa0Hr_-M-=hzy5XewWCgG7JPqNs1Wg@mail.gmail.com>
References: <CAN=WzV25TdtiEW99gBYRoGj2V_-j9YZKZ39b=G=f4Nh6637ORg@mail.gmail.com>
	<CAB4Tn6NuR-5Cqqz27HE0jYj1KqYLvujkruR9xW6crV7v2HgXRA@mail.gmail.com>
	<CAN=WzV0OdNjnqpVi2q0UTa0Hr_-M-=hzy5XewWCgG7JPqNs1Wg@mail.gmail.com>
Message-ID: <CAN=WzV0F0+_tp6OhmoTvMPAbdNnKGSZuTy_SZ3ZYVay-ZU7B6A@mail.gmail.com>

On 6 November 2011 10:12, Nginx User <nginx at nginxuser.net> wrote:
> My code (work in progress) is now
> -************************
> ? ? ? ? ?START
> -************************
> local ngx_cgi = require "cgilua_module"
> if ngx.var.request_method == "POST" then
> ? ? ? ngx.req.read_body()
> ? ? ? local post_args = ngx_cgi.get_post_args()
> ? ? ? for my_key, v in pairs( post_args ) do
> ? ? ? ? ? ? ? if type(v) == "table" then
> ? ? ? ? ? ? ? ? ? ? ? my_arg = table.concat(v, " ")
> ? ? ? ? ? ? ? else
> ? ? ? ? ? ? ? ? ? ? ? my_arg = v
> ? ? ? ? ? ? ? end
> ? ? ? ? ? ? ? regex_rules.log_alert("Alert:- " .. my_key .. " ::: " .. my_arg)
> ? ? ? end
> end
> -************************
> ? ? ? ? ?END
> -************************

UPDATED

-************************
         START
-************************
if ngx.var.request_method == "POST" then
      -- ngx.req.read_body() -- Not needed as cgilua handles this
      local ngx_cgi = require "cgilua_module"
      local post_args = ngx_cgi.get_post_args()
      for my_key, v in pairs( post_args ) do
              if type(v) == "table" then
                      my_arg = table.concat(v, " ")
              else
                      my_arg = v
              end
              regex_rules.log_alert("Alert:- " .. my_key .. " ::: " .. my_arg)
      end
end
-************************
         END
-************************


From kgorlo at gmail.com  Sun Nov  6 20:48:08 2011
From: kgorlo at gmail.com (Kamil Gorlo)
Date: Sun, 6 Nov 2011 21:48:08 +0100
Subject: Log response header from upstream
Message-ID: <CAHzX=KBmF1nu5bN1B6yWoh+J-bTwJE93FpywpKRzxPu+CFEZGw@mail.gmail.com>

Hi,

I have some problem with access log combined with X-Accel-Redirect
requests. In my case I have Nginx set up as load-balancer to group of
application servers. These servers return some special header in every
request - I need to log value of this special header (lets call it
'X-user') in access log - also I do not want to expose this header to
the world (proxy_hide_header helps here).

Everything seems to work, but when there is X-Accel-Redirect request I
have empty field in access log because of subrequest
($upstream_http_x_special is cleared because of subrequest, if I
understand this mechanism correctly). How to make this work for every
request?

Here is my config:

http {
  log_format extended '$request $upstream_http_x_user';
  access_log /var/log/nginx/access.log extended;

  ...

  server {
    listen 80;

    location / {
      proxy_pass http://backend;
      proxy_hide_header X-User;
    }

    location /files {
      internal;
      proxy_pass http://filestore;
    }
  }
}


Cheers,

--
Kamil Gorlo


From mdounin at mdounin.ru  Sun Nov  6 21:15:04 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Mon, 7 Nov 2011 01:15:04 +0400
Subject: Log response header from upstream
In-Reply-To: <CAHzX=KBmF1nu5bN1B6yWoh+J-bTwJE93FpywpKRzxPu+CFEZGw@mail.gmail.com>
References: <CAHzX=KBmF1nu5bN1B6yWoh+J-bTwJE93FpywpKRzxPu+CFEZGw@mail.gmail.com>
Message-ID: <20111106211504.GY95664@mdounin.ru>

Hello!

On Sun, Nov 06, 2011 at 09:48:08PM +0100, Kamil Gorlo wrote:

> Hi,
> 
> I have some problem with access log combined with X-Accel-Redirect
> requests. In my case I have Nginx set up as load-balancer to group of
> application servers. These servers return some special header in every
> request - I need to log value of this special header (lets call it
> 'X-user') in access log - also I do not want to expose this header to
> the world (proxy_hide_header helps here).
> 
> Everything seems to work, but when there is X-Accel-Redirect request I
> have empty field in access log because of subrequest
> ($upstream_http_x_special is cleared because of subrequest, if I
> understand this mechanism correctly). How to make this work for every
> request?
> 
> Here is my config:
> 
> http {
>   log_format extended '$request $upstream_http_x_user';
>   access_log /var/log/nginx/access.log extended;
> 
>   ...
> 
>   server {
>     listen 80;
> 
>     location / {
>       proxy_pass http://backend;
>       proxy_hide_header X-User;
>     }
> 
>     location /files {
>       internal;
>       proxy_pass http://filestore;

Workaround is to use

        set $x_user $upstream_http_x_user;

here (and to log $x_user instead).

>     }
>   }
> }


Maxim Dounin


From nginx-forum at nginx.us  Mon Nov  7 05:34:13 2011
From: nginx-forum at nginx.us (Mark)
Date: Mon, 07 Nov 2011 00:34:13 -0500
Subject: Rewrite /index.html to /
Message-ID: <25358168f10d160cb15bb45891cbfc78.NginxMailingListEnglish@forum.nginx.org>

Developing static files on the local filesystem I use index.html instead
of / for the home page.
When I push them to the remote server I want
http://domain.com/index.html to be redirected to http://domain.com/ for
SEO reasons.
I've Googled but the best I seem to be able to achieve is a redirect
loop. What do I need to do?

Here are my failed rules :

#  rewrite ^index.html$ $scheme://domain.com/ permanent;
#  rewrite /index.html $scheme://domain.com/ permanent;
#  location = /index.html {
#    rewrite ^(.*) $scheme://domain.com/ permanent;
#  }

Thanks.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,217899,217899#msg-217899


From kgorlo at gmail.com  Mon Nov  7 05:42:09 2011
From: kgorlo at gmail.com (Kamil Gorlo)
Date: Mon, 7 Nov 2011 06:42:09 +0100
Subject: Log response header from upstream
In-Reply-To: <20111106211504.GY95664@mdounin.ru>
References: <CAHzX=KBmF1nu5bN1B6yWoh+J-bTwJE93FpywpKRzxPu+CFEZGw@mail.gmail.com>
	<20111106211504.GY95664@mdounin.ru>
Message-ID: <CAHzX=KDiJFfCxZWsrkV2_D8oZZSdVn=9u2u+BxU77_dwxm00Cg@mail.gmail.com>

Yeah, I have tried this, but this way non x-accel-redirect requests must be
handled in old way which means I have to use 2 fields in access log. Is
there any way to use only one field?
06-11-2011 22:15 u?ytkownik "Maxim Dounin" <mdounin at mdounin.ru> napisa?:

> Hello!
>
> On Sun, Nov 06, 2011 at 09:48:08PM +0100, Kamil Gorlo wrote:
>
> > Hi,
> >
> > I have some problem with access log combined with X-Accel-Redirect
> > requests. In my case I have Nginx set up as load-balancer to group of
> > application servers. These servers return some special header in every
> > request - I need to log value of this special header (lets call it
> > 'X-user') in access log - also I do not want to expose this header to
> > the world (proxy_hide_header helps here).
> >
> > Everything seems to work, but when there is X-Accel-Redirect request I
> > have empty field in access log because of subrequest
> > ($upstream_http_x_special is cleared because of subrequest, if I
> > understand this mechanism correctly). How to make this work for every
> > request?
> >
> > Here is my config:
> >
> > http {
> >   log_format extended '$request $upstream_http_x_user';
> >   access_log /var/log/nginx/access.log extended;
> >
> >   ...
> >
> >   server {
> >     listen 80;
> >
> >     location / {
> >       proxy_pass http://backend;
> >       proxy_hide_header X-User;
> >     }
> >
> >     location /files {
> >       internal;
> >       proxy_pass http://filestore;
>
> Workaround is to use
>
>        set $x_user $upstream_http_x_user;
>
> here (and to log $x_user instead).
>
> >     }
> >   }
> > }
>
>
> 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/20111107/3789c291/attachment.html>

From igor at sysoev.ru  Mon Nov  7 07:24:36 2011
From: igor at sysoev.ru (Igor Sysoev)
Date: Mon, 7 Nov 2011 11:24:36 +0400
Subject: Rewrite /index.html to /
In-Reply-To: <25358168f10d160cb15bb45891cbfc78.NginxMailingListEnglish@forum.nginx.org>
References: <25358168f10d160cb15bb45891cbfc78.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20111107072436.GA75036@nginx.com>

On Mon, Nov 07, 2011 at 12:34:13AM -0500, Mark wrote:
> Developing static files on the local filesystem I use index.html instead
> of / for the home page.
> When I push them to the remote server I want
> http://domain.com/index.html to be redirected to http://domain.com/ for
> SEO reasons.
> I've Googled but the best I seem to be able to achieve is a redirect
> loop. What do I need to do?
> 
> Here are my failed rules :
> 
> #  rewrite ^index.html$ $scheme://domain.com/ permanent;
> #  rewrite /index.html $scheme://domain.com/ permanent;
> #  location = /index.html {
> #    rewrite ^(.*) $scheme://domain.com/ permanent;
> #  }
> 
> Thanks.

Either

location = / {
    try_files   /index.html  =404;
}

location = /index.html {
    internal;
    error_page  404  =301  $scheme://domain.com/;
}

or

location = / {
    index  index.html;
}

location = /index.html {
    internal;
    error_page  404  =301  $scheme://domain.com/;
}


-- 
Igor Sysoev


From nginx-forum at nginx.us  Mon Nov  7 07:55:22 2011
From: nginx-forum at nginx.us (mikiso)
Date: Mon, 07 Nov 2011 02:55:22 -0500
Subject: Calculating the max. clients by worker_connections
In-Reply-To: <819481.24136.qm@web120511.mail.ne1.yahoo.com>
References: <819481.24136.qm@web120511.mail.ne1.yahoo.com>
Message-ID: <05599f0f70e8bc3be50bb4cac31a4166.NginxMailingListEnglish@forum.nginx.org>

Hi, 

Regarding max clients , I understand it's normal to have 4 in reverse
proxy
since  browser  uses 2 connections in HTTP 1.1.
How about mail reverse proxy?
It seems POP command from telnet client uses 4 also, although I couldn't
find 
POP3 uses more than 1 connection in RFC 1939.

Regards,
Soichiro

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,171776,217904#msg-217904


From nginx-forum at nginx.us  Mon Nov  7 08:48:53 2011
From: nginx-forum at nginx.us (roger.moffatt)
Date: Mon, 07 Nov 2011 03:48:53 -0500
Subject: nginx config: multiple locations, authentication in one,
	triggered for both?
Message-ID: <1800a639381b1c7a0d427faad62c9f16.NginxMailingListEnglish@forum.nginx.org>

I originally posted this question on SO, but it might of course be more
logical to ask here;

http://stackoverflow.com/questions/8031471/nginx-location-directive-authentication-happening-in-wrong-location-block

I'm flummoxed.

I have a server that is primarily running couchdb over ssl (using nginx
to proxy the ssl connection) but also has to serve some apache stuff.

Basically I want everything that DOESN'T start /www to be sent to the
couchdb backend. If a url DOES start /www then it should be mapped to
the local apache server on port 8080.

My config below works with the exception that I'm getting prompted for
authentication on the /www paths as well. I'm a bit more used to
configuring Apache than nginx, so I suspect I'm mis-understanding
something, but if anyone can see what is wrong from my configuration
(below) I'd be most grateful.

To clarify my use scenario;

https://my-domain.com/www/script.cgi should be proxied to
http://localhost:8080/script.cgi
https://my-domain.com/anythingelse should be proxied to
http://localhost:5984/anythingelse

ONLY the second should require authentication. It is the authentication
issue that is causing problems - as I mentioned, I am being challenged
on https://my-domain.com/www/anything as well :-(

Here's the config, thanks for any insight.

server {
        listen   443;
        ssl on;

        # Any url starting /www needs to be mapped to the root
        # of the back end application server on 8080

        location ^~ /www/ {
        proxy_pass http://localhost:8080/;
        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;

        }

        # Everything else has to be sent to the couchdb server running
on
        # port 5984 and for security, this is protected with auth_basic
        # authentication.

        location / {

        auth_basic "Restricted";
        auth_basic_user_file /path-to-passwords;

        proxy_pass http://localhost:5984;
        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;
        proxy_set_header X-Forwarded-Ssl on;

        }
    }

Thanks for some pointers - I'm not sure how I can resolve this
correctly.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,217906,217906#msg-217906


From nginx-forum at nginx.us  Mon Nov  7 09:34:41 2011
From: nginx-forum at nginx.us (attiks)
Date: Mon, 07 Nov 2011 04:34:41 -0500
Subject: Finer grained control on caching
Message-ID: <f2c2f8888d846d698230f869d0f86296.NginxMailingListEnglish@forum.nginx.org>

I posted and idea last week to Ideas and Feature Requests: "Finer
grained control on caching" (http://forum.nginx.org/read.php?10,217748),
but since I'm new to this forum/mailing list I have no idea if anybody
saw this.

I want to find out if this is an acceptable feature request and if
anybody else sees the benefits of this.

Short summary:
Add a new directive fastcgi_cache_key_suffix
Construct the caching filename like this md5(fastcgi_cache_key) . '_' .
fastcgi_cache_key_suffix

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,217909,217909#msg-217909


From mdounin at mdounin.ru  Mon Nov  7 10:19:03 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Mon, 7 Nov 2011 14:19:03 +0400
Subject: Log response header from upstream
In-Reply-To: <CAHzX=KDiJFfCxZWsrkV2_D8oZZSdVn=9u2u+BxU77_dwxm00Cg@mail.gmail.com>
References: <CAHzX=KBmF1nu5bN1B6yWoh+J-bTwJE93FpywpKRzxPu+CFEZGw@mail.gmail.com>
	<20111106211504.GY95664@mdounin.ru>
	<CAHzX=KDiJFfCxZWsrkV2_D8oZZSdVn=9u2u+BxU77_dwxm00Cg@mail.gmail.com>
Message-ID: <20111107101902.GD95664@mdounin.ru>

Hello!

On Mon, Nov 07, 2011 at 06:42:09AM +0100, Kamil Gorlo wrote:

> Yeah, I have tried this, but this way non x-accel-redirect requests must be
> handled in old way which means I have to use 2 fields in access log. Is
> there any way to use only one field?

E.g. you may use different log format for internal location.

Maxim Dounin

> 06-11-2011 22:15 u?ytkownik "Maxim Dounin" <mdounin at mdounin.ru> napisa?:
> 
> > Hello!
> >
> > On Sun, Nov 06, 2011 at 09:48:08PM +0100, Kamil Gorlo wrote:
> >
> > > Hi,
> > >
> > > I have some problem with access log combined with X-Accel-Redirect
> > > requests. In my case I have Nginx set up as load-balancer to group of
> > > application servers. These servers return some special header in every
> > > request - I need to log value of this special header (lets call it
> > > 'X-user') in access log - also I do not want to expose this header to
> > > the world (proxy_hide_header helps here).
> > >
> > > Everything seems to work, but when there is X-Accel-Redirect request I
> > > have empty field in access log because of subrequest
> > > ($upstream_http_x_special is cleared because of subrequest, if I
> > > understand this mechanism correctly). How to make this work for every
> > > request?
> > >
> > > Here is my config:
> > >
> > > http {
> > >   log_format extended '$request $upstream_http_x_user';
> > >   access_log /var/log/nginx/access.log extended;
> > >
> > >   ...
> > >
> > >   server {
> > >     listen 80;
> > >
> > >     location / {
> > >       proxy_pass http://backend;
> > >       proxy_hide_header X-User;
> > >     }
> > >
> > >     location /files {
> > >       internal;
> > >       proxy_pass http://filestore;
> >
> > Workaround is to use
> >
> >        set $x_user $upstream_http_x_user;
> >
> > here (and to log $x_user instead).
> >
> > >     }
> > >   }
> > > }
> >
> >
> > 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  Mon Nov  7 10:25:26 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Mon, 7 Nov 2011 14:25:26 +0400
Subject: nginx config: multiple locations, authentication in one,
	triggered for both?
In-Reply-To: <1800a639381b1c7a0d427faad62c9f16.NginxMailingListEnglish@forum.nginx.org>
References: <1800a639381b1c7a0d427faad62c9f16.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20111107102526.GE95664@mdounin.ru>

Hello!

On Mon, Nov 07, 2011 at 03:48:53AM -0500, roger.moffatt wrote:

> I originally posted this question on SO, but it might of course be more
> logical to ask here;
> 
> http://stackoverflow.com/questions/8031471/nginx-location-directive-authentication-happening-in-wrong-location-block
> 
> I'm flummoxed.
> 
> I have a server that is primarily running couchdb over ssl (using nginx
> to proxy the ssl connection) but also has to serve some apache stuff.
> 
> Basically I want everything that DOESN'T start /www to be sent to the
> couchdb backend. If a url DOES start /www then it should be mapped to
> the local apache server on port 8080.
> 
> My config below works with the exception that I'm getting prompted for
> authentication on the /www paths as well. I'm a bit more used to
> configuring Apache than nginx, so I suspect I'm mis-understanding
> something, but if anyone can see what is wrong from my configuration
> (below) I'd be most grateful.
> 
> To clarify my use scenario;
> 
> https://my-domain.com/www/script.cgi should be proxied to
> http://localhost:8080/script.cgi
> https://my-domain.com/anythingelse should be proxied to
> http://localhost:5984/anythingelse
> 
> ONLY the second should require authentication. It is the authentication
> issue that is causing problems - as I mentioned, I am being challenged
> on https://my-domain.com/www/anything as well :-(

Most likely, the authentication request appears due to your 
browser doing automatic requests to /favicon.ico or something 
like.  Try adding

    location = /favicon.ico {
        return 404;
    }

to see if it helps.

> 
> Here's the config, thanks for any insight.
> 
> server {
>         listen   443;
>         ssl on;
> 
>         # Any url starting /www needs to be mapped to the root
>         # of the back end application server on 8080
> 
>         location ^~ /www/ {
>         proxy_pass http://localhost:8080/;
>         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;
> 
>         }
> 
>         # Everything else has to be sent to the couchdb server running
> on
>         # port 5984 and for security, this is protected with auth_basic
>         # authentication.
> 
>         location / {
> 
>         auth_basic "Restricted";
>         auth_basic_user_file /path-to-passwords;
> 
>         proxy_pass http://localhost:5984;
>         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;
>         proxy_set_header X-Forwarded-Ssl on;
> 
>         }
>     }
> 
> Thanks for some pointers - I'm not sure how I can resolve this
> correctly.

Config looks correct and should work.  Try testing it by hand 
(e.g. nc/telnet/fetch/wget/curl) to see if it actually works.  See 
above for a possible cause of the authentication request.

Maxim Dounin


From kgorlo at gmail.com  Mon Nov  7 10:57:15 2011
From: kgorlo at gmail.com (Kamil Gorlo)
Date: Mon, 7 Nov 2011 11:57:15 +0100
Subject: Log response header from upstream
In-Reply-To: <20111107101902.GD95664@mdounin.ru>
References: <CAHzX=KBmF1nu5bN1B6yWoh+J-bTwJE93FpywpKRzxPu+CFEZGw@mail.gmail.com>
	<20111106211504.GY95664@mdounin.ru>
	<CAHzX=KDiJFfCxZWsrkV2_D8oZZSdVn=9u2u+BxU77_dwxm00Cg@mail.gmail.com>
	<20111107101902.GD95664@mdounin.ru>
Message-ID: <CAHzX=KAC7iwJ2cZap_KQ+ORSjhOmq06qT5_vkiVyYciyWm-Dew@mail.gmail.com>

OK, thanks - it works.

One more question: I have to use this variables in quotes because
sometimes it resolves to just empty string and there is problem with
parsing such logs.

Why sometimes empty variables ends in single hyphen ("-", this is what
I allways want) in access log and sometimes its just empty string
("")?

--
Kamil Gorlo



On Mon, Nov 7, 2011 at 11:19 AM, Maxim Dounin <mdounin at mdounin.ru> wrote:
> Hello!
>
> On Mon, Nov 07, 2011 at 06:42:09AM +0100, Kamil Gorlo wrote:
>
>> Yeah, I have tried this, but this way non x-accel-redirect requests must be
>> handled in old way which means I have to use 2 fields in access log. Is
>> there any way to use only one field?
>
> E.g. you may use different log format for internal location.
>
> Maxim Dounin
>
>> 06-11-2011 22:15 u?ytkownik "Maxim Dounin" <mdounin at mdounin.ru> napisa?:
>>
>> > Hello!
>> >
>> > On Sun, Nov 06, 2011 at 09:48:08PM +0100, Kamil Gorlo wrote:
>> >
>> > > Hi,
>> > >
>> > > I have some problem with access log combined with X-Accel-Redirect
>> > > requests. In my case I have Nginx set up as load-balancer to group of
>> > > application servers. These servers return some special header in every
>> > > request - I need to log value of this special header (lets call it
>> > > 'X-user') in access log - also I do not want to expose this header to
>> > > the world (proxy_hide_header helps here).
>> > >
>> > > Everything seems to work, but when there is X-Accel-Redirect request I
>> > > have empty field in access log because of subrequest
>> > > ($upstream_http_x_special is cleared because of subrequest, if I
>> > > understand this mechanism correctly). How to make this work for every
>> > > request?
>> > >
>> > > Here is my config:
>> > >
>> > > http {
>> > > ? log_format extended '$request $upstream_http_x_user';
>> > > ? access_log /var/log/nginx/access.log extended;
>> > >
>> > > ? ...
>> > >
>> > > ? server {
>> > > ? ? listen 80;
>> > >
>> > > ? ? location / {
>> > > ? ? ? proxy_pass http://backend;
>> > > ? ? ? proxy_hide_header X-User;
>> > > ? ? }
>> > >
>> > > ? ? location /files {
>> > > ? ? ? internal;
>> > > ? ? ? proxy_pass http://filestore;
>> >
>> > Workaround is to use
>> >
>> > ? ? ? ?set $x_user $upstream_http_x_user;
>> >
>> > here (and to log $x_user instead).
>> >
>> > > ? ? }
>> > > ? }
>> > > }
>> >
>> >
>> > 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 nginx-forum at nginx.us  Mon Nov  7 11:21:05 2011
From: nginx-forum at nginx.us (imran)
Date: Mon, 07 Nov 2011 06:21:05 -0500
Subject: Can I log HTTP 400s to a different log file instead of my default
	access logs
Message-ID: <ef07edb86e3d3d3f7c88967231475f10.NginxMailingListEnglish@forum.nginx.org>

Hello

I'd like to know if its possible for me to log HTTP 400 errors to a
different log file. Right now they go in the access log. 

To give some background, we've got this problem where if a single
resource is requested from chrome, after the resource request is served,
a http 400 error is logged in the access logs. This I've understood to
be is due to the nature of connections opened by Chrome (2 for every
request) and one of the connections not been used before its closed. And
nginx reports a 400 in this instance. I've gathered this based on a
different forum entry. And when this 400 is logged, the
$request_filename is /etc/nginx//html

So I'm looking at any possibility where the 400s can be logged else
where. I tried the following with no luck as well;

server {
  location / {
      if ($request_filename ~* /etc/nginx) {
         access_log /someotherlog.log myformat
      }
  }
}

Any help/advice is much appreciated. Thanks!!

Cheers
-- Imran

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,217913,217913#msg-217913


From mdounin at mdounin.ru  Mon Nov  7 11:29:11 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Mon, 7 Nov 2011 15:29:11 +0400
Subject: Log response header from upstream
In-Reply-To: <CAHzX=KAC7iwJ2cZap_KQ+ORSjhOmq06qT5_vkiVyYciyWm-Dew@mail.gmail.com>
References: <CAHzX=KBmF1nu5bN1B6yWoh+J-bTwJE93FpywpKRzxPu+CFEZGw@mail.gmail.com>
	<20111106211504.GY95664@mdounin.ru>
	<CAHzX=KDiJFfCxZWsrkV2_D8oZZSdVn=9u2u+BxU77_dwxm00Cg@mail.gmail.com>
	<20111107101902.GD95664@mdounin.ru>
	<CAHzX=KAC7iwJ2cZap_KQ+ORSjhOmq06qT5_vkiVyYciyWm-Dew@mail.gmail.com>
Message-ID: <20111107112911.GI95664@mdounin.ru>

Hello!

On Mon, Nov 07, 2011 at 11:57:15AM +0100, Kamil Gorlo wrote:

> OK, thanks - it works.
> 
> One more question: I have to use this variables in quotes because
> sometimes it resolves to just empty string and there is problem with
> parsing such logs.
> 
> Why sometimes empty variables ends in single hyphen ("-", this is what
> I allways want) in access log and sometimes its just empty string
> ("")?

Hyphen is used when variable isn't found.  Empty string - if it's 
found, but empty.

Maxim Dounin

> 
> --
> Kamil Gorlo
> 
> 
> 
> On Mon, Nov 7, 2011 at 11:19 AM, Maxim Dounin <mdounin at mdounin.ru> wrote:
> > Hello!
> >
> > On Mon, Nov 07, 2011 at 06:42:09AM +0100, Kamil Gorlo wrote:
> >
> >> Yeah, I have tried this, but this way non x-accel-redirect requests must be
> >> handled in old way which means I have to use 2 fields in access log. Is
> >> there any way to use only one field?
> >
> > E.g. you may use different log format for internal location.
> >
> > Maxim Dounin
> >
> >> 06-11-2011 22:15 u?ytkownik "Maxim Dounin" <mdounin at mdounin.ru> napisa?:
> >>
> >> > Hello!
> >> >
> >> > On Sun, Nov 06, 2011 at 09:48:08PM +0100, Kamil Gorlo wrote:
> >> >
> >> > > Hi,
> >> > >
> >> > > I have some problem with access log combined with X-Accel-Redirect
> >> > > requests. In my case I have Nginx set up as load-balancer to group of
> >> > > application servers. These servers return some special header in every
> >> > > request - I need to log value of this special header (lets call it
> >> > > 'X-user') in access log - also I do not want to expose this header to
> >> > > the world (proxy_hide_header helps here).
> >> > >
> >> > > Everything seems to work, but when there is X-Accel-Redirect request I
> >> > > have empty field in access log because of subrequest
> >> > > ($upstream_http_x_special is cleared because of subrequest, if I
> >> > > understand this mechanism correctly). How to make this work for every
> >> > > request?
> >> > >
> >> > > Here is my config:
> >> > >
> >> > > http {
> >> > > ? log_format extended '$request $upstream_http_x_user';
> >> > > ? access_log /var/log/nginx/access.log extended;
> >> > >
> >> > > ? ...
> >> > >
> >> > > ? server {
> >> > > ? ? listen 80;
> >> > >
> >> > > ? ? location / {
> >> > > ? ? ? proxy_pass http://backend;
> >> > > ? ? ? proxy_hide_header X-User;
> >> > > ? ? }
> >> > >
> >> > > ? ? location /files {
> >> > > ? ? ? internal;
> >> > > ? ? ? proxy_pass http://filestore;
> >> >
> >> > Workaround is to use
> >> >
> >> > ? ? ? ?set $x_user $upstream_http_x_user;
> >> >
> >> > here (and to log $x_user instead).
> >> >
> >> > > ? ? }
> >> > > ? }
> >> > > }
> >> >
> >> >
> >> > 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 nginx-forum at nginx.us  Mon Nov  7 11:33:06 2011
From: nginx-forum at nginx.us (Mark)
Date: Mon, 07 Nov 2011 06:33:06 -0500
Subject: Rewrite /index.html to /
In-Reply-To: <20111107072436.GA75036@nginx.com>
References: <20111107072436.GA75036@nginx.com>
Message-ID: <17abaaff9bd33c5e96dd8651f9e7d368.NginxMailingListEnglish@forum.nginx.org>

Perfect!

Many thanks.

??????? ???????.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,217899,217915#msg-217915


From nginx-forum at nginx.us  Mon Nov  7 11:38:01 2011
From: nginx-forum at nginx.us (imran)
Date: Mon, 07 Nov 2011 06:38:01 -0500
Subject: Can I log HTTP 400s to a different log file instead of my default
	access logs
In-Reply-To: <ef07edb86e3d3d3f7c88967231475f10.NginxMailingListEnglish@forum.nginx.org>
References: <ef07edb86e3d3d3f7c88967231475f10.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <a6ad5afa26f540ac1becad3c33e7399f.NginxMailingListEnglish@forum.nginx.org>

Also to add on a bit more about the chrome connection issue which is
causing the 400, I turned on info logging for error logs and when I get
the 400, it logs a 'client closed prematurely connection while reading
client request...' in the error logs.

Cheers
-- Imran

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,217913,217916#msg-217916


From appa at perusio.net  Mon Nov  7 11:38:54 2011
From: appa at perusio.net (=?UTF-8?B?QW50w7NuaW8=?= P. P. Almeida)
Date: Mon, 07 Nov 2011 11:38:54 +0000
Subject: Can I log HTTP 400s to a different log file instead of my
	default	access logs
In-Reply-To: <ef07edb86e3d3d3f7c88967231475f10.NginxMailingListEnglish@forum.nginx.org>
References: <ef07edb86e3d3d3f7c88967231475f10.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <87bosonovl.wl%appa@perusio.net>

On 7 Nov 2011 11h21 WET, nginx-forum at nginx.us wrote:

> Hello
>
> I'd like to know if its possible for me to log HTTP 400 errors to a
> different log file. Right now they go in the access log. 
>
> To give some background, we've got this problem where if a single
> resource is requested from chrome, after the resource request is
> served, a http 400 error is logged in the access logs. This I've
> understood to be is due to the nature of connections opened by
> Chrome (2 for every request) and one of the connections not been
> used before its closed. And nginx reports a 400 in this
> instance. I've gathered this based on a different forum entry. And
> when this 400 is logged, the $request_filename is /etc/nginx//html

Try (untested):

at the http level:

map $request_filename $is_400 {
   default 0;
   /etc/nginx/html 1; 
}

on the vhost config:

error_page 400 @log-400;

location @log-400 {

   access_log /path/to/400.log;

}

Caveat emptor: I've never played with the access_log directive that
much. 

--- appa


From nginx-forum at nginx.us  Mon Nov  7 11:48:59 2011
From: nginx-forum at nginx.us (roger.moffatt)
Date: Mon, 07 Nov 2011 06:48:59 -0500
Subject: nginx config: multiple locations, authentication in one,
	triggered for both?
In-Reply-To: <20111107102526.GE95664@mdounin.ru>
References: <20111107102526.GE95664@mdounin.ru>
Message-ID: <60cd72290ee1886f5f936a65de4d1920.NginxMailingListEnglish@forum.nginx.org>

Doh! Of course ... I had a note on my list about favicon showing the
wrong thing, and of course it was showing the wrong thing on my logged
in system precisely because of this!

>Most likely, the authentication request appears due to your
>browser doing automatic requests to /favicon.ico or something
>like.

I can't test it at present but I'm certain this will be the problem, It
makes me think that perhaps my config is a little dangerous so now I
know the approach is correct, I'll perhaps swap things around so that I
can keep / unprotected completely just in case and add the auth for
known paths to the couch back end. That should work fine in my case as I
only have a couple of databases to secure.

Many thanks Maxim!

Roger

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,217906,217918#msg-217918


From appa at perusio.net  Mon Nov  7 12:04:22 2011
From: appa at perusio.net (=?UTF-8?B?QW50w7NuaW8=?= P. P. Almeida)
Date: Mon, 07 Nov 2011 12:04:22 +0000
Subject: Can I log HTTP 400s to a different log file instead of
	my	default	access logs
In-Reply-To: <87bosonovl.wl%appa@perusio.net>
References: <ef07edb86e3d3d3f7c88967231475f10.NginxMailingListEnglish@forum.nginx.org>
	<87bosonovl.wl%appa@perusio.net>
Message-ID: <878vnsnnp5.wl%appa@perusio.net>

On 7 Nov 2011 11h38 WET, appa at perusio.net wrote:

> On 7 Nov 2011 11h21 WET, nginx-forum at nginx.us wrote:
>
>> Hello
>>
>> I'd like to know if its possible for me to log HTTP 400 errors to a
>> different log file. Right now they go in the access log. 
>>
>> To give some background, we've got this problem where if a single
>> resource is requested from chrome, after the resource request is
>> served, a http 400 error is logged in the access logs. This I've
>> understood to be is due to the nature of connections opened by
>> Chrome (2 for every request) and one of the connections not been
>> used before its closed. And nginx reports a 400 in this
>> instance. I've gathered this based on a different forum entry. And
>> when this 400 is logged, the $request_filename is /etc/nginx//html
>
> Try (untested):
>
> at the http level:
>
> map $request_filename $is_400 {
> default 0;
> /etc/nginx/html 1; 
> }

Oops. Solly I haven't my coffe yet :)

Try (untested):

at the http level:

map $request_filename $is_400 {
   default 0;
   /etc/nginx/html 1; 
}

On the vhost:

if ($is_400) {
  return 302 @log-400;
}

location @log-400 {
   access_log /path/to/400.log;
}

--- appa


From nginx-forum at nginx.us  Mon Nov  7 13:03:10 2011
From: nginx-forum at nginx.us (imran)
Date: Mon, 07 Nov 2011 08:03:10 -0500
Subject: Can I log HTTP 400s to a different log file instead of my default
	access logs
In-Reply-To: <a6ad5afa26f540ac1becad3c33e7399f.NginxMailingListEnglish@forum.nginx.org>
References: <ef07edb86e3d3d3f7c88967231475f10.NginxMailingListEnglish@forum.nginx.org>
	<a6ad5afa26f540ac1becad3c33e7399f.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <1ca59b34305b7f9c4a0bae770c7d2736.NginxMailingListEnglish@forum.nginx.org>

Hi

Thanks for your reply. I tried this but didn't have any luck with it.
Based on the error reported on the error log (client closed permaturely
connection while reading client request line), I'm wondering if i can
instead ignore logging those errors in the access logs. Would you if
that's possible? Thanks!!

Cheers
-- Imran

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,217913,217921#msg-217921


From appa at perusio.net  Mon Nov  7 14:19:34 2011
From: appa at perusio.net (=?UTF-8?B?QW50w7NuaW8=?= P. P. Almeida)
Date: Mon, 07 Nov 2011 14:19:34 +0000
Subject: Can I log HTTP 400s to a different log file instead of my
	default	access logs
In-Reply-To: <1ca59b34305b7f9c4a0bae770c7d2736.NginxMailingListEnglish@forum.nginx.org>
References: <ef07edb86e3d3d3f7c88967231475f10.NginxMailingListEnglish@forum.nginx.org>
	<a6ad5afa26f540ac1becad3c33e7399f.NginxMailingListEnglish@forum.nginx.org>
	<1ca59b34305b7f9c4a0bae770c7d2736.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <877h3cnhft.wl%appa@perusio.net>

On 7 Nov 2011 13h03 WET, nginx-forum at nginx.us wrote:

Well you can try this:

location @is-400 {
  access_log off;
}

If you want to re-purpose the 400 code you have to use the error_page
directive.

--- appa


From nginx-forum at nginx.us  Mon Nov  7 14:43:43 2011
From: nginx-forum at nginx.us (mikolajj)
Date: Mon, 07 Nov 2011 09:43:43 -0500
Subject: problem with url http://x.x.x.x/folder_name/index.php/css/column.css
Message-ID: <7646443abbb1e9ceb9028a179f308ca6.NginxMailingListEnglish@forum.nginx.org>

Problematic URL is http://x.x.x.x/folder_name/index.php/css/column.css

Why apache and lighttpd can serve this url passing "/css/column.css" as
REQUEST_URI to index.php file.
Both in default configuration (right after clean apt-get installation on
debian)

...and nginx is throwing 404 not found error
What rewrite rules or other settings do I have to add/change to achieve
the same functionality? (because nginx is really much faster ;-) no
doubt about this) 

Regards
Mikolaj

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,217926,217926#msg-217926


From nginx-forum at nginx.us  Mon Nov  7 16:17:14 2011
From: nginx-forum at nginx.us (mikolajj)
Date: Mon, 07 Nov 2011 11:17:14 -0500
Subject: Rewrite problem http://x.x.x.x/folder_name/index.php/css/first.css
Message-ID: <974ef8502ef4834d08bd7bfd138fc699.NginxMailingListEnglish@forum.nginx.org>

Hello,
I made test against lighttpd and ngninx is 50% faster...

So I tried to use it with my already made application. (postgres &
php).
After setting evrything I found the URL to be a problem.

Apache and lighttpd can use URL
http://x.x.x.x/folder_name/index.php/css/first.css as
http://x.x.x.x/folder_name/index.php with REQUEST_URI part =
"/css/first.css" - this is handle by PHP and processed along...

Nginx is trying to open index.php as folder an of course 404 is
thrown...

I was trying to rewrite this URL but no success 

my /etc/nginx/sites-available/default
[code]
...
server {
                listen           80;
#               listen           443 ssl;
                server_name jakasnazwa.localhost;

                ssl off;
                ssl_certificate /etc/nginx/server.crt;
                ssl_certificate_key /etc/nginx/server.key;

                root /home/main-www;
                index index.php index.htm index.html;

                error_log /var/log/nginx/website.error_log notice;
                #rewrite_log on;

                location ~ .php$ {
                  fastcgi_pass   127.0.0.1:9000;
                  fastcgi_index  index.php;
                  fastcgi_param  SCRIPT_FILENAME
/home/main-www$fastcgi_script_name;
                  include fastcgi_params;
                }

                location / {
                  autoindex on;
                  if (-f $request_filename) {
                        break;
                  }

                  if (!-e $request_filename) {
                         rewrite ^(.*)/index.php/(.+)$ $1/index.php/$2
last;
                        break;
                  }
                }
       }
[/code]

Of course rewrite don't change anything (cut and paste, part of url in
the same place) but I have no idea how to force nginx to fire php-fpm na
index.php file passing everything what is after this as parameter?

Can you point me to some clues or solution?

PS I know that URL should look different, but application is ready, I
don't want to change it in this moment .

Regards
Mikolaj

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,217923,217923#msg-217923


From nginx-forum at nginx.us  Mon Nov  7 16:17:28 2011
From: nginx-forum at nginx.us (olan)
Date: Mon, 07 Nov 2011 11:17:28 -0500
Subject: port_in_redirect not working?
Message-ID: <81d98cc4e630e880491f56113535bf85.NginxMailingListEnglish@forum.nginx.org>

Hi all,

I posted this question on SO but haven't received much attention, so I
figured here would be better.

Basically I am attempting to run varnish on port 80 and my nginx server
on 8080, however any request to 80 redirects and appends the port to the
URL. 

i.e. http://site.com/ -> http://site.com:8080/

>From everything I've read it seems that 'port_in_redirect off' should
disable this, however it does not appear to be working. I've used a
workaround for php by using "fastcgi_param SERVER_PORT 80;", which
works, but requests to my "location /" are getting the port appended.

My server setup is:

server { 
    listen 8080 default;
    server_name "" xxx.xxx.xxx.xxx;  #just using IP here (no domain
yet)

    port_in_redirect off;
    server_name_in_redirect off;

    access_log  /var/log/nginx/localhost.access.log;


    location / {
            root   /var/www/site/html/;
            index  index index.php;
            try_files $uri/ $uri /index.php?q=$uri&$args;
    }
}
upstream backend {
    server 127.0.0.1:9000;
}

I have tried port_in_redirect in both the "location /" and the server
blocks, but neither works.

Can anyone shed any light on this problem? I'm currently using nginx
v1.0.9 on a ubuntu server.

[Link to SO question (full .conf there):
http://stackoverflow.com/questions/8026763/nginx-port-in-redirect-not-working]

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,217928,217928#msg-217928


From nginx-forum at nginx.us  Mon Nov  7 16:36:10 2011
From: nginx-forum at nginx.us (CarlWang)
Date: Mon, 07 Nov 2011 11:36:10 -0500
Subject: How to handle NGX_AGAIN returned by
	ngx_http_read_client_request_body() within handler module?
In-Reply-To: <15d512b5e71760db17d3b689b07a7fee.NginxMailingList@forum.nginx.org>
References: <593ae63ad4bdca3873ae92148badf3a4.NginxMailingList@forum.nginx.org>
	<15d512b5e71760db17d3b689b07a7fee.NginxMailingList@forum.nginx.org>
Message-ID: <801e5194299b4667b56f5d6ac28791e7.NginxMailingListEnglish@forum.nginx.org>

I have a similar problem here.
My handler module in Nginx is supposed to intercept POST requests,
analyse it and send response( the response body maybe larger than 64k).
Of course, I used ngx_http_read_client_request_body to register my
request_body_handler.
The request_body_handler does some computing and calls
ngx_http_output_filter(r,out) to send the response.
Here is the problem. If this response is less than 64k, it works fine.
But if this response is larger than 64k, the ngx_http_output_filter will
return NGX_AGAIN. However, since it's in the request_body_handler
function and it has nothing to return. I can't figure out how to handle
this NGX_AGAIN.

Can anyone help me out? Thanks a lot.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,3545,217932#msg-217932


From mdounin at mdounin.ru  Mon Nov  7 17:26:34 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Mon, 7 Nov 2011 21:26:34 +0400
Subject: port_in_redirect not working?
In-Reply-To: <81d98cc4e630e880491f56113535bf85.NginxMailingListEnglish@forum.nginx.org>
References: <81d98cc4e630e880491f56113535bf85.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20111107172634.GR95664@mdounin.ru>

Hello!

On Mon, Nov 07, 2011 at 11:17:28AM -0500, olan wrote:

> Hi all,
> 
> I posted this question on SO but haven't received much attention, so I
> figured here would be better.
> 
> Basically I am attempting to run varnish on port 80 and my nginx server
> on 8080, however any request to 80 redirects and appends the port to the
> URL. 
> 
> i.e. http://site.com/ -> http://site.com:8080/
> 
> From everything I've read it seems that 'port_in_redirect off' should
> disable this, however it does not appear to be working. I've used a
> workaround for php by using "fastcgi_param SERVER_PORT 80;", which
> works, but requests to my "location /" are getting the port appended.
> 
> My server setup is:
> 
> server { 
>     listen 8080 default;
>     server_name "" xxx.xxx.xxx.xxx;  #just using IP here (no domain
> yet)
> 
>     port_in_redirect off;
>     server_name_in_redirect off;
> 
>     access_log  /var/log/nginx/localhost.access.log;
> 
> 
>     location / {
>             root   /var/www/site/html/;
>             index  index index.php;
>             try_files $uri/ $uri /index.php?q=$uri&$args;
>     }
> }
> upstream backend {
>     server 127.0.0.1:9000;
> }
> 
> I have tried port_in_redirect in both the "location /" and the server
> blocks, but neither works.
> 
> Can anyone shed any light on this problem? I'm currently using nginx
> v1.0.9 on a ubuntu server.
> 
> [Link to SO question (full .conf there):
> http://stackoverflow.com/questions/8026763/nginx-port-in-redirect-not-working]

The "port_in_redirect" directive only alters behaviour of nginx 
itself (for trailing slash redirects, rewrites and so on), and 
it's expected that it does nothing for redirects returned by php.

With "fastcgi_param SERVER_PORT 80;" php should be fine, and I 
suspect that problem you see is actually related to your browser 
(or varnish) cache sitting here from previous configurations.  

If clearing cache doesn't help, please provide debug log (see 
http://wiki.nginx.org/Debugging), it will show what goes on here.

Maxim Dounin


From mdounin at mdounin.ru  Mon Nov  7 17:47:59 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Mon, 7 Nov 2011 21:47:59 +0400
Subject: How to handle NGX_AGAIN returned by
	ngx_http_read_client_request_body() within handler module?
In-Reply-To: <801e5194299b4667b56f5d6ac28791e7.NginxMailingListEnglish@forum.nginx.org>
References: <593ae63ad4bdca3873ae92148badf3a4.NginxMailingList@forum.nginx.org>
	<15d512b5e71760db17d3b689b07a7fee.NginxMailingList@forum.nginx.org>
	<801e5194299b4667b56f5d6ac28791e7.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20111107174759.GT95664@mdounin.ru>

Hello!

On Mon, Nov 07, 2011 at 11:36:10AM -0500, CarlWang wrote:

> I have a similar problem here.
> My handler module in Nginx is supposed to intercept POST requests,
> analyse it and send response( the response body maybe larger than 64k).
> Of course, I used ngx_http_read_client_request_body to register my
> request_body_handler.
> The request_body_handler does some computing and calls
> ngx_http_output_filter(r,out) to send the response.
> Here is the problem. If this response is less than 64k, it works fine.
> But if this response is larger than 64k, the ngx_http_output_filter will
> return NGX_AGAIN. However, since it's in the request_body_handler
> function and it has nothing to return. I can't figure out how to handle
> this NGX_AGAIN.
> 
> Can anyone help me out? Thanks a lot.

You have to finalize request as usual, 

    ngx_http_finalize_request(r, ngx_http_output_filter(r, out));

will do all work as long as you have full response in the "out" 
chain.

Maxim Dounin


From nginx-forum at nginx.us  Mon Nov  7 18:25:44 2011
From: nginx-forum at nginx.us (olan)
Date: Mon, 07 Nov 2011 13:25:44 -0500
Subject: port_in_redirect not working?
In-Reply-To: <20111107172634.GR95664@mdounin.ru>
References: <20111107172634.GR95664@mdounin.ru>
Message-ID: <54dbeaf04d18eba2fa2b0b178cb9de48.NginxMailingListEnglish@forum.nginx.org>

Hi Maxim, thanks for getting back to me.

I've enabled debugging and the output is here:
http://pastebin.com/Dmw2nJMY. I've cleared all caches and restarted
varnish, nginx and php5-fpm and mysql.

I didn't explain myself very well earlier. The "fastcgi_param
SERVER_PORT 80" does work for php pages (it's in my location ~ .php$
block) but the problem I'm having is with my server root.

http://site.com/page.php <-- works fine
http://site.com/ redirects to http://site.com:8080
http://site.com/index.php redirects to http://site.com:8080

One thing I noticed is "location /" block is used first, the index.php
is tried, and then the "location ~ .php$" block processes this page. I
would have thought that the SERVER_PORT would have rewritten the port
used at this stage...?

Any help is greatly appreciated!
Thanks,

Olan

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,217928,217939#msg-217939


From francis at daoine.org  Mon Nov  7 19:42:11 2011
From: francis at daoine.org (Francis Daly)
Date: Mon, 7 Nov 2011 19:42:11 +0000
Subject: Rewrite problem http://x.x.x.x/folder_name/index.php/css/first.css
In-Reply-To: <974ef8502ef4834d08bd7bfd138fc699.NginxMailingListEnglish@forum.nginx.org>
References: <974ef8502ef4834d08bd7bfd138fc699.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20111107194211.GM27078@craic.sysops.org>

On Mon, Nov 07, 2011 at 11:17:14AM -0500, mikolajj wrote:

Hi there,

> Apache and lighttpd can use URL
> http://x.x.x.x/folder_name/index.php/css/first.css as
> http://x.x.x.x/folder_name/index.php with REQUEST_URI part =
> "/css/first.css" - this is handle by PHP and processed along...
> 
> Nginx is trying to open index.php as folder an of course 404 is
> thrown...

Yes, that's how it is frequently configured by default.

You'll want to configure it to match your application.

>                 location ~ .php$ {
>                   fastcgi_pass   127.0.0.1:9000;
>                   fastcgi_index  index.php;
>                   fastcgi_param  SCRIPT_FILENAME
> /home/main-www$fastcgi_script_name;
>                   include fastcgi_params;
>                 }

That location matches urls that end in "php". Your url is
/folder_name/index.php/css/first.css, which doesn't end in php,
so you'll want to use some location setting that matches that url --
perhaps "starts with /folder_name/index.php/" will be best.

For testing, you could use "includes php", which would be

  location ~ php {}

but that is unlikely to be good for the live site.

Once you have the location definition correct, you'll want to use
fastcgi_split_path_info

http://wiki.nginx.org/HttpFcgiModule#fastcgi_split_path_info

and then you'll probably want to set PATH_INFO or REQUEST_URI or whatever
your application requires, like in the example there.

> Can you point me to some clues or solution?

Hopefully the above makes sense.

Either change your current "php" location to match all urls you want
handled by php; or make a new one that matches this location.

And use fastcgi_split_path_info.

And the debug log can be very useful if you get lost.

Good luck,

	f
-- 
Francis Daly        francis at daoine.org


From nginx at nginxuser.net  Mon Nov  7 19:44:43 2011
From: nginx at nginxuser.net (Nginx User)
Date: Mon, 7 Nov 2011 22:44:43 +0300
Subject: port_in_redirect not working?
In-Reply-To: <54dbeaf04d18eba2fa2b0b178cb9de48.NginxMailingListEnglish@forum.nginx.org>
References: <20111107172634.GR95664@mdounin.ru>
	<54dbeaf04d18eba2fa2b0b178cb9de48.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <CAN=WzV3ji=5TNUfqmENqyDGJd8FQbMhRJkdxf7PD5JnoGBZbgA@mail.gmail.com>

On 7 November 2011 21:25, olan <nginx-forum at nginx.us> wrote:
> Hi Maxim, thanks for getting back to me.
>
> I've enabled debugging and the output is here:
> http://pastebin.com/Dmw2nJMY. I've cleared all caches and restarted
> varnish, nginx and php5-fpm and mysql.

Line 245 of your debug shows the issue. php is sending a header - http
fastcgi header: "Location: http://XXX.XXX.XXX.XXX:8080/". The
proxy_redirect directive handles this for proxy_pass setups but there
isn't an equivalent for fastcgi.

You can try tricking php by putting $_SERVER["SERVER_PORT"] right at
the top of your index php (or use autoprepend in php.ini so it applies
to all php files).

I just had a problem with this with a proxy setup as the
proxy_redirect only accepts variables in the redirect part such that I
had to do ....

proxy_redirect   http://example-1.com:8080/    http://example-1.com/;
proxy_redirect   http://example-2.com:8080/    http://example-2.com/;
proxy_redirect   http://example-3.com:8080/    http://example-3.com/;
...
proxy_redirect   http://example-n.com:8080/    http://example-n.com/;

If the directive accepted variables in the original uri part, a single line ...

proxy_redirect   http://$host:8080/    http://$host/;

... would have done the job and given flexibility. I have to remember
to add every domain I create to the list.

Anyway, I digress. Try the trick on PHP. Nothing can be done in Nginx
until a fastcgi_redirect directive is introduced.


From nginx at nginxuser.net  Mon Nov  7 19:46:03 2011
From: nginx at nginxuser.net (Nginx User)
Date: Mon, 7 Nov 2011 22:46:03 +0300
Subject: port_in_redirect not working?
In-Reply-To: <CAN=WzV3ji=5TNUfqmENqyDGJd8FQbMhRJkdxf7PD5JnoGBZbgA@mail.gmail.com>
References: <20111107172634.GR95664@mdounin.ru>
	<54dbeaf04d18eba2fa2b0b178cb9de48.NginxMailingListEnglish@forum.nginx.org>
	<CAN=WzV3ji=5TNUfqmENqyDGJd8FQbMhRJkdxf7PD5JnoGBZbgA@mail.gmail.com>
Message-ID: <CAN=WzV0qjh_8ZDSNtVtMrMwq_YpguNJezO2ZcunCmP4P01fCMw@mail.gmail.com>

On 7 November 2011 22:44, Nginx User <nginx at nginxuser.net> wrote:
> You can try tricking php by putting $_SERVER["SERVER_PORT"] right at
> the top of your index php (or use autoprepend in php.ini so it applies
> to all php files).

That should read - You can try tricking php by putting
$_SERVER["SERVER_PORT"] = "80";


From varia at e-healthexpert.org  Mon Nov  7 20:53:12 2011
From: varia at e-healthexpert.org (Mark Alan)
Date: Mon, 7 Nov 2011 20:53:12 +0000
Subject: error: too many arguments to function 'ngx_time_update'
Message-ID: <20111107205312.3eb23869@e-healthexpert.org>

Hello,

While trying to compile with:
  pdebuild ../nginx_1.1.7-0.ubuntu.2.dsc
the latest nginx code using an Ubuntu 11.10 machine
with these parameters:
 ./configure \\\
\t    --prefix=/etc/nginx \\\
\t    --conf-path=/etc/nginx/nginx.conf \\\
\t    --error-log-path=/var/log/nginx/error.log \\\
\t    --http-client-body-temp-path=/var/lib/nginx/body \\\
\t    --http-fastcgi-temp-path=/var/lib/nginx/fastcgi \\\
\t    --http-log-path=/var/log/nginx/access.log \\\
\t    --http-proxy-temp-path=/var/lib/nginx/proxy \\\
\t    --lock-path=/var/lock/nginx.lock \\\
\t    --pid-path=/var/run/nginx.pid \\\
\t    --with-http_gzip_static_module \\\
\t    --with-http_ssl_module \\\
\t    --without-http-cache \\\
\t    --without-http_browser_module \\\
\t    --without-http_geo_module \\\
\t    --without-http_limit_req_module \\\
\t    --without-http_limit_zone_module \\\
\t    --without-http_map_module \\\
\t    --without-http_memcached_module \\\
\t    --without-http_referer_module \\\
\t    --without-http_scgi_module \\\
\t    --without-http_split_clients_module \\\
\t    --without-http_ssi_module \\\
\t    --without-http_upstream_keepalive_module \\\
\t    --without-http_userid_module \\\
\t    --without-http_uwsgi_module \\\
            $CONFIGURE_OPTS >$@ \
\ttouch $@ 

Can you help me to avoid the following errors?

M.

(...)
gcc -c -Wall -g -O2   -I src/core -I src/event -I src/event/modules -I
src/os/unix -I /usr/include/libxml2 -I objs -I src/http -I src/http/modules -I src/mail \
   -o objs/addon/src/ngx_http_echo_timer.o \
  /tmp/buildd/nginx-1.1.7/debian/modules/nginx-echo/src/ngx_http_echo_timer.c

/tmp/buildd/nginx-1.1.7/debian/modules/nginx-echo/src/ngx_http_echo_timer.c:
In function
'ngx_http_echo_timer_elapsed_variable':

/tmp/buildd/nginx-1.1.7/debian/modules/nginx-echo/src/ngx_http_echo_timer.c:32:5: error: too many arguments to function 'ngx_time_update'
src/core/ngx_times.h:23:6: note: declared here

/tmp/buildd/nginx-1.1.7/debian/modules/nginx-echo/src/ngx_http_echo_timer.c:
In function
'ngx_http_echo_exec_echo_reset_timer':

/tmp/buildd/nginx-1.1.7/debian/modules/nginx-echo/src/ngx_http_echo_timer.c:70:5: error: too many arguments to function 'ngx_time_update'
src/core/ngx_times.h:23:6: note: declared here

make[2]: *** [objs/addon/src/ngx_http_echo_timer.o] Error 1
make[2]: Leaving directory `/tmp/buildd/nginx-1.1.7/debian/build-full'
make[1]: *** [build] Error 2
make[1]: Leaving directory `/tmp/buildd/nginx-1.1.7/debian/build-full'
make: *** [build-arch.full] Error 2

dpkg-buildpackage: error: debian/rules build gave error exit status 2
E: Failed autobuilding of package
I: unmounting /var/cache/pbuilder/ccache filesystem
I: unmounting dev/pts filesystem
I: unmounting proc filesystem
I: cleaning the build env
I: removing directory /var/cache/pbuilder/build//23240 and its subdirectories


From nginx-forum at nginx.us  Mon Nov  7 20:54:24 2011
From: nginx-forum at nginx.us (token)
Date: Mon, 07 Nov 2011 15:54:24 -0500
Subject: Multipule Map modules
Message-ID: <848848260804f79dc05def8e8126ab0c.NginxMailingListEnglish@forum.nginx.org>

What i would like to do is like Apache is to use multiple map files and
get the vars form them

i have tried the following and can only get one value `$variable1`
what is a rewrite to get `$variable1` and `$variable2` in one rewrite

rewrite ^(^\/*)/(.*)$  /index.php?key1=$variable1&key2=$variable2 last;


map $uri $variable1 {
default 11;
/sub 7;

}
map $uri $variable2 {
default 78;
/pep 23;
}


Regards, David

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,217936,217936#msg-217936


From appa at perusio.net  Mon Nov  7 22:07:17 2011
From: appa at perusio.net (=?UTF-8?B?QW50w7NuaW8=?= P. P. Almeida)
Date: Mon, 07 Nov 2011 22:07:17 +0000
Subject: Multipule Map modules
In-Reply-To: <848848260804f79dc05def8e8126ab0c.NginxMailingListEnglish@forum.nginx.org>
References: <848848260804f79dc05def8e8126ab0c.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <871utjoacq.wl%appa@perusio.net>

On 7 Nov 2011 20h54 WET, nginx-forum at nginx.us wrote:

> What i would like to do is like Apache is to use multiple map files
> and get the vars form them
>
> i have tried the following and can only get one value `$variable1`
> what is a rewrite to get `$variable1` and `$variable2` in one
> rewrite
>
> rewrite ^(^\/*)/(.*)$ /index.php?key1=$variable1&key2=$variable2
> last;
>

What are you trying to accomplish?

> map $uri $variable1 {
> default 11;
> /sub 7;

If your URI is /sub $variable1 becomes 7.

> }
> map $uri $variable2 {
> default 78;
> /pep 23;
>}

If your URI is /sub $variable2 becomes 23.

It works as it should. I don't see any issue. Also your rewrite is
unnecessary. Try this:

return 301 /index.php?key1=$variable1&key2=$variable2;

Do you mean that the request URI can contain both /pep and /sub?Like
this:

http://example.com/sub/pep/other-stuff-if-any

If so then your matching must be done with a regex:

map $uri $variable1 {
  default 11;
  ~/sub 7;
}

map $uri $variable2 {
  default 78;
  ~/pep 23;
}

--- appa


From francis at daoine.org  Mon Nov  7 22:12:37 2011
From: francis at daoine.org (Francis Daly)
Date: Mon, 7 Nov 2011 22:12:37 +0000
Subject: Multipule Map modules
In-Reply-To: <848848260804f79dc05def8e8126ab0c.NginxMailingListEnglish@forum.nginx.org>
References: <848848260804f79dc05def8e8126ab0c.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20111107221237.GN27078@craic.sysops.org>

On Mon, Nov 07, 2011 at 03:54:24PM -0500, token wrote:

Hi there,

> What i would like to do is like Apache is to use multiple map files and
> get the vars form them
> 
> i have tried the following and can only get one value `$variable1`
> what is a rewrite to get `$variable1` and `$variable2` in one rewrite
> 
> rewrite ^(^\/*)/(.*)$  /index.php?key1=$variable1&key2=$variable2 last;

I confess I'm not sure exactly what it is you are trying to do.

For certain urls, grab parts of the url and send them as-is within
QUERY_STRING to a php-processor? (In which case: doing the grabbing in
the location definition is probably easiest.)

Or grab parts of the url, and set other values in QUERY_STRING based
on the url parts? (In which case, matching on $uri at server level and
using the map-ped value within the location block is probably easiest.)

Or maybe something else?

> map $uri $variable1 {
> default 11;
> /sub 7;
> 
> }
> map $uri $variable2 {
> default 78;
> /pep 23;
> }

The map documentation is at http://wiki.nginx.org/HttpMapModule#map
and there's an example of if/set/map in the thread at
http://forum.nginx.org/read.php?2,194480

And there are examples of php-without-rewrite around as well -- in the
location block, set "fastcgi_param SCRIPT_FILENAME" explicitly, as well
as fastcgi_pass.

Good luck,

	f
-- 
Francis Daly        francis at daoine.org


From nginx-forum at nginx.us  Mon Nov  7 22:22:29 2011
From: nginx-forum at nginx.us (nchainani)
Date: Mon, 07 Nov 2011 17:22:29 -0500
Subject: Path to a directory should return 404
Message-ID: <a26b4d7cf33845957d2189b66345962b.NginxMailingListEnglish@forum.nginx.org>

Under the document root, I have few directories like
<root>/1/file1.html
<root>/1/file2.html
<root>/2/file1.html
<root>/2/file2.html

If I go to the url http://localhost:8080/1, I get back a 301 with the
new location http://localhost/1/     autoindexing is off, and so I end
up with a 403 Forbidden page.

Can I change this default behavior such that all the requests for the
directories get the universal 404 page i.e. instead of a redirection, it
should simply return the configured 404 page for location /.

Sorry, I couldn't find a similar post on the forums. Thanks for any
pointers.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,217950,217950#msg-217950


From xiaohanhoho at gmail.com  Tue Nov  8 02:08:31 2011
From: xiaohanhoho at gmail.com (=?GB2312?B?0KS6rQ==?=)
Date: Tue, 8 Nov 2011 10:08:31 +0800
Subject: 504 Gateway Time out Nginx 1.0.4: Upstream time out
In-Reply-To: <073867d2c99fa075d96095ce39dd647c@ruby-forum.com>
References: <ac4573fd6465e0b060bcdcfada0b5e61@ruby-forum.com>
	<073867d2c99fa075d96095ce39dd647c@ruby-forum.com>
Message-ID: <CAJ-u7x61BLe9rzSjVM=ho-07CvjE-Qoc9HKZY5w81tXbLwQcMQ@mail.gmail.com>

proxy_connect_timeout   10m;
    proxy_send_timeout      10m;
     proxy_read_timeout      8m;

Is it too long?

2011/11/5 Roger Gue <lists at ruby-forum.com>

> HERE IS MY NGINX.CONF FILE DETAIL:
>
> user nginx;
> worker_processes  1;
>
> error_log  /var/log/nginx/error.log;
> pid        /var/run/nginx/nginx.pid;
>
> events {
>    worker_connections  2048;
>    # 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   3;
>    tcp_nodelay        off;
>    ignore_invalid_headers on;
>
>    gzip  on;
>    gzip_disable "MSIE [1-6]\.(?!.*SV1)";
>
>    gzip_comp_level 6;
>    gzip_static on;
>    gzip_min_length 2200;
>    gzip_buffers 16 8k;
>    gzip_http_version 1.1;
>    gzip_types text/plain text/css application/x-javascript image/x-icon
> text/xml application/xml application/xml+rss text/javascript;
>
>    include /etc/nginx/conf.d/*.conf;
>    include /etc/nginx/sites-enabled/*;
> }
>
>
> AND HERE IS MY PROXY.CONF FILE DETAIL
>
> #proxy.conf
> #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_body_buffer_size 256k;
>     proxy_connect_timeout   10m;
>     proxy_send_timeout      10m;
>      proxy_read_timeout      8m;
>     client_header_timeout   10m;
>   proxy_buffers           32 4k;
>     client_body_timeout     800;
>    proxy_ignore_client_abort on;
>
> client_header_buffer_size     1k;
> large_client_header_buffers 8 8k;
> client_max_body_size          2g;
>
>            postpone_output 1460;
>
> --
> Posted via http://www.ruby-forum.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/20111108/c2be1642/attachment-0001.html>

From nginx-forum at nginx.us  Tue Nov  8 02:42:20 2011
From: nginx-forum at nginx.us (wangbin579)
Date: Mon, 07 Nov 2011 21:42:20 -0500
Subject: Tcpcopy,an online request replication tool fit for nginx
In-Reply-To: <141f5701aa1bdc50e4d7a29c237508a3.NginxMailingListEnglish@forum.nginx.org>
References: <141f5701aa1bdc50e4d7a29c237508a3.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <df818e79c31cb252c78102640f358f02.NginxMailingListEnglish@forum.nginx.org>

Name:
    tcpcopy
    It is a request replication tool and is mainly for testing tasks
using netlink and raw sockets


Description:
    It can help you find bugs in your online project without actually
being online. 
    And it can also be used to test the stress that a system could
endure.

    For example, if your system already has "memcached" subsystem and
you want to 
    use "membase" to replace it, tcpcopy can assist you to test
"membase". While your 
    old memcached system is still running online, tcpcopy could copy the
flow of packets 
    from memcached to membase. From the point view of membase, the flow
is accessing 
    membase(just like membase online), and it will not affect memcached
at all except 
    network bandwidth and a little cpu load.


Functionalities?
    1) Distributed Stress Test
        You can use online data to test the stress that your target
machine can endure. It is better 
        than apache ab tool and you can find bugs that only occur during
high-stress situations.
    2) Hot Backup
        It is very suitable for backup tasks if connections are
short-lived and the request loss rate
        is very low(1/100000).
    3) Normal Online Test
        You can find whether the new system is stable and find bugs that
only occur in actual 
        online environments.
    4) Comparison Test
        For example, you can use tcpcopy to compare the performances of
apache and nginx.


Characteristics?
    1)real time
    2)realistic
    3)efficient
    4)easy to use
    5)distributed
    6)significant

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,217680,217954#msg-217954


From nginx-forum at nginx.us  Tue Nov  8 02:45:27 2011
From: nginx-forum at nginx.us (wangbin579)
Date: Mon, 07 Nov 2011 21:45:27 -0500
Subject: nginx_hmux_module - support hmux protocol proxy with Nginx
In-Reply-To: <2d7909af72c23eb07f1385d88d6ae095.NginxMailingListEnglish@forum.nginx.org>
References: <2d7909af72c23eb07f1385d88d6ae095.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <767a7623e78585d135e98eafb97bd091.NginxMailingListEnglish@forum.nginx.org>

this module is running stable in a large project for more than 2 months

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,217681,217955#msg-217955


From nginx-forum at nginx.us  Tue Nov  8 02:49:12 2011
From: nginx-forum at nginx.us (wangbin579)
Date: Mon, 07 Nov 2011 21:49:12 -0500
Subject: nginx_hmux_module - support hmux protocol proxy with Nginx
In-Reply-To: <2d7909af72c23eb07f1385d88d6ae095.NginxMailingListEnglish@forum.nginx.org>
References: <2d7909af72c23eb07f1385d88d6ae095.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <34eb287fcb3598c48d68c88648144899.NginxMailingListEnglish@forum.nginx.org>

this module has been running stably in a large project for more than 2
months

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,217681,217956#msg-217956


From nginx-forum at nginx.us  Tue Nov  8 04:31:04 2011
From: nginx-forum at nginx.us (CarlWang)
Date: Mon, 07 Nov 2011 23:31:04 -0500
Subject: How to handle NGX_AGAIN returned by
	ngx_http_read_client_request_body() within handler module?
In-Reply-To: <20111107174759.GT95664@mdounin.ru>
References: <20111107174759.GT95664@mdounin.ru>
Message-ID: <8cd48bd50b12c0e8ea5d0bb1812c66eb.NginxMailingListEnglish@forum.nginx.org>

Here are my code:
static void v8_embed_handler ( ngx_http_request_t * r ) 
{
	...// generating out chain.
	rc = ngx_http_output_filter ( r , out );
	while( rc == NGX_AGAIN ) {
		if( out->next == NULL )
			break;
		rc = ngx_http_output_filter ( r , out->next );
		out = out->next;
	}
	ngx_http_finalize_request ( r , rc );
}
static ngx_int_t ngx_http_v8_handler_request(ngx_http_request_t *r)
{
	ngx_int_t rc = NGX_DONE ; 
	rc = ngx_http_read_client_request_body ( r , v8_embed_handler ) ; //
call the v8_embed_handler handler to process the post data
	if ( rc >= NGX_HTTP_SPECIAL_RESPONSE )
		return rc;
	return NGX_DONE;
}

Then I changed the code as :
static void v8_embed_handler ( ngx_http_request_t * r ) 
{
	...// generating out chain.
	ngx_http_finalize_request ( r , ngx_http_output_filter ( r , out ) );
}
However, it doesn't change the test result. If the response is larger
than 64Kb, the problem is still there.
I'm using curl to test it. It firstly output 64k response and  it says
"curl: (18) transfer closed with 19013 bytes remaining to read". Then it
output a little part of the remaining response. I don't really
understand why.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,3545,217958#msg-217958


From nginx-forum at nginx.us  Tue Nov  8 05:20:21 2011
From: nginx-forum at nginx.us (token)
Date: Tue, 08 Nov 2011 00:20:21 -0500
Subject: Multipule Map modules
In-Reply-To: <848848260804f79dc05def8e8126ab0c.NginxMailingListEnglish@forum.nginx.org>
References: <848848260804f79dc05def8e8126ab0c.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <9339d4215213908c3f55b8ac38ecadbc.NginxMailingListEnglish@forum.nginx.org>

Hi, Ant?nio that is exactly wath is was trying to accomplish  id didn't
know i had to use the "~" at the beginning of the map key.
Thank you

>Do you mean that the request URI can contain both /pep and /sub?Like 
> this:

> http://example.com/sub/pep/other-stuff-if-any

> If so then your matching must be done with a regex:

> map $uri $variable1 {
> default 11;
> ~/sub 7;
> }

> map $uri $variable2 {
> default 78;
> ~/pep 23;
> }

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,217936,217959#msg-217959


From agentzh at gmail.com  Tue Nov  8 06:56:17 2011
From: agentzh at gmail.com (agentzh)
Date: Tue, 8 Nov 2011 14:56:17 +0800
Subject: How to handle NGX_AGAIN returned by
	ngx_http_read_client_request_body() within handler module?
In-Reply-To: <8cd48bd50b12c0e8ea5d0bb1812c66eb.NginxMailingListEnglish@forum.nginx.org>
References: <20111107174759.GT95664@mdounin.ru>
	<8cd48bd50b12c0e8ea5d0bb1812c66eb.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <CAB4Tn6N_edvaiCKPKOEgcpkBbEhHLMUdz3+CR-RRHHMKYUpA5Q@mail.gmail.com>

On Tue, Nov 8, 2011 at 12:31 PM, CarlWang <nginx-forum at nginx.us> wrote:
> Here are my code:
> static void v8_embed_handler ( ngx_http_request_t * r )
> {
> ? ? ? ?...// generating out chain.
> ? ? ? ?rc = ngx_http_output_filter ( r , out );
> ? ? ? ?while( rc == NGX_AGAIN ) {
> ? ? ? ? ? ? ? ?if( out->next == NULL )
> ? ? ? ? ? ? ? ? ? ? ? ?break;
> ? ? ? ? ? ? ? ?rc = ngx_http_output_filter ( r , out->next );
> ? ? ? ? ? ? ? ?out = out->next;
> ? ? ? ?}
> ? ? ? ?ngx_http_finalize_request ( r , rc );
> }

Please note that you shouldn't call ngx_http_output_filter on your
data again when NGX_AGAIN is returned. The underlying copy and writer
filters will buffer the output chains when the system send buffer is
full. Furthermore, the ngx_http_finalize_request function will
automatically register the ngx_http_writer as the write event handler
to actually emit the outputs at every write event for you. And that's
why Maxim suggested the simple call

    ngx_http_finalize_request(r, ngx_http_output_filter(r, out))

> static ngx_int_t ngx_http_v8_handler_request(ngx_http_request_t *r)
> {
> ? ? ? ?ngx_int_t rc = NGX_DONE ;
> ? ? ? ?rc = ngx_http_read_client_request_body ( r , v8_embed_handler ) ; //
> call the v8_embed_handler handler to process the post data
> ? ? ? ?if ( rc >= NGX_HTTP_SPECIAL_RESPONSE )
> ? ? ? ? ? ? ? ?return rc;
> ? ? ? ?return NGX_DONE;
> }
>

Which phase is your ngx_http_v8_handler_request function running in?
If it's running in the access or rewrite phase, then the coding
structure can be quite different here.

I suggest you take a look at how our ngx_lua module handle all of this
in access, write, and content phases:

    http://wiki.nginx.org/HttpLuaModule

BTW, taking a close look at the error.log when building nginx with
--with-debug and enabling the debug error log level will be very
helpful to debug such issues ;)

Regards,
-agentzh


From mdounin at mdounin.ru  Tue Nov  8 08:08:20 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Tue, 8 Nov 2011 12:08:20 +0400
Subject: error: too many arguments to function 'ngx_time_update'
In-Reply-To: <20111107205312.3eb23869@e-healthexpert.org>
References: <20111107205312.3eb23869@e-healthexpert.org>
Message-ID: <20111108080820.GW95664@mdounin.ru>

Hello!

On Mon, Nov 07, 2011 at 08:53:12PM +0000, Mark Alan wrote:

[...]

> Can you help me to avoid the following errors?

[...]

> /tmp/buildd/nginx-1.1.7/debian/modules/nginx-echo/src/ngx_http_echo_timer.c:32:5: error: too many arguments to function 'ngx_time_update'
> src/core/ngx_times.h:23:6: note: declared here

Errors are caused by 3rd party module you are trying to use.  
Either do not compile it, or update it (likely new version of the 
module exists which resolves this).

Maxim Dounin


From ru at nginx.com  Tue Nov  8 08:35:29 2011
From: ru at nginx.com (Ruslan Ermilov)
Date: Tue, 8 Nov 2011 08:35:29 +0000
Subject: Path to a directory should return 404
In-Reply-To: <a26b4d7cf33845957d2189b66345962b.NginxMailingListEnglish@forum.nginx.org>
References: <a26b4d7cf33845957d2189b66345962b.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20111108083529.GA19051@lo0.su>

On Mon, Nov 07, 2011 at 05:22:29PM -0500, nchainani wrote:
> Under the document root, I have few directories like
> <root>/1/file1.html
> <root>/1/file2.html
> <root>/2/file1.html
> <root>/2/file2.html
> 
> If I go to the url http://localhost:8080/1, I get back a 301 with the
> new location http://localhost/1/     autoindexing is off, and so I end
> up with a 403 Forbidden page.
> 
> Can I change this default behavior such that all the requests for the
> directories get the universal 404 page i.e. instead of a redirection, it
> should simply return the configured 404 page for location /.
> 
> Sorry, I couldn't find a similar post on the forums. Thanks for any
> pointers.

Something like that maybe?

: location / {
:     #error_page 404 ...;
:     location ~ /$ { return 404; }
: }


From agentzh at gmail.com  Tue Nov  8 08:33:51 2011
From: agentzh at gmail.com (agentzh)
Date: Tue, 8 Nov 2011 16:33:51 +0800
Subject: error: too many arguments to function 'ngx_time_update'
In-Reply-To: <20111107205312.3eb23869@e-healthexpert.org>
References: <20111107205312.3eb23869@e-healthexpert.org>
Message-ID: <CAB4Tn6NPiGUNtW-3J4268Yv6yaQeD6fV0mB7MTseZcjUafJhUw@mail.gmail.com>

On Tue, Nov 8, 2011 at 4:53 AM, Mark Alan <varia at e-healthexpert.org> wrote:
> /tmp/buildd/nginx-1.1.7/debian/modules/nginx-echo/src/ngx_http_echo_timer.c:
> In function
> 'ngx_http_echo_timer_elapsed_variable':
>
> /tmp/buildd/nginx-1.1.7/debian/modules/nginx-echo/src/ngx_http_echo_timer.c:32:5: error: too many arguments to function 'ngx_time_update'
> src/core/ngx_times.h:23:6: note: declared here
>
> /tmp/buildd/nginx-1.1.7/debian/modules/nginx-echo/src/ngx_http_echo_timer.c:
> In function
> 'ngx_http_echo_exec_echo_reset_timer':
>
> /tmp/buildd/nginx-1.1.7/debian/modules/nginx-echo/src/ngx_http_echo_timer.c:70:5: error: too many arguments to function 'ngx_time_update'
> src/core/ngx_times.h:23:6: note: declared here
>

There was an incompatible API change in 0.7.66 that the
ngx_time_update macro then takes a different number of arguments. This
issue was already fixed in ngx_echo v0.33+, see

    http://wiki.nginx.org/HttpEchoModule#v0.33

I suggest you upgrade your ngx_echo module to the latest v0.37rc7
release, which can be downloaded here:

    https://github.com/agentzh/echo-nginx-module/tags

Regards,
-agentzh


From nginx-forum at nginx.us  Tue Nov  8 09:25:57 2011
From: nginx-forum at nginx.us (abcomp01)
Date: Tue, 08 Nov 2011 04:25:57 -0500
Subject: php-fpm and nginx return "The page you are looking for is temporarily
	unavailable. Please try again later. "
Message-ID: <9636a5d8da2b172192e56ee838fe791e.NginxMailingListEnglish@forum.nginx.org>

php-fpm and nginx return "The page you are looking for is temporarily
unavailable. Please try again later. "

call http://173.212.207.146/upload/info.php

		location ~ \.php$ {
		  root   /home/admin/18loadnet;
		  fastcgi_pass  127.0.0.1:9000;
		  fastcgi_index index.php;
		  fastcgi_param  SCRIPT_FILENAME  /home/admin/info.php;
		  include /etc/nginx/fastcgi_params;
		}

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,217967,217967#msg-217967


From igor at sysoev.ru  Tue Nov  8 09:47:21 2011
From: igor at sysoev.ru (Igor Sysoev)
Date: Tue, 8 Nov 2011 13:47:21 +0400
Subject: php-fpm and nginx return "The page you are looking for is
	temporarily unavailable. Please try again later. "
In-Reply-To: <9636a5d8da2b172192e56ee838fe791e.NginxMailingListEnglish@forum.nginx.org>
References: <9636a5d8da2b172192e56ee838fe791e.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20111108094721.GB9232@nginx.com>

On Tue, Nov 08, 2011 at 04:25:57AM -0500, abcomp01 wrote:
> php-fpm and nginx return "The page you are looking for is temporarily
> unavailable. Please try again later. "
> 
> call http://173.212.207.146/upload/info.php
> 
> 		location ~ \.php$ {
> 		  root   /home/admin/18loadnet;
> 		  fastcgi_pass  127.0.0.1:9000;
> 		  fastcgi_index index.php;
> 		  fastcgi_param  SCRIPT_FILENAME  /home/admin/info.php;
> 		  include /etc/nginx/fastcgi_params;
> 		}

Please look in error_log for reason of this issue.


-- 
Igor Sysoev


From mdounin at mdounin.ru  Tue Nov  8 09:47:42 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Tue, 8 Nov 2011 13:47:42 +0400
Subject: php-fpm and nginx return "The page you are looking for is
	temporarily unavailable. Please try again later. "
In-Reply-To: <9636a5d8da2b172192e56ee838fe791e.NginxMailingListEnglish@forum.nginx.org>
References: <9636a5d8da2b172192e56ee838fe791e.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20111108094742.GY95664@mdounin.ru>

Hello!

On Tue, Nov 08, 2011 at 04:25:57AM -0500, abcomp01 wrote:

> php-fpm and nginx return "The page you are looking for is temporarily
> unavailable. Please try again later. "
> 
> call http://173.212.207.146/upload/info.php
> 
> 		location ~ \.php$ {
> 		  root   /home/admin/18loadnet;
> 		  fastcgi_pass  127.0.0.1:9000;
> 		  fastcgi_index index.php;
> 		  fastcgi_param  SCRIPT_FILENAME  /home/admin/info.php;
> 		  include /etc/nginx/fastcgi_params;
> 		}

Try looking into error_log, it likely has explanation of the error 
returned.

Maxim Dounin


From nginx-forum at nginx.us  Tue Nov  8 09:54:57 2011
From: nginx-forum at nginx.us (abcomp01)
Date: Tue, 08 Nov 2011 04:54:57 -0500
Subject: php-fpm and nginx return "The page you are looking for is
	temporarily unavailable. Please try again later. "
In-Reply-To: <9636a5d8da2b172192e56ee838fe791e.NginxMailingListEnglish@forum.nginx.org>
References: <9636a5d8da2b172192e56ee838fe791e.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <9a8b677624d53347a2f3ef680decb9ba.NginxMailingListEnglish@forum.nginx.org>

2011/11/08 02:53:31 [error] 3983#0: *513879 recv() failed (104:
Connection reset by peer) while reading response header from upstream,
client: 171.98.8.155, server: _, request: "GET /upload/info.php
HTTP/1.1", upstream: "fastcgi://127.0.0.1:9000", host:
"173.212.207.146"

?

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,217967,217973#msg-217973


From mdounin at mdounin.ru  Tue Nov  8 10:00:23 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Tue, 8 Nov 2011 14:00:23 +0400
Subject: php-fpm and nginx return "The page you are looking for is
	temporarily unavailable. Please try again later. "
In-Reply-To: <9a8b677624d53347a2f3ef680decb9ba.NginxMailingListEnglish@forum.nginx.org>
References: <9636a5d8da2b172192e56ee838fe791e.NginxMailingListEnglish@forum.nginx.org>
	<9a8b677624d53347a2f3ef680decb9ba.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20111108100023.GZ95664@mdounin.ru>

Hello!

On Tue, Nov 08, 2011 at 04:54:57AM -0500, abcomp01 wrote:

> 2011/11/08 02:53:31 [error] 3983#0: *513879 recv() failed (104:
> Connection reset by peer) while reading response header from upstream,
> client: 171.98.8.155, server: _, request: "GET /upload/info.php
> HTTP/1.1", upstream: "fastcgi://127.0.0.1:9000", host:
> "173.212.207.146"
> 
> ?

Backend (php-fpm) sent RST (i.e. terminated connection abnormally) 
before replying anything to nginx.  Most likely it died.  Try 
looking into php-fpm / system logs for more details.

In any case it doesn't look like nginx issue.

Maxim Dounin


From nginx-forum at nginx.us  Tue Nov  8 10:01:07 2011
From: nginx-forum at nginx.us (etrader)
Date: Tue, 08 Nov 2011 05:01:07 -0500
Subject: Running Shell Script from html webpage
Message-ID: <ce0a55728e7377210fd1b4df1b8a5338.NginxMailingListEnglish@forum.nginx.org>

I have a Nginx installed on Linux (Ubuntu or Centos) without any
scripting language such as PHP or Python (connected to the webserver).
Can I run Shell commands from a webpage (e.g. html)?

I mean is it possible to use nginx without script languages by shell
scripts?

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,217975,217975#msg-217975


From varia at e-healthexpert.org  Tue Nov  8 10:58:31 2011
From: varia at e-healthexpert.org (Mark Alan)
Date: Tue, 8 Nov 2011 10:58:31 +0000
Subject: solved: error: too many arguments to function 'ngx_time_update'
In-Reply-To: <20111108080820.GW95664@mdounin.ru>
References: <20111107205312.3eb23869@e-healthexpert.org>
	<20111108080820.GW95664@mdounin.ru>
Message-ID: <20111108105831.5c6613c5@e-healthexpert.org>

On Tue, 8 Nov 2011 12:08:20 +0400, Maxim Dounin <mdounin at mdounin.ru>
wrote:

> > /tmp/buildd/nginx-1.1.7/debian/modules/nginx-echo/src/ngx_http_echo_timer.c:32:5:
> > error: too many arguments to function 'ngx_time_update'
> > src/core/ngx_times.h:23:6: note: declared here
> 
> Errors are caused by 3rd party module you are trying to use.  
> Either do not compile it, or update it (likely new version of the 
> module exists which resolves this).

Thank you. As usual you were right.

A simple sed -i '/MODULESDIR/d' debian/rules took care of the
offending 'nginx-echo' module and of all the other modules that bloat
the current ubuntu/debian version of Nginx code.

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


From varia at e-healthexpert.org  Tue Nov  8 11:05:07 2011
From: varia at e-healthexpert.org (Mark Alan)
Date: Tue, 8 Nov 2011 11:05:07 +0000
Subject: Is the 607418-ipv6-addresses.diff patch still needed?
Message-ID: <20111108110507.1a0bb3ae@e-healthexpert.org>

Hello,

The current version of the debianized Nginx code applies the
attached 607418-ipv6-addresses.diff patch.

Do you still see any need to apply this patch to the current
nginx_1.1.7 code?

Regards,

M.


########  607418-ipv6-addresses.diff patch follows  ########
Description: $host variable mis-parses IPv6 literal addresses from HTTP
Author: Steven Chamberlain <steven at pyro.eu.org>
Debian-Bug: http://bugs.debian.org/607418
Last-Update: 2010-12-30

Index: trunk/src/http/ngx_http_request.c
===================================================================
--- trunk.orig/src/http/ngx_http_request.c	2010-12-30
01:46:10.308926973 -0600 +++ trunk/src/http/ngx_http_request.c
2010-12-30 01:48:21.638927393 -0600 @@ -1650,11 +1650,12 @@
 {
     u_char      *h, ch;
     size_t       i, last;
-    ngx_uint_t   dot;
+    ngx_uint_t   dot, in_brackets;
 
     last = len;
     h = *host;
     dot = 0;
+    in_brackets = 0;
 
     for (i = 0; i < len; i++) {
         ch = h[i];
@@ -1670,11 +1671,27 @@
 
         dot = 0;
 
-        if (ch == ':') {
+        if (ch == '[' && i == 0) {
+            /* start of literal IPv6 address */
+            in_brackets = 1;
+            continue;
+        }
+
+        /*
+         * Inside square brackets, the colon is a delimeter for an
IPv6 address.
+         * Otherwise it comes before the port number, so remove it.
+         */
+        if (ch == ':' && !in_brackets) {
             last = i;
             continue;
         }
 
+        if (ch == ']') {
+            /* end of literal IPv6 address */
+            in_brackets = 0;
+            continue;
+        }
+
         if (ngx_path_separator(ch) || ch == '\0') {
             return 0;
         }
@@ -1684,6 +1701,11 @@
         }
     }
 
+    if (in_brackets) {
+        /* missing the closing square bracket for IPv6 address */
+        return 0;
+    }
+
     if (dot) {
         last--;
     }


From cyril.lavier at davromaniak.eu  Tue Nov  8 11:17:43 2011
From: cyril.lavier at davromaniak.eu (Cyril LAVIER)
Date: Tue, 08 Nov 2011 12:17:43 +0100
Subject: Is the 607418-ipv6-addresses.diff patch still  needed?
In-Reply-To: <20111108110507.1a0bb3ae@e-healthexpert.org>
References: <20111108110507.1a0bb3ae@e-healthexpert.org>
Message-ID: <b20eb320896d2bc79c652b78fc2bff42@davromaniak.eu>

 Hi Mark.

 I think it can be removed.

 I'm compiling the 1.1.7 packages without the patch, and I will test it 
 after.

 If this patch is not needed anymore, I will apply the changes in the 
 SVN.

 Thanks for this remark.

 On Tue, 8 Nov 2011 11:05:07 +0000, Mark Alan wrote:
> Hello,
>
> The current version of the debianized Nginx code applies the
> attached 607418-ipv6-addresses.diff patch.
>
> Do you still see any need to apply this patch to the current
> nginx_1.1.7 code?
>
> Regards,
>
> M.
>
>
> ########  607418-ipv6-addresses.diff patch follows  ########
> Description: $host variable mis-parses IPv6 literal addresses from 
> HTTP
> Author: Steven Chamberlain <steven at pyro.eu.org>
> Debian-Bug: http://bugs.debian.org/607418
> Last-Update: 2010-12-30
>
> Index: trunk/src/http/ngx_http_request.c
> ===================================================================
> --- trunk.orig/src/http/ngx_http_request.c	2010-12-30
> 01:46:10.308926973 -0600 +++ trunk/src/http/ngx_http_request.c
> 2010-12-30 01:48:21.638927393 -0600 @@ -1650,11 +1650,12 @@
>  {
>      u_char      *h, ch;
>      size_t       i, last;
> -    ngx_uint_t   dot;
> +    ngx_uint_t   dot, in_brackets;
>
>      last = len;
>      h = *host;
>      dot = 0;
> +    in_brackets = 0;
>
>      for (i = 0; i < len; i++) {
>          ch = h[i];
> @@ -1670,11 +1671,27 @@
>
>          dot = 0;
>
> -        if (ch == ':') {
> +        if (ch == '[' && i == 0) {
> +            /* start of literal IPv6 address */
> +            in_brackets = 1;
> +            continue;
> +        }
> +
> +        /*
> +         * Inside square brackets, the colon is a delimeter for an
> IPv6 address.
> +         * Otherwise it comes before the port number, so remove it.
> +         */
> +        if (ch == ':' && !in_brackets) {
>              last = i;
>              continue;
>          }
>
> +        if (ch == ']') {
> +            /* end of literal IPv6 address */
> +            in_brackets = 0;
> +            continue;
> +        }
> +
>          if (ngx_path_separator(ch) || ch == '\0') {
>              return 0;
>          }
> @@ -1684,6 +1701,11 @@
>          }
>      }
>
> +    if (in_brackets) {
> +        /* missing the closing square bracket for IPv6 address */
> +        return 0;
> +    }
> +
>      if (dot) {
>          last--;
>      }
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx

-- 
 Cyril "Davromaniak" Lavier


From kgorlo at gmail.com  Tue Nov  8 11:18:29 2011
From: kgorlo at gmail.com (Kamil Gorlo)
Date: Tue, 8 Nov 2011 12:18:29 +0100
Subject: Log response header from upstream
In-Reply-To: <20111107112911.GI95664@mdounin.ru>
References: <CAHzX=KBmF1nu5bN1B6yWoh+J-bTwJE93FpywpKRzxPu+CFEZGw@mail.gmail.com>
	<20111106211504.GY95664@mdounin.ru>
	<CAHzX=KDiJFfCxZWsrkV2_D8oZZSdVn=9u2u+BxU77_dwxm00Cg@mail.gmail.com>
	<20111107101902.GD95664@mdounin.ru>
	<CAHzX=KAC7iwJ2cZap_KQ+ORSjhOmq06qT5_vkiVyYciyWm-Dew@mail.gmail.com>
	<20111107112911.GI95664@mdounin.ru>
Message-ID: <CAHzX=KBxT2zfnbYFM0jVkdbox1cxzy_zPfSbZ6Y-3yKDk+fOOQ@mail.gmail.com>

> Hyphen is used when variable isn't found. ?Empty string - if it's
> found, but empty.
>

I see that variable defined by user is always empty (no hyphen) even
if it is not defined in location section connected with request (but
defined somewhere else in nginx config - other location section not
some section which is inherited). Is it normal?

Cheers,
Kamil Gorlo


From masterkorp at gmail.com  Tue Nov  8 11:22:38 2011
From: masterkorp at gmail.com (Alfredo Palhares)
Date: Tue, 8 Nov 2011 11:22:38 +0000
Subject: Running Shell Script from html webpage
In-Reply-To: <ce0a55728e7377210fd1b4df1b8a5338.NginxMailingListEnglish@forum.nginx.org>
References: <ce0a55728e7377210fd1b4df1b8a5338.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <CAOhncHHzPpp6m=NWaJCSatvk-ZTX28YSJLSeZaq5SfRDxP8cVA@mail.gmail.com>

Hello, well you can use a web anti-framework, that will enable you to use
shell power. You can also use a RC templata to place the code into the html.
Here is a link to werc http://www.werc.org/

-- 
Cumprimentos,
Alfredo Palhares
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20111108/de42f2e1/attachment.html>

From cyril.lavier at davromaniak.eu  Tue Nov  8 11:24:40 2011
From: cyril.lavier at davromaniak.eu (Cyril LAVIER)
Date: Tue, 08 Nov 2011 12:24:40 +0100
Subject: Is the 607418-ipv6-addresses.diff patch still   needed?
In-Reply-To: <b20eb320896d2bc79c652b78fc2bff42@davromaniak.eu>
References: <20111108110507.1a0bb3ae@e-healthexpert.org>
	<b20eb320896d2bc79c652b78fc2bff42@davromaniak.eu>
Message-ID: <7a11c714169e41c5faf133af0f69d71f@davromaniak.eu>

 I just tested it.

 It seems to work without the patch.

 Thanks for this usefull remark.

 On Tue, 08 Nov 2011 12:17:43 +0100, Cyril LAVIER wrote:
> Hi Mark.
>
> I think it can be removed.
>
> I'm compiling the 1.1.7 packages without the patch, and I will test 
> it after.
>
> If this patch is not needed anymore, I will apply the changes in the 
> SVN.
>
> Thanks for this remark.
>
> On Tue, 8 Nov 2011 11:05:07 +0000, Mark Alan wrote:
>> Hello,
>>
>> The current version of the debianized Nginx code applies the
>> attached 607418-ipv6-addresses.diff patch.
>>
>> Do you still see any need to apply this patch to the current
>> nginx_1.1.7 code?
>>
>> Regards,
>>
>> M.
>>
>>
>> ########  607418-ipv6-addresses.diff patch follows  ########
>> Description: $host variable mis-parses IPv6 literal addresses from 
>> HTTP
>> Author: Steven Chamberlain <steven at pyro.eu.org>
>> Debian-Bug: http://bugs.debian.org/607418
>> Last-Update: 2010-12-30
>>
>> Index: trunk/src/http/ngx_http_request.c
>> ===================================================================
>> --- trunk.orig/src/http/ngx_http_request.c	2010-12-30
>> 01:46:10.308926973 -0600 +++ trunk/src/http/ngx_http_request.c
>> 2010-12-30 01:48:21.638927393 -0600 @@ -1650,11 +1650,12 @@
>>  {
>>      u_char      *h, ch;
>>      size_t       i, last;
>> -    ngx_uint_t   dot;
>> +    ngx_uint_t   dot, in_brackets;
>>
>>      last = len;
>>      h = *host;
>>      dot = 0;
>> +    in_brackets = 0;
>>
>>      for (i = 0; i < len; i++) {
>>          ch = h[i];
>> @@ -1670,11 +1671,27 @@
>>
>>          dot = 0;
>>
>> -        if (ch == ':') {
>> +        if (ch == '[' && i == 0) {
>> +            /* start of literal IPv6 address */
>> +            in_brackets = 1;
>> +            continue;
>> +        }
>> +
>> +        /*
>> +         * Inside square brackets, the colon is a delimeter for an
>> IPv6 address.
>> +         * Otherwise it comes before the port number, so remove it.
>> +         */
>> +        if (ch == ':' && !in_brackets) {
>>              last = i;
>>              continue;
>>          }
>>
>> +        if (ch == ']') {
>> +            /* end of literal IPv6 address */
>> +            in_brackets = 0;
>> +            continue;
>> +        }
>> +
>>          if (ngx_path_separator(ch) || ch == '\0') {
>>              return 0;
>>          }
>> @@ -1684,6 +1701,11 @@
>>          }
>>      }
>>
>> +    if (in_brackets) {
>> +        /* missing the closing square bracket for IPv6 address */
>> +        return 0;
>> +    }
>> +
>>      if (dot) {
>>          last--;
>>      }
>>
>> _______________________________________________
>> nginx mailing list
>> nginx at nginx.org
>> http://mailman.nginx.org/mailman/listinfo/nginx

-- 
 Cyril "Davromaniak" Lavier


From varia at e-healthexpert.org  Tue Nov  8 11:52:11 2011
From: varia at e-healthexpert.org (Mark Alan)
Date: Tue, 8 Nov 2011 11:52:11 +0000
Subject: Is the 607418-ipv6-addresses.diff patch still   needed?
In-Reply-To: <7a11c714169e41c5faf133af0f69d71f@davromaniak.eu>
References: <20111108110507.1a0bb3ae@e-healthexpert.org>
	<b20eb320896d2bc79c652b78fc2bff42@davromaniak.eu>
	<7a11c714169e41c5faf133af0f69d71f@davromaniak.eu>
Message-ID: <20111108115211.5310be68@e-healthexpert.org>

On Tue, 08 Nov 2011 12:24:40 +0100, Cyril LAVIER
<cyril.lavier at davromaniak.eu> wrote:

>  I just tested it.
>  It seems to work without the patch.
>  Thanks for this usefull remark.

You are welcome.

Although unrelated, I should note that there seems to be a market for
a high security nginx-ultralight version.
Such a version would be especially suited for those websites (most of
the current CMS?) that process loads of fastcgi using php5-fpm and
php-apc.

I am attaching the debian/rules section that I have been using to get
such 'nginx-ultralight'

config.status.ultralight: config.env.ultralight config.sub config.guess
\ cd $(BUILDDIR_ultralight) && ./configure \
    --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 \
    --lock-path=/var/lock/nginx.lock \
    --pid-path=/var/run/nginx.pid \
    --with-http_gzip_static_module \
    --with-http_ssl_module \
    --without-http-cache \
    --without-http_browser_module \
    --without-http_geo_module \
    --without-http_limit_req_module \
    --without-http_limit_zone_module \
    --without-http_map_module \
    --without-http_memcached_module \
    --without-http_referer_module \
    --without-http_scgi_module \
    --without-http_split_clients_module \
    --without-http_ssi_module \
    --without-http_upstream_keepalive_module \
    --without-http_userid_module \
    --without-http_uwsgi_module \
            $(CONFIGURE_OPTS) >$@\
touch $@' 

Regards,
M.


From cyril.lavier at davromaniak.eu  Tue Nov  8 12:00:39 2011
From: cyril.lavier at davromaniak.eu (Cyril LAVIER)
Date: Tue, 08 Nov 2011 13:00:39 +0100
Subject: Is the 607418-ipv6-addresses.diff patch still    needed?
In-Reply-To: <20111108115211.5310be68@e-healthexpert.org>
References: <20111108110507.1a0bb3ae@e-healthexpert.org>
	<b20eb320896d2bc79c652b78fc2bff42@davromaniak.eu>
	<7a11c714169e41c5faf133af0f69d71f@davromaniak.eu>
	<20111108115211.5310be68@e-healthexpert.org>
Message-ID: <1c5231cd03c59e01dcc437d1ef37276f@davromaniak.eu>

 On Tue, 8 Nov 2011 11:52:11 +0000, Mark Alan wrote:
> On Tue, 08 Nov 2011 12:24:40 +0100, Cyril LAVIER
> <cyril.lavier at davromaniak.eu> wrote:
>
>>  I just tested it.
>>  It seems to work without the patch.
>>  Thanks for this usefull remark.
>
> You are welcome.
>
> Although unrelated, I should note that there seems to be a market for
> a high security nginx-ultralight version.
> Such a version would be especially suited for those websites (most of
> the current CMS?) that process loads of fastcgi using php5-fpm and
> php-apc.
>
> I am attaching the debian/rules section that I have been using to get
> such 'nginx-ultralight'
>
> config.status.ultralight: config.env.ultralight config.sub 
> config.guess
> \ cd $(BUILDDIR_ultralight) && ./configure \
>     --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 \
>     --lock-path=/var/lock/nginx.lock \
>     --pid-path=/var/run/nginx.pid \
>     --with-http_gzip_static_module \
>     --with-http_ssl_module \
>     --without-http-cache \
>     --without-http_browser_module \
>     --without-http_geo_module \
>     --without-http_limit_req_module \
>     --without-http_limit_zone_module \
>     --without-http_map_module \
>     --without-http_memcached_module \
>     --without-http_referer_module \
>     --without-http_scgi_module \
>     --without-http_split_clients_module \
>     --without-http_ssi_module \
>     --without-http_upstream_keepalive_module \
>     --without-http_userid_module \
>     --without-http_uwsgi_module \
>             $(CONFIGURE_OPTS) >$@\
> touch $@'
>

 Could you open a Debian Bug on the source package nginx for this 
 purpose ?

 I think it's a good idea to have a special nginx build for all CMS 
 related uses, but it needs to be discussed between the maintainers.

 Thanks.

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

-- 
 Cyril "Davromaniak" Lavier


From mdounin at mdounin.ru  Tue Nov  8 13:01:16 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Tue, 8 Nov 2011 17:01:16 +0400
Subject: Is the 607418-ipv6-addresses.diff patch still needed?
In-Reply-To: <20111108110507.1a0bb3ae@e-healthexpert.org>
References: <20111108110507.1a0bb3ae@e-healthexpert.org>
Message-ID: <20111108130116.GD95664@mdounin.ru>

Hello!

On Tue, Nov 08, 2011 at 11:05:07AM +0000, Mark Alan wrote:

> The current version of the debianized Nginx code applies the
> attached 607418-ipv6-addresses.diff patch.
> 
> Do you still see any need to apply this patch to the current
> nginx_1.1.7 code?

See http://trac.nginx.org/nginx/ticket/1.

Maxim Dounin


From jelledejong at powercraft.nl  Tue Nov  8 13:15:07 2011
From: jelledejong at powercraft.nl (Jelle de Jong)
Date: Tue, 08 Nov 2011 14:15:07 +0100
Subject: Is the 607418-ipv6-addresses.diff patch still    needed?
In-Reply-To: <1c5231cd03c59e01dcc437d1ef37276f@davromaniak.eu>
References: <20111108110507.1a0bb3ae@e-healthexpert.org>
	<b20eb320896d2bc79c652b78fc2bff42@davromaniak.eu>
	<7a11c714169e41c5faf133af0f69d71f@davromaniak.eu>
	<20111108115211.5310be68@e-healthexpert.org>
	<1c5231cd03c59e01dcc437d1ef37276f@davromaniak.eu>
Message-ID: <4EB92B5B.4040208@powercraft.nl>

On 08-11-11 13:00, Cyril LAVIER wrote:
> Could you open a Debian Bug on the source package nginx for this purpose ?
> I think it's a good idea to have a special nginx build for all CMS
> related uses, but it needs to be discussed between the maintainers.

+1 for wanting such a package :)

Kind regards,

Thanks in advance,

Jelle de Jong


-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 316 bytes
Desc: OpenPGP digital signature
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20111108/55c200c9/attachment.bin>

From mdounin at mdounin.ru  Tue Nov  8 13:31:15 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Tue, 8 Nov 2011 17:31:15 +0400
Subject: Log response header from upstream
In-Reply-To: <CAHzX=KBxT2zfnbYFM0jVkdbox1cxzy_zPfSbZ6Y-3yKDk+fOOQ@mail.gmail.com>
References: <CAHzX=KBmF1nu5bN1B6yWoh+J-bTwJE93FpywpKRzxPu+CFEZGw@mail.gmail.com>
	<20111106211504.GY95664@mdounin.ru>
	<CAHzX=KDiJFfCxZWsrkV2_D8oZZSdVn=9u2u+BxU77_dwxm00Cg@mail.gmail.com>
	<20111107101902.GD95664@mdounin.ru>
	<CAHzX=KAC7iwJ2cZap_KQ+ORSjhOmq06qT5_vkiVyYciyWm-Dew@mail.gmail.com>
	<20111107112911.GI95664@mdounin.ru>
	<CAHzX=KBxT2zfnbYFM0jVkdbox1cxzy_zPfSbZ6Y-3yKDk+fOOQ@mail.gmail.com>
Message-ID: <20111108133115.GE95664@mdounin.ru>

Hello!

On Tue, Nov 08, 2011 at 12:18:29PM +0100, Kamil Gorlo wrote:

> > Hyphen is used when variable isn't found. ?Empty string - if it's
> > found, but empty.
> >
> 
> I see that variable defined by user is always empty (no hyphen) even
> if it is not defined in location section connected with request (but
> defined somewhere else in nginx config - other location section not
> some section which is inherited). Is it normal?

Yes.

Maxim Dounin


From me at keithfernie.co.uk  Tue Nov  8 14:13:03 2011
From: me at keithfernie.co.uk (Keith Fernie)
Date: Tue, 08 Nov 2011 14:13:03 -0000
Subject: Running Shell Script from html webpage
In-Reply-To: <ce0a55728e7377210fd1b4df1b8a5338.NginxMailingListEnglish@forum.nginx.org>
References: <ce0a55728e7377210fd1b4df1b8a5338.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <op.v4msf1phdv45qa@desktop.me4.it>

I'am doing this with Debian Squeeze & fcgiwrap installed. I've seen  
fcgiwrap present in Ubuntu.

Example script (Must be executable)

#!/bin/sh
# -*- coding: utf-8 -*-
NAME=`"cpuinfo"`
echo "Content-type:text/html\r\n"
echo "<html><head>"
echo "<title>$NAME</title>"
echo '<meta name="description" content="'$NAME'">'
echo '<meta name="keywords" content="'$NAME'">'
echo '<meta http-equiv="Content-type" content="text/html;charset=UTF-8">'
echo '<meta name="ROBOTS" content="noindex">'
echo "</head><body><pre>"
date
echo "\nuname -a"
uname -a
echo "\ncpuinfo"
cat /proc/cpuinfo
echo "</pre></body></html>"

Run it here http://newhost.qaq.me/cpuinfo.sh

Also using this as an include file, not restricted to only shell scripts.

location ~ (\.cgi|\.py|\.sh|\.pl|\.lua)$ {
gzip off;
root /var/www/$server_name;
autoindex on;
fastcgi_pass unix:/var/run/fcgiwrap.socket;
include /etc/nginx/fastcgi_params;
fastcgi_param DOCUMENT_ROOT /var/www/$server_name;
fastcgi_param SCRIPT_FILENAME /var/www/$server_name$fastcgi_script_name;
}

On Tue, 08 Nov 2011 10:01:07 -0000, etrader <nginx-forum at nginx.us> wrote:

> I have a Nginx installed on Linux (Ubuntu or Centos) without any
> scripting language such as PHP or Python (connected to the webserver).
> Can I run Shell commands from a webpage (e.g. html)?
>
> I mean is it possible to use nginx without script languages by shell
> scripts?
>
> Posted at Nginx Forum: 
> http://forum.nginx.org/read.php?2,217975,217975#msg-217975
>
> _______________________________________________
> 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/20111108/00775907/attachment.html>

From nginx-forum at nginx.us  Tue Nov  8 15:23:58 2011
From: nginx-forum at nginx.us (CarlWang)
Date: Tue, 08 Nov 2011 10:23:58 -0500
Subject: How to handle NGX_AGAIN returned by
	ngx_http_read_client_request_body() within handler module?
In-Reply-To: <CAB4Tn6N_edvaiCKPKOEgcpkBbEhHLMUdz3+CR-RRHHMKYUpA5Q@mail.gmail.com>
References: <CAB4Tn6N_edvaiCKPKOEgcpkBbEhHLMUdz3+CR-RRHHMKYUpA5Q@mail.gmail.com>
Message-ID: <19de791219ef7a970c5c38a0be50c9cc.NginxMailingListEnglish@forum.nginx.org>

Thanks for your help.

I just followed Emiller's Guide To Nginx Module Development and build my
Non-proxying Handler. I'm not quite sure which phase is the handler
function running in. Maybe NGX_HTTP_ACCESS_PHASE or
NGX_HTTP_CONTENT_PHASE.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,3545,218016#msg-218016


From nginx-forum at nginx.us  Tue Nov  8 16:03:03 2011
From: nginx-forum at nginx.us (etrader)
Date: Tue, 08 Nov 2011 11:03:03 -0500
Subject: Speed up static content
Message-ID: <37afec7fa5501b7d6f876ac308769295.NginxMailingListEnglish@forum.nginx.org>

Is there a special configuration for static content like image files? I
have no problem with nginx, and the reason that I am asking is to know
if there is a different nginx conf to load static files faster.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218021,218021#msg-218021


From daniel.carrillo at gmail.com  Tue Nov  8 16:23:13 2011
From: daniel.carrillo at gmail.com (Daniel Carrillo)
Date: Tue, 8 Nov 2011 17:23:13 +0100
Subject: Speed up static content
In-Reply-To: <37afec7fa5501b7d6f876ac308769295.NginxMailingListEnglish@forum.nginx.org>
References: <37afec7fa5501b7d6f876ac308769295.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <CAK6GuyRHze4TEHrB6ehdmgb5VJPxkK3fOjAFHhVq43nWZMifYg@mail.gmail.com>

2011/11/8 etrader <nginx-forum at nginx.us>:
> Is there a special configuration for static content like image files? I
> have no problem with nginx, and the reason that I am asking is to know
> if there is a different nginx conf to load static files faster.

Hard to say without know your hardware, OS, processor, etc, but look at this:

http://mailman.nginx.org/pipermail/nginx/2010-October/023025.html

Hope it helps.


From nginx-forum at nginx.us  Tue Nov  8 16:44:26 2011
From: nginx-forum at nginx.us (CarlWang)
Date: Tue, 08 Nov 2011 11:44:26 -0500
Subject: How to handle NGX_AGAIN returned by
	ngx_http_read_client_request_body() within handler module?
In-Reply-To: <19de791219ef7a970c5c38a0be50c9cc.NginxMailingListEnglish@forum.nginx.org>
References: <CAB4Tn6N_edvaiCKPKOEgcpkBbEhHLMUdz3+CR-RRHHMKYUpA5Q@mail.gmail.com>
	<19de791219ef7a970c5c38a0be50c9cc.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <316a9980abe0fd5ee19249b475960982.NginxMailingListEnglish@forum.nginx.org>

I checked the error.log and found the ngx_http_v8_handler_request called
in the post access phase.
Then how should  the coding structure be?

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,3545,218023#msg-218023


From nginx-forum at nginx.us  Tue Nov  8 16:56:53 2011
From: nginx-forum at nginx.us (lpugoy)
Date: Tue, 08 Nov 2011 11:56:53 -0500
Subject: Proxy cache for php site
In-Reply-To: <CAKJ4GFVF_qQdZ9V-Hw+J0KCwAO4kepi-cGfm-Pm4kR7D95U2Rw@mail.gmail.com>
References: <CAKJ4GFVF_qQdZ9V-Hw+J0KCwAO4kepi-cGfm-Pm4kR7D95U2Rw@mail.gmail.com>
Message-ID: <205f048aeb2995eebccc12918de5a285.NginxMailingListEnglish@forum.nginx.org>

I'm sorry, but I don't understand the effect of cookies on proxy
caching. I'm trying to implement the same configuration as the one
referred to, and I'm also having trouble in that it doesn't seem to be
caching. I'm also using httperf. For every request the backend does
return a different cookie. Is this what is preventing the caching from
happening? 

In the debug log, I found some entries that refer to the cache. They are
below:

2011/11/08 20:51:45 [debug] 21434#0: *1 http cache key:
"httpwww.site.comGET/"
2011/11/08 20:51:45 [debug] 21434#0: *1 http script var: ""
2011/11/08 20:51:45 [debug] 21434#0: *1 add cleanup: 091D56A4
2011/11/08 20:51:45 [debug] 21434#0: shmtx lock
2011/11/08 20:51:45 [debug] 21434#0: slab alloc: 76 slot: 4
2011/11/08 20:51:45 [debug] 21434#0: slab alloc: B6F6B000
2011/11/08 20:51:45 [debug] 21434#0: shmtx unlock
2011/11/08 20:51:45 [debug] 21434#0: *1 http file cache exists: -5 e:0
2011/11/08 20:51:45 [debug] 21434#0: *1 cache file:
"/var/cache/nginx/5/86/629d786ceca8ff787779a3e4ccdc8865"
2011/11/08 20:51:45 [debug] 21434#0: *1 add cleanup: 091D56E8
2011/11/08 20:51:45 [debug] 21434#0: *1 http upstream cache: -5

What do they mean? Thank you.

Jon Bennett Wrote:
-------------------------------------------------------
> Hi Appa,
> 
> got it sorted thanks, had cookies being written
> all the time and that
> was stopping it. Now in production, many thanks!
> 
> Jon

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,216756,217993#msg-217993


From nginx-forum at nginx.us  Tue Nov  8 17:06:08 2011
From: nginx-forum at nginx.us (etrader)
Date: Tue, 08 Nov 2011 12:06:08 -0500
Subject: Speed up static content
In-Reply-To: <CAK6GuyRHze4TEHrB6ehdmgb5VJPxkK3fOjAFHhVq43nWZMifYg@mail.gmail.com>
References: <CAK6GuyRHze4TEHrB6ehdmgb5VJPxkK3fOjAFHhVq43nWZMifYg@mail.gmail.com>
Message-ID: <9c4be9f7898f71bf17580caf07153850.NginxMailingListEnglish@forum.nginx.org>

Thanks Daniel! I run nginx (1.0.4 or 1.0.5) on different servers, but in
general, they are based on Centos or Ubuntu running php via php-fpm.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218021,218026#msg-218026


From ianevans at digitalhit.com  Tue Nov  8 17:59:54 2011
From: ianevans at digitalhit.com (Ian M. Evans)
Date: Tue, 8 Nov 2011 12:59:54 -0500
Subject: Mailing list software that plays nice with nginx
Message-ID: <9ff9e5633d4d8a7efa6229b329e6f092.squirrel@www.digitalhit.com>

Hi everyone,

Looking for s script (preferably PHP) that plays nice with nginx "out of
the box" i.e. doesn't require rewriting .htaccess files into nginx rules,
etc.

Just lloking for something that will let me send out an announcement-only
list, handle people adding/deleting themselves, etc. Scheduled emails
would be a blessing.

Anything out there fit the bill?


From ianevans at digitalhit.com  Tue Nov  8 18:08:31 2011
From: ianevans at digitalhit.com (Ian M. Evans)
Date: Tue, 8 Nov 2011 13:08:31 -0500
Subject: Mailing list software that plays nice with nginx
In-Reply-To: <9ff9e5633d4d8a7efa6229b329e6f092.squirrel@www.digitalhit.com>
References: <9ff9e5633d4d8a7efa6229b329e6f092.squirrel@www.digitalhit.com>
Message-ID: <f807aa40d457323e2fcbad138f73789b.squirrel@www.digitalhit.com>

On Tue, November 8, 2011 12:59 pm, Ian M. Evans wrote:
> Looking for s script (preferably PHP) that plays nice with nginx "out of
> the box" i.e. doesn't require rewriting .htaccess files into nginx rules,
> etc.
>

Okay, just discovered http://wiki.nginx.org/PHPList minutes after sending
this email. If only I had waited. :-)

Anyway, the settings there seem to be if the list software is installed at
the document root.

How would I change the locations if the software is installed at /lists?
[I suck at locations and haven't had much sleep!]

Thanks.


From ionathan at gmail.com  Tue Nov  8 18:36:14 2011
From: ionathan at gmail.com (Jonathan Leibiusky)
Date: Tue, 8 Nov 2011 15:36:14 -0300
Subject: @fallback is not working in version 1.0.4?
Message-ID: <CAO6fm-SAD0+_r+S_3o_MRPr4b9QoeQ8kapv9hOBhMKSE0TD1Qg@mail.gmail.com>

Hi! I'm using nginx 1.0.4 and I have the following configuration:

server {
    listen 8080;
    server_name xxx.yyy.com;
    location / {
    error_page 500 502 503 @fallback;
        proxy_pass http://upstream1;
    }
    location @fallback {
        proxy_pass http://upstream2;
    }
}

And for some reason I don't get to upstream2 even if requests from
upstream1 are 500.
Am I doing something wrong?

Thanks!

Jonathan
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20111108/8e5e3093/attachment.html>

From igor at sysoev.ru  Tue Nov  8 18:50:42 2011
From: igor at sysoev.ru (Igor Sysoev)
Date: Tue, 8 Nov 2011 22:50:42 +0400
Subject: @fallback is not working in version 1.0.4?
In-Reply-To: <CAO6fm-SAD0+_r+S_3o_MRPr4b9QoeQ8kapv9hOBhMKSE0TD1Qg@mail.gmail.com>
References: <CAO6fm-SAD0+_r+S_3o_MRPr4b9QoeQ8kapv9hOBhMKSE0TD1Qg@mail.gmail.com>
Message-ID: <20111108185042.GA21062@nginx.com>

On Tue, Nov 08, 2011 at 03:36:14PM -0300, Jonathan Leibiusky wrote:
> Hi! I'm using nginx 1.0.4 and I have the following configuration:
> 
> server {
>     listen 8080;
>     server_name xxx.yyy.com;
>     location / {
>     error_page 500 502 503 @fallback;
>         proxy_pass http://upstream1;

          proxy_intercept_errors  on;

>     }
>     location @fallback {
>         proxy_pass http://upstream2;
>     }
> }
> 
> And for some reason I don't get to upstream2 even if requests from
> upstream1 are 500.
> Am I doing something wrong?


-- 
Igor Sysoev


From ionathan at gmail.com  Tue Nov  8 18:58:44 2011
From: ionathan at gmail.com (Jonathan Leibiusky)
Date: Tue, 8 Nov 2011 15:58:44 -0300
Subject: @fallback is not working in version 1.0.4?
In-Reply-To: <20111108185042.GA21062@nginx.com>
References: <CAO6fm-SAD0+_r+S_3o_MRPr4b9QoeQ8kapv9hOBhMKSE0TD1Qg@mail.gmail.com>
	<20111108185042.GA21062@nginx.com>
Message-ID: <CAO6fm-SU5jxSZ5T=kPP_huw8pX_bLvop7zOk9jXr+u_FbhUCLQ@mail.gmail.com>

awesome! thanks a lot!

On Tue, Nov 8, 2011 at 3:50 PM, Igor Sysoev <igor at sysoev.ru> wrote:

> On Tue, Nov 08, 2011 at 03:36:14PM -0300, Jonathan Leibiusky wrote:
> > Hi! I'm using nginx 1.0.4 and I have the following configuration:
> >
> > server {
> >     listen 8080;
> >     server_name xxx.yyy.com;
> >     location / {
> >     error_page 500 502 503 @fallback;
> >         proxy_pass http://upstream1;
>
>           proxy_intercept_errors  on;
>
> >     }
> >     location @fallback {
> >         proxy_pass http://upstream2;
> >     }
> > }
> >
> > And for some reason I don't get to upstream2 even if requests from
> > upstream1 are 500.
> > Am I doing something wrong?
>
>
> --
> 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/20111108/569b1b3f/attachment.html>

From francis at daoine.org  Tue Nov  8 21:29:04 2011
From: francis at daoine.org (Francis Daly)
Date: Tue, 8 Nov 2011 21:29:04 +0000
Subject: Proxy cache for php site
In-Reply-To: <205f048aeb2995eebccc12918de5a285.NginxMailingListEnglish@forum.nginx.org>
References: <CAKJ4GFVF_qQdZ9V-Hw+J0KCwAO4kepi-cGfm-Pm4kR7D95U2Rw@mail.gmail.com>
	<205f048aeb2995eebccc12918de5a285.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20111108212904.GO27078@craic.sysops.org>

On Tue, Nov 08, 2011 at 11:56:53AM -0500, lpugoy wrote:

Hi there,

> I'm sorry, but I don't understand the effect of cookies on proxy
> caching.

Caching involves fetching something once from the backend, and then
sending the same response to many requests.

A cookie is something that is set differently per request.

If you have one, you don't have the other.

By default.

> I'm trying to implement the same configuration as the one
> referred to, and I'm also having trouble in that it doesn't seem to be
> caching. I'm also using httperf. For every request the backend does
> return a different cookie. Is this what is preventing the caching from
> happening? 

Probably.

There are (rfc) rules on caching and cacheability of a http response --
but be aware that not all of them apply to a reverse proxy like nginx.

Look at the http headers sent from your backend. If they indicate that the
response is not cacheable, nginx won't cache it -- unless you configure
it to. Setting a cookie is one way to make a response non-cacheable.

If you don't want the cookie set, don't set it.

Out-of-the-box, nginx obeys the rules. Get your backend to obey the
rules too, and it will all work fine.

If you won't do that, then you can (probably) configure nginx to break the
rules in the specific way that you want it to when proxying your backend.

For example, if the backend sets a cookie and you want nginx to cache
that, do you want it to set the same cookie for all requests; or to set
no cookie for all requests; or something else?

You're likely much better off fixing the backend.

Good luck,

	f
-- 
Francis Daly        francis at daoine.org


From mdounin at mdounin.ru  Tue Nov  8 21:30:52 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Wed, 9 Nov 2011 01:30:52 +0400
Subject: Proxy cache for php site
In-Reply-To: <205f048aeb2995eebccc12918de5a285.NginxMailingListEnglish@forum.nginx.org>
References: <CAKJ4GFVF_qQdZ9V-Hw+J0KCwAO4kepi-cGfm-Pm4kR7D95U2Rw@mail.gmail.com>
	<205f048aeb2995eebccc12918de5a285.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20111108213052.GK95664@mdounin.ru>

Hello!

On Tue, Nov 08, 2011 at 11:56:53AM -0500, lpugoy wrote:

> I'm sorry, but I don't understand the effect of cookies on proxy
> caching. I'm trying to implement the same configuration as the one
> referred to, and I'm also having trouble in that it doesn't seem to be
> caching. I'm also using httperf. For every request the backend does
> return a different cookie. Is this what is preventing the caching from
> happening? 

Yes, nginx won't cache responses with cookies unless you 
specifically ask it to via proxy_ignore_headers directive[1].

Please also note that if you do so, you may want to also instruct 
nginx to hide returned cookies from clients (via proxy_hide_header 
directive[2]), or you'll end up with multiple clients with the 
same cookie.

Example configuration should look like:

    proxy_ignore_headers Set-Cookie;
    proxy_hide_header Set-Cookie;

Depending on the actual headers your backend returns you may also 
need to ignore other headers as well for cache to work, notably 
Expires and Cache-Control.

Alternatively, you may want to instruct your backend to not return 
headers which prevent caching.

[1] http://wiki.nginx.org/HttpProxyModule#proxy_ignore_headers
[2] http://wiki.nginx.org/HttpProxyModule#proxy_hide_header

> In the debug log, I found some entries that refer to the cache. They are
> below:
> 
> 2011/11/08 20:51:45 [debug] 21434#0: *1 http cache key:
> "httpwww.site.comGET/"
> 2011/11/08 20:51:45 [debug] 21434#0: *1 http script var: ""
> 2011/11/08 20:51:45 [debug] 21434#0: *1 add cleanup: 091D56A4
> 2011/11/08 20:51:45 [debug] 21434#0: shmtx lock
> 2011/11/08 20:51:45 [debug] 21434#0: slab alloc: 76 slot: 4
> 2011/11/08 20:51:45 [debug] 21434#0: slab alloc: B6F6B000
> 2011/11/08 20:51:45 [debug] 21434#0: shmtx unlock
> 2011/11/08 20:51:45 [debug] 21434#0: *1 http file cache exists: -5 e:0
> 2011/11/08 20:51:45 [debug] 21434#0: *1 cache file:
> "/var/cache/nginx/5/86/629d786ceca8ff787779a3e4ccdc8865"
> 2011/11/08 20:51:45 [debug] 21434#0: *1 add cleanup: 091D56E8
> 2011/11/08 20:51:45 [debug] 21434#0: *1 http upstream cache: -5
> 
> What do they mean? Thank you.

This debug lines correspond to lookup of the request in cache, and 
they show that no cached response exists.

Maxim Dounin


From keith at scott-land.net  Wed Nov  9 01:27:47 2011
From: keith at scott-land.net (Keith)
Date: Wed, 09 Nov 2011 01:27:47 +0000
Subject: php-fpm and nginx return "The page you are looking for is
	temporarily unavailable. Please try again later. "
In-Reply-To: <9636a5d8da2b172192e56ee838fe791e.NginxMailingListEnglish@forum.nginx.org>
References: <9636a5d8da2b172192e56ee838fe791e.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <4EB9D713.9040902@scott-land.net>

Hi,
         I was getting the same error earlier today. I eventually 
figured out that for me it was the mssql (NOT MySql) extension that was 
messing things up. A phpinfo() page would work fine but the mssql data 
import script that I was writing wouldn't work through a browser 
request. I noticed some Segfault errors in one of the logs.

My script ran fine from the command line but just not through php-fpm. 
Didn't try fast-cgi though.

Keith.

On 08/11/2011 09:25, abcomp01 wrote:
> php-fpm and nginx return "The page you are looking for is temporarily
> unavailable. Please try again later. "
>
> call http://173.212.207.146/upload/info.php
>
> 		location ~ \.php$ {
> 		  root   /home/admin/18loadnet;
> 		  fastcgi_pass  127.0.0.1:9000;
> 		  fastcgi_index index.php;
> 		  fastcgi_param  SCRIPT_FILENAME  /home/admin/info.php;
> 		  include /etc/nginx/fastcgi_params;
> 		}
>
> Posted at Nginx Forum: http://forum.nginx.org/read.php?2,217967,217967#msg-217967
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx


From agentzh at gmail.com  Wed Nov  9 01:28:48 2011
From: agentzh at gmail.com (agentzh)
Date: Wed, 9 Nov 2011 09:28:48 +0800
Subject: How to handle NGX_AGAIN returned by
	ngx_http_read_client_request_body() within handler module?
In-Reply-To: <316a9980abe0fd5ee19249b475960982.NginxMailingListEnglish@forum.nginx.org>
References: <CAB4Tn6N_edvaiCKPKOEgcpkBbEhHLMUdz3+CR-RRHHMKYUpA5Q@mail.gmail.com>
	<19de791219ef7a970c5c38a0be50c9cc.NginxMailingListEnglish@forum.nginx.org>
	<316a9980abe0fd5ee19249b475960982.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <CAB4Tn6PHGGpidTTc=WukjP4XEhov4hoMbiWOk5g59Zsmeni+KA@mail.gmail.com>

On Wed, Nov 9, 2011 at 12:44 AM, CarlWang <nginx-forum at nginx.us> wrote:
> I checked the error.log and found the ngx_http_v8_handler_request called
> in the post access phase.
> Then how should ?the coding structure be?
>

I've recommended our ngx_lua module for your reference:

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

Please check its source out.

Best,
-agentzh


From btm at loftninjas.org  Wed Nov  9 01:30:50 2011
From: btm at loftninjas.org (Bryan McLellan)
Date: Tue, 8 Nov 2011 20:30:50 -0500
Subject: SOLVED: Timeout when sending over 16k of data with UTF-8 characters
Message-ID: <CAMojFHD6XHzDEp979CQOe=X9ZiZp+fxbBXySHeUvBR+v2CQQZQ@mail.gmail.com>

On Tue, Nov 1, 2011 at 9:56 AM, Maxim Dounin <mdounin at mdounin.ru> wrote:
> Here 16384 bytes has been read from client, and one byte remains
> ("rest 1"). ?OpenSSL doesn't provide any more bytes and claims it
> needs more network input (2, SSL_ERROR_WANT_READ).
>
> 2011/11/01 12:44:24 [debug] 13689#0: *129 event timer del: 12: 1596563798
> 2011/11/01 12:44:24 [debug] 13689#0: *129 event timer add: 12: 60000:1596564354
> 2011/11/01 12:45:24 [debug] 13689#0: *129 event timer del: 12: 1596564354
> 2011/11/01 12:45:24 [debug] 13689#0: *129 http run request: "/organizations/opscode-btm/nodes/broken?"
> 2011/11/01 12:45:24 [debug] 13689#0: *129 http finalize request: 408, "/organizations/opscode-btm/nodes/broken?" a:1, c:1
>
> Though client fails to provide one more byte.
>
> For me, it looks like problem in client. ?It's either calculate
> content length incorrectly or fails to properly flush ssl buffers
> on it's side.

Thanks! I confirmed that it was likely on the client side SSL by
creating a duplicate nginx configuration that lacked SSL, which did
not present the symptoms.

I then narrowed it down to Ruby's SSL implementation and found a
recent bug fix that resolves the issue.

http://redmine.ruby-lang.org/issues/5233

Bryan


From nginx-forum at nginx.us  Wed Nov  9 02:14:04 2011
From: nginx-forum at nginx.us (dbanks)
Date: Tue, 08 Nov 2011 21:14:04 -0500
Subject: gzip - unexplained side effects
Message-ID: <48ec591068896e53321f58945d5a8836.NginxMailingListEnglish@forum.nginx.org>

Hi,

I'm running nginx 1.0.0 in front of a FCGI backend.  We've been running
in production for about 4 months, and have really been impressed with
the performance and stability of nginx.

We run a medium-volume appliction: 1000 to 4000 requests/sec spread over
2 instances by an upstream round robin load balancer.  We use
keepalives, but keep them rather short given the request volume to keep
the number of open connections manageable.  

Recently, I was working to improve our gzip settings.  The largest
change was that I added an explicit gzip buffer line (find my config
below - the three lines that are commented out seem to be correlated
with this issue)

After the change, the volume of outbound traffic decreased measurably,
as if gzip was not originally doing much due to inadequate buffer space.
 Great news!, or so I thought.  What also changed was that the number of
active connections fell by about 75% - the gzip change was somehow
causing keepalives to be closed prematurely.  Also, our volume of
incoming requests decreased a bit: as if some requests were being
aborted (though accepts == handled).   The request volume makes
debugging this particular issue somewhat troublesome, since I have yet
to replicate it in a quiet instance.

The guts of my configuration appear below.  This is such an unexpected
issue that I'm not doing a great job of setting up my question well. 
What I think I'd like to know is how could a change to the gzip buffers
(or the other two commented changes) impact keepalives or overall
connection negotiation?  Also, any suggestions as to how to go about
debugging it?

    sendfile        off;  
    tcp_nodelay on;  
    ignore_invalid_headers  on;  
    if_modified_since off;

    gzip on;
    gzip_comp_level 9;
    gzip_types text/javascript text/plain application/x-javascript;
    gzip_disable "MSIE [1-6]\.(?!.*SV1)"  
    #gzip_buffers 512 4k;
    #gzip_min_length  1100;  #if it fits in one packet, no worries
    #gzip_http_version 1.1;

    keepalive_timeout  6;  
    keepalive_requests 4;  

Cheers,

Dean

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218044,218044#msg-218044


From mike503 at gmail.com  Wed Nov  9 04:59:56 2011
From: mike503 at gmail.com (Michael Shadle)
Date: Tue, 8 Nov 2011 20:59:56 -0800
Subject: t-shirts
In-Reply-To: <1307994899.2744.465.camel@portable-evil>
References: <1307994899.2744.465.camel@portable-evil>
Message-ID: <CAKa35F7jK_tx-trPuSq4xBC9Qhvas3+Gp_SP61DAOx4ONKvMxQ@mail.gmail.com>

anyone still know of any T-shirt options?

On Mon, Jun 13, 2011 at 12:54 PM, Cliff Wells <cliff at develix.com> wrote:
> For anyone interested, the guy who printed the Nginx t-shirts a couple
> of years ago finally put the remainder up on eBay:
>
> http://cgi.ebay.com/NGINX-t-shirt-/170654481676?pt=US_Mens_Tshirts&var=&hash=item6d700b391a#ht_500wt_1156
>
> He went out of business, so this is it.
>
> Regards,
> Cliff
>
>
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://nginx.org/mailman/listinfo/nginx
>


From amoiz.shine at gmail.com  Wed Nov  9 06:07:32 2011
From: amoiz.shine at gmail.com (Sharl.Jimh.Tsin)
Date: Wed, 09 Nov 2011 14:07:32 +0800
Subject: t-shirts
In-Reply-To: <CAKa35F7jK_tx-trPuSq4xBC9Qhvas3+Gp_SP61DAOx4ONKvMxQ@mail.gmail.com>
References: <1307994899.2744.465.camel@portable-evil>
	<CAKa35F7jK_tx-trPuSq4xBC9Qhvas3+Gp_SP61DAOx4ONKvMxQ@mail.gmail.com>
Message-ID: <1320818852.2227.1.camel@sharl-desktop>

? 2011-11-08?? 20:59 -0800?Michael Shadle???
> anyone still know of any T-shirt options?
> 
> On Mon, Jun 13, 2011 at 12:54 PM, Cliff Wells <cliff at develix.com> wrote:
> > For anyone interested, the guy who printed the Nginx t-shirts a couple
> > of years ago finally put the remainder up on eBay:
> >
> > http://cgi.ebay.com/NGINX-t-shirt-/170654481676?pt=US_Mens_Tshirts&var=&hash=item6d700b391a#ht_500wt_1156
> >
> > He went out of business, so this is it.
> >
> > Regards,
> > Cliff
> >
> >
> >
> > _______________________________________________
> > nginx mailing list
> > nginx at nginx.org
> > http://nginx.org/mailman/listinfo/nginx
> >
> 
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx

is this t-shirt sold by NGINX official?

-- 
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/20111109/c9ad50a4/attachment.bin>

From mike503 at gmail.com  Wed Nov  9 06:11:07 2011
From: mike503 at gmail.com (Michael Shadle)
Date: Tue, 8 Nov 2011 22:11:07 -0800
Subject: t-shirts
In-Reply-To: <1320818852.2227.1.camel@sharl-desktop>
References: <1307994899.2744.465.camel@portable-evil>
	<CAKa35F7jK_tx-trPuSq4xBC9Qhvas3+Gp_SP61DAOx4ONKvMxQ@mail.gmail.com>
	<1320818852.2227.1.camel@sharl-desktop>
Message-ID: <CAKa35F6tH8zgp6FRbxE3nRAoF3_Mw62UBnZVgAudJpfd4_U65Q@mail.gmail.com>

No I don't believe they were ever "official" - I don't believe there
are official ones.

On Tue, Nov 8, 2011 at 10:07 PM, Sharl.Jimh.Tsin <amoiz.shine at gmail.com> wrote:

> is this t-shirt sold by NGINX official?


From amoiz.shine at gmail.com  Wed Nov  9 06:25:06 2011
From: amoiz.shine at gmail.com (Sharl.Jimh.Tsin)
Date: Wed, 09 Nov 2011 14:25:06 +0800
Subject: t-shirts
In-Reply-To: <CAKa35F6tH8zgp6FRbxE3nRAoF3_Mw62UBnZVgAudJpfd4_U65Q@mail.gmail.com>
References: <1307994899.2744.465.camel@portable-evil>
	<CAKa35F7jK_tx-trPuSq4xBC9Qhvas3+Gp_SP61DAOx4ONKvMxQ@mail.gmail.com>
	<1320818852.2227.1.camel@sharl-desktop>
	<CAKa35F6tH8zgp6FRbxE3nRAoF3_Mw62UBnZVgAudJpfd4_U65Q@mail.gmail.com>
Message-ID: <1320819906.2227.5.camel@sharl-desktop>

? 2011-11-08?? 22:11 -0800?Michael Shadle???
> No I don't believe they were ever "official" - I don't believe there
> are official ones.
> 
> On Tue, Nov 8, 2011 at 10:07 PM, Sharl.Jimh.Tsin <amoiz.shine at gmail.com> wrote:
> 
> > is this t-shirt sold by NGINX official?
> 
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx

well,if it is official one,i am willing to buy one for support.

but for someone's personal action,i will worry about the quality of this
clothes.

-- 
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/20111109/3baf9a04/attachment.bin>

From cliff at develix.com  Wed Nov  9 07:18:55 2011
From: cliff at develix.com (Cliff Wells)
Date: Tue, 08 Nov 2011 23:18:55 -0800
Subject: t-shirts
In-Reply-To: <1320819906.2227.5.camel@sharl-desktop>
References: <1307994899.2744.465.camel@portable-evil>
	<CAKa35F7jK_tx-trPuSq4xBC9Qhvas3+Gp_SP61DAOx4ONKvMxQ@mail.gmail.com>
	<1320818852.2227.1.camel@sharl-desktop>
	<CAKa35F6tH8zgp6FRbxE3nRAoF3_Mw62UBnZVgAudJpfd4_U65Q@mail.gmail.com>
	<1320819906.2227.5.camel@sharl-desktop>
Message-ID: <1320823135.12168.4.camel@portable-evil>

I can ask if he has any more and see if he will re-list them if he does.

As far as quality, the screen printing is good (my wife and I have had
ours for a few years now).  The shirts themselves are fairly heavy
material (for a t-shirt).  The biggest issue is that white t-shirts get
not-white over time.

Regards,
Cliff

On Wed, 2011-11-09 at 14:25 +0800, Sharl.Jimh.Tsin wrote:
> ? 2011-11-08?? 22:11 -0800?Michael Shadle???
> > No I don't believe they were ever "official" - I don't believe there
> > are official ones.
> > 
> > On Tue, Nov 8, 2011 at 10:07 PM, Sharl.Jimh.Tsin <amoiz.shine at gmail.com> wrote:
> > 
> > > is this t-shirt sold by NGINX official?
> > 
> > _______________________________________________
> > nginx mailing list
> > nginx at nginx.org
> > http://mailman.nginx.org/mailman/listinfo/nginx
> 
> well,if it is official one,i am willing to buy one for support.
> 
> but for someone's personal action,i will worry about the quality of this
> clothes.
> 
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx



From nginx-forum at nginx.us  Wed Nov  9 11:02:52 2011
From: nginx-forum at nginx.us (lpugoy)
Date: Wed, 09 Nov 2011 06:02:52 -0500
Subject: Proxy cache for php site
In-Reply-To: <20111108213052.GK95664@mdounin.ru>
References: <20111108213052.GK95664@mdounin.ru>
Message-ID: <6a7d9d2ee046df8d9c9cfdee121e16d8.NginxMailingListEnglish@forum.nginx.org>

Maxim Dounin Wrote:
-------------------------------------------------------
> Yes, nginx won't cache responses with cookies
> unless you 
> specifically ask it to via proxy_ignore_headers
> directive[1].
> 
> Please also note that if you do so, you may want
> to also instruct 
> nginx to hide returned cookies from clients (via
> proxy_hide_header 
> directive[2]), or you'll end up with multiple
> clients with the 
> same cookie.
> 
> Example configuration should look like:
> 
>     proxy_ignore_headers Set-Cookie;
>     proxy_hide_header Set-Cookie;
> 
> Depending on the actual headers your backend
> returns you may also 
> need to ignore other headers as well for cache to
> work, notably 
> Expires and Cache-Control.
> 
> Alternatively, you may want to instruct your
> backend to not return 
> headers which prevent caching.
> 
> [1]
> http://wiki.nginx.org/HttpProxyModule#proxy_ignore
> _headers
> [2]
> http://wiki.nginx.org/HttpProxyModule#proxy_hide_h
> eader
> 

Hello.

Thank you for the explanation. I now understand the problem and it is
now working.

I want to add a condition where if a cookie with the name sess_id is
included in the request, then the cache would be bypassed. I tried the
following 

if ($http_cookie ~* "sess_id") {
  set $no_cache "1";
}

proxy_no_cache $no_cache;
proxy_cache_bypass $no_cache;

but it doesn't seem to work. What is the correct way to do this? Thank
you very much.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,216756,218066#msg-218066


From brian at akins.org  Wed Nov  9 11:56:07 2011
From: brian at akins.org (Brian Akins)
Date: Wed, 9 Nov 2011 06:56:07 -0500
Subject: gzip - unexplained side effects
In-Reply-To: <48ec591068896e53321f58945d5a8836.NginxMailingListEnglish@forum.nginx.org>
References: <48ec591068896e53321f58945d5a8836.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <82C7407F-B8AB-41F9-BE78-CB6F19198710@akins.org>


On Nov 8, 2011, at 9:14 PM, dbanks wrote:
> 
>    gzip_comp_level 9;
>   

Not really related to your question, but this uses much more cpu than the default for very little gain.  We always run this at "1"  Unless you have some special requirements (with benchmarks), I'd do the same.

--Brian

From igor at sysoev.ru  Wed Nov  9 12:41:33 2011
From: igor at sysoev.ru (Igor Sysoev)
Date: Wed, 9 Nov 2011 16:41:33 +0400
Subject: gzip - unexplained side effects
In-Reply-To: <48ec591068896e53321f58945d5a8836.NginxMailingListEnglish@forum.nginx.org>
References: <48ec591068896e53321f58945d5a8836.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20111109124133.GB42247@nginx.com>

On Tue, Nov 08, 2011 at 09:14:04PM -0500, dbanks wrote:
> Hi,
> 
> I'm running nginx 1.0.0 in front of a FCGI backend.  We've been running
> in production for about 4 months, and have really been impressed with
> the performance and stability of nginx.
> 
> We run a medium-volume appliction: 1000 to 4000 requests/sec spread over
> 2 instances by an upstream round robin load balancer.  We use
> keepalives, but keep them rather short given the request volume to keep
> the number of open connections manageable.  
> 
> Recently, I was working to improve our gzip settings.  The largest
> change was that I added an explicit gzip buffer line (find my config
> below - the three lines that are commented out seem to be correlated
> with this issue)
> 
> After the change, the volume of outbound traffic decreased measurably,
> as if gzip was not originally doing much due to inadequate buffer space.
>  Great news!, or so I thought.  What also changed was that the number of
> active connections fell by about 75% - the gzip change was somehow
> causing keepalives to be closed prematurely.  Also, our volume of
> incoming requests decreased a bit: as if some requests were being
> aborted (though accepts == handled).   The request volume makes
> debugging this particular issue somewhat troublesome, since I have yet
> to replicate it in a quiet instance.
> 
> The guts of my configuration appear below.  This is such an unexpected
> issue that I'm not doing a great job of setting up my question well. 
> What I think I'd like to know is how could a change to the gzip buffers
> (or the other two commented changes) impact keepalives or overall
> connection negotiation?  Also, any suggestions as to how to go about
> debugging it?
> 
>     sendfile        off;  
>     tcp_nodelay on;  
>     ignore_invalid_headers  on;  
>     if_modified_since off;
> 
>     gzip on;
>     gzip_comp_level 9;
>     gzip_types text/javascript text/plain application/x-javascript;
>     gzip_disable "MSIE [1-6]\.(?!.*SV1)"  
>     #gzip_buffers 512 4k;
>     #gzip_min_length  1100;  #if it fits in one packet, no worries
>     #gzip_http_version 1.1;
> 
>     keepalive_timeout  6;  
>     keepalive_requests 4;  

If you comment gzip_buffers, you see the previous site state ?
What is typical uncompressed and compressed response size ?
The default gzip_buffers are "32 4k", so they can keep up to 128K.

And as it was already suggested it's better to use default
gzip_comp_level 1.


-- 
Igor Sysoev


From nginx-forum at nginx.us  Wed Nov  9 13:18:55 2011
From: nginx-forum at nginx.us (lpugoy)
Date: Wed, 09 Nov 2011 08:18:55 -0500
Subject: Proxy cache for php site
In-Reply-To: <6a7d9d2ee046df8d9c9cfdee121e16d8.NginxMailingListEnglish@forum.nginx.org>
References: <20111108213052.GK95664@mdounin.ru>
	<6a7d9d2ee046df8d9c9cfdee121e16d8.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <865ea5535942c65c4bd29e049108398d.NginxMailingListEnglish@forum.nginx.org>

lpugoy Wrote:
> Hello.
> 
> Thank you for the explanation. I now understand
> the problem and it is now working.
> 
> I want to add a condition where if a cookie with
> the name sess_id is included in the request, then
> the cache would be bypassed. I tried the following
> 
> 
> if ($http_cookie ~* "sess_id") {
>   set $no_cache "1";
> }
> 
> proxy_no_cache $no_cache;
> proxy_cache_bypass $no_cache;
> 
> but it doesn't seem to work. What is the correct
> way to do this? Thank you very much.

An update. I don't think that the if statement is being entered. There
are these messages in the debug log


2011/11/09 21:14:03 [debug] 6397#0: *3 http script var
2011/11/09 21:14:03 [debug] 6397#0: *3 http script regex: "sess_id"
2011/11/09 21:14:03 [notice] 6397#0: *3 "sess_id" does not match "",
client: 10.214.99.16, server: *.site.com, request: "GET / HTTP/1.1",
host: "www.site.com"
2011/11/09 21:14:03 [debug] 6397#0: *3 http script if
2011/11/09 21:14:03 [debug] 6397#0: *3 http script if: false

Is there another way I can test for existence of cookies?

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,216756,218084#msg-218084


From zjay1987 at gmail.com  Wed Nov  9 14:30:57 2011
From: zjay1987 at gmail.com (li zJay)
Date: Wed, 9 Nov 2011 22:30:57 +0800
Subject: Nginx HttpAccessModule command invalid when rewrite command existed
Message-ID: <CABPr_TZwq5vHXBobkWBRYNi-OsQyivtD58gZw5KFXAsg+xr7wg@mail.gmail.com>

In the following simple case:

location /entry1 {
    allow 127.0.0.1;

}

-- 
??
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20111109/7679efcd/attachment.html>

From zjay1987 at gmail.com  Wed Nov  9 14:35:44 2011
From: zjay1987 at gmail.com (li zJay)
Date: Wed, 9 Nov 2011 22:35:44 +0800
Subject: Nginx HttpAccessModule command invalid when rewrite command
	existed
In-Reply-To: <CABPr_TZwq5vHXBobkWBRYNi-OsQyivtD58gZw5KFXAsg+xr7wg@mail.gmail.com>
References: <CABPr_TZwq5vHXBobkWBRYNi-OsQyivtD58gZw5KFXAsg+xr7wg@mail.gmail.com>
Message-ID: <CABPr_TbvvBOYXefvmrFfOukduH8P22hECPYSxL4GOB4FaBEo4g@mail.gmail.com>

In the following simple case:

location /entry1 {
     allow 127.0.0.1;
     deny all;
     rewrite ***;
}

the allow/deny command has no effect. Is that because rewrite command works
in the earlier phase?

Thanks!

On Wed, Nov 9, 2011 at 10:30 PM, li zJay <zjay1987 at gmail.com> wrote:

> In the following simple case:
>
> location /entry1 {
>     allow 127.0.0.1;
>
> }
>
> --
>
>
>


--
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20111109/e6c2b544/attachment.html>

From igor at sysoev.ru  Wed Nov  9 14:42:59 2011
From: igor at sysoev.ru (Igor Sysoev)
Date: Wed, 9 Nov 2011 18:42:59 +0400
Subject: Nginx HttpAccessModule command invalid when rewrite command
	existed
In-Reply-To: <CABPr_TbvvBOYXefvmrFfOukduH8P22hECPYSxL4GOB4FaBEo4g@mail.gmail.com>
References: <CABPr_TZwq5vHXBobkWBRYNi-OsQyivtD58gZw5KFXAsg+xr7wg@mail.gmail.com>
	<CABPr_TbvvBOYXefvmrFfOukduH8P22hECPYSxL4GOB4FaBEo4g@mail.gmail.com>
Message-ID: <20111109144259.GF42247@nginx.com>

On Wed, Nov 09, 2011 at 10:35:44PM +0800, li zJay wrote:
> In the following simple case:
> 
> location /entry1 {
>      allow 127.0.0.1;
>      deny all;
>      rewrite ***;
> }
> 
> the allow/deny command has no effect. Is that because rewrite command works
> in the earlier phase?

Yes, rewrites run before allow/deny.


-- 
Igor Sysoev


From ianevans at digitalhit.com  Wed Nov  9 14:50:05 2011
From: ianevans at digitalhit.com (Ian Evans)
Date: Wed, 09 Nov 2011 09:50:05 -0500
Subject: Mailing list software that plays nice with nginx - locations help
In-Reply-To: <f807aa40d457323e2fcbad138f73789b.squirrel@www.digitalhit.com>
References: <9ff9e5633d4d8a7efa6229b329e6f092.squirrel@www.digitalhit.com>
	<f807aa40d457323e2fcbad138f73789b.squirrel@www.digitalhit.com>
Message-ID: <4EBA931D.8070301@digitalhit.com>

On 08/11/2011 1:08 PM, Ian M. Evans wrote:
> Anyway, the settings there seem to be if the list software is installed at
> the document root.
> How would I change the locations if the software is installed at /lists?
> [I suck at locations and haven't had much sleep!]

Here are the locations on the wiki:

location ~* \.(txt|log|inc)$ {

location ~* \.(js|css|png|jpg|jpeg|gif|ico)$ {

location /config {

location ~* 
(index\.php|upload\.php|connector\.php|dl\.php|ut\.php|lt\.php|download\.php)$ 
{
fastcgi_split_path_info ^(.|\.php)(/.+)$;
fastcgi_param SCRIPT_FILENAME $document_root$fastcgi_script_name;
[etc]
}

location ~ \.php$ {

If phplists is being installed in /lists and not the root as in the wiki 
example (http://wiki.nginx.org/PHPList), how would I change these 
locations?

Would it be location ~* /lists\.(txt|log|inc)$ {
etc?

And how would I change fastcgi_split_path_info ^(.|\.php)(/.+)$; and 
fastcgi_param SCRIPT_FILENAME $document_root$fastcgi_script_name;

Thanks.


From zjay1987 at gmail.com  Wed Nov  9 14:58:13 2011
From: zjay1987 at gmail.com (li zJay)
Date: Wed, 9 Nov 2011 22:58:13 +0800
Subject: Nginx HttpAccessModule command invalid when rewrite command
	existed
In-Reply-To: <20111109144259.GF42247@nginx.com>
References: <CABPr_TZwq5vHXBobkWBRYNi-OsQyivtD58gZw5KFXAsg+xr7wg@mail.gmail.com>
	<CABPr_TbvvBOYXefvmrFfOukduH8P22hECPYSxL4GOB4FaBEo4g@mail.gmail.com>
	<20111109144259.GF42247@nginx.com>
Message-ID: <CABPr_TYcRfZwSGv6G3aWnxpQXHFV3o1ch_p9F4YH5jaWcOrOyA@mail.gmail.com>

Thanks Igor. That is OK, and I had a think and use the following commands
instead:

location /entry1 {
    if ( ! $remote_addr ~ "^(127\.0\.0|10\.10\.10)" ) { return 403; }
    rewrite ***
}

On Wed, Nov 9, 2011 at 10:42 PM, Igor Sysoev <igor at sysoev.ru> wrote:

> On Wed, Nov 09, 2011 at 10:35:44PM +0800, li zJay wrote:
> > In the following simple case:
> >
> > location /entry1 {
> >      allow 127.0.0.1;
> >      deny all;
> >      rewrite ***;
> > }
> >
> > the allow/deny command has no effect. Is that because rewrite command
> works
> > in the earlier phase?
>
> Yes, rewrites run before allow/deny.
>
>
> --
> 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/20111109/ca830659/attachment.html>

From nginx-forum at nginx.us  Wed Nov  9 19:09:29 2011
From: nginx-forum at nginx.us (dbanks)
Date: Wed, 09 Nov 2011 14:09:29 -0500
Subject: gzip - unexplained side effects
In-Reply-To: <20111109124133.GB42247@nginx.com>
References: <20111109124133.GB42247@nginx.com>
Message-ID: <aaea1bebbe64ddbac96b91a29f18a851.NginxMailingListEnglish@forum.nginx.org>

Hi Igor, thanks for your response!

>>If you comment gzip_buffers, you see the previous site state ?

If I comment gzip_buffers, gzip_min_length, and gzip_http_version, I see
the previous (desired) behavior.  

>>What is typical uncompressed and compressed response size ?

Responses are 6.8k or smaller uncompressed, 2.6k compressed
(gzip_comp_level=9), 2.8k compressed (gzip_comp_level=1).  About half of
the responses are this size, and the other half are less than 1k
uncompressed.

We currently have more than adequate CPU and want to minimize bandwidth
costs, so I had assumed that more compression was better.  Is there
another reason that I should stay with the default gzip_comp_level=1? 
(I'm happy to try it--just curious.)

Cheers,

Dean

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218044,218117#msg-218117


From nginx-forum at nginx.us  Thu Nov 10 07:13:17 2011
From: nginx-forum at nginx.us (wangbin579)
Date: Thu, 10 Nov 2011 02:13:17 -0500
Subject: Tcpcopy,an online request replication tool fit for nginx
In-Reply-To: <141f5701aa1bdc50e4d7a29c237508a3.NginxMailingListEnglish@forum.nginx.org>
References: <141f5701aa1bdc50e4d7a29c237508a3.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <18a32992e0352db3ca2a68e1370790c6.NginxMailingListEnglish@forum.nginx.org>

tcpcopy-0.2.1.tar.gz is now available.
Try it and I believe you will like this tool

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,217680,218138#msg-218138


From nginx-forum at nginx.us  Thu Nov 10 07:50:03 2011
From: nginx-forum at nginx.us (lpugoy)
Date: Thu, 10 Nov 2011 02:50:03 -0500
Subject: Proxy cache for php site
In-Reply-To: <865ea5535942c65c4bd29e049108398d.NginxMailingListEnglish@forum.nginx.org>
References: <20111108213052.GK95664@mdounin.ru>
	<6a7d9d2ee046df8d9c9cfdee121e16d8.NginxMailingListEnglish@forum.nginx.org>
	<865ea5535942c65c4bd29e049108398d.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <ff79a646d262e533f922abfb5f10942f.NginxMailingListEnglish@forum.nginx.org>

> An update. I don't think that the if statement is
> being entered. There are these messages in the
> debug log
> 
> 
> 2011/11/09 21:14:03 [debug] 6397#0: *3 http script
> var
> 2011/11/09 21:14:03 [debug] 6397#0: *3 http script
> regex: "sess_id"
> 2011/11/09 21:14:03 [notice] 6397#0: *3 "sess_id"
> does not match "", client: 10.214.99.16, server:
> *.site.com, request: "GET / HTTP/1.1", host:
> "www.site.com"
> 2011/11/09 21:14:03 [debug] 6397#0: *3 http script
> if
> 2011/11/09 21:14:03 [debug] 6397#0: *3 http script
> if: false
> 
> Is there another way I can test for existence of
> cookies?

The problem is apparently with my method of passing cookies. curl -b
<filename> does not work. curl -b sess_id=... does. It's now working
again.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,216756,218143#msg-218143


From nginx-forum at nginx.us  Thu Nov 10 11:12:54 2011
From: nginx-forum at nginx.us (etrader)
Date: Thu, 10 Nov 2011 06:12:54 -0500
Subject: Reducing Connect time for static files
Message-ID: <dc0e9e00693297b7879152c7a60b008c.NginxMailingListEnglish@forum.nginx.org>

Exploring the webpage load by pingdom service, I just found that the
main delay of page load is due to slow load of images. The images come
from a cookieless domain. I changed the nginx conf for that domain as

	location ~*
^.+.(jpg|jpeg|gif|png|ico|css|zip|tgz|gz|rar|bz2|doc|xls|exe|pdf|ppt|txt|tar|mid|midi|wav|bmp|rtf|js|swf|wma|wmv)$
{

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;
		access_log		off;
	}

It did speed up the load of images a lot; but still, the load of images
is not fast enough

Connect: 100 ms
Send: 1 ms
Wait 200 ms
Receive: 100 ms

Receive is controlled by the data transfer, but I believe it is possible
to reduce the length of connect and wait time by nginx configuration.
Any idea?

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218159,218159#msg-218159


From r at roze.lv  Thu Nov 10 14:27:14 2011
From: r at roze.lv (Reinis Rozitis)
Date: Thu, 10 Nov 2011 16:27:14 +0200
Subject: proxy_store HEAD
Message-ID: <F1792CBBB37A48349751E3BF70A2DE58@DD21>

Hello,
is there a (simple) way to make 'proxy_store on' store the files also on HEAD requests instead of just returning the response from 
upstream?

I understand the reasons behind the current behaviour but I'm trying to build some sort of synchronisation between few nginx boxes 
without the need to actually download the files to the client - so far the only option seems to use 'proxy_ignore_client_abort' make 
a GET request and forcefully abort it (since even ranged 1 byte requests don't store the file on the "proxying" host).



p.s. it seems that wiki is a bit outdated and 'proxy_cache_methods' is deprecated/notexisting at least in 1.1.x tree

rr 


From mdounin at mdounin.ru  Thu Nov 10 15:55:42 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Thu, 10 Nov 2011 19:55:42 +0400
Subject: proxy_store HEAD
In-Reply-To: <F1792CBBB37A48349751E3BF70A2DE58@DD21>
References: <F1792CBBB37A48349751E3BF70A2DE58@DD21>
Message-ID: <20111110155542.GU95664@mdounin.ru>

Hello!

On Thu, Nov 10, 2011 at 04:27:14PM +0200, Reinis Rozitis wrote:

> Hello,
> is there a (simple) way to make 'proxy_store on' store the files
> also on HEAD requests instead of just returning the response from
> upstream?
> 
> I understand the reasons behind the current behaviour but I'm trying
> to build some sort of synchronisation between few nginx boxes
> without the need to actually download the files to the client - so
> far the only option seems to use 'proxy_ignore_client_abort' make a
> GET request and forcefully abort it (since even ranged 1 byte
> requests don't store the file on the "proxying" host).

You may try

   proxy_method GET;

in a server/location dedicated for synchronization.  It will cause 
nginx to use GET requests to upstream (though it won't sent 
anything to client as long as original request was HEAD one).  
This is basically identical to what proxy_cache does with HEAD 
requests.

BTW, proxy_ignore_client_abort is meaningless with proxy_store, 
as proxy_store activates the same logic unconditionally.

> p.s. it seems that wiki is a bit outdated and 
> 'proxy_cache_methods' is deprecated/notexisting at least in 
> 1.1.x tree

The "proxy_cache_methods" directive exists and not deprecated.  
It's for proxy_cache though, not proxy_store.

Maxim Dounin


From r at roze.lv  Thu Nov 10 16:20:02 2011
From: r at roze.lv (Reinis Rozitis)
Date: Thu, 10 Nov 2011 18:20:02 +0200
Subject: proxy_store HEAD
In-Reply-To: <20111110155542.GU95664@mdounin.ru>
References: <F1792CBBB37A48349751E3BF70A2DE58@DD21>
	<20111110155542.GU95664@mdounin.ru>
Message-ID: <201D4C9150264FEAA3CEE70940A13360@DD21>

>You may try
>  proxy_method GET;

Thx Maxim, works like a charm


> The "proxy_cache_methods" directive exists and not deprecated.  It's for proxy_cache though, not proxy_store.

My fault, I had my own typical nginx istallation with --without-http-cache also quick look at Igors docs 
http://nginx.org/ru/docs/http/ngx_http_proxy_module.html didn't show directive so came to wrong conclusion after experimenting with 
the config and nginx throwing 'unknown directive' at startup.


rr 


From nginx-forum at nginx.us  Thu Nov 10 16:47:52 2011
From: nginx-forum at nginx.us (ceh329)
Date: Thu, 10 Nov 2011 11:47:52 -0500
Subject: Alter Config On Startup
Message-ID: <a26a3867c7b747eddea0ab04530b8e64.NginxMailingListEnglish@forum.nginx.org>

Hello,
I have a SSL section in my nginx.conf file and I want to start up nginx
without SSL enabled. Is this possible without removing the SSL info in
the conf file?

Thanks,
Charlie

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218180,218180#msg-218180


From lvella at gmail.com  Thu Nov 10 17:17:32 2011
From: lvella at gmail.com (Lucas Clemente Vella)
Date: Thu, 10 Nov 2011 15:17:32 -0200
Subject: Trac authentication failing
Message-ID: <CAGCathwPM=PtfYXzpeppJiYu77Nm9GkYS+abrdAZe3Aj9J7qmA@mail.gmail.com>

I am running Trac inside uWSGI, with nginx as front-end server. I have
also set up basic http authentication in nginx. Now, theoretically,
when authenticated in nginx, Trac should recognize the user and
display its name, identify the new tickets, etc. But that is not
happening.

How can make sure the authentication information is being sent to
uWSGI and received by Trac?

-- 
Lucas Clemente Vella
lvella at gmail.com


From roberto at unbit.it  Thu Nov 10 18:45:54 2011
From: roberto at unbit.it (Roberto De Ioris)
Date: Thu, 10 Nov 2011 19:45:54 +0100 (CET)
Subject: Trac authentication failing
In-Reply-To: <CAGCathwPM=PtfYXzpeppJiYu77Nm9GkYS+abrdAZe3Aj9J7qmA@mail.gmail.com>
References: <CAGCathwPM=PtfYXzpeppJiYu77Nm9GkYS+abrdAZe3Aj9J7qmA@mail.gmail.com>
Message-ID: <7a243fb4ebd93ec437f66af9b028b74e.squirrel@manage.unbit.it>


> I am running Trac inside uWSGI, with nginx as front-end server. I have
> also set up basic http authentication in nginx. Now, theoretically,
> when authenticated in nginx, Trac should recognize the user and
> display its name, identify the new tickets, etc. But that is not
> happening.
>
> How can make sure the authentication information is being sent to
> uWSGI and received by Trac?

Be sure to have

uwsgi_param REMOTE_USER $remote_user

in your config or includes file




-- 
Roberto De Ioris
http://unbit.it


From lvella at gmail.com  Thu Nov 10 23:30:13 2011
From: lvella at gmail.com (Lucas Clemente Vella)
Date: Thu, 10 Nov 2011 21:30:13 -0200
Subject: Trac authentication failing
In-Reply-To: <7a243fb4ebd93ec437f66af9b028b74e.squirrel@manage.unbit.it>
References: <CAGCathwPM=PtfYXzpeppJiYu77Nm9GkYS+abrdAZe3Aj9J7qmA@mail.gmail.com>
	<7a243fb4ebd93ec437f66af9b028b74e.squirrel@manage.unbit.it>
Message-ID: <CAGCathzcPANvA=EvL2u4vYM-3TTpeMRmnKEmcQCbrV52Thh1DA@mail.gmail.com>

2011/11/10 Roberto De Ioris <roberto at unbit.it>:
> Be sure to have
>
> uwsgi_param REMOTE_USER $remote_user
>
> in your config or includes file

Like a charm.

Thanks!

-- 
Lucas Clemente Vella
lvella at gmail.com


From appa at perusio.net  Fri Nov 11 00:31:01 2011
From: appa at perusio.net (=?UTF-8?B?QW50w7NuaW8=?= P. P. Almeida)
Date: Fri, 11 Nov 2011 00:31:01 +0000
Subject: Alter Config On Startup
In-Reply-To: <a26a3867c7b747eddea0ab04530b8e64.NginxMailingListEnglish@forum.nginx.org>
References: <a26a3867c7b747eddea0ab04530b8e64.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <878vnnmrei.wl%appa@perusio.net>

On 10 Nov 2011 16h47 WET, nginx-forum at nginx.us wrote:

> Hello, I have a SSL section in my nginx.conf file and I want to
> start up nginx without SSL enabled. Is this possible without
> removing the SSL info in the conf file?

Yes. You'll have to:

1. Create a script that enables the SSL part. The most viable way
   seems to be through the include directive.

2. Reload nginx.

3. Done.

Using sed supposing that you start Nginx without SSL enabled. The SSL
config is in a file sslhost.conf. In the file you have the regular
HTTP config:

server {
  # HTTP (no SSL)
}

##SSL include sslhost.conf;


sed -i 's/##SSL //' <vhost config file>

service nginx reload

Now the SSL config is active.

--- appa


From nginx-forum at nginx.us  Fri Nov 11 08:56:09 2011
From: nginx-forum at nginx.us (zhenwei)
Date: Fri, 11 Nov 2011 03:56:09 -0500
Subject: error: deadlock happened in ngx_http_file_cache_expire()
Message-ID: <c82ee4609e567a9ee77c20c73c735956.NginxMailingListEnglish@forum.nginx.org>

{{{
proxy_cache_path  /nginx/cache/three levels=1:1:2 
keys_zone=three:1000m;
proxy_cache_valid  200 304 302  60m;
}}}

Nginx version: 1.0.1
nginx: configure arguments: --with-pcre --group=admin --user=admin
--prefix=/var/www/nginx --with-http_ssl_module --with-http_perl_module
--with-http_stub_status_module
--with-perl_modules_path=./src/http/modules/perl/
--add-module=../ngx_cache_purge

The cache entries timeout is set as 60 minutes, on production
environment(for website hosting) service became unavailable after
running several hours. After dive into backtraces and source code, it's
a deadlock issue on reload nginx(reload occurs every several minutes),
here is the bt information.

{{{
root     28779  0.0  0.1 1124788 6464 ?        Ss   11:15   0:00 nginx:
master process /var/www/nginx//sbin/nginx
nobody   28781 48.8  0.2 1126424 8268 ?        R    11:15  77:41 nginx:
worker process is shutting down
nobody   28784 47.2  0.0 1122576 3460 ?        R    11:15  75:05 nginx:
cache manager process
nobody   28785 47.6  0.0 1122724 3496 ?        R    11:15  75:45 nginx:
cache loader process
nobody   30156 50.8  0.2 1128800 9648 ?        R    11:43  66:27 nginx:
worker process     
nobody   30157 54.1  0.2 1128800 9648 ?        R    11:43  70:46 nginx:
worker process     
nobody   30158 54.9  0.2 1128800 9572 ?        R    11:43  71:52 nginx:
worker process     
nobody   30159 55.3  0.2 1128800 9648 ?        R    11:43  72:22 nginx:
worker process     
nobody   30160  8.5 26.3 1124788 1069964 ?     R    11:43  11:12 nginx:
cache manager process
}}}

>From the ps list Nginx is under reloading, when all worker process and a
cache manager process is spinlock to get the cache, and another cache
manager acquired the lock and fall into a while loop.

{{{
#0  ngx_http_file_cache_expire (cache=0x1ea7dc00) at
src/http/ngx_http_file_cache.c:1103
1103            if (fcn->deleting) {
(gdb) bt
#0  ngx_http_file_cache_expire (cache=0x1ea7dc00) at
src/http/ngx_http_file_cache.c:1103
#1  0x0000000000474e8b in ngx_http_file_cache_manager (data=0x1ea7dc00)
at src/http/ngx_http_file_cache.c:1193
#2  0x0000000000439e9d in ngx_cache_manager_process_handler
(ev=0x7fff5f5f3120) at src/os/unix/ngx_process_cycle.c:1346
#3  0x00000000004303f0 in ngx_event_expire_timers () at
src/event/ngx_event_timer.c:149
#4  0x000000000042e537 in ngx_process_events_and_timers
(cycle=0x1e9b5a70) at src/event/ngx_event.c:261
#5  0x0000000000439d6e in ngx_cache_manager_process_cycle
(cycle=0x1e9b5a70, data=0x6c4440) at
src/os/unix/ngx_process_cycle.c:1328
#6  0x00000000004361bd in ngx_spawn_process (cycle=0x1e9b5a70,
proc=0x439bdc <ngx_cache_manager_process_cycle>, data=0x6c4440,
name=0x4abc88 "cache manager process", 
    respawn=-4) at src/os/unix/ngx_process.c:196
#7  0x00000000004382ca in ngx_start_cache_manager_processes
(cycle=0x1e9b5a70, respawn=1) at src/os/unix/ngx_process_cycle.c:398
#8  0x0000000000437dc9 in ngx_master_process_cycle (cycle=0x1e9b5a70) at
src/os/unix/ngx_process_cycle.c:251
#9  0x000000000040e36c in main (argc=1, argv=0x7fff5f5f36d8) at
src/core/nginx.c:405
(gdb) p fcn->deleting
$1 = 1
(gdb) p wait
$2 = -67
}}}

Source code
{{{
1075     for ( ;; ) {
1076 
1077         if (ngx_queue_empty(&cache->sh->queue)) {
1078             wait = 10;
1079             break;
1080         }
1081 
1082         q = ngx_queue_last(&cache->sh->queue);
1083 
1084         fcn = ngx_queue_data(q, ngx_http_file_cache_node_t,
queue);
1085 
1086         wait = fcn->expire - now;
1087 
1088         if (wait > 0) {
1089             wait = wait > 10 ? 10 : wait;
1090             break;
1091         }
.....
}}}

The for loop never exit if "wait" is a negative number and unfortunately
it becomes -67, I hope decrease the timeout to 10m would work around it,
and it failed, either.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218202,218202#msg-218202


From mdounin at mdounin.ru  Fri Nov 11 10:27:47 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Fri, 11 Nov 2011 14:27:47 +0400
Subject: error: deadlock happened in ngx_http_file_cache_expire()
In-Reply-To: <c82ee4609e567a9ee77c20c73c735956.NginxMailingListEnglish@forum.nginx.org>
References: <c82ee4609e567a9ee77c20c73c735956.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20111111102747.GA95664@mdounin.ru>

Hello!

On Fri, Nov 11, 2011 at 03:56:09AM -0500, zhenwei wrote:

> Nginx version: 1.0.1

[...]

> #0  ngx_http_file_cache_expire (cache=0x1ea7dc00) at
> src/http/ngx_http_file_cache.c:1103
> 1103            if (fcn->deleting) {

[...]

Please upgreade, this was fixed in 1.0.5:

    *) Bugfix: worker processes may got caught in an endless loop during
       reconfiguration, if a caching was used; the bug had appeared in
       0.8.48.

Maxim Dounin


From nginx-forum at nginx.us  Fri Nov 11 11:28:00 2011
From: nginx-forum at nginx.us (zhenwei)
Date: Fri, 11 Nov 2011 06:28:00 -0500
Subject: error: deadlock happened in ngx_http_file_cache_expire()
In-Reply-To: <20111111102747.GA95664@mdounin.ru>
References: <20111111102747.GA95664@mdounin.ru>
Message-ID: <4e56ac8fb1a6d11ac985bb9dcb2c3bc2.NginxMailingListEnglish@forum.nginx.org>

Wonderful, thanks a lot!

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218202,218205#msg-218205


From al-nginx at none.at  Fri Nov 11 12:16:25 2011
From: al-nginx at none.at (Aleksandar Lazic)
Date: Fri, 11 Nov 2011 13:16:25 +0100
Subject: Q: about try_files and regex location
Message-ID: <48dc5cf04b3d067ab54f333ab78fdef4@none.at>

Dear all,

please can you help me to fix the issue with try_files an regex 
location, thank you.

I use

###
sbin/nginx -V
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: --with-debug --with-libatomic 
--without-http_ssi_module --without-http_uwsgi_module 
--without-http_scgi_module --without-http_memcached_module 
--with-http_ssl_module --user=nginx --group=nginx --prefix=server/nginx 
--with-http_stub_status_module
###

with the following config.

###
http {
   server {
...
         location ~ ^/(share|alfresco)(/res)?(.*$)  {
           alias /home/alfresco/alfresco-4.0.b/tomcat/webapps;
           try_files $uri /$1$2 /$1$3 @alfresco;
         }

         location @alfresco {
           proxy_pass   http://alfresco;
         }
   }
}
###

I get the following error:

###
2011/11/11 12:58:09 [error] 5618#0: *33782 open()
"/home/alfresco/alfresco-4.0.b/tomcat/webapps/share/components/document-details/document-link/home/al" 
failed (2: No such file or directory), client: xxx, server: xxx,
request: "GET /share/components/document-details/document-links-min.js 
HTTP/1.1", host: "xxx", referrer: 
"https://xxx/share/page/site/aleks-glossar/document-details?nodeRef=workspace://SpacesStore/985297e8-ff1b-4423-b6e9-8dacc0011196"
###

Could it be that try_files match on '/$1$2' but the open call get 
'/$1$3' or something else?

attached the debug log.

BR
Aleks

#### debug
2011/11/11 12:58:09 [debug] 5618#0: *33782 posix_memalign: 
000000000CD7D350:4096 @16
2011/11/11 12:58:09 [debug] 5618#0: *33782 http process request line
2011/11/11 12:58:09 [debug] 5618#0: *33782 http request line: "GET 
/share/components/document-details/document-links-min.js HTTP/1.1"
2011/11/11 12:58:09 [debug] 5618#0: *33782 http uri: 
"/share/components/document-details/document-links-min.js"
2011/11/11 12:58:09 [debug] 5618#0: *33782 http args: ""
2011/11/11 12:58:09 [debug] 5618#0: *33782 http exten: "js"
2011/11/11 12:58:09 [debug] 5618#0: *33782 http process request header 
line
2011/11/11 12:58:09 [debug] 5618#0: *33782 http header: "Host: XXXX"
2011/11/11 12:58:09 [debug] 5618#0: *33782 http header: "User-Agent: 
Mozilla/5.0 (Windows NT 5.1; rv:6.0.2) Gecko/20100101 Firefox/6.0.2"
2011/11/11 12:58:09 [debug] 5618#0: *33782 http header: "Accept: */*"
2011/11/11 12:58:09 [debug] 5618#0: *33782 http header: 
"Accept-Language: de-de,de;q=0.8,en;q=0.5,en-us;q=0.3"
2011/11/11 12:58:09 [debug] 5618#0: *33782 http header: 
"Accept-Encoding: gzip, deflate"
2011/11/11 12:58:09 [debug] 5618#0: *33782 http header: 
"Accept-Charset: UTF-8,*"
2011/11/11 12:58:09 [debug] 5618#0: *33782 http header: "DNT: 1"
2011/11/11 12:58:09 [debug] 5618#0: *33782 http header: "Referer: 
https://xxx/share/page/site/aleks-glossar/document-details?nodeRef=workspace://SpacesStore/985297e8-ff1b-4423-b6e9-8dacc0011196"
2011/11/11 12:58:09 [debug] 5618#0: *33782 http header: "Cookie: 
JSESSIONID=C3A5E3E0985A055C732E884981544337; alfLogin=1321012572; 
alfUsername2="YWxyaXQ=""
2011/11/11 12:58:09 [debug] 5618#0: *33782 http header: "Connection: 
keep-alive"
2011/11/11 12:58:09 [debug] 5618#0: *33782 http header done
2011/11/11 12:58:09 [debug] 5618#0: *33782 event timer del: 7: 
1321012754907
2011/11/11 12:58:09 [debug] 5618#0: *33782 rewrite phase: 0
2011/11/11 12:58:09 [debug] 5618#0: *33782 test location: "/redmine"
2011/11/11 12:58:09 [debug] 5618#0: *33782 test location: ~ 
"^/(share|alfresco)(/res)?(.*$)"
2011/11/11 12:58:09 [debug] 5618#0: *33782 using configuration 
"^/(share|alfresco)(/res)?(.*$)"
2011/11/11 12:58:09 [debug] 5618#0: *33782 http cl:-1 max:52428800
2011/11/11 12:58:09 [debug] 5618#0: *33782 rewrite phase: 2
2011/11/11 12:58:09 [debug] 5618#0: *33782 post rewrite phase: 3
2011/11/11 12:58:09 [debug] 5618#0: *33782 generic phase: 4
2011/11/11 12:58:09 [debug] 5618#0: *33782 generic phase: 5
2011/11/11 12:58:09 [debug] 5618#0: *33782 access phase: 6
2011/11/11 12:58:09 [debug] 5618#0: *33782 access phase: 7
2011/11/11 12:58:09 [debug] 5618#0: *33782 post access phase: 8
2011/11/11 12:58:09 [debug] 5618#0: *33782 try files phase: 9
2011/11/11 12:58:09 [debug] 5618#0: *33782 http script copy: 
"/home/alfresco/alfresco-4.0.b/tomcat/webapps"
2011/11/11 12:58:09 [debug] 5618#0: *33782 http script var: 
"/share/components/document-details/document-links-min.js"




2011/11/11 12:58:09 [debug] 5618#0: *33782 trying to use file: 
"/share/components/document-details/document-links-min.js" 
"/home/alfresco/alfresco-4.0.b/tomcat/webapps/share/components/document-details/document-links-min.js"



2011/11/11 12:58:09 [debug] 5618#0: *33782 try file uri: 
"/share/components/document-details/document-links-min.js"
2011/11/11 12:58:09 [debug] 5618#0: *33782 content phase: 10
2011/11/11 12:58:09 [debug] 5618#0: *33782 content phase: 11
2011/11/11 12:58:09 [debug] 5618#0: *33782 content phase: 12
2011/11/11 12:58:09 [debug] 5618#0: *33782 http script copy: 
"/home/alfresco/alfresco-4.0.b/tomcat/webapps"


2011/11/11 12:58:09 [debug] 5618#0: *33782 http filename: 
"/home/alfresco/alfresco-4.0.b/tomcat/webapps/share/components/document-details/document-link/home/al"


2011/11/11 12:58:09 [debug] 5618#0: *33782 add cleanup: 
000000000CD7DF38
2011/11/11 12:58:09 [error] 5618#0: *33782 open() 
"/home/alfresco/alfresco-4.0.b/tomcat/webapps/share/components/document-details/document-link/home/al" 
failed (2: No such file or directory), client: xxx, server: xxx, 
request: "GET /share/components/document-details/document-links-min.js 
HTTP/1.1", host: "xxx", referrer: 
"https://external.none.at/share/page/site/aleks-glossar/document-details?nodeRef=workspace://SpacesStore/985297e8-ff1b-4423-b6e9-8dacc0011196"
2011/11/11 12:58:09 [debug] 5618#0: *33782 http finalize request: 404, 
"/share/components/document-details/document-link/home/al?" a:1, c:1
2011/11/11 12:58:09 [debug] 5618#0: *33782 http special response: 404, 
"/share/components/document-details/document-link/home/al?"
2011/11/11 12:58:09 [debug] 5618#0: *33782 http set discard body
2011/11/11 12:58:09 [debug] 5618#0: *33782 HTTP/1.1 404 Not Found
Server: nginx/1.1.4
Date: Fri, 11 Nov 2011 11:58:09 GMT
Content-Type: text/html
Content-Length: 168
Connection: keep-alive

2011/11/11 12:58:09 [debug] 5618#0: *33782 write new buf t:1 f:0 
000000000CD7DFB8, pos 000000000CD7DFB8, size: 154 file: 0, size: 0
2011/11/11 12:58:09 [debug] 5618#0: *33782 http write filter: l:0 f:0 
s:154
2011/11/11 12:58:09 [debug] 5618#0: *33782 http output filter 
"/share/components/document-details/document-link/home/al?"
2011/11/11 12:58:09 [debug] 5618#0: *33782 http copy filter: 
"/share/components/document-details/document-link/home/al?"
2011/11/11 12:58:09 [debug] 5618#0: *33782 write old buf t:1 f:0 
000000000CD7DFB8, pos 000000000CD7DFB8, size: 154 file: 0, size: 0
2011/11/11 12:58:09 [debug] 5618#0: *33782 write new buf t:0 f:0 
0000000000000000, pos 00000000006802C0, size: 116 file: 0, size: 0
####


From ne at vbart.ru  Fri Nov 11 13:15:02 2011
From: ne at vbart.ru (Valentin V. Bartenev)
Date: Fri, 11 Nov 2011 17:15:02 +0400
Subject: Q: about try_files and regex location
In-Reply-To: <48dc5cf04b3d067ab54f333ab78fdef4@none.at>
References: <48dc5cf04b3d067ab54f333ab78fdef4@none.at>
Message-ID: <201111111715.02934.ne@vbart.ru>

On Friday 11 November 2011 16:16:25 Aleksandar Lazic wrote:
[...]
> ###
> http {
>    server {
> ...
>          location ~ ^/(share|alfresco)(/res)?(.*$)  {
 - location ~ ^/(share|alfresco)(/res)?(.*$)  {
 + location ~ ^/(share|alfresco)(/res)?(.*)$  {

>            alias /home/alfresco/alfresco-4.0.b/tomcat/webapps;
[...]

http://nginx.org/en/docs/http/ngx_http_core_module.html#alias

The "alias" directive in your config replaces request part of path to file after 
successful "try_files" check.

wbr, Valentin V. Bartenev

 


From mdounin at mdounin.ru  Fri Nov 11 13:18:27 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Fri, 11 Nov 2011 17:18:27 +0400
Subject: Q: about try_files and regex location
In-Reply-To: <48dc5cf04b3d067ab54f333ab78fdef4@none.at>
References: <48dc5cf04b3d067ab54f333ab78fdef4@none.at>
Message-ID: <20111111131827.GG95664@mdounin.ru>

Hello!

On Fri, Nov 11, 2011 at 01:16:25PM +0100, Aleksandar Lazic wrote:

> Dear all,
> 
> please can you help me to fix the issue with try_files an regex
> location, thank you.
> 
> I use
> 
> ###
> sbin/nginx -V
> 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: --with-debug --with-libatomic
> --without-http_ssi_module --without-http_uwsgi_module
> --without-http_scgi_module --without-http_memcached_module
> --with-http_ssl_module --user=nginx --group=nginx
> --prefix=server/nginx --with-http_stub_status_module
> ###
> 
> with the following config.
> 
> ###
> http {
>   server {
> ...
>         location ~ ^/(share|alfresco)(/res)?(.*$)  {
>           alias /home/alfresco/alfresco-4.0.b/tomcat/webapps;
>           try_files $uri /$1$2 /$1$3 @alfresco;
>         }
> 
>         location @alfresco {
>           proxy_pass   http://alfresco;
>         }
>   }
> }
> ###
> 
> I get the following error:
> 
> ###
> 2011/11/11 12:58:09 [error] 5618#0: *33782 open()
> "/home/alfresco/alfresco-4.0.b/tomcat/webapps/share/components/document-details/document-link/home/al"
> failed (2: No such file or directory), client: xxx, server: xxx,
> request: "GET
> /share/components/document-details/document-links-min.js HTTP/1.1",
> host: "xxx", referrer: "https://xxx/share/page/site/aleks-glossar/document-details?nodeRef=workspace://SpacesStore/985297e8-ff1b-4423-b6e9-8dacc0011196"
> ###
> 
> Could it be that try_files match on '/$1$2' but the open call get
> '/$1$3' or something else?
> 
> attached the debug log.

This is the bug in alias and try_files interaction.  Or, more 
strictly, such configuration should be rejected during testing 
configuration as it's not really make sense: alias within regex 
location specifies full path to a file, and try_files is 
meaningless here.

You probably mean to use "root" instead, i.e.

    location ~ ^/(share|alfresco)(/res)?(.*$) {
        root /home/alfresco/alfresco-4.0.b/tomcat/webapps;
        try_files $uri /$1$2 /$1$3 @alfresco;
    }

Maxim Dounin


From tseveendorj at gmail.com  Fri Nov 11 13:17:04 2011
From: tseveendorj at gmail.com (Tseveendorj Ochirlantuu)
Date: Fri, 11 Nov 2011 21:17:04 +0800
Subject: Rewrite conversion
In-Reply-To: <4E2CC27A.2090705@gmail.com>
References: <CAAbfzq04KTvy2T_m-JFq8q31m8ttwo=dx9C3HtD0OPagqHHBbA@mail.gmail.com>
	<20110723071826.GC73233@sysoev.ru>
	<CAAbfzq1euQqr=Se-z+XV6cTDUCkL+2nhrCnMBy2ZVtGCphT9Yw@mail.gmail.com>
	<CAAbfzq0JH2WNUjDmnRHnGViqPcYz=Y45JQBXsS7piTwckSDWMQ@mail.gmail.com>
	<CAAbfzq1qLj+2p8tSJX+gv0h_vZRsCptAW8Zpcg6b4TZN-v-XOw@mail.gmail.com>
	<CAAtReCnE+1MM=ze8BHhNnD-0wZcBEDU5L8Hn+WK_vMVQzNh2rg@mail.gmail.com>
	<4E2CC27A.2090705@gmail.com>
Message-ID: <CAAbfzq2Y-a2L0FthinYxCaSzRNR1i2YW8ijqe5rDRfsPGiqgXA@mail.gmail.com>

Dear Igor,

You have converted apache rewrite to nginx. I appreciated very much for you.
But I need redirecting request www to non-www in my rewrite. Please
see your converted rewrite below.

Any help will be appreciated.

server {
   ...

   root  /path/to/files;

   error_page 403  /index.php?do=/public/error/403/;
   error_page 404  /index.php?do=/public/error/404/;

   location /file/ {
   }

   location /install/ {
   }

   location /design/ {
   }

   location /plugins/ {
   }

   location = /robots.txt {
   }

   location = /favicon.ico {
   }

   location / {
       fastcgi_pass   ...
       include        fastcgi_params;
       fastcgi_param  SCRIPT_FILENAME  /path/to/files/index.php;
       fastcgi_param  QUERY_STRING     do=$uri;
   }

   location /index.php {

       location = /index.php {
           fastcgi_pass   ...
           include        fastcgi_params;
           fastcgi_param  SCRIPT_FILENAME  /path/to/files/index.php;
           fastcgi_param  QUERY_STRING     $args;
       }

       location ~ ^/index.php(/.*)$ {
           fastcgi_pass   ...
           include        fastcgi_params;
           fastcgi_param  SCRIPT_FILENAME  /path/to/files/index.php;
           fastcgi_param  QUERY_STRING     do=$1;
       }

       return 404;
   }


On 7/25/11, Tseveendorj <tseveendorj at gmail.com> wrote:
> On 11.07.23 23:05, Edho Arief wrote:
>> On Sat, Jul 23, 2011 at 9:33 PM, Tseveendorj Ochirlantuu
>> <tseveendorj at gmail.com>  wrote:
>>> Dear Igor,
>>> I just tested rewrite but one thing did not work. When I'm
>>> accessing
>>> http://www.xac.mn/index.php?do=/mytunes/view/song_40/module_popout/ this
>>> popup but I got
>>>
>>> 404 Not Found
>>>
>>> ________________________________
>>> nginx/0.7.65
>>> on the screen. Above url is working on Apache with rewrite.
>>> I do not know difference between these two rewrites.
>>> Apache
>>> RewriteRule ^(.*)$ /index.php?do=/$1 [L]
>>> Nginx location ~ ^/index.php(/.*)$ {           fastcgi_pass   backend;
>>>       include        fastcgi_params;            fastcgi_param
>>>   SCRIPT_FILENAME  /var/www/xac/index.php;
>>>             fastcgi_param  QUERY_STRING     do=$1;       }
>>>
>> the url
>> http://www.xac.mn/index.php?do=/mytunes/view/song_40/module_popout/
>> is handled by this location block:
>>
>>         location = /index.php {
>>             fastcgi_pass   ...
>>             include        fastcgi_params;
>>             fastcgi_param  SCRIPT_FILENAME  /path/to/files/index.php;
>>             fastcgi_param  QUERY_STRING     $args;
>>         }
>>
>> _______________________________________________
>> nginx mailing list
>> nginx at nginx.org
>> http://mailman.nginx.org/mailman/listinfo/nginx
>
> Really, I thought above rewrite handled by another. What do you think
> this popup does not work ?
>
>
>
>


From al-nginx at none.at  Fri Nov 11 23:13:42 2011
From: al-nginx at none.at (Aleksandar Lazic)
Date: Sat, 12 Nov 2011 00:13:42 +0100
Subject: Q: about try_files and regex location
In-Reply-To: <20111111131827.GG95664@mdounin.ru>
References: <48dc5cf04b3d067ab54f333ab78fdef4@none.at>
	<20111111131827.GG95664@mdounin.ru>
Message-ID: <f584fd197ce2042c8fc95deea0ede8bb@none.at>

Thanks this was the solution.
Also thanks to Valentin.

BR
Aleks
On 11.11.2011 14:18, Maxim Dounin wrote:
> Hello!
>
> On Fri, Nov 11, 2011 at 01:16:25PM +0100, Aleksandar Lazic wrote:
>
>> Dear all,
>>
>> please can you help me to fix the issue with try_files an regex
>> location, thank you.
>>
>> I use
>>
>> ###
>> sbin/nginx -V
>> 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: --with-debug --with-libatomic
>> --without-http_ssi_module --without-http_uwsgi_module
>> --without-http_scgi_module --without-http_memcached_module
>> --with-http_ssl_module --user=nginx --group=nginx
>> --prefix=server/nginx --with-http_stub_status_module
>> ###
>>
>> with the following config.
>>
>> ###
>> http {
>>   server {
>> ...
>>         location ~ ^/(share|alfresco)(/res)?(.*$)  {
>>           alias /home/alfresco/alfresco-4.0.b/tomcat/webapps;
>>           try_files $uri /$1$2 /$1$3 @alfresco;
>>         }
>>
>>         location @alfresco {
>>           proxy_pass   http://alfresco;
>>         }
>>   }
>> }
>> ###
>>
>> I get the following error:
>>
>> ###
>> 2011/11/11 12:58:09 [error] 5618#0: *33782 open()
>> 
>> "/home/alfresco/alfresco-4.0.b/tomcat/webapps/share/components/document-details/document-link/home/al"
>> failed (2: No such file or directory), client: xxx, server: xxx,
>> request: "GET
>> /share/components/document-details/document-links-min.js HTTP/1.1",
>> host: "xxx", referrer: 
>> "https://xxx/share/page/site/aleks-glossar/document-details?nodeRef=workspace://SpacesStore/985297e8-ff1b-4423-b6e9-8dacc0011196"
>> ###
>>
>> Could it be that try_files match on '/$1$2' but the open call get
>> '/$1$3' or something else?
>>
>> attached the debug log.
>
> This is the bug in alias and try_files interaction.  Or, more
> strictly, such configuration should be rejected during testing
> configuration as it's not really make sense: alias within regex
> location specifies full path to a file, and try_files is
> meaningless here.
>
> You probably mean to use "root" instead, i.e.
>
>     location ~ ^/(share|alfresco)(/res)?(.*$) {
>         root /home/alfresco/alfresco-4.0.b/tomcat/webapps;
>         try_files $uri /$1$2 /$1$3 @alfresco;
>     }
>
> Maxim Dounin
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx


From nginx-forum at nginx.us  Sat Nov 12 12:05:40 2011
From: nginx-forum at nginx.us (TECK)
Date: Sat, 12 Nov 2011 07:05:40 -0500
Subject: Nginx 1.0.9 reporting wrong version in PHP 5.3.8
Message-ID: <85ae2cb4a0f8b6be2a5f6ab73923ca00.NginxMailingListEnglish@forum.nginx.org>

Hi,

I installed Nginx 1.0.9 and when I look at the phpinfo() details, it
reports as version 1.0.4.
Anyone else has this issue?

Regards,

Floren Munteanu

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218241,218241#msg-218241


From nginx-forum at nginx.us  Sat Nov 12 12:08:13 2011
From: nginx-forum at nginx.us (TECK)
Date: Sat, 12 Nov 2011 07:08:13 -0500
Subject: Nginx 1.0.9 reporting wrong version in PHP 5.3.8
In-Reply-To: <85ae2cb4a0f8b6be2a5f6ab73923ca00.NginxMailingListEnglish@forum.nginx.org>
References: <85ae2cb4a0f8b6be2a5f6ab73923ca00.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <d59078d20f8478dd63156e290a2f9b30.NginxMailingListEnglish@forum.nginx.org>

Never mind, I was on the wrong server. Sorry about that.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218241,218242#msg-218242


From nginx-forum at nginx.us  Sat Nov 12 19:47:44 2011
From: nginx-forum at nginx.us (xin)
Date: Sat, 12 Nov 2011 14:47:44 -0500
Subject: Rails App, Nginx, Virtual Hosts and bandwidth shaping
In-Reply-To: <noneya@business.com>
References: <a42b2c16fb320a285d06245963adb12e@ruby-forum.com>
	<noneya@business.com>
Message-ID: <aba798dc2668330ba40ae91dba6071b8.NginxMailingListEnglish@forum.nginx.org>

Anton Yuzhaninov Hi! This means that EVERY virtualhost have that limit
or all virtualhosts together? How I could do that to limit for example
10 virtualhosts to have all together 25mbps limited bandwidth, is it
possible? 10 virtualhosts together = 25mbps limit not each..

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,265,218247#msg-218247


From ilan at time4learning.com  Sun Nov 13 01:59:04 2011
From: ilan at time4learning.com (Ilan Berkner)
Date: Sat, 12 Nov 2011 20:59:04 -0500
Subject: Loading a PHP file without a query string parameter downlo
Message-ID: <CAOoWpPq96zpAWRar3KgY1nt1CbJu+K3JRVgeNL4NHgzvrqXX9g@mail.gmail.com>

We're upgrading our web server (physical machine).  Our site works fine.

However, I noticed that going to a PHP file, i.e. /index.php causes the
file to download.  Going to /index.php?asdfasdf causes the file to work
correctly.  Help?

In our Nginx configuration we have:

                location ~ \.php$
                {
                        fastcgi_param HTTPS on;
                        include fcgi;
                        fastcgi_pass joomphp;
                }

In our old server, this works fine and we don't have this issue.


-- 

Ilan Berkner
Chief Technology Officer
Time4Learning.com

6300 NE 1st Ave., Suite 203
Ft. Lauderdale, FL 33334
(954) 771-0914

<http://www.twitter.com/time4learning> <http://www.facebook.com/Time4Learning>


Time4Learning.com - Online interactive curriculum for home use, PreK-8th
Grade.
Time4Writing.com - Online writing tutorials for high, middle, and
elementary school students.
Time4Learning.net - A forum to chat with parents online about kids,
education, parenting and more.
spellingcity.com - Online vocabulary and spelling activities for teachers,
parents and students.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20111112/f536d068/attachment.html>

From mdounin at mdounin.ru  Sun Nov 13 02:35:40 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Sun, 13 Nov 2011 06:35:40 +0400
Subject: Loading a PHP file without a query string parameter downlo
In-Reply-To: <CAOoWpPq96zpAWRar3KgY1nt1CbJu+K3JRVgeNL4NHgzvrqXX9g@mail.gmail.com>
References: <CAOoWpPq96zpAWRar3KgY1nt1CbJu+K3JRVgeNL4NHgzvrqXX9g@mail.gmail.com>
Message-ID: <20111113023534.GL95664@mdounin.ru>

Hello!

On Sat, Nov 12, 2011 at 08:59:04PM -0500, Ilan Berkner wrote:

> We're upgrading our web server (physical machine).  Our site works fine.
> 
> However, I noticed that going to a PHP file, i.e. /index.php causes the
> file to download.  Going to /index.php?asdfasdf causes the file to work
> correctly.  Help?
> 
> In our Nginx configuration we have:
> 
>                 location ~ \.php$
>                 {
>                         fastcgi_param HTTPS on;
>                         include fcgi;
>                         fastcgi_pass joomphp;
>                 }
> 
> In our old server, this works fine and we don't have this issue.

Browser cache?

Maxim Dounin


From ilan at time4learning.com  Sun Nov 13 02:41:43 2011
From: ilan at time4learning.com (Ilan Berkner)
Date: Sat, 12 Nov 2011 21:41:43 -0500
Subject: Loading a PHP file without a query string parameter downlo
In-Reply-To: <20111113023534.GL95664@mdounin.ru>
References: <CAOoWpPq96zpAWRar3KgY1nt1CbJu+K3JRVgeNL4NHgzvrqXX9g@mail.gmail.com>
	<20111113023534.GL95664@mdounin.ru>
Message-ID: <CAOoWpPp4NUKiELJhqoxNhqvs6Y3Ye6Nb2EGjgNkwpRERA3=QOQ@mail.gmail.com>

That's what it was, thanks.

On Sat, Nov 12, 2011 at 9:35 PM, Maxim Dounin <mdounin at mdounin.ru> wrote:

> Hello!
>
> On Sat, Nov 12, 2011 at 08:59:04PM -0500, Ilan Berkner wrote:
>
> > We're upgrading our web server (physical machine).  Our site works fine.
> >
> > However, I noticed that going to a PHP file, i.e. /index.php causes the
> > file to download.  Going to /index.php?asdfasdf causes the file to work
> > correctly.  Help?
> >
> > In our Nginx configuration we have:
> >
> >                 location ~ \.php$
> >                 {
> >                         fastcgi_param HTTPS on;
> >                         include fcgi;
> >                         fastcgi_pass joomphp;
> >                 }
> >
> > In our old server, this works fine and we don't have this issue.
>
> Browser cache?
>
> Maxim Dounin
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx
>



-- 

Ilan Berkner
Chief Technology Officer
Time4Learning.com

6300 NE 1st Ave., Suite 203
Ft. Lauderdale, FL 33334
(954) 771-0914

<http://www.twitter.com/time4learning> <http://www.facebook.com/Time4Learning>


Time4Learning.com - Online interactive curriculum for home use, PreK-8th
Grade.
Time4Writing.com - Online writing tutorials for high, middle, and
elementary school students.
Time4Learning.net - A forum to chat with parents online about kids,
education, parenting and more.
spellingcity.com - Online vocabulary and spelling activities for teachers,
parents and students.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20111112/ba44d242/attachment.html>

From ilan at time4learning.com  Sun Nov 13 06:58:39 2011
From: ilan at time4learning.com (Ilan Berkner)
Date: Sun, 13 Nov 2011 01:58:39 -0500
Subject: Can anyone explain what is nginx.old?
Message-ID: <CAOoWpPrCzywiOk2YSy3H6Jh7C8GzmfPi6F2NT0h1Ti4iSOvr6g@mail.gmail.com>

I'm seeing some warnings in root log files regarding excessive processes
for nginx.old...

-- 

Ilan Berkner
Chief Technology Officer
Time4Learning.com

6300 NE 1st Ave., Suite 203
Ft. Lauderdale, FL 33334
(954) 771-0914

<http://www.twitter.com/time4learning> <http://www.facebook.com/Time4Learning>


Time4Learning.com - Online interactive curriculum for home use, PreK-8th
Grade.
Time4Writing.com - Online writing tutorials for high, middle, and
elementary school students.
Time4Learning.net - A forum to chat with parents online about kids,
education, parenting and more.
spellingcity.com - Online vocabulary and spelling activities for teachers,
parents and students.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20111113/dc5dd537/attachment.html>

From ne at vbart.ru  Sun Nov 13 08:12:07 2011
From: ne at vbart.ru (Valentin V. Bartenev)
Date: Sun, 13 Nov 2011 12:12:07 +0400
Subject: Can anyone explain what is nginx.old?
In-Reply-To: <CAOoWpPrCzywiOk2YSy3H6Jh7C8GzmfPi6F2NT0h1Ti4iSOvr6g@mail.gmail.com>
References: <CAOoWpPrCzywiOk2YSy3H6Jh7C8GzmfPi6F2NT0h1Ti4iSOvr6g@mail.gmail.com>
Message-ID: <201111131212.07416.ne@vbart.ru>

On Sunday 13 November 2011 10:58:39 Ilan Berkner wrote:
> I'm seeing some warnings in root log files regarding excessive processes
> for nginx.old...

When you do "make install" to some destination directory, and installation script 
finds out that there is an old "nginx" binary, then it renames the old binary to 
the "nginx.old", before copying the new one.

wbr, Valentin V. Bartenev


From nginx-forum at nginx.us  Sun Nov 13 12:06:02 2011
From: nginx-forum at nginx.us (Long Wan)
Date: Sun, 13 Nov 2011 07:06:02 -0500
Subject: nginx worker process hang,cpu load 100%
Message-ID: <fe1c35dd759d61202e61b784ce4e5825.NginxMailingListEnglish@forum.nginx.org>

Hi, 
I have faced a trouble with nginx runs as a http revers proxy server,the
worker process sometimes hanging there, cpu usage up to 100%,it's never
recovey until i kill the process,below is the detail informations: 

system environment:
[root at host-22 ~]# lsb_release -a
LSB Version:   
:core-3.1-amd64:core-3.1-ia32:core-3.1-noarch:graphics-3.1-amd64:graphics-3.1-ia32:graphics-3.1-noarch
Distributor ID: CentOS
Description:    CentOS release 5.5 (Final)
Release:        5.5
Codename:       Final
[root at host-22 ~]# 
[root at host-22 ~]# uname -a
Linux host-22 2.6.18-194.el5 #1 SMP Fri Apr 2 14:58:14 EDT 2010 x86_64
x86_64 x86_64 GNU/Linux



nginx version:
[root at host-22 ~]# /usr/local/nginx/sbin/nginx  -V
nginx: nginx version: nginx/1.0.4
nginx: built by gcc 4.1.2 20080704 (Red Hat 4.1.2-46)
nginx: TLS SNI support disabled
nginx: configure arguments: --user=www --group=www
--prefix=/usr/local/nginx --with-http_stub_status_module
--with-http_ssl_module --with-openssl-opt=enable-tlsext
--with-http_sub_module --with-cc-opt=-O2 --with-cpu-opt=opteron
[root at host-22 ~]# 
(also tested under 1.0.6 and 1.0.9,have the same problem)


nginx config( nginx runs as a http revers proxy server):
worker_processes  8;

events {
    use epoll;
    worker_connections  5120;
}

http {
sendfile        on;
keepalive_timeout  15;
...

upstream 2012_servers {
                server 10.0.7.5:80  max_fails=2  fail_timeout=30s;
                server 10.0.7.6:80  max_fails=2  fail_timeout=30s;
                server 10.0.7.7:80 max_fails=2  fail_timeout=30s;
                server 10.0.7.8:80 max_fails=2  fail_timeout=30s;
        }


    server {
        listen       80;
        server_name  test.2012.com ;
        ...

        location / {
                include proxy.conf;
                proxy_pass    http://2012_servers;
        }
...

}



trouble:
[root at host-22 ~]# ps aux|grep -e CPU -e nginx
USER       PID %CPU %MEM    VSZ   RSS TTY      STAT START   TIME
COMMAND
root       936  0.0  0.0  49328  7572 ?        Ss   Nov11   2:37 nginx:
master process /usr/local/nginx/sbin/nginx
www       1130 99.9  0.0  55764 13472 ?        R    Nov11 2664:28 nginx:
worker process      
www       1216 99.9  0.0  53668 11092 ?        R    Nov11 2660:23 nginx:
worker process      
www      31057  0.0  0.0  50816  8820 ?        S    19:40   0:00 nginx:
worker process      
www      31058  0.0  0.0  50816  8820 ?        S    19:40   0:00 nginx:
worker process      
www      31059  0.0  0.0  50816  8820 ?        S    19:40   0:00 nginx:
worker process      
www      31060  0.0  0.0  50816  8820 ?        S    19:40   0:00 nginx:
worker process      
www      31061  0.0  0.0  50816  8820 ?        S    19:40   0:00 nginx:
worker process      
www      31062  0.8  0.0  50816  8820 ?        S    19:40   0:00 nginx:
worker process      
www      31063  0.1  0.0  50816  9012 ?        S    19:40   0:00 nginx:
worker process      
www      31064  0.2  0.0  50816  8820 ?        S    19:40   0:00 nginx:
worker process      

two nginx worker processes(pid 1130,1216) are hanging. there is nothing
significant message i can found in error.log or strace (-p 1130|1216).

Grateful for any advice.

thanks.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218259,218259#msg-218259


From mdounin at mdounin.ru  Sun Nov 13 12:35:30 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Sun, 13 Nov 2011 16:35:30 +0400
Subject: nginx worker process hang,cpu load 100%
In-Reply-To: <fe1c35dd759d61202e61b784ce4e5825.NginxMailingListEnglish@forum.nginx.org>
References: <fe1c35dd759d61202e61b784ce4e5825.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20111113123530.GM95664@mdounin.ru>

Hello!

On Sun, Nov 13, 2011 at 07:06:02AM -0500, Long Wan wrote:

> I have faced a trouble with nginx runs as a http revers proxy server,the
> worker process sometimes hanging there, cpu usage up to 100%,it's never
> recovey until i kill the process,below is the detail informations: 

[...]

> trouble:
> [root at host-22 ~]# ps aux|grep -e CPU -e nginx
> USER       PID %CPU %MEM    VSZ   RSS TTY      STAT START   TIME
> COMMAND
> root       936  0.0  0.0  49328  7572 ?        Ss   Nov11   2:37 nginx:
> master process /usr/local/nginx/sbin/nginx
> www       1130 99.9  0.0  55764 13472 ?        R    Nov11 2664:28 nginx:
> worker process      
> www       1216 99.9  0.0  53668 11092 ?        R    Nov11 2660:23 nginx:
> worker process      
> www      31057  0.0  0.0  50816  8820 ?        S    19:40   0:00 nginx:
> worker process      

[...]

> two nginx worker processes(pid 1130,1216) are hanging. there is nothing
> significant message i can found in error.log or strace (-p 1130|1216).

Please try attaching to a runaway process with gdb and check where 
it loops, i.e. 

gdb /path/to/nginx <pid>
bt
n
... (repeat 'n' several times to see loop)

Maxim Dounin


From nginx-forum at nginx.us  Sun Nov 13 15:10:29 2011
From: nginx-forum at nginx.us (zhenwei)
Date: Sun, 13 Nov 2011 10:10:29 -0500
Subject: what's the difference between proxy_store and proxy_cache?
Message-ID: <b3f52c538e797dbd7048bbaf4fab82e8.NginxMailingListEnglish@forum.nginx.org>

>From Nginx official site there are two method to store remote upstream
on local file system, "proxy_cache" and "proxy_store", the document
describes clearly on how to configure, but still I have problem on the
differences. Months ago I did some test over the both and found
proxy_cache owns far better performance, and I thought proxy_cache is
"cache in memory", recently after diving into source code, I recognized
both are file based, so i'm not sure if it could afford high connection
where disk probably becomes the bottleneck.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218263,218263#msg-218263


From ilan at time4learning.com  Sun Nov 13 15:57:53 2011
From: ilan at time4learning.com (Ilan Berkner)
Date: Sun, 13 Nov 2011 10:57:53 -0500
Subject: Match all requests
Message-ID: <CAOoWpPpB24T4EZ4CFnfqbJNvvjdc4bV00bO-0KWpy2mDhpaMFQ@mail.gmail.com>

I have this location configuration:

        location /
        {
                index  maintenance.htm;
                error_page 404 = maintenance.htm;
                log_not_found off;
        }

which I thought captures all requests, however, entering "/index.php" for
example, causes the file to be downloaded instead of going to the
"maintenance.htm" file.  How can I capture all requests?

-- 

Ilan Berkner
Chief Technology Officer
Time4Learning.com

6300 NE 1st Ave., Suite 203
Ft. Lauderdale, FL 33334
(954) 771-0914

<http://www.twitter.com/time4learning> <http://www.facebook.com/Time4Learning>


Time4Learning.com - Online interactive curriculum for home use, PreK-8th
Grade.
Time4Writing.com - Online writing tutorials for high, middle, and
elementary school students.
Time4Learning.net - A forum to chat with parents online about kids,
education, parenting and more.
spellingcity.com - Online vocabulary and spelling activities for teachers,
parents and students.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20111113/7f3225dd/attachment.html>

From lvella at gmail.com  Sun Nov 13 16:30:44 2011
From: lvella at gmail.com (Lucas Clemente Vella)
Date: Sun, 13 Nov 2011 14:30:44 -0200
Subject: Match all requests
In-Reply-To: <CAOoWpPpB24T4EZ4CFnfqbJNvvjdc4bV00bO-0KWpy2mDhpaMFQ@mail.gmail.com>
References: <CAOoWpPpB24T4EZ4CFnfqbJNvvjdc4bV00bO-0KWpy2mDhpaMFQ@mail.gmail.com>
Message-ID: <CAGCathzrodSqPuDCwckPhJF_LjbpJwSzLqDy0z9SzkyYcM+==g@mail.gmail.com>

2011/11/13 Ilan Berkner <ilan at time4learning.com>

> I have this location configuration:
>
>         location /
>         {
>                 index  maintenance.htm;
>                 error_page 404 = maintenance.htm;
>                 log_not_found off;
>         }
>
> which I thought captures all requests, however, entering "/index.php" for
> example, causes the file to be downloaded instead of going to the
> "maintenance.htm" file.  How can I capture all requests?
>

Maybe if you put like this before the other locations:

 location ~ .*$
        {
                index  maintenance.htm;
                error_page 404 = maintenance.htm;
                log_not_found off;
        }


-- 
Lucas Clemente Vella
lvella at gmail.com
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20111113/f1739234/attachment.html>

From ne at vbart.ru  Sun Nov 13 16:32:46 2011
From: ne at vbart.ru (Valentin V. Bartenev)
Date: Sun, 13 Nov 2011 20:32:46 +0400
Subject: Match all requests
In-Reply-To: <CAOoWpPpB24T4EZ4CFnfqbJNvvjdc4bV00bO-0KWpy2mDhpaMFQ@mail.gmail.com>
References: <CAOoWpPpB24T4EZ4CFnfqbJNvvjdc4bV00bO-0KWpy2mDhpaMFQ@mail.gmail.com>
Message-ID: <201111132032.46316.ne@vbart.ru>

On Sunday 13 November 2011 19:57:53 Ilan Berkner wrote:
> I have this location configuration:
> 
>         location /
>         {

This:
>                 index  maintenance.htm;
only captures requests to directories (i.e., ends with a /). 

This:
>                 error_page 404 = maintenance.htm;
only captures 404 responses.

>                 log_not_found off;
>         }
> 
> which I thought captures all requests, however, entering "/index.php" for
> example, causes the file to be downloaded instead of going to the
> "maintenance.htm" file.  How can I capture all requests?

Probably, you want something like this:

 error_page 404 = /maintenance.htm;

 location / {
   return 404;
 }

 location = /maintenance.htm {}

wbr, Valentin V. Bartenev


From igor at sysoev.ru  Sun Nov 13 16:44:04 2011
From: igor at sysoev.ru (Igor Sysoev)
Date: Sun, 13 Nov 2011 20:44:04 +0400
Subject: Match all requests
In-Reply-To: <CAOoWpPpB24T4EZ4CFnfqbJNvvjdc4bV00bO-0KWpy2mDhpaMFQ@mail.gmail.com>
References: <CAOoWpPpB24T4EZ4CFnfqbJNvvjdc4bV00bO-0KWpy2mDhpaMFQ@mail.gmail.com>
Message-ID: <20111113164404.GA96350@nginx.com>

On Sun, Nov 13, 2011 at 10:57:53AM -0500, Ilan Berkner wrote:
> I have this location configuration:
> 
>         location /
>         {
>                 index  maintenance.htm;
>                 error_page 404 = maintenance.htm;
>                 log_not_found off;
>         }
> 
> which I thought captures all requests, however, entering "/index.php" for
> example, causes the file to be downloaded instead of going to the
> "maintenance.htm" file.  How can I capture all requests?

location / {
    try_files  /maintenance.htm  =404;
}


-- 
Igor Sysoev


From mdounin at mdounin.ru  Sun Nov 13 17:04:33 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Sun, 13 Nov 2011 21:04:33 +0400
Subject: what's the difference between proxy_store and proxy_cache?
In-Reply-To: <b3f52c538e797dbd7048bbaf4fab82e8.NginxMailingListEnglish@forum.nginx.org>
References: <b3f52c538e797dbd7048bbaf4fab82e8.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20111113170433.GN95664@mdounin.ru>

Hello!

On Sun, Nov 13, 2011 at 10:10:29AM -0500, zhenwei wrote:

> From Nginx official site there are two method to store remote upstream
> on local file system, "proxy_cache" and "proxy_store", the document
> describes clearly on how to configure, but still I have problem on the
> differences. Months ago I did some test over the both and found
> proxy_cache owns far better performance, and I thought proxy_cache is
> "cache in memory", recently after diving into source code, I recognized
> both are file based, so i'm not sure if it could afford high connection
> where disk probably becomes the bottleneck.

"proxy_cache" is general-purpose cache with automatic lookups 
before proxy_pass, expiration support and so on.  It is usually 
what you need if you need caching capabilities.

"proxy_store" is just a method to store proxied files on disk.  It 
may be used to construct cache-like setups (usually involving 
try_files and/or error_page-based fallback), though it's up to you 
to implement any required logic.

Maxim Dounin


From vbart at nginx.com  Sun Nov 13 16:50:05 2011
From: vbart at nginx.com (Valentin V. Bartenev)
Date: Sun, 13 Nov 2011 20:50:05 +0400
Subject: what's the difference between proxy_store and proxy_cache?
In-Reply-To: <b3f52c538e797dbd7048bbaf4fab82e8.NginxMailingListEnglish@forum.nginx.org>
References: <b3f52c538e797dbd7048bbaf4fab82e8.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <201111132050.06153.vbart@nginx.com>

On Sunday 13 November 2011 19:10:29 zhenwei wrote:
> From Nginx official site there are two method to store remote upstream
> on local file system, "proxy_cache" and "proxy_store", the document
> describes clearly on how to configure, but still I have problem on the
> differences.

The "proxy_store" just stores backend's responses to a defined path. It's totally 
up to you, what to do with these files after they were stored.

The "proxy_cache" alone doesn't do anything. But with other proxy_cache_* 
directives, you can setup a file cache with key, life time, etc.

wbr, Valentin V. Bartenev


From ilan at time4learning.com  Sun Nov 13 21:38:53 2011
From: ilan at time4learning.com (Ilan Berkner)
Date: Sun, 13 Nov 2011 16:38:53 -0500
Subject: Match all requests
In-Reply-To: <20111113164404.GA96350@nginx.com>
References: <CAOoWpPpB24T4EZ4CFnfqbJNvvjdc4bV00bO-0KWpy2mDhpaMFQ@mail.gmail.com>
	<20111113164404.GA96350@nginx.com>
Message-ID: <CAOoWpPqznT2ujQYhwTromFR_KeS5CkBO0_6tWoKPyyrziwdohQ@mail.gmail.com>

Igor,

Thanks, but the code you provided still causes the
domain.com/index.phpfile to be downloaded instead of the maintenance
page to show up?

On Sun, Nov 13, 2011 at 11:44 AM, Igor Sysoev <igor at sysoev.ru> wrote:

> On Sun, Nov 13, 2011 at 10:57:53AM -0500, Ilan Berkner wrote:
> > I have this location configuration:
> >
> >         location /
> >         {
> >                 index  maintenance.htm;
> >                 error_page 404 = maintenance.htm;
> >                 log_not_found off;
> >         }
> >
> > which I thought captures all requests, however, entering "/index.php" for
> > example, causes the file to be downloaded instead of going to the
> > "maintenance.htm" file.  How can I capture all requests?
>
> location / {
>    try_files  /maintenance.htm  =404;
> }
>
>
> --
> Igor Sysoev
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx
>



-- 

Ilan Berkner
Chief Technology Officer
Time4Learning.com

6300 NE 1st Ave., Suite 203
Ft. Lauderdale, FL 33334
(954) 771-0914

<http://www.twitter.com/time4learning> <http://www.facebook.com/Time4Learning>


Time4Learning.com - Online interactive curriculum for home use, PreK-8th
Grade.
Time4Writing.com - Online writing tutorials for high, middle, and
elementary school students.
Time4Learning.net - A forum to chat with parents online about kids,
education, parenting and more.
spellingcity.com - Online vocabulary and spelling activities for teachers,
parents and students.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20111113/4cca3ebc/attachment.html>

From jabberuser at gmail.com  Sun Nov 13 21:46:53 2011
From: jabberuser at gmail.com (Piotr Karbowski)
Date: Sun, 13 Nov 2011 22:46:53 +0100
Subject: Match all requests
In-Reply-To: <CAOoWpPqznT2ujQYhwTromFR_KeS5CkBO0_6tWoKPyyrziwdohQ@mail.gmail.com>
References: <CAOoWpPpB24T4EZ4CFnfqbJNvvjdc4bV00bO-0KWpy2mDhpaMFQ@mail.gmail.com>
	<20111113164404.GA96350@nginx.com>
	<CAOoWpPqznT2ujQYhwTromFR_KeS5CkBO0_6tWoKPyyrziwdohQ@mail.gmail.com>
Message-ID: <4EC03ACD.7080500@gmail.com>

On 13.11.2011 22:38, Ilan Berkner wrote:
> Thanks, but the code you provided still causes the
> domain.com/index.phpfile to be downloaded instead of the maintenance
> page to show up?

Because you use location .php$ for php, I presume so you need add there 
something like 'try_files /maintenance.htm $uri =404;

There = to the php's location block.

You may need read http://wiki.nginx.org/NginxHttpCoreModule#location

-- Piotr.


From nginx-forum at nginx.us  Mon Nov 14 01:28:49 2011
From: nginx-forum at nginx.us (Long Wan)
Date: Sun, 13 Nov 2011 20:28:49 -0500
Subject: nginx worker process hang,cpu load 100%
In-Reply-To: <fe1c35dd759d61202e61b784ce4e5825.NginxMailingListEnglish@forum.nginx.org>
References: <fe1c35dd759d61202e61b784ce4e5825.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <8be64fff352bfff2ab00fc62646fab46.NginxMailingListEnglish@forum.nginx.org>

Hello Maxim,Thanks for your reply, I tried gdb as you tolde me , it
reported something :


[root at host-22 ~]# gdb /usr/local/nginx/sbin/nginx 1130
GNU gdb (GDB) Red Hat Enterprise Linux (7.0.1-23.el5)
Copyright (C) 2009 Free Software Foundation, Inc.
License GPLv3+: GNU GPL version 3 or later
<http://gnu.org/licenses/gpl.html>
This is free software: you are free to change and redistribute it.
There is NO WARRANTY, to the extent permitted by law.  Type "show
copying"
and "show warranty" for details.
This GDB was configured as "x86_64-redhat-linux-gnu".
For bug reporting instructions, please see:
<http://www.gnu.org/software/gdb/bugs/>...
Reading symbols from /usr/local/nginx/sbin/nginx...done.
Attaching to program: /usr/local/nginx/sbin/nginx, process 1130
Reading symbols from /lib64/libpthread.so.0...(no debugging symbols
found)...done.
[Thread debugging using libthread_db enabled]
Loaded symbols for /lib64/libpthread.so.0
Reading symbols from /lib64/libcrypt.so.1...(no debugging symbols
found)...done.
Loaded symbols for /lib64/libcrypt.so.1
Reading symbols from /lib64/libpcre.so.0...(no debugging symbols
found)...done.
Loaded symbols for /lib64/libpcre.so.0
Reading symbols from /lib64/libssl.so.6...(no debugging symbols
found)...done.
Loaded symbols for /lib64/libssl.so.6
Reading symbols from /lib64/libcrypto.so.6...(no debugging symbols
found)...done.
Loaded symbols for /lib64/libcrypto.so.6
Reading symbols from /lib64/libdl.so.2...(no debugging symbols
found)...done.
Loaded symbols for /lib64/libdl.so.2
Reading symbols from /usr/lib64/libz.so.1...(no debugging symbols
found)...done.
Loaded symbols for /usr/lib64/libz.so.1
Reading symbols from /lib64/libc.so.6...(no debugging symbols
found)...done.
Loaded symbols for /lib64/libc.so.6
Reading symbols from /lib64/ld-linux-x86-64.so.2...(no debugging symbols
found)...done.
Loaded symbols for /lib64/ld-linux-x86-64.so.2
Reading symbols from /usr/lib64/libgssapi_krb5.so.2...(no debugging
symbols found)...done.
Loaded symbols for /usr/lib64/libgssapi_krb5.so.2
Reading symbols from /usr/lib64/libkrb5.so.3...(no debugging symbols
found)...done.
Loaded symbols for /usr/lib64/libkrb5.so.3
Reading symbols from /lib64/libcom_err.so.2...(no debugging symbols
found)...done.
Loaded symbols for /lib64/libcom_err.so.2
Reading symbols from /usr/lib64/libk5crypto.so.3...(no debugging symbols
found)...done.
Loaded symbols for /usr/lib64/libk5crypto.so.3
Reading symbols from /usr/lib64/libkrb5support.so.0...(no debugging
symbols found)...done.
Loaded symbols for /usr/lib64/libkrb5support.so.0
Reading symbols from /lib64/libkeyutils.so.1...(no debugging symbols
found)...done.
Loaded symbols for /lib64/libkeyutils.so.1
Reading symbols from /lib64/libresolv.so.2...(no debugging symbols
found)...done.
Loaded symbols for /lib64/libresolv.so.2
Reading symbols from /lib64/libselinux.so.1...(no debugging symbols
found)...done.
Loaded symbols for /lib64/libselinux.so.1
Reading symbols from /lib64/libsepol.so.1...(no debugging symbols
found)...done.
Loaded symbols for /lib64/libsepol.so.1
Reading symbols from /lib64/libnss_files.so.2...(no debugging symbols
found)...done.
Loaded symbols for /lib64/libnss_files.so.2
ngx_http_upstream_get_round_robin_peer (pc=0x78166f0, data=<value
optimized out>) at src/http/ngx_http_upstream_round_robin.c:413
413     src/http/ngx_http_upstream_round_robin.c: No such file or
directory.
        in src/http/ngx_http_upstream_round_robin.c
(gdb) bt
#0  ngx_http_upstream_get_round_robin_peer (pc=0x78166f0, data=<value
optimized out>) at src/http/ngx_http_upstream_round_robin.c:413
#1  0x000000000041a8fc in ngx_event_connect_peer (pc=0x78166f0) at
src/event/ngx_event_connect.c:24
#2  0x000000000043d1e8 in ngx_http_upstream_connect (r=0x7cf6310,
u=0x78166e0) at src/http/ngx_http_upstream.c:1089
#3  0x000000000043ea3a in ngx_http_upstream_init_request (r=0x7cf6310)
at src/http/ngx_http_upstream.c:628
#4  0x0000000000435185 in ngx_http_read_client_request_body
(r=0x7cf6310, post_handler=0x43eec0 <ngx_http_upstream_init>)
    at src/http/ngx_http_request_body.c:153
#5  0x0000000000456b46 in ngx_http_proxy_handler (r=0x7cf6310) at
src/http/modules/ngx_http_proxy_module.c:617
#6  0x000000000042b15c in ngx_http_core_content_phase (r=0x7cf6310,
ph=0x7b5a4e0) at src/http/ngx_http_core_module.c:1339
#7  0x0000000000426817 in ngx_http_core_run_phases (r=0x7cf6310) at
src/http/ngx_http_core_module.c:837
#8  0x000000000042f6d6 in ngx_http_process_request (r=0x7cf6310) at
src/http/ngx_http_request.c:1650
#9  0x0000000000430314 in ngx_http_process_request_line (rev=0x7c65578)
at src/http/ngx_http_request.c:893
#10 0x0000000000420a04 in ngx_epoll_process_events (cycle=<value
optimized out>, timer=<value optimized out>, flags=<value optimized
out>)
    at src/event/modules/ngx_epoll_module.c:635
#11 0x0000000000419bad in ngx_process_events_and_timers
(cycle=0x784e770) at src/event/ngx_event.c:245
#12 0x000000000041f528 in ngx_worker_process_cycle (cycle=0x784e770,
data=<value optimized out>) at src/os/unix/ngx_process_cycle.c:800
#13 0x000000000041dc89 in ngx_spawn_process (cycle=0x784e770,
proc=0x41f470 <ngx_worker_process_cycle>, data=0x0, name=0x4652e1
"worker process", 
    respawn=-4) at src/os/unix/ngx_process.c:196
#14 0x000000000041eb0b in ngx_start_worker_processes (cycle=0x784e770,
n=8, type=-4) at src/os/unix/ngx_process_cycle.c:360
#15 0x000000000041fea8 in ngx_master_process_cycle (cycle=0x784e770) at
src/os/unix/ngx_process_cycle.c:249
#16 0x0000000000406069 in main (argc=1, argv=<value optimized out>) at
src/core/nginx.c:405
(gdb) n

there is no output when type 'n', should i recompile nginx with
'--with-debug' configure option ?

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218259,218280#msg-218280


From nginx-forum at nginx.us  Mon Nov 14 01:37:58 2011
From: nginx-forum at nginx.us (zhenwei)
Date: Sun, 13 Nov 2011 20:37:58 -0500
Subject: what's the difference between proxy_store and proxy_cache?
In-Reply-To: <201111132050.06153.vbart@nginx.com>
References: <201111132050.06153.vbart@nginx.com>
Message-ID: <87d9fadc36d30fe8b02560bd1c9720d8.NginxMailingListEnglish@forum.nginx.org>

thanks, and  how about the performance?

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218263,218281#msg-218281


From ne at vbart.ru  Mon Nov 14 05:48:42 2011
From: ne at vbart.ru (Valentin V. Bartenev)
Date: Mon, 14 Nov 2011 09:48:42 +0400
Subject: what's the difference between proxy_store and proxy_cache?
In-Reply-To: <87d9fadc36d30fe8b02560bd1c9720d8.NginxMailingListEnglish@forum.nginx.org>
References: <201111132050.06153.vbart@nginx.com>
	<87d9fadc36d30fe8b02560bd1c9720d8.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <201111140948.43044.ne@vbart.ru>

On Monday 14 November 2011 05:37:58 zhenwei wrote:
> thanks, and  how about the performance?

So, if you care about disk performance and have enough RAM, probably, you may 
need to tune kernel disk cache or even consider to put nginx cache on "/dev/shm".

Also, you can use the memcached module for caching.

wbr, Valentin V. Bartenev


From nginx-forum at nginx.us  Mon Nov 14 06:26:43 2011
From: nginx-forum at nginx.us (amastr)
Date: Mon, 14 Nov 2011 01:26:43 -0500
Subject: Handle connection abort
Message-ID: <040feea69e166cf13bfadea620db5a50.NginxMailingListEnglish@forum.nginx.org>

Hello,

I have an nginx module that receives http request, treats it and sends
response to the client host. 
I need to handle situation when connection between server and client is
aborted. How can I do it?

Thanks in advance

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218285,218285#msg-218285


From kasatkinnv at gmail.com  Mon Nov 14 09:14:45 2011
From: kasatkinnv at gmail.com (kasatkinnv at gmail.com)
Date: Mon, 14 Nov 2011 13:14:45 +0400
Subject: Nginx and httpfs2
Message-ID: <CACXN2RS4=T1Ewbr8UoJuG+2Q+9bk0oXgf_Ut=o+c03QmgSDmnQ@mail.gmail.com>

Hi, List!

Has anyone successfully used httpfs2 with nginx?
I'm getting Input/Output error when trying to mount file system over
HTTP from nginx server.
The requirements for httpfs2: "The server must be able to send byte
ranges". Does nginx have support for this?
The mounting over HTTP works with Apache and lighttpd but slow.

Thank you!

-- 
Nikolay Kasatkin


From mdounin at mdounin.ru  Mon Nov 14 10:19:40 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Mon, 14 Nov 2011 14:19:40 +0400
Subject: nginx worker process hang,cpu load 100%
In-Reply-To: <8be64fff352bfff2ab00fc62646fab46.NginxMailingListEnglish@forum.nginx.org>
References: <fe1c35dd759d61202e61b784ce4e5825.NginxMailingListEnglish@forum.nginx.org>
	<8be64fff352bfff2ab00fc62646fab46.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20111114101940.GP95664@mdounin.ru>

Hello!

On Sun, Nov 13, 2011 at 08:28:49PM -0500, Long Wan wrote:

> Hello Maxim,Thanks for your reply, I tried gdb as you tolde me , it
> reported something :

[...]

> (gdb) bt
> #0  ngx_http_upstream_get_round_robin_peer (pc=0x78166f0, data=<value
> optimized out>) at src/http/ngx_http_upstream_round_robin.c:413
> #1  0x000000000041a8fc in ngx_event_connect_peer (pc=0x78166f0) at
> src/event/ngx_event_connect.c:24
> #2  0x000000000043d1e8 in ngx_http_upstream_connect (r=0x7cf6310,
> u=0x78166e0) at src/http/ngx_http_upstream.c:1089
[...]
> (gdb) n
> 
> there is no output when type 'n', should i recompile nginx with
> '--with-debug' configure option ?

This looks very similar to this problem, fixed in 1.1.1/1.0.7:

    *) Bugfix: nginx hogged CPU if all servers in an upstream were marked as
       "down".

Are you sure you see the same problem in 1.0.9?

Maxim Dounin


From ssrini_vasan at hotmail.com  Mon Nov 14 13:04:39 2011
From: ssrini_vasan at hotmail.com (Srinivasan Subramanian)
Date: Mon, 14 Nov 2011 18:34:39 +0530
Subject: Nginx Load balancer mode for JBoss / Icefaces application
Message-ID: <SNT144-W248F760B63BF7E1E138037E7C00@phx.gbl>




Hello
We have setup nginx as a Loadbalancer on Centos 5.2 x64.  nginx is acting as a LB for a web application developed using Java servlets and Icefaces (1.8.2).  The web application is deployed on JBoss 5.1.   After configuration nginx is able to redirect the queries to the upstream servers properly.  However the session information is not being passed through or is being modified.  So the Icefaces servlet is repeatedly refreshing the login page every few seconds and keeps creating new sessions.
Please advise on any additional settings that need to be made.  The current settings are:
(nginx is running on 192.168.1.137)
upstream int-lb {        server 192.168.1.139:8080;        server 192.168.1.138:8080;}
server {    listen       80;    server_name  int-lb;
    #charset koi8-r;    access_log  /var/log/nginx/host.access.log  main;    error_log  /var/log/nginx/host.error.log  debug;    root /usr/app/jboss5/server/default/deploy/admin-console.war;
    location / {        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_pass http://int-lb;    }}
Thanks in advance for all assistance.

Regards
Srini 		 	   		  
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20111114/125e8476/attachment.html>

From nginx-forum at nginx.us  Mon Nov 14 14:37:29 2011
From: nginx-forum at nginx.us (Long Wan)
Date: Mon, 14 Nov 2011 09:37:29 -0500
Subject: nginx worker process hang,cpu load 100%
In-Reply-To: <fe1c35dd759d61202e61b784ce4e5825.NginxMailingListEnglish@forum.nginx.org>
References: <fe1c35dd759d61202e61b784ce4e5825.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <7ed1abd36d44e812afca391274f9e925.NginxMailingListEnglish@forum.nginx.org>

Hello,Maxim. Thanks for you help.

I reproduce the problem in nginx-1.0.9,

[root at host-22 ~]# /usr/local/nginx/sbin/nginx -V
nginx: nginx version: nginx/1.0.9
nginx: built by gcc 4.1.2 20080704 (Red Hat 4.1.2-51)
nginx: TLS SNI support disabled
nginx: configure arguments: --user=www --group=www
--prefix=/usr/local/nginx --with-http_stub_status_module
--with-http_ssl_module --with-openssl-opt=enable-tlsext
--with-http_sub_module --with-cc-opt=-O2 --with-cpu-opt=opteron
--add-module=../ngx_cache_purge-1.4
[root at host-22 ~]# 
[root at host-22 ~]# 
[root at host-22 ~]# ps aux|grep -e CPU -e nginx
USER       PID %CPU %MEM    VSZ   RSS TTY      STAT START   TIME
COMMAND
root       569  0.0  0.0  49016  6884 ?        Ss   21:59   0:00 nginx:
master process /usr/local/nginx/sbin/nginx
www        587 97.3  0.0  49380  7556 ?        R    22:00  10:16 nginx:
worker process      
www        588 93.8  0.0  49380  7556 ?        R    22:00   9:54 nginx:
worker process      
www        614 43.1  0.0  49412  7584 ?        T    22:01   4:07 nginx:
worker process      
root       781  0.0  0.1  95984 19684 pts/0    S+   22:06   0:00 gdb
/usr/local/nginx/sbin/nginx 614
www        876  0.0  0.0  50504  8464 ?        S    22:10   0:00 nginx:
worker process      
www        877  0.0  0.0  50504  8464 ?        S    22:10   0:00 nginx:
worker process      
www        878  0.0  0.0  50504  8464 ?        S    22:10   0:00 nginx:
worker process      
www        879  0.0  0.0  50504  8660 ?        S    22:10   0:00 nginx:
worker process      
www        880  0.0  0.0  50504  8464 ?        S    22:10   0:00 nginx:
worker process      
www        881  0.0  0.0  50504  8464 ?        S    22:10   0:00 nginx:
worker process      
www        882  0.0  0.0  50504  8464 ?        S    22:10   0:00 nginx:
worker process      
www        883  0.0  0.0  50504  8464 ?        S    22:10   0:00 nginx:
worker process      
root       954  0.0  0.0  61168   788 pts/1    S+   22:10   0:00 grep -e
CPU -e nginx
[root at host-22 ~]# 



[root at host-22 ~]# gdb /usr/local/nginx/sbin/nginx 614
GNU gdb (GDB) Red Hat Enterprise Linux (7.0.1-23.el5)
Copyright (C) 2009 Free Software Foundation, Inc.
License GPLv3+: GNU GPL version 3 or later
<http://gnu.org/licenses/gpl.html>
This is free software: you are free to change and redistribute it.
There is NO WARRANTY, to the extent permitted by law.  Type "show
copying"
and "show warranty" for details.
This GDB was configured as "x86_64-redhat-linux-gnu".
For bug reporting instructions, please see:
<http://www.gnu.org/software/gdb/bugs/>...
Reading symbols from /usr/local/nginx/sbin/nginx...done.
Attaching to program: /usr/local/nginx/sbin/nginx, process 614
Reading symbols from /lib64/libpthread.so.0...(no debugging symbols
found)...done.
[Thread debugging using libthread_db enabled]
Loaded symbols for /lib64/libpthread.so.0
Reading symbols from /lib64/libcrypt.so.1...(no debugging symbols
found)...done.
Loaded symbols for /lib64/libcrypt.so.1
Reading symbols from /lib64/libpcre.so.0...(no debugging symbols
found)...done.
Loaded symbols for /lib64/libpcre.so.0
Reading symbols from /lib64/libssl.so.6...(no debugging symbols
found)...done.
Loaded symbols for /lib64/libssl.so.6
Reading symbols from /lib64/libcrypto.so.6...(no debugging symbols
found)...done.
Loaded symbols for /lib64/libcrypto.so.6
Reading symbols from /lib64/libdl.so.2...(no debugging symbols
found)...done.
Loaded symbols for /lib64/libdl.so.2
Reading symbols from /usr/lib64/libz.so.1...(no debugging symbols
found)...done.
Loaded symbols for /usr/lib64/libz.so.1
Reading symbols from /lib64/libc.so.6...(no debugging symbols
found)...done.
Loaded symbols for /lib64/libc.so.6
Reading symbols from /lib64/ld-linux-x86-64.so.2...(no debugging symbols
found)...done.
Loaded symbols for /lib64/ld-linux-x86-64.so.2
Reading symbols from /usr/lib64/libgssapi_krb5.so.2...(no debugging
symbols found)...done.
Loaded symbols for /usr/lib64/libgssapi_krb5.so.2
Reading symbols from /usr/lib64/libkrb5.so.3...(no debugging symbols
found)...done.
Loaded symbols for /usr/lib64/libkrb5.so.3
Reading symbols from /lib64/libcom_err.so.2...(no debugging symbols
found)...done.
Loaded symbols for /lib64/libcom_err.so.2
Reading symbols from /usr/lib64/libk5crypto.so.3...(no debugging symbols
found)...done.
Loaded symbols for /usr/lib64/libk5crypto.so.3
Reading symbols from /usr/lib64/libkrb5support.so.0...(no debugging
symbols found)...done.
Loaded symbols for /usr/lib64/libkrb5support.so.0
Reading symbols from /lib64/libkeyutils.so.1...(no debugging symbols
found)...done.
Loaded symbols for /lib64/libkeyutils.so.1
Reading symbols from /lib64/libresolv.so.2...(no debugging symbols
found)...done.
Loaded symbols for /lib64/libresolv.so.2
Reading symbols from /lib64/libselinux.so.1...(no debugging symbols
found)...done.
Loaded symbols for /lib64/libselinux.so.1
Reading symbols from /lib64/libsepol.so.1...(no debugging symbols
found)...done.
Loaded symbols for /lib64/libsepol.so.1
Reading symbols from /lib64/libnss_files.so.2...(no debugging symbols
found)...done.
Loaded symbols for /lib64/libnss_files.so.2
ngx_http_upstream_get_peer (pc=0x54ae260, data=<value optimized out>) at
src/http/ngx_http_upstream_round_robin.c:632
632             if (reset++) {
(gdb) bt
#0  ngx_http_upstream_get_peer (pc=0x54ae260, data=<value optimized
out>) at src/http/ngx_http_upstream_round_robin.c:632
#1  ngx_http_upstream_get_round_robin_peer (pc=0x54ae260, data=<value
optimized out>) at src/http/ngx_http_upstream_round_robin.c:425
#2  0x000000000041a99c in ngx_event_connect_peer (pc=0x54ae960) at
src/event/ngx_event_connect.c:24
#3  0x000000000043d5a8 in ngx_http_upstream_connect (r=0x54c3b30,
u=0x54ae250) at src/http/ngx_http_upstream.c:1103
#4  0x000000000043ee0a in ngx_http_upstream_init_request (r=0x54c3b30)
at src/http/ngx_http_upstream.c:631
#5  0x00000000004354a5 in ngx_http_read_client_request_body
(r=0x54c3b30, post_handler=0x43f310 <ngx_http_upstream_init>)
    at src/http/ngx_http_request_body.c:154
#6  0x00000000004572d6 in ngx_http_proxy_handler (r=0x54c3b30) at
src/http/modules/ngx_http_proxy_module.c:617
#7  0x000000000042b47c in ngx_http_core_content_phase (r=0x54c3b30,
ph=0x583ffd8) at src/http/ngx_http_core_module.c:1365
#8  0x0000000000426967 in ngx_http_core_run_phases (r=0x54c3b30) at
src/http/ngx_http_core_module.c:861
#9  0x000000000042fa66 in ngx_http_process_request (r=0x54c3b30) at
src/http/ngx_http_request.c:1665
#10 0x00000000004306a4 in ngx_http_process_request_line (rev=0x5843fc0)
at src/http/ngx_http_request.c:911
#11 0x0000000000419e86 in ngx_event_process_posted (cycle=<value
optimized out>, posted=0x68bd88) at src/event/ngx_event_posted.c:39
#12 0x000000000041f608 in ngx_worker_process_cycle (cycle=0x560f670,
data=<value optimized out>) at src/os/unix/ngx_process_cycle.c:801
#13 0x000000000041dd69 in ngx_spawn_process (cycle=0x560f670,
proc=0x41f550 <ngx_worker_process_cycle>, data=0x0, name=0x466581
"worker process", 
    respawn=-4) at src/os/unix/ngx_process.c:196
#14 0x000000000041ebeb in ngx_start_worker_processes (cycle=0x560f670,
n=8, type=-4) at src/os/unix/ngx_process_cycle.c:360
#15 0x000000000041ff88 in ngx_master_process_cycle (cycle=0x560f670) at
src/os/unix/ngx_process_cycle.c:249
#16 0x00000000004060d9 in main (argc=1, argv=<value optimized out>) at
src/core/nginx.c:405
(gdb) n
425                     rrp->current =
ngx_http_upstream_get_peer(rrp->peers);
(gdb) n
435                     if (!(rrp->tried[n] & m)) {
(gdb) n
460                     if (pc->tries == 0) {
(gdb) n
464                     if (--i == 0) {
(gdb) n
425                     rrp->current =
ngx_http_upstream_get_peer(rrp->peers);
(gdb) n
435                     if (!(rrp->tried[n] & m)) {
(gdb) n
460                     if (pc->tries == 0) {
(gdb) n
464                     if (--i == 0) {
(gdb) n
425                     rrp->current =
ngx_http_upstream_get_peer(rrp->peers);
(gdb) n
435                     if (!(rrp->tried[n] & m)) {
(gdb) n
460                     if (pc->tries == 0) {
(gdb) n
464                     if (--i == 0) {
(gdb) 





I found i made a mistake in nginx.conf, i include a virtual host
configuation like this:
        upstream test_servers {
                #server 10.0.7.4:80 ;
                server 10.0.7.5:80 backup;
                #server 10.0.7.6:80 ;
                #server 10.0.7.7:80 ;
        }

    server {
        listen       80;
        server_name  test.org ;

        access_log  /data1/logs/$host.access.log  main;

        location / {
                include proxy.conf;
                proxy_pass    http://test_servers;
        }


    }

there was only one server in upstream,which marked 'backup'. after some
test,i found this is the reason.

but when i test the nginx.conf syntax by using nginx -t,the result is
ok.

[root at host-22 ~]# /usr/local/nginx/sbin/nginx -t
nginx: the configuration file /usr/local/nginx/conf/nginx.conf syntax is
ok
nginx: configuration file /usr/local/nginx/conf/nginx.conf test is
successful
[root at host-22 ~]#

i think nginx should warn me when that situation,haha...

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218259,218315#msg-218315


From igor at sysoev.ru  Mon Nov 14 15:44:03 2011
From: igor at sysoev.ru (Igor Sysoev)
Date: Mon, 14 Nov 2011 19:44:03 +0400
Subject: nginx-1.1.8
Message-ID: <20111114154403.GF33940@nginx.com>

Changes with nginx 1.1.8                                         14 Nov 2011

    *) Change: the ngx_http_limit_zone_module was renamed to the
       ngx_http_limit_conn_module.

    *) Change: the "limit_zone" directive was superseded by the
       "limit_conn_zone" directive with a new syntax.

    *) Feature: support for multiple "limit_conn" limits on the same level.

    *) Feature: the "image_filter_sharpen" directive.

    *) Bugfix: a segmentation fault might occur in a worker process if
       resolver got a big DNS response.
       Thanks to Ben Hawkes.

    *) Bugfix: in cache key calculation if internal MD5 implementation was
       used; the bug had appeared in 1.0.4.

    *) Bugfix: the "If-Modified-Since", "If-Range", etc. client request
       header lines might be passed to backend while caching; or not passed
       without caching if caching was enabled in another part of the
       configuration.

    *) Bugfix: the module ngx_http_mp4_module sent incorrect
       "Content-Length" response header line if the "start" argument was
       used.
       Thanks to Piotr Sikora.


-- 
Igor Sysoev


From mdounin at mdounin.ru  Mon Nov 14 15:58:27 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Mon, 14 Nov 2011 19:58:27 +0400
Subject: nginx worker process hang,cpu load 100%
In-Reply-To: <7ed1abd36d44e812afca391274f9e925.NginxMailingListEnglish@forum.nginx.org>
References: <fe1c35dd759d61202e61b784ce4e5825.NginxMailingListEnglish@forum.nginx.org>
	<7ed1abd36d44e812afca391274f9e925.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20111114155827.GR95664@mdounin.ru>

Hello!

On Mon, Nov 14, 2011 at 09:37:29AM -0500, Long Wan wrote:

[...]

> I found i made a mistake in nginx.conf, i include a virtual host
> configuation like this:
>         upstream test_servers {
>                 #server 10.0.7.4:80 ;
>                 server 10.0.7.5:80 backup;
>                 #server 10.0.7.6:80 ;
>                 #server 10.0.7.7:80 ;
>         }

[...]

> there was only one server in upstream,which marked 'backup'. after some
> test,i found this is the reason.

Yes, thank you for report.  This is somewhat known issue, 'backup' 
handling needs attention.

Maxim Dounin


From nginx-forum at nginx.us  Tue Nov 15 01:56:21 2011
From: nginx-forum at nginx.us (Long Wan)
Date: Mon, 14 Nov 2011 20:56:21 -0500
Subject: nginx worker process hang,cpu load 100%
In-Reply-To: <fe1c35dd759d61202e61b784ce4e5825.NginxMailingListEnglish@forum.nginx.org>
References: <fe1c35dd759d61202e61b784ce4e5825.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <800ad31679e45e06363a87a924be77e6.NginxMailingListEnglish@forum.nginx.org>

Do  you plan to fix this issue in next release?

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218259,218342#msg-218342


From nginx-forum at nginx.us  Tue Nov 15 02:30:27 2011
From: nginx-forum at nginx.us (dannynoonan)
Date: Mon, 14 Nov 2011 21:30:27 -0500
Subject: compile ngx_resty to statically link some libs?
Message-ID: <c67cbde241dd2d8b55c8c3ef13905489.NginxMailingListEnglish@forum.nginx.org>

Hey agentzh, could you provide tips on how I could cut down this list of
dynamically linked libs:

[david at dev-3 ngx_openresty-1.0.8.26]$ ldd
/usr/local/encap/nginx-resty-1.0.8.26/nginx/sbin/nginx 
	linux-vdso.so.1 =>  (0x00007fff079fd000)
	libpthread.so.0 => /lib64/libpthread.so.0 (0x0000003aac200000)
	libcrypt.so.1 => /lib64/libcrypt.so.1 (0x0000003e08a00000)
	libssl.so.10 => /usr/lib64/libssl.so.10 (0x00000033d9c00000)
	libdrizzle.so.0 => /usr/lib64/libdrizzle.so.0 (0x00007fd4bcdf1000)
	libm.so.6 => /lib64/libm.so.6 (0x0000003e12000000)
	libpcre.so.0 => /lib64/libpcre.so.0 (0x0000003710800000)
	libcrypto.so.10 => /lib64/libcrypto.so.10 (0x0000003e13400000)
	libdl.so.2 => /lib64/libdl.so.2 (0x0000003e11000000)
	libz.so.1 => /lib64/libz.so.1 (0x0000003e11c00000)
	libc.so.6 => /lib64/libc.so.6 (0x0000003e10c00000)
	/lib64/ld-linux-x86-64.so.2 (0x0000003e10800000)
	libfreebl3.so => /lib64/libfreebl3.so (0x0000003e08e00000)
	libgssapi_krb5.so.2 => /lib64/libgssapi_krb5.so.2 (0x00000033d9400000)
	libkrb5.so.3 => /lib64/libkrb5.so.3 (0x00000033d9800000)
	libcom_err.so.2 => /lib64/libcom_err.so.2 (0x00000033d9000000)
	libk5crypto.so.3 => /lib64/libk5crypto.so.3 (0x000000374a800000)
	libkrb5support.so.0 => /lib64/libkrb5support.so.0 (0x000000374a000000)
	libkeyutils.so.1 => /lib64/libkeyutils.so.1 (0x0000003e13800000)
	libresolv.so.2 => /lib64/libresolv.so.2 (0x0000003e12c00000)
	libselinux.so.1 => /lib64/libselinux.so.1 (0x0000003749c00000)


Specifically, I'd really like to remove libdrizzle as a dependency on
the target machine and just have the compile pull it in at link time.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218343,218343#msg-218343


From nbubingo at gmail.com  Tue Nov 15 02:47:53 2011
From: nbubingo at gmail.com (=?GB2312?B?0qbOsLHz?=)
Date: Tue, 15 Nov 2011 10:47:53 +0800
Subject: Nginx Load balancer mode for JBoss / Icefaces application
In-Reply-To: <SNT144-W248F760B63BF7E1E138037E7C00@phx.gbl>
References: <SNT144-W248F760B63BF7E1E138037E7C00@phx.gbl>
Message-ID: <CAGxuLK1C-3+yFGs+BxOG4BR0bn3P=HJwFwNnv0kfrt9ZdVb2vw@mail.gmail.com>

?However the session information is not being passed through or is being
modified.?

How do you know that? I think the problem is the session is sent to be the
wrong jboss server.

If the Jboss uses the similar session sticky way as Tomcat/Resin, you can
use my nginx_jvm_route_module:
http://code.google.com/p/nginx-upstream-jvm-route/

2011/11/14 Srinivasan Subramanian <ssrini_vasan at hotmail.com>

>
>
> Hello
>
> We have setup nginx as a Loadbalancer on Centos 5.2 x64.  nginx is acting
> as a LB for a web application developed using Java servlets and Icefaces
> (1.8.2).  The web application is deployed on JBoss 5.1.   After
> configuration nginx is able to redirect the queries to the upstream servers
> properly.  However the session information is not being passed through or
> is being modified.  So the Icefaces servlet is repeatedly refreshing the
> login page every few seconds and keeps creating new sessions.
>
> Please advise on any additional settings that need to be made.  The
> current settings are:
>
> (nginx is running on 192.168.1.137)
>
> upstream int-lb {
>         server 192.168.1.139:8080;
>         server 192.168.1.138:8080;
> }
>
> server {
>     listen       80;
>     server_name  int-lb;
>
>     #charset koi8-r;
>     access_log  /var/log/nginx/host.access.log  main;
>     error_log  /var/log/nginx/host.error.log  debug;
>     root /usr/app/jboss5/server/default/deploy/admin-console.war;
>
>     location / {
>         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_pass http://int-lb;
>     }
> }
>
> Thanks in advance for all assistance.
>
> Regards
> Srini
>
> _______________________________________________
> 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/20111115/dcfd3675/attachment-0001.html>

From ssrini_vasan at hotmail.com  Tue Nov 15 03:18:36 2011
From: ssrini_vasan at hotmail.com (Srinivasan Subramanian)
Date: Tue, 15 Nov 2011 08:48:36 +0530
Subject: Nginx Load balancer mode for JBoss / Icefaces application
In-Reply-To: <CAGxuLK1C-3+yFGs+BxOG4BR0bn3P=HJwFwNnv0kfrt9ZdVb2vw@mail.gmail.com>
References: <SNT144-W248F760B63BF7E1E138037E7C00@phx.gbl>,
	<CAGxuLK1C-3+yFGs+BxOG4BR0bn3P=HJwFwNnv0kfrt9ZdVb2vw@mail.gmail.com>
Message-ID: <SNT144-W494E2C963B3DB17FB151BFE7C10@phx.gbl>


Hi
You were bang on!  Thanks.  That was the issue.  For now i introduced the ip_hask; key for now in the upstream section and that fixed it.  
I will try your module.  We want to actually also implement the HttpUpstreamFairModule to improve the distribution.  Will the sticky session module that you linked to work in conjunction with that?
Thanks for your timely help, greatly appreciated.
Regards




Date: Tue, 15 Nov 2011 10:47:53 +0800
Subject: Re: Nginx Load balancer mode for JBoss / Icefaces application
From: nbubingo at gmail.com
To: nginx at nginx.org

?However the session information is not being passed through or is being modified.?

How do you know that? I think the problem is the session is sent to be the wrong jboss server.
If the Jboss uses the similar session sticky way as Tomcat/Resin, you can use my nginx_jvm_route_module: http://code.google.com/p/nginx-upstream-jvm-route/  

2011/11/14 Srinivasan Subramanian <ssrini_vasan at hotmail.com>








Hello
We have setup nginx as a Loadbalancer on Centos 5.2 x64.  nginx is acting as a LB for a web application developed using Java servlets and Icefaces (1.8.2).  The web application is deployed on JBoss 5.1.   After configuration nginx is able to redirect the queries to the upstream servers properly.  However the session information is not being passed through or is being modified.  So the Icefaces servlet is repeatedly refreshing the login page every few seconds and keeps creating new sessions.

Please advise on any additional settings that need to be made.  The current settings are:

(nginx is running on 192.168.1.137)
upstream int-lb {
        server 192.168.1.139:8080;        server 192.168.1.138:8080;
}
server {    listen       80;
    server_name  int-lb;
    #charset koi8-r;    access_log  /var/log/nginx/host.access.log  main;
    error_log  /var/log/nginx/host.error.log  debug;    root /usr/app/jboss5/server/default/deploy/admin-console.war;

    location / {        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_pass http://int-lb;
    }}
Thanks in advance for all assistance.


Regards
Srini 		 	   		  

_______________________________________________

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/20111115/ebb49c38/attachment.html>

From nginx-forum at nginx.us  Tue Nov 15 03:36:08 2011
From: nginx-forum at nginx.us (archon810)
Date: Mon, 14 Nov 2011 22:36:08 -0500
Subject: Bug while using "proxy_cache_use_stale updating"
In-Reply-To: <a5f9ac61da7ec7f5a24d8b07f08a9f15.NginxMailingListEnglish@forum.nginx.org>
References: <20090901132100.GG98063@rambler-co.ru>
	<a5f9ac61da7ec7f5a24d8b07f08a9f15.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <da2d3cfaf90618fa31b1976975bdf5c5.NginxMailingListEnglish@forum.nginx.org>

I have 0.8.54, and I just hit this issue, which resulted in the RSS feed
not updating for a full day. Due to the way my server is configured,
because of my cookies, I was getting straight to the feed bypassing the
cache and only found the problem when using incognito mode without
cookies. A server restart seemed to fix the problem for now.

Igor, is this fixed in some version >0.8.54?

Thanks,
Artem

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,5225,218347#msg-218347


From nginx-forum at nginx.us  Tue Nov 15 03:56:50 2011
From: nginx-forum at nginx.us (fengguang)
Date: Mon, 14 Nov 2011 22:56:50 -0500
Subject: Green screen with New MP4 module
Message-ID: <c0f36c24edeee98ad9bb4c077b1c518b.NginxMailingListEnglish@forum.nginx.org>

When pseudo-streaming with new mp4 module, sometimes the flash player
will display a green screen for a second. Anyone has the same problem?
I'm sure the mp4 file was processed correctly. The metadata was inserted
at the beginning of the file.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218348,218348#msg-218348


From agentzh at gmail.com  Tue Nov 15 04:24:01 2011
From: agentzh at gmail.com (agentzh)
Date: Tue, 15 Nov 2011 12:24:01 +0800
Subject: compile ngx_resty to statically link some libs?
In-Reply-To: <c67cbde241dd2d8b55c8c3ef13905489.NginxMailingListEnglish@forum.nginx.org>
References: <c67cbde241dd2d8b55c8c3ef13905489.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <CAB4Tn6PrgtPoC0+Q-TP+fOZ8f3k3AS0eyAOA2aKS9fAOSngFgQ@mail.gmail.com>

On Tue, Nov 15, 2011 at 10:30 AM, dannynoonan <nginx-forum at nginx.us> wrote:
> Hey agentzh, could you provide tips on how I could cut down this list of
> dynamically linked libs:
>
[snip]
>
> Specifically, I'd really like to remove libdrizzle as a dependency on
> the target machine and just have the compile pull it in at link time.
>

Try

    ./configure --with-ld-opt="-static" ...

while building ngx_openresty :)

Regards,
-agentzh


From nginx-forum at nginx.us  Tue Nov 15 05:37:43 2011
From: nginx-forum at nginx.us (zhenwei)
Date: Tue, 15 Nov 2011 00:37:43 -0500
Subject: what's the difference between proxy_store and proxy_cache?
In-Reply-To: <201111140948.43044.ne@vbart.ru>
References: <201111140948.43044.ne@vbart.ru>
Message-ID: <a84e6981e835c3acf29b9186368c9c70.NginxMailingListEnglish@forum.nginx.org>

thanks for your advise, using pseudo disk via memory or memcached are
good approaches.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218263,218355#msg-218355


From nginx-forum at nginx.us  Tue Nov 15 07:32:10 2011
From: nginx-forum at nginx.us (faskiri.devel)
Date: Tue, 15 Nov 2011 02:32:10 -0500
Subject: Nginx fails to accept new connection if active worker crashes
Message-ID: <9506f66614285b9ecfddda8261d94984.NginxMailingListEnglish@forum.nginx.org>

Hi All

I use nginx configured with multiple workers. I also have an nginx
module that crashed due to an error when I noticed that the module crash
leaves nginx in a state where it cannot accept new calls.

Removing my module and killing the "active" worker (the one which seems
to take the new request) with a SIGHUP again caused nginx to hang.
Killing the other worker(s) seem to be working just fine.

Further investigations(with nginx at debug level) showed that all
threads are fine but none of the workers are getting the
ngx_accept_mutex_lock.

Master tries to release the ngx_accept_mutex_lock if the dead process
was holding it [
 https://svn.nginx.org/nginx/browser/nginx/trunk/src/os/unix/ngx_process.c?annotate=blame#L503]
but doesnt look like the value is set anywhere.

I have been using nginx only for a couple of months now so I am not very
sure of the diagnosis, please feel free to correct.


uname -a:	 Linux faskiri-pc 2.6.32-24-generic #43-Ubuntu SMP Thu Sep 16
14:58:24 UTC 2010 x86_64 GNU/Linux

nginx -V:	 nginx: nginx version: nginx/1.0.5 nginx: built by gcc 4.4.5
(Ubuntu/Linaro 4.4.4-14ubuntu5) nginx: configure arguments:
--without-http_ssi_module --without-http_geo_module
--without-http_fastcgi_module --without-http_uwsgi_module
--without-http_scgi_module --without-http_memcached_module
--without-mail_pop3_module --without-mail_imap_module
--without-mail_smtp_module --with-pcre --with-debug

I will be grateful for any advice.

Best Regards
+Fasih

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218359,218359#msg-218359


From andrew at nginx.com  Tue Nov 15 09:15:50 2011
From: andrew at nginx.com (Andrew Alexeev)
Date: Tue, 15 Nov 2011 13:15:50 +0400
Subject: Nginx fails to accept new connection if active worker crashes
In-Reply-To: <9506f66614285b9ecfddda8261d94984.NginxMailingListEnglish@forum.nginx.org>
References: <9506f66614285b9ecfddda8261d94984.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <CF8CC5EE-CF5D-4039-A9E8-4261EA5106FB@nginx.com>

Fasih,

On Nov 15, 2011, at 11:32 AM, faskiri.devel wrote:

> Hi All
> 
> I use nginx configured with multiple workers. I also have an nginx
> module that crashed due to an error when I noticed that the module crash
> leaves nginx in a state where it cannot accept new calls.
> 
> Removing my module and killing the "active" worker (the one which seems
> to take the new request) with a SIGHUP again caused nginx to hang.
> Killing the other worker(s) seem to be working just fine.
> 
> Further investigations(with nginx at debug level) showed that all
> threads are fine but none of the workers are getting the
> ngx_accept_mutex_lock.
> 
> Master tries to release the ngx_accept_mutex_lock if the dead process
> was holding it [
> https://svn.nginx.org/nginx/browser/nginx/trunk/src/os/unix/ngx_process.c?annotate=blame#L503]
> but doesnt look like the value is set anywhere.
> 
> I have been using nginx only for a couple of months now so I am not very
> sure of the diagnosis, please feel free to correct.

Thanks for spotting this one. It's kind of a known issue and we're working on a fix currently. In the meanwhile you can switch accept mutex off as a workaround (the only downside could potentially be in minor increase of CPU utilization).

> uname -a:	 Linux faskiri-pc 2.6.32-24-generic #43-Ubuntu SMP Thu Sep 16
> 14:58:24 UTC 2010 x86_64 GNU/Linux
> 
> nginx -V:	 nginx: nginx version: nginx/1.0.5 nginx: built by gcc 4.4.5
> (Ubuntu/Linaro 4.4.4-14ubuntu5) nginx: configure arguments:
> --without-http_ssi_module --without-http_geo_module
> --without-http_fastcgi_module --without-http_uwsgi_module
> --without-http_scgi_module --without-http_memcached_module
> --without-mail_pop3_module --without-mail_imap_module
> --without-mail_smtp_module --with-pcre --with-debug
> 
> I will be grateful for any advice.
> 
> Best Regards
> +Fasih
> 
> Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218359,218359#msg-218359
> 
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx
> 


From igor at sysoev.ru  Tue Nov 15 09:42:57 2011
From: igor at sysoev.ru (Igor Sysoev)
Date: Tue, 15 Nov 2011 13:42:57 +0400
Subject: nginx-1.0.10
Message-ID: <20111115094257.GB58136@nginx.com>

Changes with nginx 1.0.10                                        15 Nov 2011

    *) Bugfix: a segmentation fault might occur in a worker process if
       resolver got a big DNS response.
       Thanks to Ben Hawkes.

    *) Bugfix: in cache key calculation if internal MD5 implementation was
       used; the bug had appeared in 1.0.4.

    *) Bugfix: the module ngx_http_mp4_module sent incorrect
       "Content-Length" response header line if the "start" argument was
       used.
       Thanks to Piotr Sikora.


-- 
Igor Sysoev


From andrew at nginx.com  Tue Nov 15 09:50:07 2011
From: andrew at nginx.com (Andrew Alexeev)
Date: Tue, 15 Nov 2011 13:50:07 +0400
Subject: DNS TTLs being ignored
In-Reply-To: <AB401AA3-211C-4401-B050-712D7CBBB054@nginx.com>
References: <CAAqHd6v_LxnOpfoE6Fq+sKebtwHiguse4d6GmmvpAUayR5eqaA@mail.gmail.com>
	<d7ffe70b7f12b278cb9ffeac1a02c358@ruby-forum.com>
	<AB401AA3-211C-4401-B050-712D7CBBB054@nginx.com>
Message-ID: <72FF6524-75CF-4123-8F83-50363C25AE21@nginx.com>

On Nov 3, 2011, at 1:50 PM, Andrew Alexeev wrote:

> Noah,
> 
> This fix/improvement be introduced in 1.1.8 which will come out around Nov 14.

Apologies, it didn't get in either 1.1.8 (yesterday) or 1.1.10 (today). It's almost ready and would hopefully get into the next dev and stable releases in a couple of weeks.

> 
> Hope this helps
> 
> On Nov 3, 2011, at 1:46 PM, Noah C. wrote:
> 
>> Thanks for the reply Andrew. Do you have any idea when it's likely to be 
>> generally available? This is a pretty big nuisance for us, and I'd like 
>> to be able to figure out if I need to look at using a new reverse proxy, 
>> at least for the time being.
>> 
>> --Noah
>> 
>> -- 
>> Posted via http://www.ruby-forum.com/.
>> 
>> _______________________________________________
>> nginx mailing list
>> nginx at nginx.org
>> http://mailman.nginx.org/mailman/listinfo/nginx


From nginx-forum at nginx.us  Tue Nov 15 10:35:16 2011
From: nginx-forum at nginx.us (dannynoonan)
Date: Tue, 15 Nov 2011 05:35:16 -0500
Subject: compile ngx_resty to statically link some libs?
In-Reply-To: <CAB4Tn6PrgtPoC0+Q-TP+fOZ8f3k3AS0eyAOA2aKS9fAOSngFgQ@mail.gmail.com>
References: <CAB4Tn6PrgtPoC0+Q-TP+fOZ8f3k3AS0eyAOA2aKS9fAOSngFgQ@mail.gmail.com>
Message-ID: <15e3a05026fcf57706fc601b2b7bb4d1.NginxMailingListEnglish@forum.nginx.org>

I'd tried that, but figured I was invoking it wrong, the error it
yields:

<snip>
checking for --with-ld-opt="-static" ... not found
./configure: error: the invalid value in --with-ld-opt="-static"

failed to run command: ./configure
--prefix=/data/local/nginx-resty-1.0.8.26/nginx \

</snip>

Reading up on the man page for ld caused me to try
--with-ld-opt="-Bstatic" which gets me past configure, but gmake still
yields an ELF binary that dynamically links out to libdrizzle. The final
gcc linking step looked like this:

gcc -o objs/nginx \
	objs/src/core/nginx.o \
	objs/src/core/ngx_log.o \
	objs/src/core/ngx_palloc.o \
	objs/src/core/ngx_array.o \
	objs/src/core/ngx_list.o \
	objs/src/core/ngx_hash.o \
	objs/src/core/ngx_buf.o \
	objs/src/core/ngx_queue.o \
	objs/src/core/ngx_output_chain.o \
	objs/src/core/ngx_string.o \
	objs/src/core/ngx_parse.o \
	objs/src/core/ngx_inet.o \
	objs/src/core/ngx_file.o \
	objs/src/core/ngx_crc32.o \
	objs/src/core/ngx_murmurhash.o \
	objs/src/core/ngx_md5.o \
	objs/src/core/ngx_rbtree.o \
	objs/src/core/ngx_radix_tree.o \
	objs/src/core/ngx_slab.o \
	objs/src/core/ngx_times.o \
	objs/src/core/ngx_shmtx.o \
	objs/src/core/ngx_connection.o \
	objs/src/core/ngx_cycle.o \
	objs/src/core/ngx_spinlock.o \
	objs/src/core/ngx_cpuinfo.o \
	objs/src/core/ngx_conf_file.o \
	objs/src/core/ngx_resolver.o \
	objs/src/core/ngx_open_file_cache.o \
	objs/src/core/ngx_crypt.o \
	objs/src/event/ngx_event.o \
	objs/src/event/ngx_event_timer.o \
	objs/src/event/ngx_event_posted.o \
	objs/src/event/ngx_event_busy_lock.o \
	objs/src/event/ngx_event_accept.o \
	objs/src/event/ngx_event_connect.o \
	objs/src/event/ngx_event_pipe.o \
	objs/src/os/unix/ngx_time.o \
	objs/src/os/unix/ngx_errno.o \
	objs/src/os/unix/ngx_alloc.o \
	objs/src/os/unix/ngx_files.o \
	objs/src/os/unix/ngx_socket.o \
	objs/src/os/unix/ngx_recv.o \
	objs/src/os/unix/ngx_readv_chain.o \
	objs/src/os/unix/ngx_udp_recv.o \
	objs/src/os/unix/ngx_send.o \
	objs/src/os/unix/ngx_writev_chain.o \
	objs/src/os/unix/ngx_channel.o \
	objs/src/os/unix/ngx_shmem.o \
	objs/src/os/unix/ngx_process.o \
	objs/src/os/unix/ngx_daemon.o \
	objs/src/os/unix/ngx_setproctitle.o \
	objs/src/os/unix/ngx_posix_init.o \
	objs/src/os/unix/ngx_user.o \
	objs/src/os/unix/ngx_process_cycle.o \
	objs/src/os/unix/ngx_linux_init.o \
	objs/src/event/modules/ngx_epoll_module.o \
	objs/src/os/unix/ngx_linux_sendfile_chain.o \
	objs/src/event/ngx_event_openssl.o \
	objs/src/core/ngx_regex.o \
	objs/src/http/ngx_http.o \
	objs/src/http/ngx_http_core_module.o \
	objs/src/http/ngx_http_special_response.o \
	objs/src/http/ngx_http_request.o \
	objs/src/http/ngx_http_parse.o \
	objs/src/http/ngx_http_header_filter_module.o \
	objs/src/http/ngx_http_write_filter_module.o \
	objs/src/http/ngx_http_copy_filter_module.o \
	objs/src/http/modules/ngx_http_log_module.o \
	objs/src/http/ngx_http_request_body.o \
	objs/src/http/ngx_http_variables.o \
	objs/src/http/ngx_http_script.o \
	objs/src/http/ngx_http_upstream.o \
	objs/src/http/ngx_http_upstream_round_robin.o \
	objs/src/http/ngx_http_parse_time.o \
	objs/src/http/modules/ngx_http_static_module.o \
	objs/src/http/modules/ngx_http_index_module.o \
	objs/src/http/modules/ngx_http_chunked_filter_module.o \
	objs/src/http/modules/ngx_http_range_filter_module.o \
	objs/src/http/modules/ngx_http_headers_filter_module.o \
	objs/src/http/modules/ngx_http_not_modified_filter_module.o \
	objs/src/http/ngx_http_busy_lock.o \
	objs/src/http/ngx_http_file_cache.o \
	objs/src/http/modules/ngx_http_gzip_filter_module.o \
	objs/src/http/ngx_http_postpone_filter_module.o \
	objs/src/http/modules/ngx_http_ssi_filter_module.o \
	objs/src/http/modules/ngx_http_charset_filter_module.o \
	objs/src/http/modules/ngx_http_userid_filter_module.o \
	objs/src/http/modules/ngx_http_autoindex_module.o \
	objs/src/http/modules/ngx_http_auth_basic_module.o \
	objs/src/http/modules/ngx_http_access_module.o \
	objs/src/http/modules/ngx_http_limit_zone_module.o \
	objs/src/http/modules/ngx_http_limit_req_module.o \
	objs/src/http/modules/ngx_http_geo_module.o \
	objs/src/http/modules/ngx_http_map_module.o \
	objs/src/http/modules/ngx_http_split_clients_module.o \
	objs/src/http/modules/ngx_http_referer_module.o \
	objs/src/http/modules/ngx_http_rewrite_module.o \
	objs/src/http/modules/ngx_http_ssl_module.o \
	objs/src/http/modules/ngx_http_proxy_module.o \
	objs/src/http/modules/ngx_http_fastcgi_module.o \
	objs/src/http/modules/ngx_http_uwsgi_module.o \
	objs/src/http/modules/ngx_http_scgi_module.o \
	objs/src/http/modules/ngx_http_memcached_module.o \
	objs/src/http/modules/ngx_http_empty_gif_module.o \
	objs/src/http/modules/ngx_http_browser_module.o \
	objs/src/http/modules/ngx_http_upstream_ip_hash_module.o \
	objs/addon/src/ndk.o \
	objs/addon/src/ngx_http_echo_module.o \
	objs/addon/src/ngx_http_echo_util.o \
	objs/addon/src/ngx_http_echo_timer.o \
	objs/addon/src/ngx_http_echo_var.o \
	objs/addon/src/ngx_http_echo_handler.o \
	objs/addon/src/ngx_http_echo_filter.o \
	objs/addon/src/ngx_http_echo_sleep.o \
	objs/addon/src/ngx_http_echo_location.o \
	objs/addon/src/ngx_http_echo_echo.o \
	objs/addon/src/ngx_http_echo_request_info.o \
	objs/addon/src/ngx_http_echo_subrequest.o \
	objs/addon/src/ngx_http_echo_foreach.o \
	objs/addon/src/ngx_http_xss_filter_module.o \
	objs/addon/src/ngx_http_xss_util.o \
	objs/addon/src/ngx_http_set_base32.o \
	objs/addon/src/ngx_http_set_default_value.o \
	objs/addon/src/ngx_http_set_hashed_upstream.o \
	objs/addon/src/ngx_http_set_quote_sql.o \
	objs/addon/src/ngx_http_set_quote_json.o \
	objs/addon/src/ngx_http_set_unescape_uri.o \
	objs/addon/src/ngx_http_set_misc_module.o \
	objs/addon/src/ngx_http_set_escape_uri.o \
	objs/addon/src/ngx_http_set_hash.o \
	objs/addon/src/ngx_http_set_local_today.o \
	objs/addon/src/ngx_http_set_hex.o \
	objs/addon/src/ngx_http_set_base64.o \
	objs/addon/src/ngx_http_set_random.o \
	objs/addon/src/ngx_http_set_hmac.o \
	objs/addon/src/ngx_http_form_input_module.o \
	objs/addon/src/ngx_http_encrypted_session_module.o \
	objs/addon/src/ngx_http_encrypted_session_cipher.o \
	objs/addon/src/ngx_http_drizzle_module.o \
	objs/addon/src/ngx_http_drizzle_handler.o \
	objs/addon/src/ngx_http_drizzle_processor.o \
	objs/addon/src/ngx_http_drizzle_upstream.o \
	objs/addon/src/ngx_http_drizzle_util.o \
	objs/addon/src/ngx_http_drizzle_output.o \
	objs/addon/src/ngx_http_drizzle_keepalive.o \
	objs/addon/src/ngx_http_drizzle_quoting.o \
	objs/addon/src/ngx_http_drizzle_checker.o \
	objs/addon/src/ngx_http_lua_script.o \
	objs/addon/src/ngx_http_lua_log.o \
	objs/addon/src/ngx_http_lua_subrequest.o \
	objs/addon/src/ngx_http_lua_ndk.o \
	objs/addon/src/ngx_http_lua_control.o \
	objs/addon/src/ngx_http_lua_time.o \
	objs/addon/src/ngx_http_lua_misc.o \
	objs/addon/src/ngx_http_lua_variable.o \
	objs/addon/src/ngx_http_lua_string.o \
	objs/addon/src/ngx_http_lua_output.o \
	objs/addon/src/ngx_http_lua_headers.o \
	objs/addon/src/ngx_http_lua_req_body.o \
	objs/addon/src/ngx_http_lua_uri.o \
	objs/addon/src/ngx_http_lua_args.o \
	objs/addon/src/ngx_http_lua_ctx.o \
	objs/addon/src/ngx_http_lua_regex.o \
	objs/addon/src/ngx_http_lua_module.o \
	objs/addon/src/ngx_http_lua_headers_out.o \
	objs/addon/src/ngx_http_lua_headers_in.o \
	objs/addon/src/ngx_http_lua_directive.o \
	objs/addon/src/ngx_http_lua_consts.o \
	objs/addon/src/ngx_http_lua_exception.o \
	objs/addon/src/ngx_http_lua_util.o \
	objs/addon/src/ngx_http_lua_cache.o \
	objs/addon/src/ngx_http_lua_conf.o \
	objs/addon/src/ngx_http_lua_contentby.o \
	objs/addon/src/ngx_http_lua_rewriteby.o \
	objs/addon/src/ngx_http_lua_accessby.o \
	objs/addon/src/ngx_http_lua_setby.o \
	objs/addon/src/ngx_http_lua_capturefilter.o \
	objs/addon/src/ngx_http_lua_clfactory.o \
	objs/addon/src/ngx_http_lua_pcrefix.o \
	objs/addon/src/ngx_http_lua_headerfilterby.o \
	objs/addon/src/ngx_http_lua_shdict.o \
	objs/addon/src/ngx_http_headers_more_filter_module.o \
	objs/addon/src/ngx_http_headers_more_headers_out.o \
	objs/addon/src/ngx_http_headers_more_headers_in.o \
	objs/addon/src/ngx_http_headers_more_util.o \
	objs/addon/src/ngx_http_srcache_filter_module.o \
	objs/addon/src/ngx_http_srcache_util.o \
	objs/addon/src/ngx_http_srcache_var.o \
	objs/addon/src/ngx_http_srcache_store.o \
	objs/addon/src/ngx_http_srcache_fetch.o \
	objs/addon/src/ngx_http_array_var_module.o \
	objs/addon/src/ngx_http_array_var_util.o \
	objs/addon/src/ngx_http_memc_module.o \
	objs/addon/src/ngx_http_memc_request.o \
	objs/addon/src/ngx_http_memc_response.o \
	objs/addon/src/ngx_http_memc_util.o \
	objs/addon/src/ngx_http_memc_handler.o \
	objs/addon/src/ngx_http_redis2_module.o \
	objs/addon/src/ngx_http_redis2_handler.o \
	objs/addon/src/ngx_http_redis2_reply.o \
	objs/addon/src/ngx_http_redis2_util.o \
	objs/addon/upstream-keepalive-nginx-module-0.3/ngx_http_upstream_keepalive_module.o
\
	objs/addon/auth-request-nginx-module-0.2/ngx_http_auth_request_module.o
\
	objs/addon/src/ngx_http_rds_json_filter_module.o \
	objs/addon/src/ngx_http_rds_json_processor.o \
	objs/addon/src/ngx_http_rds_json_util.o \
	objs/addon/src/ngx_http_rds_json_output.o \
	objs/addon/src/ngx_http_rds_json_handler.o \
	objs/addon/src/ngx_http_rds_csv_filter_module.o \
	objs/addon/src/ngx_http_rds_csv_processor.o \
	objs/addon/src/ngx_http_rds_csv_util.o \
	objs/addon/src/ngx_http_rds_csv_output.o \
	objs/ngx_modules.o \
	-Bstatic -Wl,-E -lpthread -lcrypt -lssl -ldrizzle
-L/home/david/src/third-party/ngx_openresty-1.0.8.26/build/lua-root/data/local/nginx-resty-1.0.8.26/lua/lib
-llua -lm -lpcre -lssl -lcrypto -ldl -lz

Here's the full configure line:
 ./configure  --prefix=/data/local/nginx-resty-loko-1.0.8.26  
--with-ld-opt="-Bstatic" --with-http_drizzle_module  

Any ideas? Is trying to statically link a never ending struggle I should
give up on early?

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218343,218372#msg-218372


From nginx-forum at nginx.us  Tue Nov 15 11:37:22 2011
From: nginx-forum at nginx.us (faskiri.devel)
Date: Tue, 15 Nov 2011 06:37:22 -0500
Subject: Nginx fails to accept new connection if active worker crashes
In-Reply-To: <CF8CC5EE-CF5D-4039-A9E8-4261EA5106FB@nginx.com>
References: <CF8CC5EE-CF5D-4039-A9E8-4261EA5106FB@nginx.com>
Message-ID: <2cd0f23b89b14d39b23cd6f97858c8a6.NginxMailingListEnglish@forum.nginx.org>

Hi Andrew

Thanks for the prompt reply.

As a temporary fix I had created a variable in the shared memory to
track which pid is holding the mutex so that the check in
[https://svn.nginx.org/nginx/browser/nginx/trunk/src/os/unix/ngx_process.c?annotate=blame#L503]
works. It works fine for me, hadnt realized I could switch it off. Are
we talking about putting "accept_mutex off" in the nginx.conf file? It
will be great if you could explain what that will actually do, as in,
how will the new requests be handed off to workers.

Best Regards
+Fasih

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218359,218373#msg-218373


From bruno.premont at restena.lu  Tue Nov 15 12:29:04 2011
From: bruno.premont at restena.lu (Bruno =?UTF-8?B?UHLDqW1vbnQ=?=)
Date: Tue, 15 Nov 2011 13:29:04 +0100
Subject: nginx-1.0.10
In-Reply-To: <20111115094257.GB58136@nginx.com>
References: <20111115094257.GB58136@nginx.com>
Message-ID: <20111115132904.2c1c47d7@pluto.restena.lu>

Hi Igor,

It would be nice if you could include tarball checksums in the release
announcements!

Bruno


On Tue, 15 Nov 2011 13:42:57 Igor Sysoev wrote:
> Changes with nginx 1.0.10                                        15 Nov 2011
> 
>     *) Bugfix: a segmentation fault might occur in a worker process if
>        resolver got a big DNS response.
>        Thanks to Ben Hawkes.
> 
>     *) Bugfix: in cache key calculation if internal MD5 implementation was
>        used; the bug had appeared in 1.0.4.
> 
>     *) Bugfix: the module ngx_http_mp4_module sent incorrect
>        "Content-Length" response header line if the "start" argument was
>        used.
>        Thanks to Piotr Sikora.


From igor at sysoev.ru  Tue Nov 15 12:50:20 2011
From: igor at sysoev.ru (Igor Sysoev)
Date: Tue, 15 Nov 2011 16:50:20 +0400
Subject: nginx-1.0.10
In-Reply-To: <20111115132904.2c1c47d7@pluto.restena.lu>
References: <20111115094257.GB58136@nginx.com>
	<20111115132904.2c1c47d7@pluto.restena.lu>
Message-ID: <20111115125019.GA63671@nginx.com>

On Tue, Nov 15, 2011 at 01:29:04PM +0100, Bruno Pr?mont wrote:
> Hi Igor,
> 
> It would be nice if you could include tarball checksums in the release
> announcements!

Is http://nginx.org/download/nginx-1.0.10.tar.gz.asc not enough ?


-- 
Igor Sysoev


From bruno.premont at restena.lu  Tue Nov 15 13:18:59 2011
From: bruno.premont at restena.lu (Bruno =?UTF-8?B?UHLDqW1vbnQ=?=)
Date: Tue, 15 Nov 2011 14:18:59 +0100
Subject: nginx-1.0.10
In-Reply-To: <20111115125019.GA63671@nginx.com>
References: <20111115094257.GB58136@nginx.com>
	<20111115132904.2c1c47d7@pluto.restena.lu>
	<20111115125019.GA63671@nginx.com>
Message-ID: <20111115141859.53d923ae@pluto.restena.lu>

On Tue, 15 Nov 2011 16:50:20 Igor Sysoev wrote:
> On Tue, Nov 15, 2011 at 01:29:04PM +0100, Bruno Pr?mont wrote:
> > Hi Igor,
> > 
> > It would be nice if you could include tarball checksums in the release
> > announcements!
> 
> Is http://nginx.org/download/nginx-1.0.10.tar.gz.asc not enough ?

It's useful though not optimal (well, depends on workflow :) ).

When reading announcements happens on one system but downloading
and compiling happens on a different system it's easier to verify
checksums on the build host and have the checksums verified via signed
e-mail.

Bruno


From andrew at nginx.com  Tue Nov 15 13:26:11 2011
From: andrew at nginx.com (Andrew Alexeev)
Date: Tue, 15 Nov 2011 17:26:11 +0400
Subject: Nginx fails to accept new connection if active worker crashes
In-Reply-To: <2cd0f23b89b14d39b23cd6f97858c8a6.NginxMailingListEnglish@forum.nginx.org>
References: <CF8CC5EE-CF5D-4039-A9E8-4261EA5106FB@nginx.com>
	<2cd0f23b89b14d39b23cd6f97858c8a6.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <965A100A-3E08-4024-A2B4-188A92AB47C1@nginx.com>

On Nov 15, 2011, at 3:37 PM, faskiri.devel wrote:

> Hi Andrew
> 
> Thanks for the prompt reply.
> 
> As a temporary fix I had created a variable in the shared memory to
> track which pid is holding the mutex so that the check in
> [https://svn.nginx.org/nginx/browser/nginx/trunk/src/os/unix/ngx_process.c?annotate=blame#L503]
> works. It works fine for me, hadnt realized I could switch it off. Are
> we talking about putting "accept_mutex off" in the nginx.conf file? It
> will be great if you could explain what that will actually do, as in,
> how will the new requests be handed off to workers.

Yes, accept_mutex off. What accept mutex does is trying to prevent workers from competing over accept from listening sockets (in the kernel). In other words, without accept mutex workers may try to simultaneously check for new events on sockets which may lead to a slight increase in CPU usage.

Depending on your OS and the event notification mechanisms the results may vary.

Actually it's quite safe to try it and we'd appreciate your feedback here!

And as I mentioned, we've been working on fixing the situation with crashed workers and mutex lock-ups.

> Best Regards
> +Fasih
> 
> Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218359,218373#msg-218373
> 
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx


From ilan at time4learning.com  Tue Nov 15 15:57:42 2011
From: ilan at time4learning.com (Ilan Berkner)
Date: Tue, 15 Nov 2011 10:57:42 -0500
Subject: Upgrade to 1.1.8 still shows 1.1.7?
Message-ID: <CAOoWpPrY1+T=xe2L9eL3GjSK25qHbGS38xXaV3SY8qEbkiM9Aw@mail.gmail.com>

We usually upgrade Nginx when a new version comes out.  I upgraded to 1.1.8
and when I do "nginx -v" from the command line, it shows 1.1.8.  When I run
phpinfo() through the php-fpm service that we use (having restarted it), it
still shows 1.1.7, why would that be the case?

How can I confirm via an online request that Nginx is running the correct
version?  The server status stub does not show it.



-- 

Ilan Berkner
Chief Technology Officer
Time4Learning.com

6300 NE 1st Ave., Suite 203
Ft. Lauderdale, FL 33334
(954) 771-0914

<http://www.twitter.com/time4learning> <http://www.facebook.com/Time4Learning>


Time4Learning.com - Online interactive curriculum for home use, PreK-8th
Grade.
Time4Writing.com - Online writing tutorials for high, middle, and
elementary school students.
Time4Learning.net - A forum to chat with parents online about kids,
education, parenting and more.
spellingcity.com - Online vocabulary and spelling activities for teachers,
parents and students.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20111115/06e528a2/attachment.html>

From francis at daoine.org  Tue Nov 15 16:09:29 2011
From: francis at daoine.org (Francis Daly)
Date: Tue, 15 Nov 2011 16:09:29 +0000
Subject: Upgrade to 1.1.8 still shows 1.1.7?
In-Reply-To: <CAOoWpPrY1+T=xe2L9eL3GjSK25qHbGS38xXaV3SY8qEbkiM9Aw@mail.gmail.com>
References: <CAOoWpPrY1+T=xe2L9eL3GjSK25qHbGS38xXaV3SY8qEbkiM9Aw@mail.gmail.com>
Message-ID: <20111115160929.GR27078@craic.sysops.org>

On Tue, Nov 15, 2011 at 10:57:42AM -0500, Ilan Berkner wrote:

Hi there,

> How can I confirm via an online request that Nginx is running the correct
> version?  The server status stub does not show it.

Probably easiest is just

curl -I http://your_server/

Look for the Server: header.

If you've configured that not to show, then perhaps pick one specific
url that you will expose the Server: header in and use that.

All the best,

	f
-- 
Francis Daly        francis at daoine.org


From ilan at time4learning.com  Tue Nov 15 16:18:25 2011
From: ilan at time4learning.com (Ilan Berkner)
Date: Tue, 15 Nov 2011 11:18:25 -0500
Subject: Upgrade to 1.1.8 still shows 1.1.7?
In-Reply-To: <20111115160929.GR27078@craic.sysops.org>
References: <CAOoWpPrY1+T=xe2L9eL3GjSK25qHbGS38xXaV3SY8qEbkiM9Aw@mail.gmail.com>
	<20111115160929.GR27078@craic.sysops.org>
Message-ID: <CAOoWpPqUma5V6xu5zMMtPvN3xaKHJniPOwVnsDdPd1iuGJZB1Q@mail.gmail.com>

I actually had to do a full restart and now it works.

Previously kill -HUP masterpid worked, this time it didn't, not sure why.

On Tue, Nov 15, 2011 at 11:09 AM, Francis Daly <francis at daoine.org> wrote:

> On Tue, Nov 15, 2011 at 10:57:42AM -0500, Ilan Berkner wrote:
>
> Hi there,
>
> > How can I confirm via an online request that Nginx is running the correct
> > version?  The server status stub does not show it.
>
> Probably easiest is just
>
> curl -I http://your_server/
>
> Look for the Server: header.
>
> If you've configured that not to show, then perhaps pick one specific
> url that you will expose the Server: header in and use that.
>
> All the best,
>
>        f
> --
> Francis Daly        francis at daoine.org
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx
>



-- 

Ilan Berkner
Chief Technology Officer
Time4Learning.com

6300 NE 1st Ave., Suite 203
Ft. Lauderdale, FL 33334
(954) 771-0914

<http://www.twitter.com/time4learning> <http://www.facebook.com/Time4Learning>


Time4Learning.com - Online interactive curriculum for home use, PreK-8th
Grade.
Time4Writing.com - Online writing tutorials for high, middle, and
elementary school students.
Time4Learning.net - A forum to chat with parents online about kids,
education, parenting and more.
spellingcity.com - Online vocabulary and spelling activities for teachers,
parents and students.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20111115/99158960/attachment.html>

From ehabkost at raisama.net  Tue Nov 15 16:38:31 2011
From: ehabkost at raisama.net (Eduardo Habkost)
Date: Tue, 15 Nov 2011 14:38:31 -0200
Subject: SIGWINCH not working on second on-the-fly binary upgrade
Message-ID: <CAFszQJhKubYCtoR4qyWNg2pHrdU5UbZbSagsUDxscYAGynH=Jg@mail.gmail.com>

Hi,

I was trying to upgrade nginx from 1.0.9 to 1.0.10 on-the-fly, using
the process described on the wiki[1], but it looks like it is ignoring
the SIGWINCH signal I send to it.

bender:~# ps ax -H | grep nginx
31603 pts/2    S+     0:00             grep nginx
32606 ?        S      0:00   nginx: master process /usr/local/nginx/sbin/nginx
32607 ?        S      9:47     nginx: worker process
31309 ?        S      0:00     nginx: master process /usr/local/nginx/sbin/nginx
31310 ?        S      0:00       nginx: worker process
bender:~# kill -WINCH 32606 # 32606 is the 1.0.9 master process, 31309
is the 1.0.10 master process
bender:~# ps ax -H | grep nginx
31605 pts/2    S+     0:00             grep nginx
32606 ?        S      0:00   nginx: master process /usr/local/nginx/sbin/nginx
32607 ?        S      9:47     nginx: worker process
31309 ?        S      0:00     nginx: master process /usr/local/nginx/sbin/nginx
31310 ?        S      0:00       nginx: worker process

After a quick look at the code, it looks like this happened because
the 1.0.9 process I am running was started during a previous
1.0.7->1.0.9 on-the-fly upgrade, and the 1.0.9 master process now
thinks it is not daemonized (ngx_daemonized is not set on
initialization if ngx_inherited is set).

I have easily reproduced it by killing nginx completely, making a
on-the-fly 1.0.10->1.0.10 executable upgrade twice. On the first try,
SIGWINCH works; after the first upgrade, SIGWINCH stops working. Full
shell session showing the bug is pasted at the end of my message.

[1] http://wiki.nginx.org/CommandLine#Upgrading_To_a_New_Binary_On_The_Fly

--
Eduardo


Shell session:

bender:~# ps ax -H | grep nginx
31678 pts/2    S+     0:00             script -t -a /tmp/nginx.script
31679 pts/2    S+     0:00               script -t -a /tmp/nginx.script
31684 pts/5    S+     0:00                   grep nginx
bender:~# /usr/local/nginx/sbin/nginx
bender:~# ps ax -H | grep nginx
31678 pts/2    S+     0:00             script -t -a /tmp/nginx.script
31679 pts/2    S+     0:00               script -t -a /tmp/nginx.script
31689 pts/5    S+     0:00                   grep nginx
31686 ?        Ss     0:00   nginx: master process /usr/local/nginx/sbin/nginx
31687 ?        S      0:00     nginx: worker process
bender:~# kill -USR2 31686
bender:~# ps ax -H | grep nginx
31678 pts/2    S+     0:00             script -t -a /tmp/nginx.script
31679 pts/2    S+     0:00               script -t -a /tmp/nginx.script
31693 pts/5    S+     0:00                   grep nginx
31686 ?        Ss     0:00   nginx: master process /usr/local/nginx/sbin/nginx
31687 ?        S      0:00     nginx: worker process
31690 ?        S      0:00     nginx: master process /usr/local/nginx/sbin/nginx
31691 ?        S      0:00       nginx: worker process
bender:~# kill -WINCH 31686
bender:~# ps ax -H | grep nginx
31678 pts/2    S+     0:00             script -t -a /tmp/nginx.script
31679 pts/2    S+     0:00               script -t -a /tmp/nginx.script
31695 pts/5    S+     0:00                   grep nginx
31686 ?        Ss     0:00   nginx: master process /usr/local/nginx/sbin/nginx
31690 ?        S      0:00     nginx: master process /usr/local/nginx/sbin/nginx
31691 ?        S      0:00       nginx: worker process
bender:~# echo SIGWINCH worked on the first process
SIGWINCH worked on the first process
bender:~# kill -QUIT 31686
bender:~# ps ax -H | grep nginx
31678 pts/2    S+     0:00             script -t -a /tmp/nginx.script
31679 pts/2    S+     0:00               script -t -a /tmp/nginx.script
31697 pts/5    S+     0:00                   grep nginx
31690 ?        S      0:00   nginx: master process /usr/local/nginx/sbin/nginx
31691 ?        S      0:00     nginx: worker process
bender:~# kill -USR2 31690
bender:~# ps ax -H | grep nginx
31678 pts/2    S+     0:00             script -t -a /tmp/nginx.script
31679 pts/2    S+     0:00               script -t -a /tmp/nginx.script
31705 pts/5    S+     0:00                   grep nginx
31690 ?        S      0:00   nginx: master process /usr/local/nginx/sbin/nginx
31691 ?        S      0:00     nginx: worker process
31702 ?        S      0:00     nginx: master process /usr/local/nginx/sbin/nginx
31703 ?        S      0:00       nginx: worker process
bender:~# kill -WINCH 31690
bender:~# ps ax -H | grep nginx
31678 pts/2    S+     0:00             script -t -a /tmp/nginx.script
31679 pts/2    S+     0:00               script -t -a /tmp/nginx.script
31709 pts/5    S+     0:00                   grep nginx
31690 ?        S      0:00   nginx: master process /usr/local/nginx/sbin/nginx
31691 ?        S      0:00     nginx: worker process
31702 ?        S      0:00     nginx: master process /usr/local/nginx/sbin/nginx
31703 ?        S      0:00       nginx: worker process
bender:~# kill -WINCH 31690
bender:~# ps ax -H | grep nginx
31678 pts/2    S+     0:00             script -t -a /tmp/nginx.script
31679 pts/2    S+     0:00               script -t -a /tmp/nginx.script
31711 pts/5    S+     0:00                   grep nginx
31690 ?        S      0:00   nginx: master process /usr/local/nginx/sbin/nginx
31691 ?        S      0:00     nginx: worker process
31702 ?        S      0:00     nginx: master process /usr/local/nginx/sbin/nginx
31703 ?        S      0:00       nginx: worker process
bender:~# echo PID 31690 is ignoring SIGWINCH signals
PID 31690 is ignoring SIGWINCH signals
bender:~# kill -QUIT 31690
bender:~# ps ax -H | grep nginx
31678 pts/2    S+     0:00             script -t -a /tmp/nginx.script
31679 pts/2    S+     0:00               script -t -a /tmp/nginx.script
31713 pts/5    S+     0:00                   grep nginx
31702 ?        S      0:00   nginx: master process /usr/local/nginx/sbin/nginx
31703 ?        S      0:00     nginx: worker process


From ilan at time4learning.com  Tue Nov 15 16:40:40 2011
From: ilan at time4learning.com (Ilan Berkner)
Date: Tue, 15 Nov 2011 11:40:40 -0500
Subject: SIGWINCH not working on second on-the-fly binary upgrade
In-Reply-To: <CAFszQJhKubYCtoR4qyWNg2pHrdU5UbZbSagsUDxscYAGynH=Jg@mail.gmail.com>
References: <CAFszQJhKubYCtoR4qyWNg2pHrdU5UbZbSagsUDxscYAGynH=Jg@mail.gmail.com>
Message-ID: <CAOoWpPrtKNP90czhX0_7m0OSOsoYL+SfEHP+JWGDQ_6L416wvA@mail.gmail.com>

Nice find, I think I was experiencing a similar issue with upgrading on the
fly from 1.1.7 to 1.1.9.

On Tue, Nov 15, 2011 at 11:38 AM, Eduardo Habkost <ehabkost at raisama.net>wrote:

> Hi,
>
> I was trying to upgrade nginx from 1.0.9 to 1.0.10 on-the-fly, using
> the process described on the wiki[1], but it looks like it is ignoring
> the SIGWINCH signal I send to it.
>
> bender:~# ps ax -H | grep nginx
> 31603 pts/2    S+     0:00             grep nginx
> 32606 ?        S      0:00   nginx: master process
> /usr/local/nginx/sbin/nginx
> 32607 ?        S      9:47     nginx: worker process
> 31309 ?        S      0:00     nginx: master process
> /usr/local/nginx/sbin/nginx
> 31310 ?        S      0:00       nginx: worker process
> bender:~# kill -WINCH 32606 # 32606 is the 1.0.9 master process, 31309
> is the 1.0.10 master process
> bender:~# ps ax -H | grep nginx
> 31605 pts/2    S+     0:00             grep nginx
> 32606 ?        S      0:00   nginx: master process
> /usr/local/nginx/sbin/nginx
> 32607 ?        S      9:47     nginx: worker process
> 31309 ?        S      0:00     nginx: master process
> /usr/local/nginx/sbin/nginx
> 31310 ?        S      0:00       nginx: worker process
>
> After a quick look at the code, it looks like this happened because
> the 1.0.9 process I am running was started during a previous
> 1.0.7->1.0.9 on-the-fly upgrade, and the 1.0.9 master process now
> thinks it is not daemonized (ngx_daemonized is not set on
> initialization if ngx_inherited is set).
>
> I have easily reproduced it by killing nginx completely, making a
> on-the-fly 1.0.10->1.0.10 executable upgrade twice. On the first try,
> SIGWINCH works; after the first upgrade, SIGWINCH stops working. Full
> shell session showing the bug is pasted at the end of my message.
>
> [1] http://wiki.nginx.org/CommandLine#Upgrading_To_a_New_Binary_On_The_Fly
>
> --
> Eduardo
>
>
> Shell session:
>
> bender:~# ps ax -H | grep nginx
> 31678 pts/2    S+     0:00             script -t -a /tmp/nginx.script
> 31679 pts/2    S+     0:00               script -t -a /tmp/nginx.script
> 31684 pts/5    S+     0:00                   grep nginx
> bender:~# /usr/local/nginx/sbin/nginx
> bender:~# ps ax -H | grep nginx
> 31678 pts/2    S+     0:00             script -t -a /tmp/nginx.script
> 31679 pts/2    S+     0:00               script -t -a /tmp/nginx.script
> 31689 pts/5    S+     0:00                   grep nginx
> 31686 ?        Ss     0:00   nginx: master process
> /usr/local/nginx/sbin/nginx
> 31687 ?        S      0:00     nginx: worker process
> bender:~# kill -USR2 31686
> bender:~# ps ax -H | grep nginx
> 31678 pts/2    S+     0:00             script -t -a /tmp/nginx.script
> 31679 pts/2    S+     0:00               script -t -a /tmp/nginx.script
> 31693 pts/5    S+     0:00                   grep nginx
> 31686 ?        Ss     0:00   nginx: master process
> /usr/local/nginx/sbin/nginx
> 31687 ?        S      0:00     nginx: worker process
> 31690 ?        S      0:00     nginx: master process
> /usr/local/nginx/sbin/nginx
> 31691 ?        S      0:00       nginx: worker process
> bender:~# kill -WINCH 31686
> bender:~# ps ax -H | grep nginx
> 31678 pts/2    S+     0:00             script -t -a /tmp/nginx.script
> 31679 pts/2    S+     0:00               script -t -a /tmp/nginx.script
> 31695 pts/5    S+     0:00                   grep nginx
> 31686 ?        Ss     0:00   nginx: master process
> /usr/local/nginx/sbin/nginx
> 31690 ?        S      0:00     nginx: master process
> /usr/local/nginx/sbin/nginx
> 31691 ?        S      0:00       nginx: worker process
> bender:~# echo SIGWINCH worked on the first process
> SIGWINCH worked on the first process
> bender:~# kill -QUIT 31686
> bender:~# ps ax -H | grep nginx
> 31678 pts/2    S+     0:00             script -t -a /tmp/nginx.script
> 31679 pts/2    S+     0:00               script -t -a /tmp/nginx.script
> 31697 pts/5    S+     0:00                   grep nginx
> 31690 ?        S      0:00   nginx: master process
> /usr/local/nginx/sbin/nginx
> 31691 ?        S      0:00     nginx: worker process
> bender:~# kill -USR2 31690
> bender:~# ps ax -H | grep nginx
> 31678 pts/2    S+     0:00             script -t -a /tmp/nginx.script
> 31679 pts/2    S+     0:00               script -t -a /tmp/nginx.script
> 31705 pts/5    S+     0:00                   grep nginx
> 31690 ?        S      0:00   nginx: master process
> /usr/local/nginx/sbin/nginx
> 31691 ?        S      0:00     nginx: worker process
> 31702 ?        S      0:00     nginx: master process
> /usr/local/nginx/sbin/nginx
> 31703 ?        S      0:00       nginx: worker process
> bender:~# kill -WINCH 31690
> bender:~# ps ax -H | grep nginx
> 31678 pts/2    S+     0:00             script -t -a /tmp/nginx.script
> 31679 pts/2    S+     0:00               script -t -a /tmp/nginx.script
> 31709 pts/5    S+     0:00                   grep nginx
> 31690 ?        S      0:00   nginx: master process
> /usr/local/nginx/sbin/nginx
> 31691 ?        S      0:00     nginx: worker process
> 31702 ?        S      0:00     nginx: master process
> /usr/local/nginx/sbin/nginx
> 31703 ?        S      0:00       nginx: worker process
> bender:~# kill -WINCH 31690
> bender:~# ps ax -H | grep nginx
> 31678 pts/2    S+     0:00             script -t -a /tmp/nginx.script
> 31679 pts/2    S+     0:00               script -t -a /tmp/nginx.script
> 31711 pts/5    S+     0:00                   grep nginx
> 31690 ?        S      0:00   nginx: master process
> /usr/local/nginx/sbin/nginx
> 31691 ?        S      0:00     nginx: worker process
> 31702 ?        S      0:00     nginx: master process
> /usr/local/nginx/sbin/nginx
> 31703 ?        S      0:00       nginx: worker process
> bender:~# echo PID 31690 is ignoring SIGWINCH signals
> PID 31690 is ignoring SIGWINCH signals
> bender:~# kill -QUIT 31690
> bender:~# ps ax -H | grep nginx
> 31678 pts/2    S+     0:00             script -t -a /tmp/nginx.script
> 31679 pts/2    S+     0:00               script -t -a /tmp/nginx.script
> 31713 pts/5    S+     0:00                   grep nginx
> 31702 ?        S      0:00   nginx: master process
> /usr/local/nginx/sbin/nginx
> 31703 ?        S      0:00     nginx: worker process
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx
>



-- 

Ilan Berkner
Chief Technology Officer
Time4Learning.com

6300 NE 1st Ave., Suite 203
Ft. Lauderdale, FL 33334
(954) 771-0914

<http://www.twitter.com/time4learning> <http://www.facebook.com/Time4Learning>


Time4Learning.com - Online interactive curriculum for home use, PreK-8th
Grade.
Time4Writing.com - Online writing tutorials for high, middle, and
elementary school students.
Time4Learning.net - A forum to chat with parents online about kids,
education, parenting and more.
spellingcity.com - Online vocabulary and spelling activities for teachers,
parents and students.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20111115/43760f91/attachment.html>

From ilan at time4learning.com  Tue Nov 15 16:40:48 2011
From: ilan at time4learning.com (Ilan Berkner)
Date: Tue, 15 Nov 2011 11:40:48 -0500
Subject: SIGWINCH not working on second on-the-fly binary upgrade
In-Reply-To: <CAOoWpPrtKNP90czhX0_7m0OSOsoYL+SfEHP+JWGDQ_6L416wvA@mail.gmail.com>
References: <CAFszQJhKubYCtoR4qyWNg2pHrdU5UbZbSagsUDxscYAGynH=Jg@mail.gmail.com>
	<CAOoWpPrtKNP90czhX0_7m0OSOsoYL+SfEHP+JWGDQ_6L416wvA@mail.gmail.com>
Message-ID: <CAOoWpPo0sOFXmp0y-OYzcyBa7R8vKWZS2T_XaNr9z8WJnUJMzQ@mail.gmail.com>

I mean 1.1.8.

On Tue, Nov 15, 2011 at 11:40 AM, Ilan Berkner <ilan at time4learning.com>wrote:

> Nice find, I think I was experiencing a similar issue with upgrading on
> the fly from 1.1.7 to 1.1.9.
>
>
> On Tue, Nov 15, 2011 at 11:38 AM, Eduardo Habkost <ehabkost at raisama.net>wrote:
>
>> Hi,
>>
>> I was trying to upgrade nginx from 1.0.9 to 1.0.10 on-the-fly, using
>> the process described on the wiki[1], but it looks like it is ignoring
>> the SIGWINCH signal I send to it.
>>
>> bender:~# ps ax -H | grep nginx
>> 31603 pts/2    S+     0:00             grep nginx
>> 32606 ?        S      0:00   nginx: master process
>> /usr/local/nginx/sbin/nginx
>> 32607 ?        S      9:47     nginx: worker process
>> 31309 ?        S      0:00     nginx: master process
>> /usr/local/nginx/sbin/nginx
>> 31310 ?        S      0:00       nginx: worker process
>> bender:~# kill -WINCH 32606 # 32606 is the 1.0.9 master process, 31309
>> is the 1.0.10 master process
>> bender:~# ps ax -H | grep nginx
>> 31605 pts/2    S+     0:00             grep nginx
>> 32606 ?        S      0:00   nginx: master process
>> /usr/local/nginx/sbin/nginx
>> 32607 ?        S      9:47     nginx: worker process
>> 31309 ?        S      0:00     nginx: master process
>> /usr/local/nginx/sbin/nginx
>> 31310 ?        S      0:00       nginx: worker process
>>
>> After a quick look at the code, it looks like this happened because
>> the 1.0.9 process I am running was started during a previous
>> 1.0.7->1.0.9 on-the-fly upgrade, and the 1.0.9 master process now
>> thinks it is not daemonized (ngx_daemonized is not set on
>> initialization if ngx_inherited is set).
>>
>> I have easily reproduced it by killing nginx completely, making a
>> on-the-fly 1.0.10->1.0.10 executable upgrade twice. On the first try,
>> SIGWINCH works; after the first upgrade, SIGWINCH stops working. Full
>> shell session showing the bug is pasted at the end of my message.
>>
>> [1]
>> http://wiki.nginx.org/CommandLine#Upgrading_To_a_New_Binary_On_The_Fly
>>
>> --
>> Eduardo
>>
>>
>> Shell session:
>>
>> bender:~# ps ax -H | grep nginx
>> 31678 pts/2    S+     0:00             script -t -a /tmp/nginx.script
>> 31679 pts/2    S+     0:00               script -t -a /tmp/nginx.script
>> 31684 pts/5    S+     0:00                   grep nginx
>> bender:~# /usr/local/nginx/sbin/nginx
>> bender:~# ps ax -H | grep nginx
>> 31678 pts/2    S+     0:00             script -t -a /tmp/nginx.script
>> 31679 pts/2    S+     0:00               script -t -a /tmp/nginx.script
>> 31689 pts/5    S+     0:00                   grep nginx
>> 31686 ?        Ss     0:00   nginx: master process
>> /usr/local/nginx/sbin/nginx
>> 31687 ?        S      0:00     nginx: worker process
>> bender:~# kill -USR2 31686
>> bender:~# ps ax -H | grep nginx
>> 31678 pts/2    S+     0:00             script -t -a /tmp/nginx.script
>> 31679 pts/2    S+     0:00               script -t -a /tmp/nginx.script
>> 31693 pts/5    S+     0:00                   grep nginx
>> 31686 ?        Ss     0:00   nginx: master process
>> /usr/local/nginx/sbin/nginx
>> 31687 ?        S      0:00     nginx: worker process
>> 31690 ?        S      0:00     nginx: master process
>> /usr/local/nginx/sbin/nginx
>> 31691 ?        S      0:00       nginx: worker process
>> bender:~# kill -WINCH 31686
>> bender:~# ps ax -H | grep nginx
>> 31678 pts/2    S+     0:00             script -t -a /tmp/nginx.script
>> 31679 pts/2    S+     0:00               script -t -a /tmp/nginx.script
>> 31695 pts/5    S+     0:00                   grep nginx
>> 31686 ?        Ss     0:00   nginx: master process
>> /usr/local/nginx/sbin/nginx
>> 31690 ?        S      0:00     nginx: master process
>> /usr/local/nginx/sbin/nginx
>> 31691 ?        S      0:00       nginx: worker process
>> bender:~# echo SIGWINCH worked on the first process
>> SIGWINCH worked on the first process
>> bender:~# kill -QUIT 31686
>> bender:~# ps ax -H | grep nginx
>> 31678 pts/2    S+     0:00             script -t -a /tmp/nginx.script
>> 31679 pts/2    S+     0:00               script -t -a /tmp/nginx.script
>> 31697 pts/5    S+     0:00                   grep nginx
>> 31690 ?        S      0:00   nginx: master process
>> /usr/local/nginx/sbin/nginx
>> 31691 ?        S      0:00     nginx: worker process
>> bender:~# kill -USR2 31690
>> bender:~# ps ax -H | grep nginx
>> 31678 pts/2    S+     0:00             script -t -a /tmp/nginx.script
>> 31679 pts/2    S+     0:00               script -t -a /tmp/nginx.script
>> 31705 pts/5    S+     0:00                   grep nginx
>> 31690 ?        S      0:00   nginx: master process
>> /usr/local/nginx/sbin/nginx
>> 31691 ?        S      0:00     nginx: worker process
>> 31702 ?        S      0:00     nginx: master process
>> /usr/local/nginx/sbin/nginx
>> 31703 ?        S      0:00       nginx: worker process
>> bender:~# kill -WINCH 31690
>> bender:~# ps ax -H | grep nginx
>> 31678 pts/2    S+     0:00             script -t -a /tmp/nginx.script
>> 31679 pts/2    S+     0:00               script -t -a /tmp/nginx.script
>> 31709 pts/5    S+     0:00                   grep nginx
>> 31690 ?        S      0:00   nginx: master process
>> /usr/local/nginx/sbin/nginx
>> 31691 ?        S      0:00     nginx: worker process
>> 31702 ?        S      0:00     nginx: master process
>> /usr/local/nginx/sbin/nginx
>> 31703 ?        S      0:00       nginx: worker process
>> bender:~# kill -WINCH 31690
>> bender:~# ps ax -H | grep nginx
>> 31678 pts/2    S+     0:00             script -t -a /tmp/nginx.script
>> 31679 pts/2    S+     0:00               script -t -a /tmp/nginx.script
>> 31711 pts/5    S+     0:00                   grep nginx
>> 31690 ?        S      0:00   nginx: master process
>> /usr/local/nginx/sbin/nginx
>> 31691 ?        S      0:00     nginx: worker process
>> 31702 ?        S      0:00     nginx: master process
>> /usr/local/nginx/sbin/nginx
>> 31703 ?        S      0:00       nginx: worker process
>> bender:~# echo PID 31690 is ignoring SIGWINCH signals
>> PID 31690 is ignoring SIGWINCH signals
>> bender:~# kill -QUIT 31690
>> bender:~# ps ax -H | grep nginx
>> 31678 pts/2    S+     0:00             script -t -a /tmp/nginx.script
>> 31679 pts/2    S+     0:00               script -t -a /tmp/nginx.script
>> 31713 pts/5    S+     0:00                   grep nginx
>> 31702 ?        S      0:00   nginx: master process
>> /usr/local/nginx/sbin/nginx
>> 31703 ?        S      0:00     nginx: worker process
>>
>> _______________________________________________
>> nginx mailing list
>> nginx at nginx.org
>> http://mailman.nginx.org/mailman/listinfo/nginx
>>
>
>
>
> --
>
> Ilan Berkner
> Chief Technology Officer
> Time4Learning.com
>
> 6300 NE 1st Ave., Suite 203
> Ft. Lauderdale, FL 33334
> (954) 771-0914
>
> <http://www.twitter.com/time4learning> <http://www.facebook.com/Time4Learning>
>
>
> Time4Learning.com - Online interactive curriculum for home use, PreK-8th
> Grade.
> Time4Writing.com - Online writing tutorials for high, middle, and
> elementary school students.
> Time4Learning.net - A forum to chat with parents online about kids,
> education, parenting and more.
> spellingcity.com - Online vocabulary and spelling activities for
> teachers, parents and students.
>
>
>


-- 

Ilan Berkner
Chief Technology Officer
Time4Learning.com

6300 NE 1st Ave., Suite 203
Ft. Lauderdale, FL 33334
(954) 771-0914

<http://www.twitter.com/time4learning> <http://www.facebook.com/Time4Learning>


Time4Learning.com - Online interactive curriculum for home use, PreK-8th
Grade.
Time4Writing.com - Online writing tutorials for high, middle, and
elementary school students.
Time4Learning.net - A forum to chat with parents online about kids,
education, parenting and more.
spellingcity.com - Online vocabulary and spelling activities for teachers,
parents and students.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20111115/4693551d/attachment-0001.html>

From b.yordanov at exsisto.com  Tue Nov 15 16:43:04 2011
From: b.yordanov at exsisto.com (Boyko Yordanov)
Date: Tue, 15 Nov 2011 18:43:04 +0200
Subject: SIGWINCH not working on second on-the-fly binary upgrade
In-Reply-To: <CAFszQJhKubYCtoR4qyWNg2pHrdU5UbZbSagsUDxscYAGynH=Jg@mail.gmail.com>
References: <CAFszQJhKubYCtoR4qyWNg2pHrdU5UbZbSagsUDxscYAGynH=Jg@mail.gmail.com>
Message-ID: <8A05CECC-CFB5-4862-8912-4B9FF10D7D0C@exsisto.com>

Verified here as well, I am upgrading by sending QUIT to the old master process as WINCH does not work.

Boyko

On Nov 15, 2011, at 6:38 PM, Eduardo Habkost wrote:

> Hi,
> 
> I was trying to upgrade nginx from 1.0.9 to 1.0.10 on-the-fly, using
> the process described on the wiki[1], but it looks like it is ignoring
> the SIGWINCH signal I send to it.


From mdounin at mdounin.ru  Tue Nov 15 17:27:32 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Tue, 15 Nov 2011 21:27:32 +0400
Subject: Upgrade to 1.1.8 still shows 1.1.7?
In-Reply-To: <CAOoWpPqUma5V6xu5zMMtPvN3xaKHJniPOwVnsDdPd1iuGJZB1Q@mail.gmail.com>
References: <CAOoWpPrY1+T=xe2L9eL3GjSK25qHbGS38xXaV3SY8qEbkiM9Aw@mail.gmail.com>
	<20111115160929.GR27078@craic.sysops.org>
	<CAOoWpPqUma5V6xu5zMMtPvN3xaKHJniPOwVnsDdPd1iuGJZB1Q@mail.gmail.com>
Message-ID: <20111115172731.GG95664@mdounin.ru>

Hello!

On Tue, Nov 15, 2011 at 11:18:25AM -0500, Ilan Berkner wrote:

> I actually had to do a full restart and now it works.
> 
> Previously kill -HUP masterpid worked, this time it didn't, not sure why.

kill -HUP isn't expected to upgrade nginx binary, upgrade 
procedure is outlined here:

http://wiki.nginx.org/CommandLine#Upgrading_To_a_New_Binary_On_The_Fly

Maxim Dounin

> 
> On Tue, Nov 15, 2011 at 11:09 AM, Francis Daly <francis at daoine.org> wrote:
> 
> > On Tue, Nov 15, 2011 at 10:57:42AM -0500, Ilan Berkner wrote:
> >
> > Hi there,
> >
> > > How can I confirm via an online request that Nginx is running the correct
> > > version?  The server status stub does not show it.
> >
> > Probably easiest is just
> >
> > curl -I http://your_server/
> >
> > Look for the Server: header.
> >
> > If you've configured that not to show, then perhaps pick one specific
> > url that you will expose the Server: header in and use that.
> >
> > All the best,
> >
> >        f
> > --
> > Francis Daly        francis at daoine.org
> >
> > _______________________________________________
> > nginx mailing list
> > nginx at nginx.org
> > http://mailman.nginx.org/mailman/listinfo/nginx
> >
> 
> 
> 
> -- 
> 
> Ilan Berkner
> Chief Technology Officer
> Time4Learning.com
> 
> 6300 NE 1st Ave., Suite 203
> Ft. Lauderdale, FL 33334
> (954) 771-0914
> 
> <http://www.twitter.com/time4learning> <http://www.facebook.com/Time4Learning>
> 
> 
> Time4Learning.com - Online interactive curriculum for home use, PreK-8th
> Grade.
> Time4Writing.com - Online writing tutorials for high, middle, and
> elementary school students.
> Time4Learning.net - A forum to chat with parents online about kids,
> education, parenting and more.
> spellingcity.com - Online vocabulary and spelling activities for teachers,
> parents and students.

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


From mdounin at mdounin.ru  Tue Nov 15 17:55:47 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Tue, 15 Nov 2011 21:55:47 +0400
Subject: SIGWINCH not working on second on-the-fly binary upgrade
In-Reply-To: <CAFszQJhKubYCtoR4qyWNg2pHrdU5UbZbSagsUDxscYAGynH=Jg@mail.gmail.com>
References: <CAFszQJhKubYCtoR4qyWNg2pHrdU5UbZbSagsUDxscYAGynH=Jg@mail.gmail.com>
Message-ID: <20111115175547.GH95664@mdounin.ru>

Hello!

On Tue, Nov 15, 2011 at 02:38:31PM -0200, Eduardo Habkost wrote:

> I was trying to upgrade nginx from 1.0.9 to 1.0.10 on-the-fly, using
> the process described on the wiki[1], but it looks like it is ignoring
> the SIGWINCH signal I send to it.

[...]

> After a quick look at the code, it looks like this happened because
> the 1.0.9 process I am running was started during a previous
> 1.0.7->1.0.9 on-the-fly upgrade, and the 1.0.9 master process now
> thinks it is not daemonized (ngx_daemonized is not set on
> initialization if ngx_inherited is set).
> 
> I have easily reproduced it by killing nginx completely, making a
> on-the-fly 1.0.10->1.0.10 executable upgrade twice. On the first try,
> SIGWINCH works; after the first upgrade, SIGWINCH stops working. Full
> shell session showing the bug is pasted at the end of my message.

Yes, thank you, this is a bug introduced in 1.1.1/1.0.9.

As a workaround SIGQUIT to old master process may be used (it 
doesn't allow to revive old master, but allow upgrade).

The following patch should fix this:

diff --git a/src/core/nginx.c b/src/core/nginx.c
--- a/src/core/nginx.c
+++ b/src/core/nginx.c
@@ -374,6 +374,10 @@ main(int argc, char *const *argv)
         ngx_daemonized = 1;
     }
 
+    if (ngx_inherited) {
+        ngx_daemonized = 1;
+    }
+
 #endif
 
     if (ngx_create_pidfile(&ccf->pid, cycle->log) != NGX_OK) {

Maxim Dounin


From james.lyons at gmail.com  Tue Nov 15 21:58:25 2011
From: james.lyons at gmail.com (James Lyons)
Date: Tue, 15 Nov 2011 13:58:25 -0800
Subject: DNS caching issue
Message-ID: <CACg1PGkMpeAt5MgeKRuSFV3Z=W+BLF6_3Pdx9aKSVs2XMO_LzA@mail.gmail.com>

So we are using nginx from a few builds back as upgrading consistently
has proven difficult for our ops team to keep up with where I work.  I
think we're on 1.0.5.

We have some hosts setup to handle subrequests using proxy_pass
directive.  When the DNS record is changed, and the host command on
the machine is reporting a *new* host ip in the dns result.  That host
does not see traffic like it should.

We have always attributed this to caching in dns and we perform "nginx
reconfigure" to reparse conf and re-read dns.  This isn't working for
us though and i'm not sure why.  We did upgrade to 1.0.5 relatively
recently and i'm wondering if we inherited a bug.  Two things are odd
though.

When we pulled the host ip *out* of the dns, and performed nginx
reconfigure, the traffic to the machine ceased.  But now that our work
is done, and we're trying to put it back into rotation, nginx
reconfigure does not seem to be working.

Doing a restart is harder, as we have to drain the machine in
question.  But is there any known behavior that might be the root
cause of this?  Far as I'm aware the TTL on the DNS is 5m.  But its
been hours since the change.

-James-


From nginx-forum at nginx.us  Tue Nov 15 22:54:11 2011
From: nginx-forum at nginx.us (artemg)
Date: Tue, 15 Nov 2011 17:54:11 -0500
Subject: abort_request callback of ngx_http_upstream_s
Message-ID: <ff21e3f8884c46099c90937cd7e923f3.NginxMailingListEnglish@forum.nginx.org>

As I understand, abort_request() callback of upstream
(ngx_http_upstream_s) is never called now. Please correct me, if I am
wrong. The place to insert this function call is in
ngx_http_upstream_check_broken_connection(), right?

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218412,218412#msg-218412


From brian at akins.org  Tue Nov 15 23:08:02 2011
From: brian at akins.org (Brian Akins)
Date: Tue, 15 Nov 2011 18:08:02 -0500
Subject: what's the difference between proxy_store and proxy_cache?
In-Reply-To: <201111140948.43044.ne@vbart.ru>
References: <201111132050.06153.vbart@nginx.com>
	<87d9fadc36d30fe8b02560bd1c9720d8.NginxMailingListEnglish@forum.nginx.org>
	<201111140948.43044.ne@vbart.ru>
Message-ID: <CAO3qTJXi7K9iNYdJ=zYgpGAZcFYw=Yuzb=rqci6ZhQ45bneNAw@mail.gmail.com>

On Mon, Nov 14, 2011 at 12:48 AM, Valentin V. Bartenev <ne at vbart.ru> wrote:

>  you may
> need to tune kernel disk cache or even consider to put nginx cache on
> "/dev/shm".
>
>
I wouldn't recommened putting it in tmpfs.  Just let the OS buffer cache
keep it in RAM.

If you happen to have some SSD's on the other hand...

--Brian
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20111115/8429b660/attachment.html>

From nginx-forum at nginx.us  Wed Nov 16 00:35:59 2011
From: nginx-forum at nginx.us (artemg)
Date: Tue, 15 Nov 2011 19:35:59 -0500
Subject: abort_request callback of ngx_http_upstream_s
In-Reply-To: <ff21e3f8884c46099c90937cd7e923f3.NginxMailingListEnglish@forum.nginx.org>
References: <ff21e3f8884c46099c90937cd7e923f3.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <73d987887ff2a0f4b7cf9e9e6185a011.NginxMailingListEnglish@forum.nginx.org>

I have made this patch, but not sure if it is correct.

diff -u nginx-1.0.6/src/http/ngx_http_upstream.c
nginx-1.0.6_/src/http/ngx_http_upstream.c
--- nginx-1.0.6/src/http/ngx_http_upstream.c    2011-08-29
05:56:09.000000000 -0700
+++ nginx-1.0.6_/src/http/ngx_http_upstream.c   2011-11-11
02:17:36.000000000 -0800
@@ -1048,6 +1048,8 @@
     ev->eof = 1;
     c->error = 1;

+    u->abort_request(r);
+
     if (!u->cacheable && u->peer.connection) {
         ngx_log_error(NGX_LOG_INFO, ev->log, err,
                       "client closed prematurely connection, "

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218412,218415#msg-218415


From agentzh at gmail.com  Wed Nov 16 01:36:40 2011
From: agentzh at gmail.com (agentzh)
Date: Wed, 16 Nov 2011 09:36:40 +0800
Subject: compile ngx_resty to statically link some libs?
In-Reply-To: <15e3a05026fcf57706fc601b2b7bb4d1.NginxMailingListEnglish@forum.nginx.org>
References: <CAB4Tn6PrgtPoC0+Q-TP+fOZ8f3k3AS0eyAOA2aKS9fAOSngFgQ@mail.gmail.com>
	<15e3a05026fcf57706fc601b2b7bb4d1.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <CAB4Tn6O9+oikUGkcKzgbd3t8ddWJprUipEPXYdBVnh99ZPX8-Q@mail.gmail.com>

On Tue, Nov 15, 2011 at 6:35 PM, dannynoonan <nginx-forum at nginx.us> wrote:
>
> Reading up on the man page for ld caused me to try
> --with-ld-opt="-Bstatic" which gets me past configure, but gmake still
> yields an ELF binary that dynamically links out to libdrizzle.

Could you confirm that the file libdrizzle.a indeed exists in your system?

>
> Any ideas? Is trying to statically link a never ending struggle I should
> give up on early?
>

What's your gcc's version? And what does your "uname -a" say? Do you
have the libdrizzle.a file in the right place?

Regards,
-agentzh


From agentzh at gmail.com  Wed Nov 16 03:31:45 2011
From: agentzh at gmail.com (agentzh)
Date: Wed, 16 Nov 2011 11:31:45 +0800
Subject: [ANN] ngx_openresty 1.0.9.10 (stable) released
In-Reply-To: <CAB4Tn6ODvopPwto24e6LSptnkfs1B9KwKoE5wOtk5Gc2Yk+4gA@mail.gmail.com>
References: <CAB4Tn6ODvopPwto24e6LSptnkfs1B9KwKoE5wOtk5Gc2Yk+4gA@mail.gmail.com>
Message-ID: <CAB4Tn6PUfZ6Q4ZgWgpDO6hD8Q8uboVLzEWo5-2BKTnBxWqMpyg@mail.gmail.com>

Hello, folks!

I'm happy to announce that the new stable release of ngx_openresty, 1.0.9.10,
has just been kicked out of door:

  http://openresty.org/#Download

This is the first stable release of ngx_openresty that is based on the
Nginx core 1.0.9.

Special thanks go to all our contributors and users for helping make this
release happen :)

Here goes the complete change log for this release, as compared to the last
stable release, 1.0.8.26, released two weeks ago:

   - upgraded the Nginx core to 1.0.9.
   - applied the epoll_check_stale_wev
patch<http://mailman.nginx.org/pipermail/nginx-devel/2011-November/001408.html>to
the Nginx 1.0.9 core. this issue affected
   PostgresNginxModule when connecting to a remote PostgreSQL server over a
   slow network. thanks @??XX <http://weibo.com/u/1878897190>.
   - bugfix: nginx-1.0.9-variable_header_ignore_no_hash.patch<https://github.com/agentzh/ngx_openresty/blob/master/patches/nginx-1.0.9-variable_header_ignore_no_hash.patch>might
introduce a memory overflow issue in multi-header variables. thanks
   Markus Linnala.
   - bugfix: fixed the error message length while the ./configure script
   fails.
   - feature: applied a
patch<https://github.com/agentzh/ngx_openresty/blob/master/patches/nginx-1.0.9-log_escape_non_ascii.patch>to
add new directives
   log_escape_non_ascii to prevent escaping non-ascii bytes in access log
   variable values. requested by @??? <http://weibo.com/egis>. It can be
   turned on and off, and default to on just as the standard Nginx version.
   - upgraded DrizzleNginxModule to 0.1.2rc4.
      - bugfix: fixed issues with poll, rtsig, and select used by the Nginx
      event model by eliminating the poll syscall performed by libdrizzle.
      This also gives rise to a nice speedup (about 10% in simple cases).
   - upgraded LuaNginxModule to 0.3.1rc28.
      - feature: added the
ngx.encode_args<http://wiki.nginx.org/HttpLuaModule#ngx.encode_args>method
to encode a Lua code to a URI query string. thanks ?? (
      0597? <http://weibo.com/shrimp0597>).
      - feature:
ngx.location.capture<http://wiki.nginx.org/HttpLuaModule#ngx.location.capture>and
      ngx.exec <http://wiki.nginx.org/HttpLuaModule#ngx.exec> now supports
      the same Lua args table format as in
ngx.encode_args<http://wiki.nginx.org/HttpLuaModule#ngx.encode_args>.
      thanks ?? (0597? <http://weibo.com/shrimp0597>).
      - bugfix: Cache-Control header modification might introduce empty
      value headers when using with the standard
ngx_headers<http://wiki.nginx.org/HttpHeadersModule>module.
      - feature: added the ctx option to
ngx.location.capture<http://wiki.nginx.org/HttpLuaModule#ngx.location.capture>:
      you can now specify a custom Lua table to pass to the subrequest as its
      ngx.ctx <http://wiki.nginx.org/HttpLuaModule#ngx.ctx>. thanks
@hugozhu<http://weibo.com/hugozhu>
      .
      - bugfix: fixed compatibility with nginx 0.8.54. thanks
@0579?<http://weibo.com/shrimp0597>
      .
   - upgraded HeadersMoreNginxModule to 0.16rc4.
      - bugfix: Cache-Control header modification might introduce empty
      value headers when using with the standard
ngx_headers<http://wiki.nginx.org/HttpHeadersModule>module.
   - upgraded PostgresNginxModule to 0.9rc2
      - bugfix: now we log an error message when the postgres_pass target
      is not found at all and returns 500 in this case instead of
returning empty
      response.
      - bugfix: we should no longer return NGX_AGAIN when the re-polling
      returns IO WAIT in case of the "connection made" state.
      - feature: added some debugging outputs which be enabled by passing
      the --with-debug option while building Nginx or OpenResty.
      - bugfix: fixed compatibility issues with Nginx 1.1.4+:
      ngx_chain_update_chains now requires a pool argument.
   - upgraded LuaRdsParserLibrary to 0.04.
      - bugfix: fixed a serious memory leak reported by bearnard.
   - upgraded XssNginxModule to 0.03rc5.
      - bugfix: the callback argument value parser did not accept
      JavaScript identifier names started with underscores. thanks Sam Mulube.

As always, you're welcome to report bugs and feature requests either here
or directly to me :)

It'll also be highly appreciated to try out the devel releases (based on
the Nginx core 1.0.10+) that are coming out later ;)

OpenResty (aka. ngx_openresty) is a full-fledged web application server by
bundling the standard Nginx core, lots of 3rd-party Nginx
modules<http://wiki.nginx.org/3rdPartyModules>,
as well as most of their external dependencies.

By taking adantage of various well-designed Nginx modules, OpenResty
effectively turns the nginx server into a powerful web app server, in which
the web developers can use the Lua programming language to script various
existing nginx C modules and Lua modules and construct extremely
high-performance web applications that is capable to handle 10K+
connections.

OpenResty aims to run your server-side web app completely in the Nginx
server, leveraging Nginx's event model to do non-blocking I/O not only with
the HTTP clients, but also with remote backends like MySQL, PostgreSQL,
Memcached, and Redis.

You can find more details on the homepage of ngx_openresty here:

    http://openresty.org

Have fun!
-agentzh
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20111116/0443bab9/attachment-0001.html>

From andrew at nginx.com  Wed Nov 16 08:23:45 2011
From: andrew at nginx.com (Andrew Alexeev)
Date: Wed, 16 Nov 2011 12:23:45 +0400
Subject: DNS caching issue
In-Reply-To: <CACg1PGkMpeAt5MgeKRuSFV3Z=W+BLF6_3Pdx9aKSVs2XMO_LzA@mail.gmail.com>
References: <CACg1PGkMpeAt5MgeKRuSFV3Z=W+BLF6_3Pdx9aKSVs2XMO_LzA@mail.gmail.com>
Message-ID: <39FCCA69-38BE-485C-B1B0-11BFE0B68069@nginx.com>

James,

On Nov 16, 2011, at 1:58 AM, James Lyons wrote:

> So we are using nginx from a few builds back as upgrading consistently
> has proven difficult for our ops team to keep up with where I work.  I
> think we're on 1.0.5.
> 
> We have some hosts setup to handle subrequests using proxy_pass
> directive.  When the DNS record is changed, and the host command on
> the machine is reporting a *new* host ip in the dns result.  That host
> does not see traffic like it should.
> 
> We have always attributed this to caching in dns and we perform "nginx
> reconfigure" to reparse conf and re-read dns.  This isn't working for
> us though and i'm not sure why.  We did upgrade to 1.0.5 relatively
> recently and i'm wondering if we inherited a bug.  Two things are odd
> though.
> 
> When we pulled the host ip *out* of the dns, and performed nginx
> reconfigure, the traffic to the machine ceased.  But now that our work
> is done, and we're trying to put it back into rotation, nginx
> reconfigure does not seem to be working.
> 
> Doing a restart is harder, as we have to drain the machine in
> question.  But is there any known behavior that might be the root
> cause of this?  Far as I'm aware the TTL on the DNS is 5m.  But its
> been hours since the change.

Can you please show the relevant portion of your proxy configuration?

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



From nginx-forum at nginx.us  Wed Nov 16 08:40:07 2011
From: nginx-forum at nginx.us (zhenwei)
Date: Wed, 16 Nov 2011 03:40:07 -0500
Subject: what's the difference between proxy_store and proxy_cache?
In-Reply-To: <CAO3qTJXi7K9iNYdJ=zYgpGAZcFYw=Yuzb=rqci6ZhQ45bneNAw@mail.gmail.com>
References: <CAO3qTJXi7K9iNYdJ=zYgpGAZcFYw=Yuzb=rqci6ZhQ45bneNAw@mail.gmail.com>
Message-ID: <d6da8f217a71781219e05a61101dcd0f.NginxMailingListEnglish@forum.nginx.org>

Quit agree with you that SSD seems the right way to get high random
write/read performs well, especially we're serving thousands of websites
for each Nginx.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218263,218424#msg-218424


From nginx-forum at nginx.us  Wed Nov 16 08:51:09 2011
From: nginx-forum at nginx.us (faskiri.devel)
Date: Wed, 16 Nov 2011 03:51:09 -0500
Subject: Nginx fails to accept new connection if active worker crashes
In-Reply-To: <965A100A-3E08-4024-A2B4-188A92AB47C1@nginx.com>
References: <965A100A-3E08-4024-A2B4-188A92AB47C1@nginx.com>
Message-ID: <335afd057b4626537c43149994c40d04.NginxMailingListEnglish@forum.nginx.org>

Thanks for your attention!

It works fine with accept_mutex off, I will run my stress test harness
over the weekend to see the impact on the performance. If there is a
significant difference in performance, will surely update the thread
with the same.

For my understanding, I had implemented a workaround to get around this
problem. Is your solution along the same line?

My patch:

diff --git a/service/nginxServer/nginx-1.0.5/src/event/ngx_event.c
b/service/nginxServer/nginx-1.0.5/src/event/ngx_event.c
index c57d37e..a6ed725 100644
--- a/service/nginxServer/nginx-1.0.5/src/event/ngx_event.c
+++ b/service/nginxServer/nginx-1.0.5/src/event/ngx_event.c
@@ -49,6 +49,10 @@ ngx_atomic_t         *ngx_connection_counter =
&connection_counter;
 
 ngx_atomic_t         *ngx_accept_mutex_ptr;
 ngx_shmtx_t           ngx_accept_mutex;
+// This is shared var protected by ngx_use_accept_mutex. Access only
when
+// ngx_accept_mutex is held. The var stores the PID of the process
currently
+// holding the mutex
+ngx_pid_t            *ngx_accept_mutex_held_by;
 ngx_uint_t            ngx_use_accept_mutex;
 ngx_uint_t            ngx_accept_events;
 ngx_uint_t            ngx_accept_mutex_held;
@@ -254,6 +258,7 @@ ngx_process_events_and_timers(ngx_cycle_t *cycle)
     }
 
     if (ngx_accept_mutex_held) {
+        *ngx_accept_mutex_held_by = 0;
         ngx_shmtx_unlock(&ngx_accept_mutex);
     }
 
@@ -526,6 +531,9 @@ ngx_event_module_init(ngx_cycle_t *cycle)
     {
         return NGX_ERROR;
     }
+    // cl = 128 bytes are available for us to use. ngx_shmtx_create
uses
+    // ngx_atomic_t bytes to assign to mutex->lock, using the memory
after that
+    ngx_accept_mutex_held_by = (ngx_pid_t*) (shared +
sizeof(ngx_atomic_t));
 
     ngx_connection_counter = (ngx_atomic_t *) (shared + 1 * cl);
 
diff --git a/service/nginxServer/nginx-1.0.5/src/event/ngx_event.h
b/service/nginxServer/nginx-1.0.5/src/event/ngx_event.h
index 778da52..f1b06d4 100644
--- a/service/nginxServer/nginx-1.0.5/src/event/ngx_event.h
+++ b/service/nginxServer/nginx-1.0.5/src/event/ngx_event.h
@@ -501,6 +501,7 @@ extern ngx_atomic_t         
*ngx_connection_counter;
 
 extern ngx_atomic_t          *ngx_accept_mutex_ptr;
 extern ngx_shmtx_t            ngx_accept_mutex;
+extern ngx_pid_t             *ngx_accept_mutex_held_by;
 extern ngx_uint_t             ngx_use_accept_mutex;
 extern ngx_uint_t             ngx_accept_events;
 extern ngx_uint_t             ngx_accept_mutex_held;
diff --git
a/service/nginxServer/nginx-1.0.5/src/event/ngx_event_accept.c
b/service/nginxServer/nginx-1.0.5/src/event/ngx_event_accept.c
index 2355d1b..feb4568 100644
--- a/service/nginxServer/nginx-1.0.5/src/event/ngx_event_accept.c
+++ b/service/nginxServer/nginx-1.0.5/src/event/ngx_event_accept.c
@@ -298,6 +298,10 @@ ngx_trylock_accept_mutex(ngx_cycle_t *cycle)
         ngx_log_debug0(NGX_LOG_DEBUG_EVENT, cycle->log, 0,
                        "accept mutex locked");
 
+        *ngx_accept_mutex_held_by = ngx_pid;
+
+        // If the mutex was already held by me and we are using
RTSIG_EVENT, no
+        // need to enable accept_events

         if (ngx_accept_mutex_held
             && ngx_accept_events == 0
             && !(ngx_event_flags & NGX_USE_RTSIG_EVENT))
@@ -306,6 +310,8 @@ ngx_trylock_accept_mutex(ngx_cycle_t *cycle)
         }
 
         if (ngx_enable_accept_events(cycle) == NGX_ERROR) {
+            // No one is holding the mutex now
+            *ngx_accept_mutex_held_by = 0;
             ngx_shmtx_unlock(&ngx_accept_mutex);
             return NGX_ERROR;
         }
@@ -317,8 +323,9 @@ ngx_trylock_accept_mutex(ngx_cycle_t *cycle)
     }
 
     ngx_log_debug1(NGX_LOG_DEBUG_EVENT, cycle->log, 0,
-                   "accept mutex lock failed: %ui",
ngx_accept_mutex_held);
+                   "accept mutex lock failed: held by: %ui",
*ngx_accept_mutex_held_by);
 
+    // If I held it earlier, but not anymore (ngx_trylock_accept_mutex
failed)
     if (ngx_accept_mutex_held) {
         if (ngx_disable_accept_events(cycle) == NGX_ERROR) {
             return NGX_ERROR;
diff --git a/service/nginxServer/nginx-1.0.5/src/os/unix/ngx_process.c
b/service/nginxServer/nginx-1.0.5/src/os/unix/ngx_process.c
index 6055587..b66d4b3 100644
--- a/service/nginxServer/nginx-1.0.5/src/os/unix/ngx_process.c
+++ b/service/nginxServer/nginx-1.0.5/src/os/unix/ngx_process.c
@@ -492,17 +492,18 @@ ngx_process_get_status(void)
         }
 
 
-        if (ngx_accept_mutex_ptr) {
-
-            /*
-             * unlock the accept mutex if the abnormally exited
process
-             * held it
-             */
-
-            ngx_atomic_cmp_set(ngx_accept_mutex_ptr, pid, 0);
+        // If the accept mutex is held by the abnormally exited
process
+        // Note: If the process holding this has died, the mutex cannot
be
+        // acquired by someone else, in which case,
ngx_accept_mutex_held is
+        // free to be accessed
+        if (ngx_accept_mutex_held_by != NULL && pid ==
*ngx_accept_mutex_held_by) {
+            ngx_log_error(NGX_LOG_INFO, ngx_cycle->log, 0,
+                "PID %P held the accept mutex. Releasing", pid);
+            // Reset the value before unlocking
+            *ngx_accept_mutex_held_by = 0;
+            ngx_shmtx_unlock(&ngx_accept_mutex);
         }
 
-
         one = 1;
         process = "unknown process";
 
-- 
1.7.1

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218359,218425#msg-218425


From kworthington at gmail.com  Wed Nov 16 11:30:35 2011
From: kworthington at gmail.com (Kevin Worthington)
Date: Wed, 16 Nov 2011 06:30:35 -0500
Subject: nginx-1.0.10
In-Reply-To: <20111115141859.53d923ae@pluto.restena.lu>
References: <20111115094257.GB58136@nginx.com>
	<20111115132904.2c1c47d7@pluto.restena.lu>
	<20111115125019.GA63671@nginx.com>
	<20111115141859.53d923ae@pluto.restena.lu>
Message-ID: <CAGo79UV5_DfPwNYBbqwVzLp46-a4yXnmy8azFVAeFfw4fPeetw@mail.gmail.com>

Hello Nginx Users,
Just released: Nginx 1.0.10 For Windows http://goo.gl/5GcCE (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

Thank you,
Kevin
--
Kevin Worthington
kworthington ~at] gmail [dot} .C0M~
http://www.kevinworthington.com/



On Tue, Nov 15, 2011 at 8:18 AM, Bruno Pr?mont <bruno.premont at restena.lu> wrote:
> On Tue, 15 Nov 2011 16:50:20 Igor Sysoev wrote:
>> On Tue, Nov 15, 2011 at 01:29:04PM +0100, Bruno Pr?mont wrote:
>> > Hi Igor,
>> >
>> > It would be nice if you could include tarball checksums in the release
>> > announcements!
>>
>> Is http://nginx.org/download/nginx-1.0.10.tar.gz.asc not enough ?
>
> It's useful though not optimal (well, depends on workflow :) ).
>
> When reading announcements happens on one system but downloading
> and compiling happens on a different system it's easier to verify
> checksums on the build host and have the checksums verified via signed
> e-mail.
>
> Bruno
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx


From andrew at nginx.com  Wed Nov 16 14:00:12 2011
From: andrew at nginx.com (Andrew Alexeev)
Date: Wed, 16 Nov 2011 18:00:12 +0400
Subject: DNS TTLs being ignored
In-Reply-To: <72FF6524-75CF-4123-8F83-50363C25AE21@nginx.com>
References: <CAAqHd6v_LxnOpfoE6Fq+sKebtwHiguse4d6GmmvpAUayR5eqaA@mail.gmail.com>
	<d7ffe70b7f12b278cb9ffeac1a02c358@ruby-forum.com>
	<AB401AA3-211C-4401-B050-712D7CBBB054@nginx.com>
	<72FF6524-75CF-4123-8F83-50363C25AE21@nginx.com>
Message-ID: <EAF2AA38-826E-434D-B9A2-AEBA553BA590@nginx.com>

On Nov 15, 2011, at 1:50 PM, Andrew Alexeev wrote:

> On Nov 3, 2011, at 1:50 PM, Andrew Alexeev wrote:
> 
>> Noah,
>> 
>> This fix/improvement be introduced in 1.1.8 which will come out around Nov 14.
> 
> Apologies, it didn't get in either 1.1.8 (yesterday) or 1.1.10 (today). It's almost ready and would hopefully get into the next dev and stable releases in a couple of weeks.

Jfyi, it went committed today

http://mailman.nginx.org/pipermail/nginx-devel/2011-November/001466.html
http://nginx.org/en/docs/http/ngx_http_core_module.html#resolver

and will be included in 1.1.9.

>> Hope this helps
>> 
>> On Nov 3, 2011, at 1:46 PM, Noah C. wrote:
>> 
>>> Thanks for the reply Andrew. Do you have any idea when it's likely to be 
>>> generally available? This is a pretty big nuisance for us, and I'd like 
>>> to be able to figure out if I need to look at using a new reverse proxy, 
>>> at least for the time being.
>>> 
>>> --Noah
>>> 
>>> -- 
>>> Posted via http://www.ruby-forum.com/.
>>> 
>>> _______________________________________________
>>> nginx mailing list
>>> nginx at nginx.org
>>> http://mailman.nginx.org/mailman/listinfo/nginx


From magicbearmo at gmail.com  Wed Nov 16 16:34:16 2011
From: magicbearmo at gmail.com (MagicBear)
Date: Thu, 17 Nov 2011 00:34:16 +0800
Subject: [Module]Nginx Regexp Plugins
Message-ID: <CAENo=s0vd3pDwdYyztVasqhj7rDzuJ7CdzXgnsUnds3gytGoPw@mail.gmail.com>

This plugin in base on ngx_cache_purge-1.4

P.S: This plugin current will blocking web request, now is only for
test. I cannot found a method to make nginx run a new thread.

Usage:
	location ~ ^/regex_purge(/.*) {
		proxy_cache_batchpurge cache_zone $1$is_args$args;
	}

download: http://m-b.cc/share/ngx_batch_purge-0.1.tgz

-- 
MagicBear


From piotr.sikora at frickle.com  Wed Nov 16 17:32:11 2011
From: piotr.sikora at frickle.com (Piotr Sikora)
Date: Wed, 16 Nov 2011 18:32:11 +0100
Subject: [Module]Nginx Regexp Plugins
In-Reply-To: <CAENo=s0vd3pDwdYyztVasqhj7rDzuJ7CdzXgnsUnds3gytGoPw@mail.gmail.com>
References: <CAENo=s0vd3pDwdYyztVasqhj7rDzuJ7CdzXgnsUnds3gytGoPw@mail.gmail.com>
Message-ID: <138D0318626C487BB0CAE851942DB3D5@Desktop>

Hey,

> P.S: This plugin current will blocking web request, now is only for test.

I know you've got good intentions, but this module will kill your box.

> I cannot found a method to make nginx run a new thread.

Just fork() the worker process.

Best regards,
Piotr Sikora < piotr.sikora at frickle.com >


From appa at perusio.net  Wed Nov 16 17:57:32 2011
From: appa at perusio.net (=?UTF-8?B?QW50w7NuaW8=?= P. P. Almeida)
Date: Wed, 16 Nov 2011 17:57:32 +0000
Subject: nginx-1.0.10
In-Reply-To: <20111115141859.53d923ae@pluto.restena.lu>
References: <20111115094257.GB58136@nginx.com>
	<20111115132904.2c1c47d7@pluto.restena.lu>
	<20111115125019.GA63671@nginx.com>
	<20111115141859.53d923ae@pluto.restena.lu>
Message-ID: <87zkfwj6gj.wl%appa@perusio.net>

On 15 Nov 2011 13h18 WET, bruno.premont at restena.lu wrote:

It's quite easy to create a script that downloads the source, the sig
and verifies it with GPG.

Here's my take on it:

https://github.com/perusio/nginx-get-source 

Feel free to use it or fork it.

--- appa


From magicbearmo at gmail.com  Wed Nov 16 18:47:31 2011
From: magicbearmo at gmail.com (Bear Magic)
Date: Thu, 17 Nov 2011 02:47:31 +0800
Subject: [Module]Nginx Regexp Plugins
In-Reply-To: <138D0318626C487BB0CAE851942DB3D5@Desktop>
References: <CAENo=s0vd3pDwdYyztVasqhj7rDzuJ7CdzXgnsUnds3gytGoPw@mail.gmail.com>
	<138D0318626C487BB0CAE851942DB3D5@Desktop>
Message-ID: <3134703292502901380@unknownmsgid>

I have tried for this, but when done the first time, it will have some
error for next. I Have thinking for spawn a process at startup or
using nginx cache manager process to do this.

Piotr Sikora <piotr.sikora at frickle.com> ? 2011-11-17 1:32 ???

> Hey,
>
>> P.S: This plugin current will blocking web request, now is only for test.
>
> I know you've got good intentions, but this module will kill your box.
>
>> I cannot found a method to make nginx run a new thread.
>
> Just fork() the worker process.
>
> Best regards,
> Piotr Sikora < piotr.sikora at frickle.com >
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx


From nginx-forum at nginx.us  Thu Nov 17 01:04:30 2011
From: nginx-forum at nginx.us (sisif)
Date: Wed, 16 Nov 2011 20:04:30 -0500
Subject: MP4 - start time is out mp4 stts samples
Message-ID: <3c2ccb0e721cdc80d2dd4bf09ec014ea.NginxMailingListEnglish@forum.nginx.org>

Hello,
About MP4 module, my files are around 100-200MB, 30-50minutes videos.
When I try to seek more then 14-20min from start, I got "video not
found" and this error in nginx log:
2011/11/17 01:17:46 [error] 15463#0: *19 start time is out mp4 stts
samples in "/home/host/download/4040292549.mp4   

How can I fix this, anyone have this error and know something about ?

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218451,218451#msg-218451


From igor at sysoev.ru  Thu Nov 17 05:06:03 2011
From: igor at sysoev.ru (Igor Sysoev)
Date: Thu, 17 Nov 2011 09:06:03 +0400
Subject: MP4 - start time is out mp4 stts samples
In-Reply-To: <3c2ccb0e721cdc80d2dd4bf09ec014ea.NginxMailingListEnglish@forum.nginx.org>
References: <3c2ccb0e721cdc80d2dd4bf09ec014ea.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20111117050603.GA10385@nginx.com>

On Wed, Nov 16, 2011 at 08:04:30PM -0500, sisif wrote:
> Hello,
> About MP4 module, my files are around 100-200MB, 30-50minutes videos.
> When I try to seek more then 14-20min from start, I got "video not
> found" and this error in nginx log:
> 2011/11/17 01:17:46 [error] 15463#0: *19 start time is out mp4 stts
> samples in "/home/host/download/4040292549.mp4   
> 
> How can I fix this, anyone have this error and know something about ?

Could you show "nginx -V" output ?


-- 
Igor Sysoev


From benlancaster at holler.co.uk  Thu Nov 17 10:11:38 2011
From: benlancaster at holler.co.uk (Ben Lancaster)
Date: Thu, 17 Nov 2011 10:11:38 +0000
Subject: Nginx (PPA stable) periodically returning headers then hanging
	connection when serving from fcgi cache
Message-ID: <EFDACDA3-CFF4-4CBD-8D65-AF1669756928@holler.co.uk>

List,

I'm experiencing some problems with FastCGI cache and nginx/1.0.4 from stable PPA on Ubuntu Lucid.

In short, it seems that sometimes the FastCGI cache is getting corrupted somehow - Nginx will serve sane headers from cache, but then the connection will seem to hang, with no body returned until the request is timed out by the client (apparently it'll hang indefinitely).

For example, headers will appear good:

HTTP/1.1 200 OK
Server: nginx
X-Cache-Status: HIT
Cache-Control: public, max_age=300
Content-Type: text/html; charset=utf-8
Date: Thu, 17 Nov 2011 09:58:26 GMT
Expires: Thu, 17 Nov 2011 10:02:46
Etag: 611a2b5dcde004cf68ffd56345584d40
Connection: close
Last-Modified: Thu, 17 Nov 2011 09:57:46
Transfer-Encoding: Identity

?but then the connection sits there without returning the body.

Once nginx returns one "bad" response (as described above), all subsequent requests for the same (cached) resource have the same problem. Other cached resources seem to work as normal, and have experienced it twice in the past 24 hours.

The only resolution I've found so far is to junk my cache folder and bounce the nginx service.

Here's what my vhost config looks like:

server {

  listen 80 default;
  server_name example.com;

  server_tokens off;

  root      /home/user/example.com/web;
  index     index.php;

  access_log  /dev/null;
  error_log   /var/log/nginx/error.log;

  location / {

    if (-f $request_filename) {
      expires 3h;
      break;
    }

    rewrite ^(.*) /index.php last;
  }

  location ~ (.*\.php)($|/) {
    set  $script   $1;
    set  $path_info  "";

    if ($uri ~ "^(.+\.php)(/.+)") {
        set  $script   $1;
        set  $path_info  $2;
    }

    fastcgi_pass 127.0.0.1:9000;


    include     fastcgi_params;
    fastcgi_read_timeout 180;
    fastcgi_param   PATH_INFO $path_info;
    fastcgi_param   SCRIPT_FILENAME $document_root$script;
    fastcgi_param   SCRIPT_NAME $script;

    fastcgi_pass_header  Set-Cookie;

    fastcgi_cache_methods GET HEAD;
    fastcgi_cache fcgi-cache;
    fastcgi_cache_key canimationlive$request_uri;
    fastcgi_cache_valid 200 1h;
    fastcgi_cache_min_uses 1;
    fastcgi_cache_use_stale error timeout http_500 updating;

    add_header X-Cache-Status $upstream_cache_status;
  }
}
## --end

Here's what the fcgi-cache definition looks like:

fastcgi_cache_path   /var/www/cache levels=1:2
           keys_zone=fcgi-cache:10m
           max_size=512m
           inactive=28d;

I've just upgraded to nginx 1.0.9 (also from PPA), and noticed in the changelog for versions 1.0.4 and 1.0.5 a "Bugfix: "stalled cache updating" alert" - is this the same problem? If so, the above may well be null and void.

Thanks in advance,

Ben


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20111117/0caee091/attachment.html>

From nginx-forum at nginx.us  Thu Nov 17 10:38:00 2011
From: nginx-forum at nginx.us (AlexXF)
Date: Thu, 17 Nov 2011 05:38:00 -0500
Subject: SSL problem failed (SSL: error:0B080074:x509 certificate
	routines:X509_check_private_key:key values mismatch)
In-Reply-To: <7227a5cde138f26c5e24e1f0f94a6485.NginxMailingListEnglish@forum.nginx.org>
References: <4D5D45A3.2030209@feurix.com>
	<7227a5cde138f26c5e24e1f0f94a6485.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <328aa20b68d6f060d65706cda96e86dd.NginxMailingListEnglish@forum.nginx.org>

I've got this problem also, but solved!

There is a two files that gandi sent to you:
site.crt
site-bundle.crt

Use site.crt instead of site-bundle.crt. Nginx requires certificate for
exactly site only. So it is not require to use chain (bundle)
certificate file.

Enjoy!

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,176163,218470#msg-218470


From mdounin at mdounin.ru  Thu Nov 17 10:40:42 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Thu, 17 Nov 2011 14:40:42 +0400
Subject: SSL problem failed (SSL: error:0B080074:x509 certificate
	routines:X509_check_private_key:key values mismatch)
In-Reply-To: <328aa20b68d6f060d65706cda96e86dd.NginxMailingListEnglish@forum.nginx.org>
References: <4D5D45A3.2030209@feurix.com>
	<7227a5cde138f26c5e24e1f0f94a6485.NginxMailingListEnglish@forum.nginx.org>
	<328aa20b68d6f060d65706cda96e86dd.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20111117104042.GO95664@mdounin.ru>

Hello!

On Thu, Nov 17, 2011 at 05:38:00AM -0500, AlexXF wrote:

> I've got this problem also, but solved!
> 
> There is a two files that gandi sent to you:
> site.crt
> site-bundle.crt
> 
> Use site.crt instead of site-bundle.crt. Nginx requires certificate for
> exactly site only. So it is not require to use chain (bundle)
> certificate file.

Both site certificate and bundle should be used.  See here for 
details:

http://nginx.org/en/docs/http/configuring_https_servers.html#chains

Maxim Dounin


From nginx-forum at nginx.us  Thu Nov 17 10:44:32 2011
From: nginx-forum at nginx.us (AlexXF)
Date: Thu, 17 Nov 2011 05:44:32 -0500
Subject: SSL problem failed (SSL: error:0B080074:x509 certificate
	routines:X509_check_private_key:key values mismatch)
In-Reply-To: <328aa20b68d6f060d65706cda96e86dd.NginxMailingListEnglish@forum.nginx.org>
References: <4D5D45A3.2030209@feurix.com>
	<7227a5cde138f26c5e24e1f0f94a6485.NginxMailingListEnglish@forum.nginx.org>
	<328aa20b68d6f060d65706cda96e86dd.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <d4bd03a19d5266508850d6547b4ee577.NginxMailingListEnglish@forum.nginx.org>

Anyway - it works after i've made that changes in nginx.conf file.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,176163,218472#msg-218472


From nginx-forum at nginx.us  Thu Nov 17 10:53:05 2011
From: nginx-forum at nginx.us (AlexXF)
Date: Thu, 17 Nov 2011 05:53:05 -0500
Subject: SSL problem failed (SSL: error:0B080074:x509 certificate
	routines:X509_check_private_key:key values mismatch)
In-Reply-To: <d4bd03a19d5266508850d6547b4ee577.NginxMailingListEnglish@forum.nginx.org>
References: <4D5D45A3.2030209@feurix.com>
	<7227a5cde138f26c5e24e1f0f94a6485.NginxMailingListEnglish@forum.nginx.org>
	<328aa20b68d6f060d65706cda96e86dd.NginxMailingListEnglish@forum.nginx.org>
	<d4bd03a19d5266508850d6547b4ee577.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <9da9434af94e62f895db53f84093291c.NginxMailingListEnglish@forum.nginx.org>

Upd. It works for concatenated cert files also.

Looks like topic starter forgot to  concatenate cert files before.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,176163,218473#msg-218473


From igor at sysoev.ru  Thu Nov 17 11:36:17 2011
From: igor at sysoev.ru (Igor Sysoev)
Date: Thu, 17 Nov 2011 15:36:17 +0400
Subject: SSL problem failed (SSL: error:0B080074:x509 certificate
	routines:X509_check_private_key:key values mismatch)
In-Reply-To: <9da9434af94e62f895db53f84093291c.NginxMailingListEnglish@forum.nginx.org>
References: <4D5D45A3.2030209@feurix.com>
	<7227a5cde138f26c5e24e1f0f94a6485.NginxMailingListEnglish@forum.nginx.org>
	<328aa20b68d6f060d65706cda96e86dd.NginxMailingListEnglish@forum.nginx.org>
	<d4bd03a19d5266508850d6547b4ee577.NginxMailingListEnglish@forum.nginx.org>
	<9da9434af94e62f895db53f84093291c.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20111117113617.GA18146@nginx.com>

On Thu, Nov 17, 2011 at 05:53:05AM -0500, AlexXF wrote:
> Upd. It works for concatenated cert files also.
> 
> Looks like topic starter forgot to  concatenate cert files before.

He might concatenate them in the wrong order or might use only
site-bundle.crt.


-- 
Igor Sysoev


From igor at sysoev.ru  Thu Nov 17 11:36:51 2011
From: igor at sysoev.ru (Igor Sysoev)
Date: Thu, 17 Nov 2011 15:36:51 +0400
Subject: SSL problem failed (SSL: error:0B080074:x509 certificate
	routines:X509_check_private_key:key values mismatch)
In-Reply-To: <d4bd03a19d5266508850d6547b4ee577.NginxMailingListEnglish@forum.nginx.org>
References: <4D5D45A3.2030209@feurix.com>
	<7227a5cde138f26c5e24e1f0f94a6485.NginxMailingListEnglish@forum.nginx.org>
	<328aa20b68d6f060d65706cda96e86dd.NginxMailingListEnglish@forum.nginx.org>
	<d4bd03a19d5266508850d6547b4ee577.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20111117113651.GB18146@nginx.com>

On Thu, Nov 17, 2011 at 05:44:32AM -0500, AlexXF wrote:
> Anyway - it works after i've made that changes in nginx.conf file.

Browsers usually store intermediate certificates which they receive
and which are signed by trusted authorities, so actively used browsers
may already have the required intermediate certificates and may not
complain about a certificate sent without a chained bundle.


-- 
Igor Sysoev


From brian at akins.org  Thu Nov 17 12:33:49 2011
From: brian at akins.org (Brian Akins)
Date: Thu, 17 Nov 2011 07:33:49 -0500
Subject: [ANN] ngx_openresty 1.0.9.10 (stable) released
In-Reply-To: <CAB4Tn6PUfZ6Q4ZgWgpDO6hD8Q8uboVLzEWo5-2BKTnBxWqMpyg@mail.gmail.com>
References: <CAB4Tn6ODvopPwto24e6LSptnkfs1B9KwKoE5wOtk5Gc2Yk+4gA@mail.gmail.com>
	<CAB4Tn6PUfZ6Q4ZgWgpDO6hD8Q8uboVLzEWo5-2BKTnBxWqMpyg@mail.gmail.com>
Message-ID: <03D3F526-CD29-4110-8F9B-A096178BF11D@akins.org>


On Nov 15, 2011, at 10:31 PM, agentzh wrote:

> 
> I'm happy to announce that the new stable release of ngx_openresty, 1.0.9.10, has just been kicked out of door:
> 

Once again, thanks for doing this!

--Brian

From magicbearmo at gmail.com  Thu Nov 17 13:22:01 2011
From: magicbearmo at gmail.com (MagicBear)
Date: Thu, 17 Nov 2011 21:22:01 +0800
Subject: [MOD] proxy_cache regexp batch purge modules 0.2
Message-ID: <CAENo=s15bjJj8C-u430kd-Aby9vtw2_U2VpLRcDdrxMfLbFQeg@mail.gmail.com>

Download
=====
http://m-b.cc/share/ngx_batch_purge-0.2.tar.gz


About
=====
nginx_cache_batchpurge is 'nginx' module which adds regex batch purge content
from 'proxy' caches, and is base on `ngx_cache_purge`.


Notice
=====
This module is use only for batch purge, if you want to purge a single object,
please install ngx_cache_purge module, this module can work with that module
exists.


Install
=====
./configure --add-module=../ngx_batch_purge-0.2


Configuration directives
========================
proxy_cache_purge
-----------------
* **syntax**: `proxy_cache_batchpurge zone_name key`
* **default**: `none`
* **context**: `location`

Sets area and key used for purging selected pages from `proxy`'s cache.



Sample configuration
====================
    http {
        proxy_cache_path  /tmp/cache  keys_zone=tmpcache:10m;

        server {
            location / {
                proxy_pass         http://127.0.0.1:8000;
                proxy_cache        tmpcache;
                proxy_cache_key    $uri$is_args$args;
            }

            location ~ /purge(/.*) {
                allow              127.0.0.1;
                deny               all;
                proxy_cache_purge  tmpcache $1$is_args$args;
            }

            location ~ /regex_purge(.*) {
                allow              127.0.0.1;
                deny               all;
                proxy_cache_batchpurge  tmpcache $1$is_args$args;
            }
        }
    }




-- 
MagicBear


From nginx-forum at nginx.us  Thu Nov 17 14:25:13 2011
From: nginx-forum at nginx.us (sisif)
Date: Thu, 17 Nov 2011 09:25:13 -0500
Subject: MP4 - start time is out mp4 stts samples
In-Reply-To: <20111117050603.GA10385@nginx.com>
References: <20111117050603.GA10385@nginx.com>
Message-ID: <5e238ac1a229e14931b6ea72c107997c.NginxMailingListEnglish@forum.nginx.org>

nginx version: nginx/1.1.8
built by gcc 4.4.4 20100726 (Red Hat 4.4.4-13) (GCC)
TLS SNI support enabled
configure arguments: --prefix=/usr --sbin-path=/usr/sbin/nginx
--conf-path=/etc/nginx/nginx.conf
--error-log-path=/var/log/nginx/error.log
--pid-path=/var/run/nginx/nginx.pid --lock-path=/var/lock/nginx.lock
--user=nginx --group=nginx --add-module=/root/nginx-accesskey-2.0.3
--with-http_ssl_module --with-http_flv_module --with-http_mp4_module
--with-http_gzip_static_module --http-log-path=/var/log/nginx/access.log
--http-client-body-temp-path=/var/tmp/nginx/client/
--http-proxy-temp-path=/var/tmp/nginx/proxy/
--http-fastcgi-temp-path=/var/tmp/nginx/fcgi/

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218451,218477#msg-218477


From magicbearmo at gmail.com  Thu Nov 17 15:57:10 2011
From: magicbearmo at gmail.com (MagicBear)
Date: Thu, 17 Nov 2011 23:57:10 +0800
Subject: [MOD] proxy_cache regexp batch purge modules 0.2
In-Reply-To: <CAENo=s15bjJj8C-u430kd-Aby9vtw2_U2VpLRcDdrxMfLbFQeg@mail.gmail.com>
References: <CAENo=s15bjJj8C-u430kd-Aby9vtw2_U2VpLRcDdrxMfLbFQeg@mail.gmail.com>
Message-ID: <CAENo=s0sGqAG+x2HqxPTVHav5Ac+o3YDQ+R5oLZCfSXEPRf0KQ@mail.gmail.com>

because nginx is not threadsafe, it will cause segment failure for
this version. Bad news. Need to continue improve.

2011/11/17 MagicBear <magicbearmo at gmail.com>:
> Download
> =====
> http://m-b.cc/share/ngx_batch_purge-0.2.tar.gz
>
>
> About
> =====
> nginx_cache_batchpurge is 'nginx' module which adds regex batch purge content
> from 'proxy' caches, and is base on `ngx_cache_purge`.
>
>
> Notice
> =====
> This module is use only for batch purge, if you want to purge a single object,
> please install ngx_cache_purge module, this module can work with that module
> exists.
>
>
> Install
> =====
> ./configure --add-module=../ngx_batch_purge-0.2
>
>
> Configuration directives
> ========================
> proxy_cache_purge
> -----------------
> * **syntax**: `proxy_cache_batchpurge zone_name key`
> * **default**: `none`
> * **context**: `location`
>
> Sets area and key used for purging selected pages from `proxy`'s cache.
>
>
>
> Sample configuration
> ====================
> ? ?http {
> ? ? ? ?proxy_cache_path ?/tmp/cache ?keys_zone=tmpcache:10m;
>
> ? ? ? ?server {
> ? ? ? ? ? ?location / {
> ? ? ? ? ? ? ? ?proxy_pass ? ? ? ? http://127.0.0.1:8000;
> ? ? ? ? ? ? ? ?proxy_cache ? ? ? ?tmpcache;
> ? ? ? ? ? ? ? ?proxy_cache_key ? ?$uri$is_args$args;
> ? ? ? ? ? ?}
>
> ? ? ? ? ? ?location ~ /purge(/.*) {
> ? ? ? ? ? ? ? ?allow ? ? ? ? ? ? ?127.0.0.1;
> ? ? ? ? ? ? ? ?deny ? ? ? ? ? ? ? all;
> ? ? ? ? ? ? ? ?proxy_cache_purge ?tmpcache $1$is_args$args;
> ? ? ? ? ? ?}
>
> ? ? ? ? ? ?location ~ /regex_purge(.*) {
> ? ? ? ? ? ? ? ?allow ? ? ? ? ? ? ?127.0.0.1;
> ? ? ? ? ? ? ? ?deny ? ? ? ? ? ? ? all;
> ? ? ? ? ? ? ? ?proxy_cache_batchpurge ?tmpcache $1$is_args$args;
> ? ? ? ? ? ?}
> ? ? ? ?}
> ? ?}
>
>
>
>
> --
> MagicBear
>



-- 
MagicBear

From dieterknopf at googlemail.com  Thu Nov 17 16:50:05 2011
From: dieterknopf at googlemail.com (Dieter Knopf)
Date: Thu, 17 Nov 2011 17:50:05 +0100
Subject: Bad perfomance with nginx and php-fpm
Message-ID: <CAHb4MBh49rant83toCWjzJG+afCL2T=CAy1Dxsnm+njWBqi62A@mail.gmail.com>

Hello,

i'm trying to install nginx with php-fpm. It works, but it's not
really fast and i have problems with multiple connections.

Versions:
nginx: nginx version: nginx/1.0.10
PHP 5.3.8-1~dotdeb.2 (fpm-fcgi) (built: Aug 25 2011 13:36:54)

Configs:
VHost:
[...]
location ~ \.php$ {
  fastcgi_index index.php;
  include /etc/nginx/fastcgi_params;
  fastcgi_param  SCRIPT_FILENAME /web$fastcgi_script_name;
  fastcgi_pass unix:/tmp/foo.socket;
}
[...]

php5-fpm:
[...]
listen = '/tmp/foo.socket'
user = foo
group = foo
pm = dynamic
pm.max_children = 10
pm.start_servers = 2
pm.min_spare_servers = 1
pm.max_spare_servers = 3
chroot = /var/www/foo/
chdir = /web/
[...]

fastcgi_params:
[...]
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;
fastcgi_intercept_errors on;
[...]

I installed a fresh Wordpress without any plugins on the page and
tried ApacheBench with -n 1000 -c 10 on it:

Time taken for tests:   25.860 seconds
Complete requests:      1000
Failed requests:        875
   (Connect: 0, Receive: 0, Length: 875, Exceptions: 0)

The same page on a standard apache setup spawns 10x php5-cgi:

Time taken for tests:   104.929 seconds
Complete requests:      1000
Failed requests:        0


Not sure what's wrong :-(

Thanks

Dieter


From frumentius at gmail.com  Thu Nov 17 17:46:13 2011
From: frumentius at gmail.com (Joe)
Date: Fri, 18 Nov 2011 00:46:13 +0700
Subject: Bad perfomance with nginx and php-fpm
In-Reply-To: <CAHb4MBh49rant83toCWjzJG+afCL2T=CAy1Dxsnm+njWBqi62A@mail.gmail.com>
References: <CAHb4MBh49rant83toCWjzJG+afCL2T=CAy1Dxsnm+njWBqi62A@mail.gmail.com>
Message-ID: <CAPihCJVtHMTqPVGjnYXGDw6poLEdzx1J+C7aoeiwp+u63BYbFQ@mail.gmail.com>

How about the logs file?


Regards,
Joe


On Thu, Nov 17, 2011 at 11:50 PM, Dieter Knopf
<dieterknopf at googlemail.com>wrote:

> Hello,
>
> i'm trying to install nginx with php-fpm. It works, but it's not
> really fast and i have problems with multiple connections.
>
> Versions:
> nginx: nginx version: nginx/1.0.10
> PHP 5.3.8-1~dotdeb.2 (fpm-fcgi) (built: Aug 25 2011 13:36:54)
>
> Configs:
> VHost:
> [...]
> location ~ \.php$ {
>  fastcgi_index index.php;
>  include /etc/nginx/fastcgi_params;
>  fastcgi_param  SCRIPT_FILENAME /web$fastcgi_script_name;
>  fastcgi_pass unix:/tmp/foo.socket;
> }
> [...]
>
> php5-fpm:
> [...]
> listen = '/tmp/foo.socket'
> user = foo
> group = foo
> pm = dynamic
> pm.max_children = 10
> pm.start_servers = 2
> pm.min_spare_servers = 1
> pm.max_spare_servers = 3
> chroot = /var/www/foo/
> chdir = /web/
> [...]
>
> fastcgi_params:
> [...]
> 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;
> fastcgi_intercept_errors on;
> [...]
>
> I installed a fresh Wordpress without any plugins on the page and
> tried ApacheBench with -n 1000 -c 10 on it:
>
> Time taken for tests:   25.860 seconds
> Complete requests:      1000
> Failed requests:        875
>   (Connect: 0, Receive: 0, Length: 875, Exceptions: 0)
>
> The same page on a standard apache setup spawns 10x php5-cgi:
>
> Time taken for tests:   104.929 seconds
> Complete requests:      1000
> Failed requests:        0
>
>
> Not sure what's wrong :-(
>
> Thanks
>
> Dieter
>
> _______________________________________________
> 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/20111118/a2fb942d/attachment.html>

From jerome at loyet.net  Thu Nov 17 17:54:24 2011
From: jerome at loyet.net (=?ISO-8859-1?B?Suly9G1lIExveWV0?=)
Date: Thu, 17 Nov 2011 18:54:24 +0100
Subject: Bad perfomance with nginx and php-fpm
In-Reply-To: <CAHb4MBh49rant83toCWjzJG+afCL2T=CAy1Dxsnm+njWBqi62A@mail.gmail.com>
References: <CAHb4MBh49rant83toCWjzJG+afCL2T=CAy1Dxsnm+njWBqi62A@mail.gmail.com>
Message-ID: <CAFK6my4gcqZU1Dzq67q_gaPjCofPRQcn04y=rN2Tx5oPEkXi_A@mail.gmail.com>

2011/11/17 Dieter Knopf <dieterknopf at googlemail.com>
>
> Hello,
>
> i'm trying to install nginx with php-fpm. It works, but it's not
> really fast and i have problems with multiple connections.
>
> Versions:
> nginx: nginx version: nginx/1.0.10
> PHP 5.3.8-1~dotdeb.2 (fpm-fcgi) (built: Aug 25 2011 13:36:54)
>
> Configs:
> VHost:
> [...]
> location ~ \.php$ {
> ?fastcgi_index index.php;
> ?include /etc/nginx/fastcgi_params;
> ?fastcgi_param ?SCRIPT_FILENAME /web$fastcgi_script_name;
> ?fastcgi_pass unix:/tmp/foo.socket;
> }
> [...]
>
> php5-fpm:
> [...]
> listen = '/tmp/foo.socket'
> user = foo
> group = foo
> pm = dynamic

can you test by setting pm = static and pm.max_children to something a
little bit hight than 10 (12 or 15).
Just to ensure the problem does not come from the dynamic PM.

> pm.max_children = 10
> pm.start_servers = 2
> pm.min_spare_servers = 1
> pm.max_spare_servers = 3
> chroot = /var/www/foo/
> chdir = /web/
> [...]
>
> fastcgi_params:
> [...]
> 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;
> fastcgi_intercept_errors on;
> [...]
>
> I installed a fresh Wordpress without any plugins on the page and
> tried ApacheBench with -n 1000 -c 10 on it:
>
> Time taken for tests: ? 25.860 seconds
> Complete requests: ? ? ?1000
> Failed requests: ? ? ? ?875
> ? (Connect: 0, Receive: 0, Length: 875, Exceptions: 0)
>
> The same page on a standard apache setup spawns 10x php5-cgi:
>
> Time taken for tests: ? 104.929 seconds
> Complete requests: ? ? ?1000
> Failed requests: ? ? ? ?0
>
>
> Not sure what's wrong :-(
>
> Thanks
>
> Dieter
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx


From dieterknopf at googlemail.com  Thu Nov 17 21:59:15 2011
From: dieterknopf at googlemail.com (Dieter Knopf)
Date: Thu, 17 Nov 2011 22:59:15 +0100
Subject: Bad perfomance with nginx and php-fpm
In-Reply-To: <CAPihCJVtHMTqPVGjnYXGDw6poLEdzx1J+C7aoeiwp+u63BYbFQ@mail.gmail.com>
References: <CAHb4MBh49rant83toCWjzJG+afCL2T=CAy1Dxsnm+njWBqi62A@mail.gmail.com>
	<CAPihCJVtHMTqPVGjnYXGDw6poLEdzx1J+C7aoeiwp+u63BYbFQ@mail.gmail.com>
Message-ID: <CAHb4MBiV_vvXWipWN4MrPmL5zTaDGEy=g-Smm3UhUotS2pJZdQ@mail.gmail.com>

2011/11/17 Joe <frumentius at gmail.com>:
> How about the logs file?

I tried a new benchmark with the old settings:
Time taken for tests:   24.939 seconds
Complete requests:      1000
Failed requests:        861
   (Connect: 0, Receive: 0, Length: 861, Exceptions: 0)


In the logfile from php5-fpm i see 1000 requests (everyone with status 200):
[...]
- -  17/Nov/2011:21:55:40 +0000GET /index.php200 /www/index.php
174.837 20224 102.96%
- -  17/Nov/2011:21:55:40 +0000GET /index.php200 /www/index.php
175.963 20224 96.61%
- -  17/Nov/2011:21:55:40 +0000GET /index.php200 /www/index.php
191.407 20224 83.59%
- -  17/Nov/2011:21:55:40 +0000GET /index.php200 /www/index.php
212.139 20224 75.42%
[...]

In the nginx-access.log the same:
[...]
46.252.24.80 - - [17/Nov/2011:22:55:37 +0100] "GET / HTTP/1.0" 200
5935 "-" "ApacheBench/2.3"
46.252.24.80 - - [17/Nov/2011:22:55:37 +0100] "GET / HTTP/1.0" 200
5931 "-" "ApacheBench/2.3"
46.252.24.80 - - [17/Nov/2011:22:55:37 +0100] "GET / HTTP/1.0" 200
5933 "-" "ApacheBench/2.3"
46.252.24.80 - - [17/Nov/2011:22:55:37 +0100] "GET / HTTP/1.0" 200
5931 "-" "ApacheBench/2.3"
[...]

Thanks

Dieter


From dieterknopf at googlemail.com  Thu Nov 17 22:02:04 2011
From: dieterknopf at googlemail.com (Dieter Knopf)
Date: Thu, 17 Nov 2011 23:02:04 +0100
Subject: Bad perfomance with nginx and php-fpm
In-Reply-To: <CAFK6my4gcqZU1Dzq67q_gaPjCofPRQcn04y=rN2Tx5oPEkXi_A@mail.gmail.com>
References: <CAHb4MBh49rant83toCWjzJG+afCL2T=CAy1Dxsnm+njWBqi62A@mail.gmail.com>
	<CAFK6my4gcqZU1Dzq67q_gaPjCofPRQcn04y=rN2Tx5oPEkXi_A@mail.gmail.com>
Message-ID: <CAHb4MBhoRd+NWuxFF+av-OdCBtEsHvvKbimJP743EsH+BiH8Ow@mail.gmail.com>

2011/11/17 J?r?me Loyet <jerome at loyet.net>:
> can you test by setting pm = static and pm.max_children to something a
> little bit hight than 10 (12 or 15).
> Just to ensure the problem does not come from the dynamic PM.

Sure. I just tested it with 20 children:
Time taken for tests:   24.580 seconds
Complete requests:      1000
Failed requests:        859
   (Connect: 0, Receive: 0, Length: 859, Exceptions: 0)

The same result :-(

This sould be far better with 20 php daemons, it must be another
problem, like a timeout or something like that?

Thanks

Dieter


From nginx-forum at nginx.us  Thu Nov 17 22:33:58 2011
From: nginx-forum at nginx.us (dannynoonan)
Date: Thu, 17 Nov 2011 17:33:58 -0500
Subject: compile ngx_resty to statically link some libs?
In-Reply-To: <CAB4Tn6O9+oikUGkcKzgbd3t8ddWJprUipEPXYdBVnh99ZPX8-Q@mail.gmail.com>
References: <CAB4Tn6O9+oikUGkcKzgbd3t8ddWJprUipEPXYdBVnh99ZPX8-Q@mail.gmail.com>
Message-ID: <b3a29aad735c3c2ceddb32777056f7a5.NginxMailingListEnglish@forum.nginx.org>

I replied to this, but it seems my reply got lost. I had some responses
and follow-up questions, but I'll wait a bit to see if this BB will just
post it after a few hours.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218343,218520#msg-218520


From nginx-forum at nginx.us  Fri Nov 18 02:20:28 2011
From: nginx-forum at nginx.us (dannynoonan)
Date: Thu, 17 Nov 2011 21:20:28 -0500
Subject: compile ngx_resty to statically link some libs?
In-Reply-To: <c67cbde241dd2d8b55c8c3ef13905489.NginxMailingListEnglish@forum.nginx.org>
References: <c67cbde241dd2d8b55c8c3ef13905489.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <fbeba90b984f42ed71c8e36620412466.NginxMailingListEnglish@forum.nginx.org>

This forum can't keep up w/ my posts.

Anyway, I figured out how to build a libdrizzle.a, now I need to figure
out how to tell the ngx resty congfigure or gmake steps to slurp it in.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218343,218534#msg-218534


From nginx-forum at nginx.us  Fri Nov 18 02:52:57 2011
From: nginx-forum at nginx.us (liuzhida)
Date: Thu, 17 Nov 2011 21:52:57 -0500
Subject: How do proxy_module response buffering options work?
In-Reply-To: <20110424212203.GO56867@mdounin.ru>
References: <20110424212203.GO56867@mdounin.ru>
Message-ID: <96ec2ea88638c78172c823d078b1e63f.NginxMailingListEnglish@forum.nginx.org>

> Additionally, there is proxy_max_temp_file_size,
> which controls how 
> much data may be written to disk.  Once temp file
> size becomes 
> bigger - nginx pauses reading data from upstream
> until data from 
> temporary file is sent to client.

do you mean if a response size is larger than all the proxy buffers,
after some part of the response been written into buffers, the rest of
data which can't be written into buffers will be written into temp_file?
for every single response? the proxy_max_temp_file_size are global or
per request?
for example the upstream response is 2000 bytes. Nginx is configured
with 4 buffers, each 100 bytes in size. is that nginx deal with this
single response with written the 400 bytes of response into buffer and
written the 1600 bytes into temp_file? and after the buffer is
completely send, Nginx will read the rest of response data from the
temp_file and written into buffer? 

Thx

liuzhida

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,193347,218535#msg-218535


From nginx-forum at nginx.us  Fri Nov 18 03:06:40 2011
From: nginx-forum at nginx.us (bigplum)
Date: Thu, 17 Nov 2011 22:06:40 -0500
Subject: How to set $ into a variable?
Message-ID: <0c871be943c329ca4015f2c36d8b6b48.NginxMailingListEnglish@forum.nginx.org>

nginx.conf: 
 location / {
    ....
    set $var "$uri=200k"
    ....

and run "curl localhost/"

The module use ngx_http_get_indexed_variable() for $var will get the
string "/=200k", but I want the value is "$uri=200k". 

So any escape character supported in nginx.conf?

THANKS.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218536,218536#msg-218536


From agentzh at gmail.com  Fri Nov 18 03:07:23 2011
From: agentzh at gmail.com (agentzh)
Date: Fri, 18 Nov 2011 11:07:23 +0800
Subject: compile ngx_resty to statically link some libs?
In-Reply-To: <fbeba90b984f42ed71c8e36620412466.NginxMailingListEnglish@forum.nginx.org>
References: <c67cbde241dd2d8b55c8c3ef13905489.NginxMailingListEnglish@forum.nginx.org>
	<fbeba90b984f42ed71c8e36620412466.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <CAB4Tn6O-DYS_ur0gRPp=YhgLag71DDMt7_p087XNwwy1T_-NcQ@mail.gmail.com>

On Fri, Nov 18, 2011 at 10:20 AM, dannynoonan <nginx-forum at nginx.us> wrote:
> This forum can't keep up w/ my posts.
>
> Anyway, I figured out how to build a libdrizzle.a, now I need to figure
> out how to tell the ngx resty congfigure or gmake steps to slurp it in.
>

When you have libdrizzle.a in the right place, then passing proper
--with-ld-opt option for your system (like -static on my system) to
the configure script should be sufficient :)

Regards,
-agentzh


From edho at myconan.net  Fri Nov 18 03:13:29 2011
From: edho at myconan.net (Edho Arief)
Date: Fri, 18 Nov 2011 10:13:29 +0700
Subject: Bad perfomance with nginx and php-fpm
In-Reply-To: <CAHb4MBhoRd+NWuxFF+av-OdCBtEsHvvKbimJP743EsH+BiH8Ow@mail.gmail.com>
References: <CAHb4MBh49rant83toCWjzJG+afCL2T=CAy1Dxsnm+njWBqi62A@mail.gmail.com>
	<CAFK6my4gcqZU1Dzq67q_gaPjCofPRQcn04y=rN2Tx5oPEkXi_A@mail.gmail.com>
	<CAHb4MBhoRd+NWuxFF+av-OdCBtEsHvvKbimJP743EsH+BiH8Ow@mail.gmail.com>
Message-ID: <CABTjkKn8_GSpR08BjpDp2oVHN-Xb8k17YycVA2gtDH9HqkC+4Q@mail.gmail.com>

On Fri, Nov 18, 2011 at 5:02 AM, Dieter Knopf
<dieterknopf at googlemail.com> wrote:
>
> The same result :-(
>
> This sould be far better with 20 php daemons, it must be another
> problem, like a timeout or something like that?
>

Here's my result on my own site:

[root at eustia ~]# ab -c 10 -n 1000 http://animebsd.net/
This is ApacheBench, Version 2.3 <$Revision: 655654 $>
Copyright 1996 Adam Twiss, Zeus Technology Ltd, http://www.zeustech.net/
Licensed to The Apache Software Foundation, http://www.apache.org/

Benchmarking animebsd.net (be patient)
Completed 100 requests
Completed 200 requests
Completed 300 requests
Completed 400 requests
Completed 500 requests
Completed 600 requests
Completed 700 requests
Completed 800 requests
Completed 900 requests
Completed 1000 requests
Finished 1000 requests


Server Software:        nginx/1.1.8
Server Hostname:        animebsd.net
Server Port:            80

Document Path:          /
Document Length:        54635 bytes

Concurrency Level:      10
Time taken for tests:   261.211 seconds
Complete requests:      1000
Failed requests:        0
Write errors:           0
Total transferred:      54893000 bytes
HTML transferred:       54635000 bytes
Requests per second:    3.83 [#/sec] (mean)
Time per request:       2612.112 [ms] (mean)
Time per request:       261.211 [ms] (mean, across all concurrent requests)
Transfer rate:          205.22 [Kbytes/sec] received

Connection Times (ms)
              min  mean[+/-sd] median   max
Connect:       54   61 134.0     54    3053
Processing:  1260 2545 304.0   2515    3884
Waiting:      456 1534 219.6   1510    2673
Total:       1315 2606 328.8   2570    5409

Percentage of the requests served within a certain time (ms)
  50%   2570
  66%   2605
  75%   2652
  80%   2690
  90%   2978
  95%   3187
  98%   3270
  99%   3902
 100%   5409 (longest request)

Not exactly fast (cpu constrained - kvm instance of 1 core/thread
e3-1270 cpu, 512MB ram, debian linux 6) but certainly completed all
requests without fail. 15 php5-cgi instances ran through supervisord
in tcp mode. There are quite a bit plugins installed, too (but no
wp-cache of some kind).

No tweaking at nginx side, some timeout and limit increase on php side.

-- 
O< ascii ribbon campaign - stop html mail - www.asciiribbon.org


From jerome at loyet.net  Fri Nov 18 07:33:32 2011
From: jerome at loyet.net (=?ISO-8859-1?B?Suly9G1lIExveWV0?=)
Date: Fri, 18 Nov 2011 08:33:32 +0100
Subject: Bad perfomance with nginx and php-fpm
In-Reply-To: <CAHb4MBhoRd+NWuxFF+av-OdCBtEsHvvKbimJP743EsH+BiH8Ow@mail.gmail.com>
References: <CAHb4MBh49rant83toCWjzJG+afCL2T=CAy1Dxsnm+njWBqi62A@mail.gmail.com>
	<CAFK6my4gcqZU1Dzq67q_gaPjCofPRQcn04y=rN2Tx5oPEkXi_A@mail.gmail.com>
	<CAHb4MBhoRd+NWuxFF+av-OdCBtEsHvvKbimJP743EsH+BiH8Ow@mail.gmail.com>
Message-ID: <CAFK6my48HF7KddEps9BCQLhhCbSK5KjxfEgkcqp0i49D3kmwcQ@mail.gmail.com>

2011/11/17 Dieter Knopf <dieterknopf at googlemail.com>:
> 2011/11/17 J?r?me Loyet <jerome at loyet.net>:
>> can you test by setting pm = static and pm.max_children to something a
>> little bit hight than 10 (12 or 15).
>> Just to ensure the problem does not come from the dynamic PM.
>
> Sure. I just tested it with 20 children:
> Time taken for tests: ? 24.580 seconds
> Complete requests: ? ? ?1000
> Failed requests: ? ? ? ?859
> ? (Connect: 0, Receive: 0, Length: 859, Exceptions: 0)
>
> The same result :-(
>
> This sould be far better with 20 php daemons, it must be another
> problem, like a timeout or something like that?

what page are you requesting ?

Start by requesting a very simple PHP page (<?php echo "toto; ?>).
It's a good start to figure out what's going on.

Here is what I have with a very short and simple page  (20 static
children on FPM  PHP 5.3 trunk and nginx 1.1.4)

fat at dev:~/web$ ab -c 10 -n 1000 http://xxxxxxxxxx/pwd.php
Server Software:        nginx/1.1.4
Server Hostname:        xxxxxx
Server Port:            80

Document Path:          /pwd.php
Document Length:        22 bytes

Concurrency Level:      10
Time taken for tests:   0.649 seconds
Complete requests:      1000
Failed requests:        0
Write errors:           0
Total transferred:      175000 bytes
HTML transferred:       22000 bytes
Requests per second:    1539.96 [#/sec] (mean)
Time per request:       6.494 [ms] (mean)
Time per request:       0.649 [ms] (mean, across all concurrent requests)
Transfer rate:          263.18 [Kbytes/sec] received

Connection Times (ms)
              min  mean[+/-sd] median   max
Connect:        0    0   0.4      0       5
Processing:     0    6   2.6      6      37
Waiting:        0    6   2.6      6      37
Total:          1    6   2.9      6      42

Percentage of the requests served within a certain time (ms)
  50%      6
  66%      6
  75%      7
  80%      7
  90%      7
  95%      7
  98%     12
  99%     19
 100%     42 (longest request)


Then increase the size the PHP page returns (ex: 64k --> <?php echo
str_pad("", 1024*64, "x"); ?>)

fat at dev:~/web$ ab -c 10 -n 1000 http://xxxxxxx/size.php
Server Software:        nginx/1.1.4
Server Hostname:        xxxxxxx
Server Port:            80

Document Path:          /size.php
Document Length:        65536 bytes

Concurrency Level:      10
Time taken for tests:   1.242 seconds
Complete requests:      1000
Failed requests:        0
Write errors:           0
Total transferred:      65689000 bytes
HTML transferred:       65536000 bytes
Requests per second:    805.17 [#/sec] (mean)
Time per request:       12.420 [ms] (mean)
Time per request:       1.242 [ms] (mean, across all concurrent requests)
Transfer rate:          51651.05 [Kbytes/sec] received

Connection Times (ms)
              min  mean[+/-sd] median   max
Connect:        0    0   0.4      0       5
Processing:     1   12   2.9     12      34
Waiting:        1   12   2.9     12      32
Total:          1   12   3.0     12      37

Percentage of the requests served within a certain time (ms)
  50%     12
  66%     13
  75%     14
  80%     14
  90%     16
  95%     17
  98%     18
  99%     20
 100%     37 (longest request)

Then increase the complexity (size, CPU, memory, fork, database requests, ...)

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


From ru at nginx.com  Fri Nov 18 08:53:46 2011
From: ru at nginx.com (Ruslan Ermilov)
Date: Fri, 18 Nov 2011 12:53:46 +0400
Subject: Upgrade to 1.1.8 still shows 1.1.7?
In-Reply-To: <20111115172731.GG95664@mdounin.ru>
References: <CAOoWpPrY1+T=xe2L9eL3GjSK25qHbGS38xXaV3SY8qEbkiM9Aw@mail.gmail.com>
	<20111115160929.GR27078@craic.sysops.org>
	<CAOoWpPqUma5V6xu5zMMtPvN3xaKHJniPOwVnsDdPd1iuGJZB1Q@mail.gmail.com>
	<20111115172731.GG95664@mdounin.ru>
Message-ID: <20111118085346.GB29345@lo0.su>

On Tue, Nov 15, 2011 at 09:27:32PM +0400, Maxim Dounin wrote:
> Hello!
> 
> On Tue, Nov 15, 2011 at 11:18:25AM -0500, Ilan Berkner wrote:
> 
> > I actually had to do a full restart and now it works.
> > 
> > Previously kill -HUP masterpid worked, this time it didn't, not sure why.
> 
> kill -HUP isn't expected to upgrade nginx binary, upgrade 
> procedure is outlined here:
> 
> http://wiki.nginx.org/CommandLine#Upgrading_To_a_New_Binary_On_The_Fly

Please also make yourself familiar with the "Controlling nginx"
article that has just appeared on the website:

http://nginx.org/en/docs/control.html


From nginx-forum at nginx.us  Fri Nov 18 11:55:40 2011
From: nginx-forum at nginx.us (janedenone)
Date: Fri, 18 Nov 2011 06:55:40 -0500
Subject: Using ab to benchmark nginx: Connection reset by peer (54)
Message-ID: <f19d913bc3ccc866681516f52d3cb7c0.NginxMailingListEnglish@forum.nginx.org>

I recently updated to nginx 1.0.8 and tried to benchmark performance for
cached dynamic pages (initially served by a Django app via proxy_pass)
and for static pages. In both cases, nginx will not serve more than 3 or
4 requests (even without concurrent connections), so ab almost
immediately reports:

Benchmarking testsite.static (be patient)...apr_socket_recv: Connection
reset by peer (54)

It is only when choosing a maximum of 4 (or fewer) requests that ab
finishes successfully.

Why is that? I tried increasing the number of worker processes (no
luck), but I assume that nginx should be capable of serving more than 4
requests without tweaking any configuration variable. Could it be that I
accidentally triggered some sort of DOS protection mechanism?

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218554,218554#msg-218554


From magicbearmo at gmail.com  Fri Nov 18 13:58:37 2011
From: magicbearmo at gmail.com (MagicBear)
Date: Fri, 18 Nov 2011 21:58:37 +0800
Subject: How to set $ into a variable?
In-Reply-To: <0c871be943c329ca4015f2c36d8b6b48.NginxMailingListEnglish@forum.nginx.org>
References: <0c871be943c329ca4015f2c36d8b6b48.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <CAENo=s3OmSC4iunK=tpb8PeQN6uqJ2_5r31WB0q+h9KvwTLMMQ@mail.gmail.com>

may be try set $var "\$uri=200k"

2011/11/18 bigplum <nginx-forum at nginx.us>:
> nginx.conf:
> ?location / {
> ? ?....
> ? ?set $var "$uri=200k"
> ? ?....
>
> and run "curl localhost/"
>
> The module use ngx_http_get_indexed_variable() for $var will get the
> string "/=200k", but I want the value is "$uri=200k".
>
> So any escape character supported in nginx.conf?
>
> THANKS.
>
> Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218536,218536#msg-218536
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx
>



-- 
MagicBear


From mdounin at mdounin.ru  Fri Nov 18 14:00:18 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Fri, 18 Nov 2011 18:00:18 +0400
Subject: How do proxy_module response buffering options work?
In-Reply-To: <96ec2ea88638c78172c823d078b1e63f.NginxMailingListEnglish@forum.nginx.org>
References: <20110424212203.GO56867@mdounin.ru>
	<96ec2ea88638c78172c823d078b1e63f.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20111118140018.GB95664@mdounin.ru>

Hello!

On Thu, Nov 17, 2011 at 09:52:57PM -0500, liuzhida wrote:

> > Additionally, there is proxy_max_temp_file_size,
> > which controls how 
> > much data may be written to disk.  Once temp file
> > size becomes 
> > bigger - nginx pauses reading data from upstream
> > until data from 
> > temporary file is sent to client.
> 
> do you mean if a response size is larger than all the proxy buffers,
> after some part of the response been written into buffers, the rest of
> data which can't be written into buffers will be written into temp_file?

Yes, as long as it's not possible to write data to client fast 
enough.

> for every single response? the proxy_max_temp_file_size are global or
> per request?

Per request.  It limits maximum size of a temporary file used to 
buffer a request.

> for example the upstream response is 2000 bytes. Nginx is configured
> with 4 buffers, each 100 bytes in size. is that nginx deal with this
> single response with written the 400 bytes of response into buffer and
> written the 1600 bytes into temp_file?

Roughly yes.  (There are some nuances though, as you can't write 
data to file without using in-memory buffers, and that's why some 
memory buffers will be reserved for reading response from upstream 
and writing it to disk.  And again, this all assumes it's not 
possible to write anything to client, while 2000 bytes usually 
just fit into socket buffer and will be passed to kernel 
immediately even if client isn't reading at all.)

> and after the buffer is
> completely send, Nginx will read the rest of response data from the
> temp_file and written into buffer? 

It will follow usual procedure to send file-backed data (the same 
one which is used for static files), i.e either it will use 
sendfile() or read data to output_buffers and send them.

Maxim Dounin


From mdounin at mdounin.ru  Fri Nov 18 14:10:32 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Fri, 18 Nov 2011 18:10:32 +0400
Subject: Using ab to benchmark nginx: Connection reset by peer (54)
In-Reply-To: <f19d913bc3ccc866681516f52d3cb7c0.NginxMailingListEnglish@forum.nginx.org>
References: <f19d913bc3ccc866681516f52d3cb7c0.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20111118141032.GC95664@mdounin.ru>

Hello!

On Fri, Nov 18, 2011 at 06:55:40AM -0500, janedenone wrote:

> I recently updated to nginx 1.0.8 and tried to benchmark performance for
> cached dynamic pages (initially served by a Django app via proxy_pass)
> and for static pages. In both cases, nginx will not serve more than 3 or
> 4 requests (even without concurrent connections), so ab almost
> immediately reports:
> 
> Benchmarking testsite.static (be patient)...apr_socket_recv: Connection
> reset by peer (54)
> 
> It is only when choosing a maximum of 4 (or fewer) requests that ab
> finishes successfully.
> 
> Why is that? I tried increasing the number of worker processes (no
> luck), but I assume that nginx should be capable of serving more than 4
> requests without tweaking any configuration variable. Could it be that I
> accidentally triggered some sort of DOS protection mechanism?

By default there are no anti-DoS protections are activated in 
nginx.  If you've activated some - it may be the reason.  Looking 
into error may shed some light on the issue.

Alternatively, it may be some other limits, e.g. your firewall 
ones.

Maxim Dounin


From mdounin at mdounin.ru  Fri Nov 18 14:16:12 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Fri, 18 Nov 2011 18:16:12 +0400
Subject: How to set $ into a variable?
In-Reply-To: <0c871be943c329ca4015f2c36d8b6b48.NginxMailingListEnglish@forum.nginx.org>
References: <0c871be943c329ca4015f2c36d8b6b48.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20111118141612.GD95664@mdounin.ru>

Hello!

On Thu, Nov 17, 2011 at 10:06:40PM -0500, bigplum wrote:

> nginx.conf: 
>  location / {
>     ....
>     set $var "$uri=200k"
>     ....
> 
> and run "curl localhost/"
> 
> The module use ngx_http_get_indexed_variable() for $var will get the
> string "/=200k", but I want the value is "$uri=200k". 
> 
> So any escape character supported in nginx.conf?

Right now there are no way to escape $ in arguments which support 
variables.  This is a bug.

Possible workaround is to use some variable defined with module 
which doesn't support variables, e.g. you may do so with geo:

    geo $dollar {
        default  "$";
    } 

    ...
    set $var "${dollar}uri=200k"
    ...

Maxim Dounin


From lists at ruby-forum.com  Fri Nov 18 14:27:44 2011
From: lists at ruby-forum.com (Noah C.)
Date: Fri, 18 Nov 2011 15:27:44 +0100
Subject: DNS TTLs being ignored
In-Reply-To: <CAAqHd6v_LxnOpfoE6Fq+sKebtwHiguse4d6GmmvpAUayR5eqaA@mail.gmail.com>
References: <CAAqHd6v_LxnOpfoE6Fq+sKebtwHiguse4d6GmmvpAUayR5eqaA@mail.gmail.com>
Message-ID: <21d1c4e19dc65e2ae861c2506d937a30@ruby-forum.com>

That's great news. Thank you very much. I'll be sure to get hold of it 
as soon as 1.1.9 is released.

--Noah

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


From nginx-forum at nginx.us  Fri Nov 18 14:39:05 2011
From: nginx-forum at nginx.us (janedenone)
Date: Fri, 18 Nov 2011 09:39:05 -0500
Subject: Conflict: index and try_files
Message-ID: <12b16bcca767bd5986409e54ad6bad9c.NginxMailingListEnglish@forum.nginx.org>

In the following simple configuration

server {
	server_name testsite.static;
	root /some/path/;
	index blog.html;
	try_files $uri $uri.html =404;
}

the try_files directive seems to interfere with the index directive: I
always get 404 for http://testsite.static/. If I remove the fallback
part of try_files, the same request results in an error:

 *990 rewrite or internal redirection cycle while internal redirect to
"/.html.html.html....

The file blog.html is present in /some/path/, so I really do not
understand why the file is not found/served for the above URL.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218562,218562#msg-218562


From nginx-forum at nginx.us  Fri Nov 18 14:41:40 2011
From: nginx-forum at nginx.us (janedenone)
Date: Fri, 18 Nov 2011 09:41:40 -0500
Subject: Using ab to benchmark nginx: Connection reset by peer (54)
In-Reply-To: <20111118141032.GC95664@mdounin.ru>
References: <20111118141032.GC95664@mdounin.ru>
Message-ID: <ebbeb6afe6c4617d063d1ee5587066dc.NginxMailingListEnglish@forum.nginx.org>

Thanks ??there is no firewall (the test is run on my local machine),
and the error.log shows nothing. I increased the log level, only to see
notifications about exiting worker and cache processes.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218554,218563#msg-218563


From igor at sysoev.ru  Fri Nov 18 14:42:02 2011
From: igor at sysoev.ru (Igor Sysoev)
Date: Fri, 18 Nov 2011 18:42:02 +0400
Subject: Conflict: index and try_files
In-Reply-To: <12b16bcca767bd5986409e54ad6bad9c.NginxMailingListEnglish@forum.nginx.org>
References: <12b16bcca767bd5986409e54ad6bad9c.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20111118144202.GA47997@nginx.com>

On Fri, Nov 18, 2011 at 09:39:05AM -0500, janedenone wrote:
> In the following simple configuration
> 
> server {
> 	server_name testsite.static;
> 	root /some/path/;
> 	index blog.html;
> 	try_files $uri $uri.html =404;
> }
> 
> the try_files directive seems to interfere with the index directive: I
> always get 404 for http://testsite.static/. If I remove the fallback
> part of try_files, the same request results in an error:
> 
>  *990 rewrite or internal redirection cycle while internal redirect to
> "/.html.html.html....
> 
> The file blog.html is present in /some/path/, so I really do not
> understand why the file is not found/served for the above URL.

 	try_files $uri $uri/ $uri.html =404;


-- 
Igor Sysoev


From nginx-forum at nginx.us  Fri Nov 18 14:46:45 2011
From: nginx-forum at nginx.us (janedenone)
Date: Fri, 18 Nov 2011 09:46:45 -0500
Subject: Conflict: index and try_files
In-Reply-To: <20111118144202.GA47997@nginx.com>
References: <20111118144202.GA47997@nginx.com>
Message-ID: <0ebad7d446696b257a008755595ce0cf.NginxMailingListEnglish@forum.nginx.org>

Excellent, this works.

Thanks a lot.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218562,218567#msg-218567


From mdounin at mdounin.ru  Fri Nov 18 14:50:27 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Fri, 18 Nov 2011 18:50:27 +0400
Subject: Conflict: index and try_files
In-Reply-To: <12b16bcca767bd5986409e54ad6bad9c.NginxMailingListEnglish@forum.nginx.org>
References: <12b16bcca767bd5986409e54ad6bad9c.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20111118145027.GG95664@mdounin.ru>

Hello!

On Fri, Nov 18, 2011 at 09:39:05AM -0500, janedenone wrote:

> In the following simple configuration
> 
> server {
> 	server_name testsite.static;
> 	root /some/path/;
> 	index blog.html;
> 	try_files $uri $uri.html =404;
> }
> 
> the try_files directive seems to interfere with the index directive: I
> always get 404 for http://testsite.static/.

If you want try_files to test directories as well (and to allow 
index directive to work), you have to explicitly specify "/" at 
the end of name.  I.e. use something like this:

   try_files $uri/ $uri $uri.html =404;

> If I remove the fallback
> part of try_files, the same request results in an error:

If you remove "=404" from the above try_files, it will become

    try_files $uri $uri.html;

i.e. if no file $uri found, it will redirect to $uri.html (/.html) 
in your case.  In it's turn it won't be found and will again 
redirect to $uri.html (/.html.html now).  The

>  *990 rewrite or internal redirection cycle while internal redirect to
> "/.html.html.html....

is expected.

> The file blog.html is present in /some/path/, so I really do not
> understand why the file is not found/served for the above URL.

That's because it tests for files, and rejects directories by 
default.  Test for directories must by explicitly requested.

See here for more information:

http://nginx.org/en/docs/http/ngx_http_core_module.html#try_files

Maxim Dounin


From agentzh at gmail.com  Fri Nov 18 15:25:56 2011
From: agentzh at gmail.com (agentzh)
Date: Fri, 18 Nov 2011 23:25:56 +0800
Subject: Using ab to benchmark nginx: Connection reset by peer (54)
In-Reply-To: <f19d913bc3ccc866681516f52d3cb7c0.NginxMailingListEnglish@forum.nginx.org>
References: <f19d913bc3ccc866681516f52d3cb7c0.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <CAB4Tn6P5Gas11en3wH00YFS0nj5kw1r9Gxiiow0uB2+++m_PDg@mail.gmail.com>

On Fri, Nov 18, 2011 at 7:55 PM, janedenone <nginx-forum at nginx.us> wrote:
> I recently updated to nginx 1.0.8 and tried to benchmark performance for
> cached dynamic pages (initially served by a Django app via proxy_pass)
> and for static pages. In both cases, nginx will not serve more than 3 or
> 4 requests (even without concurrent connections), so ab almost
> immediately reports:
>
> Benchmarking testsite.static (be patient)...apr_socket_recv: Connection
> reset by peer (54)
>

What kind of system are you in? ab is known to have such issues in
certain Mac OS X boxes to my knowledge :)

Regards,
-agentzh


From nginx-forum at nginx.us  Fri Nov 18 15:36:25 2011
From: nginx-forum at nginx.us (janedenone)
Date: Fri, 18 Nov 2011 10:36:25 -0500
Subject: Using ab to benchmark nginx: Connection reset by peer (54)
In-Reply-To: <CAB4Tn6P5Gas11en3wH00YFS0nj5kw1r9Gxiiow0uB2+++m_PDg@mail.gmail.com>
References: <CAB4Tn6P5Gas11en3wH00YFS0nj5kw1r9Gxiiow0uB2+++m_PDg@mail.gmail.com>
Message-ID: <91ffb52587e559a9a14884aa20c25e4a.NginxMailingListEnglish@forum.nginx.org>

agentzh Wrote:
> What kind of system are you in? ab is known to
> have such issues in
> certain Mac OS X boxes to my knowledge :)

Well ? Mac OS X. :) Are these issues to the ab version that comes with
OS X, or are they related to the OS in general? In the former case, I'd
just recompile ab manually.

Thanks!

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218554,218571#msg-218571


From agentzh at gmail.com  Fri Nov 18 15:41:09 2011
From: agentzh at gmail.com (agentzh)
Date: Fri, 18 Nov 2011 23:41:09 +0800
Subject: Using ab to benchmark nginx: Connection reset by peer (54)
In-Reply-To: <91ffb52587e559a9a14884aa20c25e4a.NginxMailingListEnglish@forum.nginx.org>
References: <CAB4Tn6P5Gas11en3wH00YFS0nj5kw1r9Gxiiow0uB2+++m_PDg@mail.gmail.com>
	<91ffb52587e559a9a14884aa20c25e4a.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <CAB4Tn6NxfRkfoKX24i6uetmspHA6ROdbLM9KQUeUAFEa3WwwvQ@mail.gmail.com>

On Fri, Nov 18, 2011 at 11:36 PM, janedenone <nginx-forum at nginx.us> wrote:
> agentzh Wrote:
>> What kind of system are you in? ab is known to
>> have such issues in
>> certain Mac OS X boxes to my knowledge :)
>
> Well ? Mac OS X. :) Are these issues to the ab version that comes with
> OS X, or are they related to the OS in general? In the former case, I'd
> just recompile ab manually.
>

On Mac OS X 10.6, I have to use "127.0.0.1" instead of "localhost"
while running the ab command, or I'll get exactly the same error
message.

And for 10.7 (Lion), people say they have to apply a patch to ab to get it work.

Hope this helps,
-agentzh


From nginx-forum at nginx.us  Fri Nov 18 15:57:04 2011
From: nginx-forum at nginx.us (ceh329)
Date: Fri, 18 Nov 2011 10:57:04 -0500
Subject: rewrite rule that alters content results
Message-ID: <61d14ebe4c4d548478be71f47354c6cf.NginxMailingListEnglish@forum.nginx.org>

Hello,
Is there a rewrite statement that I can use to rewrite the html content
that is being sent back to the client? Some of the content has urls in
it and they have the wrong port I would like to change it as it's being
sent back to the client.

Thanks,
Charles

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218573,218573#msg-218573


From brian at akins.org  Fri Nov 18 16:38:30 2011
From: brian at akins.org (Brian Akins)
Date: Fri, 18 Nov 2011 11:38:30 -0500
Subject: Filter_cache module released
Message-ID: <CAO3qTJUyJ0nTd7wEVXOrzYY7f5gcoLPwOu5vX5FpK5CWxqUDvA@mail.gmail.com>

https://github.com/bakins/ngx_http_filter_cache

Tested with up to nginx 1.0.6

Note: this currently doesn't work when the ctx gets reset on named
locations.  The quickest fix I have is to patch the request struct to add a
filter_cache pointer.

Filter cache caches full responses after the results of filters such as SSI
and gzip.

It was the first nginx module we ever wrote and it grew organically, so the
code is krufty.

--Brian
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20111118/84e8baf3/attachment.html>

From nginx-forum at nginx.us  Fri Nov 18 16:50:53 2011
From: nginx-forum at nginx.us (dbanks)
Date: Fri, 18 Nov 2011 11:50:53 -0500
Subject: gzip - unexplained side effects
In-Reply-To: <48ec591068896e53321f58945d5a8836.NginxMailingListEnglish@forum.nginx.org>
References: <48ec591068896e53321f58945d5a8836.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <4ae63147b1d2837011736899194e07a6.NginxMailingListEnglish@forum.nginx.org>

I spent more time testing this particular issue.  I believe that what
appeared to be lost traffic is simply due to the shortened keepalives
and the load balancer favoring keepalive connections over new
connections.  However, there still seems to be a link between gzip
settings and the number of open connections (keepalives). 

gzip on;
gzip_comp_level 1;
gzip_types text/javascript text/plain application/x-javascript;
gzip_disable "MSIE [1-6]\.(?!.*SV1)"
#gzip_buffers 64 4k;
#gzip_min_length 1100; #if it fits in one packet, no worries
#gzip_http_version 1.1;

I have confirmed that if any one of the three config lines at the bottom
(the ones that are commented out) are present, the number of open
connections drops from ~7600 to ~1300.  Removing that line from the
config and reloading restores normal operation.  Strange.

Cheers,

Dean

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218044,218575#msg-218575


From mdounin at mdounin.ru  Fri Nov 18 17:40:06 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Fri, 18 Nov 2011 21:40:06 +0400
Subject: gzip - unexplained side effects
In-Reply-To: <4ae63147b1d2837011736899194e07a6.NginxMailingListEnglish@forum.nginx.org>
References: <48ec591068896e53321f58945d5a8836.NginxMailingListEnglish@forum.nginx.org>
	<4ae63147b1d2837011736899194e07a6.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20111118174006.GH95664@mdounin.ru>

Hello!

On Fri, Nov 18, 2011 at 11:50:53AM -0500, dbanks wrote:

> I spent more time testing this particular issue.  I believe that what
> appeared to be lost traffic is simply due to the shortened keepalives
> and the load balancer favoring keepalive connections over new
> connections.  However, there still seems to be a link between gzip
> settings and the number of open connections (keepalives). 
> 
> gzip on;
> gzip_comp_level 1;
> gzip_types text/javascript text/plain application/x-javascript;
> gzip_disable "MSIE [1-6]\.(?!.*SV1)"

It looks like you've missed ";" here.

This is probably a reason for all "strange" effects you observe - 
missing ";" causes "gzip_disable" to eat next directive in config.

> #gzip_buffers 64 4k;
> #gzip_min_length 1100; #if it fits in one packet, no worries
> #gzip_http_version 1.1;
> 
> I have confirmed that if any one of the three config lines at the bottom
> (the ones that are commented out) are present, the number of open
> connections drops from ~7600 to ~1300.  Removing that line from the
> config and reloading restores normal operation.  Strange.

Maxim Dounin


From nginx-forum at nginx.us  Fri Nov 18 17:44:55 2011
From: nginx-forum at nginx.us (ceh329)
Date: Fri, 18 Nov 2011 12:44:55 -0500
Subject: Alter Config On Startup
In-Reply-To: <a26a3867c7b747eddea0ab04530b8e64.NginxMailingListEnglish@forum.nginx.org>
References: <a26a3867c7b747eddea0ab04530b8e64.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <8b3cfc0ca735733f4c1530570821f181.NginxMailingListEnglish@forum.nginx.org>

Thanks, was not what I was hoping for but I'll build it in to a startup
script.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218180,218577#msg-218577


From nginx-forum at nginx.us  Fri Nov 18 19:00:44 2011
From: nginx-forum at nginx.us (dbanks)
Date: Fri, 18 Nov 2011 14:00:44 -0500
Subject: gzip - unexplained side effects
In-Reply-To: <4ae63147b1d2837011736899194e07a6.NginxMailingListEnglish@forum.nginx.org>
References: <48ec591068896e53321f58945d5a8836.NginxMailingListEnglish@forum.nginx.org>
	<4ae63147b1d2837011736899194e07a6.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <745955042a91e67e21ea83f48d8c74db.NginxMailingListEnglish@forum.nginx.org>

Hello Maxim!

Mystery solved.  I cannot believe that I missed that!

Cheers,

Dean

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218044,218580#msg-218580


From nginx-forum at nginx.us  Fri Nov 18 23:31:51 2011
From: nginx-forum at nginx.us (Salem)
Date: Fri, 18 Nov 2011 18:31:51 -0500
Subject: php and urls with /?xxx
Message-ID: <ccd73c97fbbbcf584db361b9f73de67e.NginxMailingListEnglish@forum.nginx.org>

Ahoi,

at the moment i use a "location \.php {}" block for opening php-files
through fastcgi_pass. But what's the right way for
calls like www.xxx.xxx/?xxx=xxx ?

For Wordpress-Urls like /index.php/xxx/ 
try_files $uri $uri/ /index.php?q=$request_uri;
works fine, but for /?

Thank you!

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218588,218588#msg-218588


From appa at perusio.net  Sat Nov 19 01:38:13 2011
From: appa at perusio.net (=?UTF-8?B?QW50w7NuaW8=?= P. P. Almeida)
Date: Sat, 19 Nov 2011 01:38:13 +0000
Subject: php and urls with /?xxx
In-Reply-To: <ccd73c97fbbbcf584db361b9f73de67e.NginxMailingListEnglish@forum.nginx.org>
References: <ccd73c97fbbbcf584db361b9f73de67e.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <87vcqgriwq.wl%appa@perusio.net>

On 18 Nov 2011 23h31 WET, nginx-forum at nginx.us wrote:

> Ahoi,
>
> at the moment i use a "location \.php {}" block for opening
> php-files through fastcgi_pass. But what's the right way for calls
> like www.xxx.xxx/?xxx=xxx ?
>
> For Wordpress-Urls like /index.php/xxx/ 
> try_files $uri $uri/ /index.php?q=$request_uri;
> works fine, but for /?

Try this:

        ## Regular PHP processing.
        location ~ ^(?<script>.+\.php)(?<path_info>.*)$ {
            include fastcgi.conf;
            ## The fastcgi_params must be redefined from the ones
            ## given in fastcgi.conf. No longer standard names
            ## but arbitrary: named patterns in regex.
            fastcgi_param SCRIPT_FILENAME $document_root$script;
            fastcgi_param SCRIPT_NAME $script;
            fastcgi_param PATH_INFO $path_info;
            ## Passing the request upstream to the FastCGI
            ## listener.
            fastcgi_pass phpcgi;
        }

Combined with:

        location / {
           try_files $uri $uri/ /index.php?q=$uri&$args;
        }

Of course you have to fill in the details on static file handling,
blocking the readme.html file, etc.

--- appa


From nginx-forum at nginx.us  Sat Nov 19 03:04:48 2011
From: nginx-forum at nginx.us (Salem)
Date: Fri, 18 Nov 2011 22:04:48 -0500
Subject: php and urls with /?xxx
In-Reply-To: <87vcqgriwq.wl%appa@perusio.net>
References: <87vcqgriwq.wl%appa@perusio.net>
Message-ID: <b5afe6b9807a540046dd429afe480799.NginxMailingListEnglish@forum.nginx.org>

Ant?nio P. P. Almeida Wrote:
-------------------------------------------------------
> On 18 Nov 2011 23h31 WET, nginx-forum at nginx.us
> wrote:

> Try this:
> 
>  [...]

Thanks, but i alway get 500 error.

I'm using php-fpm with chroot-config, not sure what's the right setting
for this.

For "location \.php" i'm using: 
fastcgi_param  SCRIPT_FILENAME /htdocs$fastcgi_script_name;

> Of course you have to fill in the details on
> static file handling,
> blocking the readme.html file, etc.

Is there really no easier way with the same behavior as apache?
/?xxx=xxx is alway /index.php?xxx=xxx :-/

Thanks

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218588,218590#msg-218590


From edho at myconan.net  Sat Nov 19 03:10:40 2011
From: edho at myconan.net (Edho Arief)
Date: Sat, 19 Nov 2011 10:10:40 +0700
Subject: php and urls with /?xxx
In-Reply-To: <b5afe6b9807a540046dd429afe480799.NginxMailingListEnglish@forum.nginx.org>
References: <87vcqgriwq.wl%appa@perusio.net>
	<b5afe6b9807a540046dd429afe480799.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <CABTjkKm7qA3+57viZMS+Gzz9f8t8x4m=T95etYWbTN5gK-SGdA@mail.gmail.com>

On Sat, Nov 19, 2011 at 10:04 AM, Salem <nginx-forum at nginx.us> wrote:
>
> Is there really no easier way with the same behavior as apache?
> /?xxx=xxx is alway /index.php?xxx=xxx :-/
>

It's always like that unless you modify your fastcgi_params.

-- 
O< ascii ribbon campaign - stop html mail - www.asciiribbon.org


From edho at myconan.net  Sat Nov 19 03:13:42 2011
From: edho at myconan.net (Edho Arief)
Date: Sat, 19 Nov 2011 10:13:42 +0700
Subject: php and urls with /?xxx
In-Reply-To: <CABTjkKm7qA3+57viZMS+Gzz9f8t8x4m=T95etYWbTN5gK-SGdA@mail.gmail.com>
References: <87vcqgriwq.wl%appa@perusio.net>
	<b5afe6b9807a540046dd429afe480799.NginxMailingListEnglish@forum.nginx.org>
	<CABTjkKm7qA3+57viZMS+Gzz9f8t8x4m=T95etYWbTN5gK-SGdA@mail.gmail.com>
Message-ID: <CABTjkKnk3ZqaX_kApYSH3YDc2Jr5CDdxj=1Eov=HksXcHVccMw@mail.gmail.com>

On Sat, Nov 19, 2011 at 10:10 AM, Edho Arief <edho at myconan.net> wrote:
> On Sat, Nov 19, 2011 at 10:04 AM, Salem <nginx-forum at nginx.us> wrote:
>>
>> Is there really no easier way with the same behavior as apache?
>> /?xxx=xxx is alway /index.php?xxx=xxx :-/
>>
>
> It's always like that unless you modify your fastcgi_params.
>

Also you should remove index.php part from your permalink structure in
wordpress.

-- 
O< ascii ribbon campaign - stop html mail - www.asciiribbon.org


From nginx-forum at nginx.us  Sat Nov 19 04:30:39 2011
From: nginx-forum at nginx.us (cowboy129)
Date: Fri, 18 Nov 2011 23:30:39 -0500
Subject: how to get uid_set by userid module in php
In-Reply-To: <a107cfe4a7cca07e19b92961ff8eca04.NginxMailingListEnglish@forum.nginx.org>
References: <d6298f322ed41aaa5e0346fa1151ba6e.NginxMailingListEnglish@forum.nginx.org>
	<a107cfe4a7cca07e19b92961ff8eca04.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <75aa9918a518a2863967c15751a24ae0.NginxMailingListEnglish@forum.nginx.org>

$bin = base64_decode('wKiAg0uMcKcZqg3nAwMDAg==');
$uid = join(array_map(function($i){return strtoupper(dechex($i));},
unpack("L*", $bin)), "");
print $uid;

see also http://forum.nginx.org/read.php?2,215035

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,58522,218594#msg-218594


From nginx-forum at nginx.us  Sat Nov 19 13:56:28 2011
From: nginx-forum at nginx.us (Salem)
Date: Sat, 19 Nov 2011 08:56:28 -0500
Subject: php and urls with /?xxx
In-Reply-To: <ccd73c97fbbbcf584db361b9f73de67e.NginxMailingListEnglish@forum.nginx.org>
References: <ccd73c97fbbbcf584db361b9f73de67e.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <152dbc32abcab0984c99bb3d7bef5ad7.NginxMailingListEnglish@forum.nginx.org>

Thanks for the help. 

Salem

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218588,218606#msg-218606


From francis at daoine.org  Sat Nov 19 15:46:54 2011
From: francis at daoine.org (Francis Daly)
Date: Sat, 19 Nov 2011 15:46:54 +0000
Subject: php and urls with /?xxx
In-Reply-To: <ccd73c97fbbbcf584db361b9f73de67e.NginxMailingListEnglish@forum.nginx.org>
References: <ccd73c97fbbbcf584db361b9f73de67e.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20111119154654.GS27078@craic.sysops.org>

On Fri, Nov 18, 2011 at 06:31:51PM -0500, Salem wrote:

Hi there,

> For Wordpress-Urls like /index.php/xxx/ 
> try_files $uri $uri/ /index.php?q=$request_uri;
> works fine, but for /?

The "location" goes from the / to just before the first ? or #.

You want this to be handled as if it were /index.php?xxx=xxx. So do
just that.

  location = / {
    fastcgi_pass  unix:php.sock;
    include fastcgi.conf;
    fastcgi_param SCRIPT_FILENAME /htdocs/index.php;
  }

The SCRIPT_FILENAME would normally be $document_root/index.php, but you
mention elsewhere that chroot is used. SCRIPT_FILENAME is "the name of
the file the php interpreter should load, from its perspective".

(If you did fastcgi_pass to a different server, the filename on
that server would be what you would send. Which may be unrelated to
$document_root.)


Maybe just adding

  index index.php;

within the server{} block instead, would be sufficient.

Good luck with it,

	f
-- 
Francis Daly        francis at daoine.org


From nginx-forum at nginx.us  Sat Nov 19 18:46:54 2011
From: nginx-forum at nginx.us (TECK)
Date: Sat, 19 Nov 2011 13:46:54 -0500
Subject: Enable SSL for a specific directory?
Message-ID: <e7ec2721b716a6e58af97a071d3162bf.NginxMailingListEnglish@forum.nginx.org>

Hi,

I'm trying to enable SSL for a specific directory only. In other words,
the / directory is not encrypted, while the /protected is.

server {
	listen				192.168.1.2:80 default_server;
	server_name			www.domain.com;
	root				/var/www/html;

	location / {
		try_files		$uri $uri/ index.php?q=$uri&$args;
	}

	location /protected/ {
		rewrite ^		https://www.domain.com$request_uri? permanent;
	}

	...
}

server {
	listen				192.168.1.2:443 ssl;
	server_name			www.domain.com;
	root				/var/www/html;

	location / {
		rewrite ^		http://www.domain.com$request_uri? permanent;
	}

	location /protected/ {
		try_files		$uri $uri/ /index.php?q=$uri&$args;
	}

	...
}

Right now, I get a slashed https into browser, with the message:
Your connection to www.domain.com is encrypted with 256-bit encryption.
However, this page includes other resources which are not secure. These
resources can be viewed by others while in transit and can be modified
by an attacker to change the behavior of the page.

The /protected directory is a "virtual" one that does not exist in
reality. It is created by /index.php with some SEO technique.
In real life, the URL format is: index.php?protected

Thanks for your help.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218610,218610#msg-218610


From mdounin at mdounin.ru  Sat Nov 19 21:01:25 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Sun, 20 Nov 2011 01:01:25 +0400
Subject: Enable SSL for a specific directory?
In-Reply-To: <e7ec2721b716a6e58af97a071d3162bf.NginxMailingListEnglish@forum.nginx.org>
References: <e7ec2721b716a6e58af97a071d3162bf.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20111119210125.GJ95664@mdounin.ru>

Hello!

On Sat, Nov 19, 2011 at 01:46:54PM -0500, TECK wrote:

[...]

> Right now, I get a slashed https into browser, with the message:
> Your connection to www.domain.com is encrypted with 256-bit encryption.
> However, this page includes other resources which are not secure. These
> resources can be viewed by others while in transit and can be modified
> by an attacker to change the behavior of the page.

This is because reply from the protected page includes links to 
unprotected resources (most likely some javascript files).  You 
have to load other resources from https as well.

Maxim Dounin


From nginx-forum at nginx.us  Sat Nov 19 22:25:46 2011
From: nginx-forum at nginx.us (anagio)
Date: Sat, 19 Nov 2011 17:25:46 -0500
Subject: Nginx mass dynamic virtual hosts like apache?
Message-ID: <6d54a2a87692a9a43d4ffcbc5e30968d.NginxMailingListEnglish@forum.nginx.org>

Hello

I'm new to nginx and was wondering if there is a configuration that is
similar to apaches mass dynamic virtual hosts
http://httpd.apache.org/docs/2.0/vhosts/mass.html

I'm having an SaaS developed where users will get a subdomain and would
like to set this up as efficiently as possible. As of now i'm using
apache but would like to move to nginx and get this setup

Thanks

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218617,218617#msg-218617


From appa at perusio.net  Sat Nov 19 22:34:02 2011
From: appa at perusio.net (=?UTF-8?B?QW50w7NuaW8=?= P. P. Almeida)
Date: Sat, 19 Nov 2011 22:34:02 +0000
Subject: Enable SSL for a specific directory?
In-Reply-To: <e7ec2721b716a6e58af97a071d3162bf.NginxMailingListEnglish@forum.nginx.org>
References: <e7ec2721b716a6e58af97a071d3162bf.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <87k46vrbc5.wl%appa@perusio.net>

On 19 Nov 2011 18h46 WET, nginx-forum at nginx.us wrote:

> Hi,
>
> I'm trying to enable SSL for a specific directory only. In other
> words, the / directory is not encrypted, while the /protected is.
>
> server {
> 	listen				192.168.1.2:80 default_server;
> 	server_name			www.domain.com;
> 	root				/var/www/html;
>
> 	location / {
> 		try_files $uri $uri/ index.php?q=$uri&$args;
> 	}
>
> 	location /protected/ {
> 		rewrite ^ https://www.domain.com$request_uri?
> 		permanent;
> 	}
>
> 	...
> }
>
> server {
> 	listen				192.168.1.2:443 ssl;
> 	server_name			www.domain.com;
> 	root				/var/www/html;
>
> 	location / {
> 		rewrite ^ http://www.domain.com$request_uri?
> 		permanent;
> 	}
>
> 	location /protected/ {
> 		try_files $uri $uri/ /index.php?q=$uri&$args;
> 	}
>
> 	...
> }

Define a vhost for SSL that has all the same locations, meaning static
resources and such. Then do on the HTTP host.

location ^~ /protected {
   return 301 http://www.domain.com$request_uri;
}

On the SSL host:

location / {
   return 301 http://www.domain.com$request_uri;
}

location ^~ /protected {
   try_files $uri $uri/ /index.php?q=$uri&$args;
}

--- appa


From zzz at zzz.org.ua  Sat Nov 19 22:54:10 2011
From: zzz at zzz.org.ua (Alexandr Gomoliako)
Date: Sun, 20 Nov 2011 00:54:10 +0200
Subject: Nginx mass dynamic virtual hosts like apache?
In-Reply-To: <6d54a2a87692a9a43d4ffcbc5e30968d.NginxMailingListEnglish@forum.nginx.org>
References: <6d54a2a87692a9a43d4ffcbc5e30968d.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <CAKb9G9i5eXdZnSv2=xAbTb_d7xCUt9sSgZNsWY4i+iGPHi+dhg@mail.gmail.com>

On Sun, Nov 20, 2011 at 12:25 AM, anagio <nginx-forum at nginx.us> wrote:
> Hello
>
> I'm new to nginx and was wondering if there is a configuration that is
> similar to apaches mass dynamic virtual hosts
> http://httpd.apache.org/docs/2.0/vhosts/mass.html
>
> I'm having an SaaS developed where users will get a subdomain and would
> like to set this up as efficiently as possible. As of now i'm using
> apache but would like to move to nginx and get this setup

Well, the idea is to set your root dynamically from the default
server block depending on the 'host' header.

You can either use $http_host as a part of your root, which
is probably not a good thing, or map it to some subdir, and
use its value instead:

    map $http_host $subdir {
        hostnames;
        default "default";
        .foo.bar.com  "foo";
        .baz.bar.com  "baz";
    }

    server {
        root /path/to/$subdir;

    }

Hope this helps.


From nginx-forum at nginx.us  Sun Nov 20 00:17:52 2011
From: nginx-forum at nginx.us (anagio)
Date: Sat, 19 Nov 2011 19:17:52 -0500
Subject: Nginx mass dynamic virtual hosts like apache?
In-Reply-To: <6d54a2a87692a9a43d4ffcbc5e30968d.NginxMailingListEnglish@forum.nginx.org>
References: <6d54a2a87692a9a43d4ffcbc5e30968d.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <e8a5abfd1535d58fd60795b2a32d713d.NginxMailingListEnglish@forum.nginx.org>

Hi,

In your example does that mean I would need to add entries for each
sub-domain as needed. If that's the case, wouldn't I have to restart
nginx each time a new entry is made? I plan to have sub-domains created
when users register. Will I have to restart?

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218617,218620#msg-218620


From zzz at zzz.org.ua  Sun Nov 20 00:33:16 2011
From: zzz at zzz.org.ua (Alexandr Gomoliako)
Date: Sun, 20 Nov 2011 02:33:16 +0200
Subject: Nginx mass dynamic virtual hosts like apache?
In-Reply-To: <e8a5abfd1535d58fd60795b2a32d713d.NginxMailingListEnglish@forum.nginx.org>
References: <6d54a2a87692a9a43d4ffcbc5e30968d.NginxMailingListEnglish@forum.nginx.org>
	<e8a5abfd1535d58fd60795b2a32d713d.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <CAKb9G9h2FzRcJgHYMoM6EYOfv51PE-=q1AoBWUDcR++MANpoAA@mail.gmail.com>

On Sun, Nov 20, 2011 at 2:17 AM, anagio <nginx-forum at nginx.us> wrote:
> Hi,
>
> In your example does that mean I would need to add entries for each
> sub-domain as needed. If that's the case, wouldn't I have to restart
> nginx each time a new entry is made? I plan to have sub-domains created
> when users register. Will I have to restart?

No, just reload, which is safe.

But if that's a problem, you can extract subdir with regex
in the same map, something like this:

    ~^(?<sub>[a-zA-Z0-9-]+)\.bar\.com$  $sub;


From nginx-forum at nginx.us  Sun Nov 20 00:41:31 2011
From: nginx-forum at nginx.us (anagio)
Date: Sat, 19 Nov 2011 19:41:31 -0500
Subject: Nginx mass dynamic virtual hosts like apache?
In-Reply-To: <e8a5abfd1535d58fd60795b2a32d713d.NginxMailingListEnglish@forum.nginx.org>
References: <6d54a2a87692a9a43d4ffcbc5e30968d.NginxMailingListEnglish@forum.nginx.org>
	<e8a5abfd1535d58fd60795b2a32d713d.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <258517f8df0d15952b255a4b15eadc47.NginxMailingListEnglish@forum.nginx.org>

Thanks now if I wanted to let a member who has a sub-domain on our
domain, "user.domain.com" map one of their own domains or sub-domains
using a cname. I would need to create a ServerAlias for their
sub-domain, in apache2 it would be something like below. So using the
regex probably wouldn't be the best method, so just adding them as you
listed in the first reply is how to go about it?

  ServerName user.domain.com
  ServerAlias *.user.domain.com
  ServerAlias usersowndomain.com

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218617,218622#msg-218622


From zzz at zzz.org.ua  Sun Nov 20 00:57:45 2011
From: zzz at zzz.org.ua (Alexandr Gomoliako)
Date: Sun, 20 Nov 2011 02:57:45 +0200
Subject: Nginx mass dynamic virtual hosts like apache?
In-Reply-To: <258517f8df0d15952b255a4b15eadc47.NginxMailingListEnglish@forum.nginx.org>
References: <6d54a2a87692a9a43d4ffcbc5e30968d.NginxMailingListEnglish@forum.nginx.org>
	<e8a5abfd1535d58fd60795b2a32d713d.NginxMailingListEnglish@forum.nginx.org>
	<258517f8df0d15952b255a4b15eadc47.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <CAKb9G9ji+3qegT22y+rhsMJh9YZaOu+4-V6B4xz4=je9_antPg@mail.gmail.com>

On Sun, Nov 20, 2011 at 2:41 AM, anagio <nginx-forum at nginx.us> wrote:
> Thanks now if I wanted to let a member who has a sub-domain on our
> domain, "user.domain.com" map one of their own domains or sub-domains
> using a cname. I would need to create a ServerAlias for their
> sub-domain, in apache2 it would be something like below. So using the
> regex probably wouldn't be the best method, so just adding them as you
> listed in the first reply is how to go about it?

Using those methods together is fine too. You can go either or both ways.


From nam.nh at nd24.net  Sun Nov 20 08:11:27 2011
From: nam.nh at nd24.net (Nguyen Hai Nam)
Date: Sun, 20 Nov 2011 15:11:27 +0700
Subject: trailing slash at end URL with nginx as a reverse proxy for multiple
	website
Message-ID: <CAA44a769YfVzvT2SFmYVJv=YS5NLr_+zhwFNvnbOKd=+FWf8vA@mail.gmail.com>

 Hi,

I'm running OpenIndiana 151a server then now I'd like to build the web
service with nginx. Briefly, my scenario is: one central server installed a
few website in virtual machines, and there is a different real machine that
stores static contents. I want nginx in central server will be the reverse
proxy (get idea from
http://tumblr.intranation.com/post/766288369/using-nginx-reverse-proxy).

For example, if someone open website a.com, it will be redirected to
10.0.0.2:8001, and if other go to b.com it will be redirected to
10.0.0.3:8001, etc ...

server {
    listen       80;
    server_name  a.com;

    location / {
        proxy_pass         http://10.0.0.2:8001;
        proxy_redirect     off;
        server_name_in_redirect off;
    }
}

Nginx works fine, but I don't know how to rewrite/redirect the URL looks
like: http://a.com/something becomes http://a.com/something/

If someone open the link http://a.com/something nginx can't serve it then
return the error message that says "http://a.com:8001/something"
unreachable.

If you have any advice, please tell. It's highly appreciated.

Thanks,
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20111120/e4391ea9/attachment.html>

From appa at perusio.net  Sun Nov 20 08:35:55 2011
From: appa at perusio.net (=?UTF-8?B?QW50w7NuaW8=?= P. P. Almeida)
Date: Sun, 20 Nov 2011 08:35:55 +0000
Subject: trailing slash at end URL with nginx as a reverse proxy for
	multiple	website
In-Reply-To: <CAA44a769YfVzvT2SFmYVJv=YS5NLr_+zhwFNvnbOKd=+FWf8vA@mail.gmail.com>
References: <CAA44a769YfVzvT2SFmYVJv=YS5NLr_+zhwFNvnbOKd=+FWf8vA@mail.gmail.com>
Message-ID: <87hb1zqjh0.wl%appa@perusio.net>

On 20 Nov 2011 08h11 WET, nam.nh at nd24.net wrote:

> Hi,
>
> I'm running OpenIndiana 151a server then now I'd like to build the
> web service with nginx. Briefly, my scenario is: one central server
> installed a few website in virtual machines, and there is a
> different real machine that stores static contents. I want nginx in
> central server will be the reverse proxy (get idea from
> http://tumblr.intranation.com/post/766288369/using-nginx-reverse-proxy).
>
> For example, if someone open website a.com, it will be redirected to
> 10.0.0.2:8001, and if other go to b.com it will be redirected to
> 10.0.0.3:8001, etc ...
>
> server {
> listen       80;
> server_name  a.com;
>
> location / {
> proxy_pass         http://10.0.0.2:8001;
> proxy_redirect     off;
> server_name_in_redirect off;
> }
> }
>
> Nginx works fine, but I don't know how to rewrite/redirect the URL
> looks like: http://a.com/something becomes http://a.com/something/
>
> If someone open the link http://a.com/something nginx can't serve it
> then return the error message that says
> "http://a.com:8001/something" unreachable.
>
> If you have any advice, please tell. It's highly appreciated.

If it's rewriting all the URIs so that they have a trailing slash then
try this:

1. At the http level:

map $request_uri $no_trailing_slash {
  default  1;
  ~.*[^/]$ 0;
} 

2. At the server level (vhost):

server {
    listen       80;
    server_name  a.com;


    location / {
        if ($no_trailing_slash) {
            return 302 $request_uri/; 
        }

        proxy_pass         http://10.0.0.2:8001;
        proxy_redirect     off;
        server_name_in_redirect off;
    }
}
 
--- appa



> Thanks,


From appa at perusio.net  Sun Nov 20 08:40:13 2011
From: appa at perusio.net (=?UTF-8?B?QW50w7NuaW8=?= P. P. Almeida)
Date: Sun, 20 Nov 2011 08:40:13 +0000
Subject: trailing slash at end URL with nginx as a reverse proxy for
	multiple	website
In-Reply-To: <CAA44a769YfVzvT2SFmYVJv=YS5NLr_+zhwFNvnbOKd=+FWf8vA@mail.gmail.com>
References: <CAA44a769YfVzvT2SFmYVJv=YS5NLr_+zhwFNvnbOKd=+FWf8vA@mail.gmail.com>
Message-ID: <87fwhjqj9u.wl%appa@perusio.net>

On 20 Nov 2011 08h11 WET, nam.nh at nd24.net wrote:

Oops it's the other way around.

If it's rewriting all the URIs so that they have a trailing slash then
try this:

1. At the http level:

map $request_uri $no_trailing_slash {
  default  0;
  ~.*[^/]$ 1;
} 

2. At the server level (vhost):

server {
    listen       80;
    server_name  a.com;


    location / {
        if ($no_trailing_slash) {
            return 302 $request_uri/; 
        }

        proxy_pass         http://10.0.0.2:8001;
        proxy_redirect     off;
        server_name_in_redirect off;
    }
}
 
--- appa


From nginx-forum at nginx.us  Sun Nov 20 09:05:50 2011
From: nginx-forum at nginx.us (anagio)
Date: Sun, 20 Nov 2011 04:05:50 -0500
Subject: Sub-domains only server for web application?
Message-ID: <a8ab8f01f83a18f8a88fd0604e0b088d.NginxMailingListEnglish@forum.nginx.org>

Our website currently runs wordpress to manage the front end site
www.example.com

When users register they get a subdomain username.example.com where they
can login to our application. The application is built on zend
framework. I'd like to have the zf project on an entirely different
server than where we run the wordpress www.example.com domain.

How would the default server config be setup since the front end site is
on another server with different IP? Virtual hosts would be setup as
usual, i'm just not sure how to configure the main server so it
redirects to www.example.com? Any help is appreciated

Thanks

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218630,218630#msg-218630


From edho at myconan.net  Sun Nov 20 09:12:05 2011
From: edho at myconan.net (Edho Arief)
Date: Sun, 20 Nov 2011 16:12:05 +0700
Subject: trailing slash at end URL with nginx as a reverse proxy for
	multiple website
In-Reply-To: <CAA44a769YfVzvT2SFmYVJv=YS5NLr_+zhwFNvnbOKd=+FWf8vA@mail.gmail.com>
References: <CAA44a769YfVzvT2SFmYVJv=YS5NLr_+zhwFNvnbOKd=+FWf8vA@mail.gmail.com>
Message-ID: <CABTjkKkHKWtE_EbOX5c9mihPD894PS1GBUMp4WUUfm5KNsojuQ@mail.gmail.com>

On Sun, Nov 20, 2011 at 3:11 PM, Nguyen Hai Nam <nam.nh at nd24.net> wrote:
> If someone open the link http://a.com/something nginx can't serve it then
> return the error message that says "http://a.com:8001/something"
> unreachable.
>

Are you sure it's nginx error and not backend's?

Also you may want to try adding these lines:

    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;


-- 
O< ascii ribbon campaign - stop html mail - www.asciiribbon.org


From igor at sysoev.ru  Sun Nov 20 09:16:34 2011
From: igor at sysoev.ru (Igor Sysoev)
Date: Sun, 20 Nov 2011 13:16:34 +0400
Subject: trailing slash at end URL with nginx as a reverse proxy for
	multiple website
In-Reply-To: <CAA44a769YfVzvT2SFmYVJv=YS5NLr_+zhwFNvnbOKd=+FWf8vA@mail.gmail.com>
References: <CAA44a769YfVzvT2SFmYVJv=YS5NLr_+zhwFNvnbOKd=+FWf8vA@mail.gmail.com>
Message-ID: <20111120091634.GC99580@nginx.com>

On Sun, Nov 20, 2011 at 03:11:27PM +0700, Nguyen Hai Nam wrote:
>  Hi,
> 
> I'm running OpenIndiana 151a server then now I'd like to build the web
> service with nginx. Briefly, my scenario is: one central server installed a
> few website in virtual machines, and there is a different real machine that
> stores static contents. I want nginx in central server will be the reverse
> proxy (get idea from
> http://tumblr.intranation.com/post/766288369/using-nginx-reverse-proxy).
> 
> For example, if someone open website a.com, it will be redirected to
> 10.0.0.2:8001, and if other go to b.com it will be redirected to
> 10.0.0.3:8001, etc ...
> 
> server {
>     listen       80;
>     server_name  a.com;
> 
>     location / {
>         proxy_pass         http://10.0.0.2:8001;
>         proxy_redirect     off;
>         server_name_in_redirect off;
>     }
> }
> 
> Nginx works fine, but I don't know how to rewrite/redirect the URL looks
> like: http://a.com/something becomes http://a.com/something/
> 
> If someone open the link http://a.com/something nginx can't serve it then
> return the error message that says "http://a.com:8001/something"
> unreachable.
> 
> If you have any advice, please tell. It's highly appreciated.
> 

-        proxy_redirect     off;
-        server_name_in_redirect off;
+        proxy_redirect     http://a.com:8001/  /;


-- 
Igor Sysoev


From appa at perusio.net  Sun Nov 20 09:22:15 2011
From: appa at perusio.net (=?UTF-8?B?QW50w7NuaW8=?= P. P. Almeida)
Date: Sun, 20 Nov 2011 09:22:15 +0000
Subject: Sub-domains only server for web application?
In-Reply-To: <a8ab8f01f83a18f8a88fd0604e0b088d.NginxMailingListEnglish@forum.nginx.org>
References: <a8ab8f01f83a18f8a88fd0604e0b088d.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <87ehx3qhbs.wl%appa@perusio.net>

On 20 Nov 2011 09h05 WET, nginx-forum at nginx.us wrote:

> Our website currently runs wordpress to manage the front end site
> www.example.com
>
> When users register they get a subdomain username.example.com where
> they can login to our application. The application is built on zend
> framework. I'd like to have the zf project on an entirely different
> server than where we run the wordpress www.example.com domain.

If I understand correctly. Try:

server { # frontend
    listen [::]:80 default_server;
    server_name www.example.com;
}

server { # backend
    listen [::]:80;
    server_name *.example.com;
}

Check: http://nginx.org/en/docs/http/server_names.html

> How would the default server config be setup since the front end
> site is on another server with different IP? Virtual hosts would be
> setup as usual, i'm just not sure how to configure the main server
> so it redirects to www.example.com? Any help is appreciated

--- appa


From nginx-forum at nginx.us  Sun Nov 20 09:42:35 2011
From: nginx-forum at nginx.us (anagio)
Date: Sun, 20 Nov 2011 04:42:35 -0500
Subject: Nginx mass dynamic virtual hosts like apache?
In-Reply-To: <6d54a2a87692a9a43d4ffcbc5e30968d.NginxMailingListEnglish@forum.nginx.org>
References: <6d54a2a87692a9a43d4ffcbc5e30968d.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <b0a8db8227e4098ac95bd0c4eb211fb5.NginxMailingListEnglish@forum.nginx.org>

Given I use this method for sub-domains

map $http_host $subdir {
hostnames;
default "default";
.foo.bar.com "foo";
.baz.bar.com "baz";
}

Is there a method to redirect any requests for sub-domains which do not
exist (404s) to a custom html/php page?

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218617,218634#msg-218634


From nam.nh at nd24.net  Sun Nov 20 13:57:39 2011
From: nam.nh at nd24.net (Nguyen Hai Nam)
Date: Sun, 20 Nov 2011 20:57:39 +0700
Subject: trailing slash at end URL with nginx as a reverse proxy for
	multiple website
In-Reply-To: <CAA44a769YfVzvT2SFmYVJv=YS5NLr_+zhwFNvnbOKd=+FWf8vA@mail.gmail.com>
References: <CAA44a769YfVzvT2SFmYVJv=YS5NLr_+zhwFNvnbOKd=+FWf8vA@mail.gmail.com>
Message-ID: <CAA44a74PnXk87iOL1g1MfeUhOn96qiteJ8tWRi9hkCH515BRUg@mail.gmail.com>

Thank you guys for helping.

I've tried following of Antonio, Edho and Igor's advices but no thing changes.

@Igor: If I add proxy_redirect http://a.com:8001/ /; it will return
error 404 Not Found.

I'm curious in why when I type http://a.com/something it's redirected
to http://a.com:8001/something, so  maybe it's wrong at reverse proxy
point.

Configuration of reverse proxy as seen:

http {

include /usr/nginx/conf/proxy.conf;

    server {
        listen       8080;
        server_name  abc.com;

 proxy_set_header   Host             $host;
    proxy_set_header   X-Real-IP        $remote_addr;
    proxy_set_header   X-Forwarded-For  $proxy_add_x_forwarded_for;
    proxy_max_temp_file_size 0;
}
}

In proxy.conf

server {
    listen       80;
    server_name  a.com;

    access_log  on;
    error_log on;

    location / {
        proxy_pass         http://10.2.176.21:8001/;
        #proxy_redirect     off;
        #server_name_in_redirect on;

        proxy_set_header   Host             $host;
        proxy_set_header   X-Real-IP        $remote_addr;
        proxy_set_header   X-Forwarded-For  $proxy_add_x_forwarded_for;
}
}

Thanks for your help.


From lists at ruby-forum.com  Sun Nov 20 15:37:23 2011
From: lists at ruby-forum.com (John Travolota)
Date: Sun, 20 Nov 2011 16:37:23 +0100
Subject: kernel: nginx[18233]: segfault at 0000000000000008 rip
	000000000043edf8 rsp 00007fff34a21fa0 err
In-Reply-To: <d6ba00b08aeeb18ecc9a0873cdd53e5e@ruby-forum.com>
References: <d6ba00b08aeeb18ecc9a0873cdd53e5e@ruby-forum.com>
Message-ID: <f6b11b2bc814f0fab4e700d8e1b3cb4a@ruby-forum.com>

This bug still occur even after upgrading to latest version.I really 
dont understood why this domain causing seg fault....And i really need 
geo ip there because it's domain where i receiving more traffic then any 
other domain.

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


From calin at presslabs.com  Sun Nov 20 19:06:49 2011
From: calin at presslabs.com (Calin Don)
Date: Sun, 20 Nov 2011 21:06:49 +0200
Subject: trailing slash at end URL with nginx as a reverse proxy for
	multiple website
In-Reply-To: <CAA44a74PnXk87iOL1g1MfeUhOn96qiteJ8tWRi9hkCH515BRUg@mail.gmail.com>
References: <CAA44a769YfVzvT2SFmYVJv=YS5NLr_+zhwFNvnbOKd=+FWf8vA@mail.gmail.com>
	<CAA44a74PnXk87iOL1g1MfeUhOn96qiteJ8tWRi9hkCH515BRUg@mail.gmail.com>
Message-ID: <CAEOe2JyJXhYOWK=sBhpKfebtSRS-=q5mnrkuhdK4rsMvmEpLpA@mail.gmail.com>

Remove the trailing / from proxy_pass directive.
On Nov 20, 2011 3:57 PM, "Nguyen Hai Nam" <nam.nh at nd24.net> wrote:

> Thank you guys for helping.
>
> I've tried following of Antonio, Edho and Igor's advices but no thing
> changes.
>
> @Igor: If I add proxy_redirect http://a.com:8001/ /; it will return
> error 404 Not Found.
>
> I'm curious in why when I type http://a.com/something it's redirected
> to http://a.com:8001/something, so  maybe it's wrong at reverse proxy
> point.
>
> Configuration of reverse proxy as seen:
>
> http {
>
> include /usr/nginx/conf/proxy.conf;
>
>    server {
>        listen       8080;
>        server_name  abc.com;
>
>  proxy_set_header   Host             $host;
>    proxy_set_header   X-Real-IP        $remote_addr;
>    proxy_set_header   X-Forwarded-For  $proxy_add_x_forwarded_for;
>    proxy_max_temp_file_size 0;
> }
> }
>
> In proxy.conf
>
> server {
>    listen       80;
>    server_name  a.com;
>
>    access_log  on;
>    error_log on;
>
>    location / {
>        proxy_pass         http://10.2.176.21:8001/;
>        #proxy_redirect     off;
>        #server_name_in_redirect on;
>
>        proxy_set_header   Host             $host;
>        proxy_set_header   X-Real-IP        $remote_addr;
>        proxy_set_header   X-Forwarded-For  $proxy_add_x_forwarded_for;
> }
> }
>
> Thanks for your 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/20111120/c991ebf2/attachment.html>

From mdounin at mdounin.ru  Sun Nov 20 20:43:39 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Mon, 21 Nov 2011 00:43:39 +0400
Subject: kernel: nginx[18233]: segfault at 0000000000000008 rip
	000000000043edf8 rsp 00007fff34a21fa0 err
In-Reply-To: <f6b11b2bc814f0fab4e700d8e1b3cb4a@ruby-forum.com>
References: <d6ba00b08aeeb18ecc9a0873cdd53e5e@ruby-forum.com>
	<f6b11b2bc814f0fab4e700d8e1b3cb4a@ruby-forum.com>
Message-ID: <20111120204339.GO95664@mdounin.ru>

Hello!

On Sun, Nov 20, 2011 at 04:37:23PM +0100, John Travolota wrote:

> This bug still occur even after upgrading to latest version.I really 
> dont understood why this domain causing seg fault....And i really need 
> geo ip there because it's domain where i receiving more traffic then any 
> other domain.

You've been already pointed to this link:
http://wiki.nginx.org/Debugging

Do you have backtrace?

Maxim Dounin


From nginx-forum at nginx.us  Sun Nov 20 20:59:13 2011
From: nginx-forum at nginx.us (anagio)
Date: Sun, 20 Nov 2011 15:59:13 -0500
Subject: Redirect arbitrary sub-domains to custom 404 when wild card A record
	is set?
Message-ID: <708fcc2b7651663aee8f7b17abd5e921.NginxMailingListEnglish@forum.nginx.org>

Is there a way to capture all requests to sub-domains which do not have
a virtual host setup, and redirect to a custom 404 page? I will have a
wild card A record setup *.example.com and all our users will have a
sub-domain username.example.com. If someone enters a sub-domain which
does not exist how can I redirect to a custom 404 page rather than have
it resolve since wild card is setup?

Thanks

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218654,218654#msg-218654


From grails at jmsd.co.uk  Sun Nov 20 21:16:35 2011
From: grails at jmsd.co.uk (John Moore)
Date: Sun, 20 Nov 2011 21:16:35 +0000
Subject: nginx serving old version of CSS file
Message-ID: <4EC96E33.6090004@jmsd.co.uk>

I have spent 2 hours trying to get to the bottom of a problem and am 
utterly stumped. I am certain it must be something I'm doing wrong, and 
NOT an nginx problem, but I can't think for the life of me what it might 
be. What is happening is that nginx is serving up a version of a CSS 
file which I overwrote a couple of hours ago. It's version 0.7.65 
running on Linux. I have a CSS file called temp.css, which I have in the 
/css folder below the root of a virtual host called kf.dev. If I look in 
the folder and view the file with less, I can see it's the correct one. 
But when I do 'wget http://kf.dev/css/temp.css' I get an old version. In 
fact, when I try in Firefox to see a page which uses this css file, it 
doesn't display properly and when I click on the CSS link in View 
Source, I see the old version with what appears to be a load of binary 
characters at the bottom (they display as little question marks against 
black diamonds). As I say, I am completely stumped. Where could nginx be 
getting this from? I have determined that there is in fact only one file 
actually called this on the hard drive, so it can't be finding the wrong 
copy of the file. I'm beginning to think this might be some weird 
operating system level corruption. Possible?


From francis at daoine.org  Sun Nov 20 22:27:38 2011
From: francis at daoine.org (Francis Daly)
Date: Sun, 20 Nov 2011 22:27:38 +0000
Subject: Redirect arbitrary sub-domains to custom 404 when wild card A
	record is set?
In-Reply-To: <708fcc2b7651663aee8f7b17abd5e921.NginxMailingListEnglish@forum.nginx.org>
References: <708fcc2b7651663aee8f7b17abd5e921.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20111120222738.GT27078@craic.sysops.org>

On Sun, Nov 20, 2011 at 03:59:13PM -0500, anagio wrote:

Hi there,

> Is there a way to capture all requests to sub-domains which do not have
> a virtual host setup, and redirect to a custom 404 page?

That's called the default server.

http://nginx.org/en/docs/http/request_processing.html

In that server, you can redirect or respond any way you choose.

	f
-- 
Francis Daly        francis at daoine.org


From nam.nh at nd24.net  Mon Nov 21 01:49:01 2011
From: nam.nh at nd24.net (Nguyen Hai Nam)
Date: Mon, 21 Nov 2011 08:49:01 +0700
Subject: trailing slash at end URL with nginx as a reverse proxy for
	multiple website
In-Reply-To: <CAA44a74PnXk87iOL1g1MfeUhOn96qiteJ8tWRi9hkCH515BRUg@mail.gmail.com>
References: <CAA44a769YfVzvT2SFmYVJv=YS5NLr_+zhwFNvnbOKd=+FWf8vA@mail.gmail.com>
	<CAA44a74PnXk87iOL1g1MfeUhOn96qiteJ8tWRi9hkCH515BRUg@mail.gmail.com>
Message-ID: <4EC9AE0D.7040400@nd24.net>

Hi Calin,

It's removed and still no change.

Thanks,


From nginx-forum at nginx.us  Mon Nov 21 03:23:38 2011
From: nginx-forum at nginx.us (scriptonian)
Date: Sun, 20 Nov 2011 22:23:38 -0500
Subject: Cant start Nginx on Mac OS X
Message-ID: <ecf722d5c6882456f2a7beeb90797ed2.NginxMailingListEnglish@forum.nginx.org>

Sorry for the newbie question. I am new to nginx. i installed it
successfully on my mac, but when i try to start it, i get this error

nginx: [alert] could not open error log file: open()
"/opt/local/var/log/nginx/error.log" failed (13: Permission denied)
2011/11/20 22:15:20 [emerg] 1620#0: open()
"/opt/local/etc/nginx/nginx.conf" failed (2: No such file or directory)

Can someone please help me resolve this? Thank you

Kofi

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218663,218663#msg-218663


From smallfish.xy at gmail.com  Mon Nov 21 03:25:47 2011
From: smallfish.xy at gmail.com (smallfish)
Date: Mon, 21 Nov 2011 11:25:47 +0800
Subject: Cant start Nginx on Mac OS X
In-Reply-To: <ecf722d5c6882456f2a7beeb90797ed2.NginxMailingListEnglish@forum.nginx.org>
References: <ecf722d5c6882456f2a7beeb90797ed2.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <CACPpB2xekz_NUqzZcyHQz8c8yfF=92qOzt1yqVke40rYe7EtLg@mail.gmail.com>

>>> Permission denied

start nginx use root or sudo
--
blog: http://chenxiaoyu.org



On Mon, Nov 21, 2011 at 11:23 AM, scriptonian <nginx-forum at nginx.us> wrote:

> Sorry for the newbie question. I am new to nginx. i installed it
> successfully on my mac, but when i try to start it, i get this error
>
> nginx: [alert] could not open error log file: open()
> "/opt/local/var/log/nginx/error.log" failed (13: Permission denied)
> 2011/11/20 22:15:20 [emerg] 1620#0: open()
> "/opt/local/etc/nginx/nginx.conf" failed (2: No such file or directory)
>
> Can someone please help me resolve this? Thank you
>
> Kofi
>
> Posted at Nginx Forum:
> http://forum.nginx.org/read.php?2,218663,218663#msg-218663
>
> _______________________________________________
> 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/20111121/01c4cf09/attachment-0001.html>

From nginx-forum at nginx.us  Mon Nov 21 04:34:17 2011
From: nginx-forum at nginx.us (scriptonian)
Date: Sun, 20 Nov 2011 23:34:17 -0500
Subject: Cant start Nginx on Mac OS X
In-Reply-To: <CACPpB2xekz_NUqzZcyHQz8c8yfF=92qOzt1yqVke40rYe7EtLg@mail.gmail.com>
References: <CACPpB2xekz_NUqzZcyHQz8c8yfF=92qOzt1yqVke40rYe7EtLg@mail.gmail.com>
Message-ID: <33521217098901ebf30c3f3e2138473f.NginxMailingListEnglish@forum.nginx.org>

super!... one error went away, now i am left with just one

nginx: [emerg] open() "/opt/local/etc/nginx/nginx.conf" failed (2: No
such file or directory)


Any ideas why that is?

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218663,218667#msg-218667


From edho at myconan.net  Mon Nov 21 04:36:54 2011
From: edho at myconan.net (Edho Arief)
Date: Mon, 21 Nov 2011 11:36:54 +0700
Subject: Cant start Nginx on Mac OS X
In-Reply-To: <33521217098901ebf30c3f3e2138473f.NginxMailingListEnglish@forum.nginx.org>
References: <CACPpB2xekz_NUqzZcyHQz8c8yfF=92qOzt1yqVke40rYe7EtLg@mail.gmail.com>
	<33521217098901ebf30c3f3e2138473f.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <CABTjkKk+_sX3j4Kebt1worr0dDYoybFAUpLc74gps1KVx_0eNA@mail.gmail.com>

On Mon, Nov 21, 2011 at 11:34 AM, scriptonian <nginx-forum at nginx.us> wrote:
> super!... one error went away, now i am left with just one
>
> nginx: [emerg] open() "/opt/local/etc/nginx/nginx.conf" failed (2: No
> such file or directory)
>
>
> Any ideas why that is?
>

....missing config file?

-- 
O< ascii ribbon campaign - stop html mail - www.asciiribbon.org


From nginx-forum at nginx.us  Mon Nov 21 04:52:28 2011
From: nginx-forum at nginx.us (scriptonian)
Date: Sun, 20 Nov 2011 23:52:28 -0500
Subject: Cant start Nginx on Mac OS X
In-Reply-To: <CABTjkKk+_sX3j4Kebt1worr0dDYoybFAUpLc74gps1KVx_0eNA@mail.gmail.com>
References: <CABTjkKk+_sX3j4Kebt1worr0dDYoybFAUpLc74gps1KVx_0eNA@mail.gmail.com>
Message-ID: <f46b77a6de4161d7d69f8f4cbaf76cd1.NginxMailingListEnglish@forum.nginx.org>

Yes but why is it missing? :) 

I installed the server :( ... shouldn't the installation put that file
there during setup?

i see only these files in the /etc/nginx/ folder

fastcgi.conf		        koi-win			          scgi_params.example
fastcgi.conf.example	mime.types		          uwsgi_params
fastcgi_params		mime.types.example	  uwsgi_params.example
fastcgi_params.example	nginx.conf.example	          win-utf
koi-utf			        scgi_params

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218663,218670#msg-218670


From nginx-forum at nginx.us  Mon Nov 21 04:53:43 2011
From: nginx-forum at nginx.us (scriptonian)
Date: Sun, 20 Nov 2011 23:53:43 -0500
Subject: Cant start Nginx on Mac OS X
In-Reply-To: <f46b77a6de4161d7d69f8f4cbaf76cd1.NginxMailingListEnglish@forum.nginx.org>
References: <CABTjkKk+_sX3j4Kebt1worr0dDYoybFAUpLc74gps1KVx_0eNA@mail.gmail.com>
	<f46b77a6de4161d7d69f8f4cbaf76cd1.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <786ce31202a3c372f37630f37b5d0f20.NginxMailingListEnglish@forum.nginx.org>

Reposting...

fastcgi.conf		
koi-win			
scgi_params.example
fastcgi.conf.example	
mime.types		
uwsgi_params
fastcgi_params		
mime.types.example	
uwsgi_params.example
fastcgi_params.example	
nginx.conf.example	
win-utf
koi-utf			
scgi_params

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218663,218671#msg-218671


From edho at myconan.net  Mon Nov 21 04:55:31 2011
From: edho at myconan.net (Edho Arief)
Date: Mon, 21 Nov 2011 11:55:31 +0700
Subject: Cant start Nginx on Mac OS X
In-Reply-To: <f46b77a6de4161d7d69f8f4cbaf76cd1.NginxMailingListEnglish@forum.nginx.org>
References: <CABTjkKk+_sX3j4Kebt1worr0dDYoybFAUpLc74gps1KVx_0eNA@mail.gmail.com>
	<f46b77a6de4161d7d69f8f4cbaf76cd1.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <CABTjkKmCVhtNM+DSB55sKVEb80m1RUUo-zUFwwntkVmboGiwZQ@mail.gmail.com>

On Mon, Nov 21, 2011 at 11:52 AM, scriptonian <nginx-forum at nginx.us> wrote:
> Yes but why is it missing? :)
>
> I installed the server :( ... shouldn't the installation put that file
> there during setup?
>

Even if it did, you should edit the config anyway. The
nginx.conf.example should help.

-- 
O< ascii ribbon campaign - stop html mail - www.asciiribbon.org


From amoiz.shine at gmail.com  Mon Nov 21 04:56:53 2011
From: amoiz.shine at gmail.com (Sharl.Jimh.Tsin)
Date: Mon, 21 Nov 2011 12:56:53 +0800
Subject: Cant start Nginx on Mac OS X
In-Reply-To: <33521217098901ebf30c3f3e2138473f.NginxMailingListEnglish@forum.nginx.org>
References: <CACPpB2xekz_NUqzZcyHQz8c8yfF=92qOzt1yqVke40rYe7EtLg@mail.gmail.com>
	<33521217098901ebf30c3f3e2138473f.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <1321851413.2289.2.camel@sharl-desktop>

? 2011-11-20?? 23:34 -0500?scriptonian???
> super!... one error went away, now i am left with just one
> 
> nginx: [emerg] open() "/opt/local/etc/nginx/nginx.conf" failed (2: No
> such file or directory)
> 
> 
> Any ideas why that is?
> 
> Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218663,218667#msg-218667
> 
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx

make sure that the path /opt/local/etc/nginx/ existed,and give it umask
775.

also,do it under ROOT.

-- 
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/20111121/829be1a3/attachment.bin>

From zhuzhaoyuan at gmail.com  Mon Nov 21 05:30:20 2011
From: zhuzhaoyuan at gmail.com (Joshua Zhu)
Date: Mon, 21 Nov 2011 13:30:20 +0800
Subject: nginx serving old version of CSS file
In-Reply-To: <4EC96E33.6090004@jmsd.co.uk>
References: <4EC96E33.6090004@jmsd.co.uk>
Message-ID: <CAOKiq2A-frUbS0410tVmtsQ22GcQp-jVsbWTCs9c+Dy3Sk0gJQ@mail.gmail.com>

Hi,

On Mon, Nov 21, 2011 at 5:16 AM, John Moore <grails at jmsd.co.uk> wrote:

> I have spent 2 hours trying to get to the bottom of a problem and am
> utterly stumped. I am certain it must be something I'm doing wrong, and NOT
> an nginx problem, but I can't think for the life of me what it might be.
> What is happening is that nginx is serving up a version of a CSS file which
> I overwrote a couple of hours ago. It's version 0.7.65 running on Linux. I
> have a CSS file called temp.css, which I have in the /css folder below the
> root of a virtual host called kf.dev. If I look in the folder and view the
> file with less, I can see it's the correct one. But when I do 'wget
> http://kf.dev/css/temp.css' I get an old version. In fact, when I try in
> Firefox to see a page which uses this css file, it doesn't display properly
> and when I click on the CSS link in View Source, I see the old version with
> what appears to be a load of binary characters at the bottom (they display
> as little question marks against black diamonds). As I say, I am completely
> stumped. Where could nginx be getting this from? I have determined that
> there is in fact only one file actually called this on the hard drive, so
> it can't be finding the wrong copy of the file. I'm beginning to think this
> might be some weird operating system level corruption. Possible?
>
>

Did you turn on open_file_cache in your configuration? It's buggy.

Regards,

-- 
Joshua Zhu
Senior Software Engineer
Server Platforms Team at Taobao
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20111121/9e24eabc/attachment.html>

From nginx-forum at nginx.us  Mon Nov 21 07:02:29 2011
From: nginx-forum at nginx.us (sevmax)
Date: Mon, 21 Nov 2011 02:02:29 -0500
Subject: Please check my nginx/php-fpm configs
Message-ID: <3412015444ba396bee98da8721b4fcb3.NginxMailingListEnglish@forum.nginx.org>

Hi dear nginx community!
Please check my ginx and php-fpm logs.
I use that configuration on high-loaded servers in startup company. 
Thank you in avance! 

nginx: http://pastebin.com/GxL8h0KD
php-fpm: http://pastebin.com/vvvhSdAj

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218678,218678#msg-218678


From nam.nh at nd24.net  Mon Nov 21 07:05:51 2011
From: nam.nh at nd24.net (Nguyen Hai Nam)
Date: Mon, 21 Nov 2011 14:05:51 +0700
Subject: trailing slash at end URL with nginx as a reverse proxy for
	multiple website
In-Reply-To: <4EC9AE0D.7040400@nd24.net>
References: <CAA44a769YfVzvT2SFmYVJv=YS5NLr_+zhwFNvnbOKd=+FWf8vA@mail.gmail.com>
	<CAA44a74PnXk87iOL1g1MfeUhOn96qiteJ8tWRi9hkCH515BRUg@mail.gmail.com>
	<4EC9AE0D.7040400@nd24.net>
Message-ID: <4EC9F84F.8060802@nd24.net>

I've found that TCP port 8001 was used as VCOM tunnel. Then change nginx 
to listen to other port works fine.

Thanks for your kind support.


From lhmwzy at gmail.com  Mon Nov 21 08:32:56 2011
From: lhmwzy at gmail.com (lhmwzy)
Date: Mon, 21 Nov 2011 16:32:56 +0800
Subject: Can Mail modules provides simple load-balancing (round-robin and
	client IP) across backend servers?
Message-ID: <CALL55ghWjrNrPnRw0RJn324G9=+xfCR08oo4J1QyEFs735-+XQ@mail.gmail.com>

Can Mail modules provides simple load-balancing (round-robin and
client IP) across backend servers Like http HttpUpstreamModule?

upstream backend {
  ip_hash;
  server   backend1.example.com;
  server   backend2.example.com;
  server   backend3.example.com  down;
  server   backend4.example.com;
}

mail

{
 ......
  proxy_pass backend;
 .....
}


From nginx-forum at nginx.us  Mon Nov 21 08:36:12 2011
From: nginx-forum at nginx.us (sibsoft)
Date: Mon, 21 Nov 2011 03:36:12 -0500
Subject: MP4 - start time is out mp4 stts samples
In-Reply-To: <5e238ac1a229e14931b6ea72c107997c.NginxMailingListEnglish@forum.nginx.org>
References: <20111117050603.GA10385@nginx.com>
	<5e238ac1a229e14931b6ea72c107997c.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <4a3fe641ebcaebadf1cc53b9aaf76c92.NginxMailingListEnglish@forum.nginx.org>

Having the same problem on 1.0.10


nginx: nginx version: nginx/1.0.10
nginx: configure arguments: --with-http_stub_status_module
--with-http_perl_module --with-http_flv_module --with-http_mp4_module
--with-debug

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218451,218685#msg-218685


From igor at sysoev.ru  Mon Nov 21 08:57:58 2011
From: igor at sysoev.ru (Igor Sysoev)
Date: Mon, 21 Nov 2011 12:57:58 +0400
Subject: MP4 - start time is out mp4 stts samples
In-Reply-To: <5e238ac1a229e14931b6ea72c107997c.NginxMailingListEnglish@forum.nginx.org>
References: <20111117050603.GA10385@nginx.com>
	<5e238ac1a229e14931b6ea72c107997c.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20111121085757.GA69567@nginx.com>

On Thu, Nov 17, 2011 at 09:25:13AM -0500, sisif wrote:
> nginx version: nginx/1.1.8
> built by gcc 4.4.4 20100726 (Red Hat 4.4.4-13) (GCC)
> TLS SNI support enabled
> configure arguments: --prefix=/usr --sbin-path=/usr/sbin/nginx
> --conf-path=/etc/nginx/nginx.conf
> --error-log-path=/var/log/nginx/error.log
> --pid-path=/var/run/nginx/nginx.pid --lock-path=/var/lock/nginx.lock
> --user=nginx --group=nginx --add-module=/root/nginx-accesskey-2.0.3
> --with-http_ssl_module --with-http_flv_module --with-http_mp4_module
> --with-http_gzip_static_module --http-log-path=/var/log/nginx/access.log
> --http-client-body-temp-path=/var/tmp/nginx/client/
> --http-proxy-temp-path=/var/tmp/nginx/proxy/
> --http-fastcgi-temp-path=/var/tmp/nginx/fcgi/

Could you create debugging log of the request ?
http://nginx.org/en/docs/debugging_log.html


-- 
Igor Sysoev


From grails at jmsd.co.uk  Mon Nov 21 09:35:25 2011
From: grails at jmsd.co.uk (John Moore)
Date: Mon, 21 Nov 2011 09:35:25 +0000
Subject: nginx serving old version of CSS file - SOLVED(ish)
In-Reply-To: <CAOKiq2A-frUbS0410tVmtsQ22GcQp-jVsbWTCs9c+Dy3Sk0gJQ@mail.gmail.com>
References: <4EC96E33.6090004@jmsd.co.uk>
	<CAOKiq2A-frUbS0410tVmtsQ22GcQp-jVsbWTCs9c+Dy3Sk0gJQ@mail.gmail.com>
Message-ID: <4ECA1B5D.4010702@jmsd.co.uk>

On 21/11/11 05:30, Joshua Zhu wrote:
>
>
> Did you turn on open_file_cache in your configuration? It's buggy.
>

No, I didn't. I actually made no change to the configuration other than 
adding the virtual host.

I have, for the moment, solved the problem by moving the root to a 
different location, with the result that the CSS is now served fine. I 
think it must be to do with a bug in VirtualBox, which I have been using 
to provide virtualization. I have a Linux virtual machine running under 
Windows 7. I had moved the virtual host's root to a folder shared from 
the host, so I could work on the website under Windows 7 (and have it 
served by a Linux server). I think there must be a bug in the shared 
folder functionality in the version of VirtualBox I'm using.


From nginx-forum at nginx.us  Mon Nov 21 11:08:09 2011
From: nginx-forum at nginx.us (vanderkerkoff)
Date: Mon, 21 Nov 2011 06:08:09 -0500
Subject: shibboleth and authorizer mode
Message-ID: <ba02f8ed73b2e3dad9d060e8c051a1be.NginxMailingListEnglish@forum.nginx.org>

Hello everyone

I need to confirm something for a work colleague

This page

https://wiki.shibboleth.net/confluence/display/SHIB2/NativeSPFastCGIConfig

 states this

"Nginx has fastcgi responder support but not the "authorizer" mode
needed for the filter. Until this support exists nginx cannot be used in
this way."

Can anyone confirm that is still the case?  If so, are there any plans
to remedy this situation?  If not, please say, I just need to let
someone else know.

Thanks

V

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218694,218694#msg-218694


From mdounin at mdounin.ru  Mon Nov 21 11:17:32 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Mon, 21 Nov 2011 15:17:32 +0400
Subject: shibboleth and authorizer mode
In-Reply-To: <ba02f8ed73b2e3dad9d060e8c051a1be.NginxMailingListEnglish@forum.nginx.org>
References: <ba02f8ed73b2e3dad9d060e8c051a1be.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20111121111732.GU95664@mdounin.ru>

Hello!

On Mon, Nov 21, 2011 at 06:08:09AM -0500, vanderkerkoff wrote:

> Hello everyone
> 
> I need to confirm something for a work colleague
> 
> This page
> 
> https://wiki.shibboleth.net/confluence/display/SHIB2/NativeSPFastCGIConfig
> 
>  states this
> 
> "Nginx has fastcgi responder support but not the "authorizer" mode
> needed for the filter. Until this support exists nginx cannot be used in
> this way."
> 
> Can anyone confirm that is still the case?  If so, are there any plans
> to remedy this situation?  If not, please say, I just need to let
> someone else know.

Confirmed: nginx doesn't support fastcgi authorizer role (as well 
as fastcgi filter role).

There are no plans to fix this.  If you need functionality mostly 
identical to fastcgi authorizers (but not bound to fastcgi), you 
may want to take a look at auth_request module,

http://mdounin.ru/hg/ngx_http_auth_request_module/

Maxim Dounin


From nginx-forum at nginx.us  Mon Nov 21 11:22:26 2011
From: nginx-forum at nginx.us (vanderkerkoff)
Date: Mon, 21 Nov 2011 06:22:26 -0500
Subject: shibboleth and authorizer mode
In-Reply-To: <20111121111732.GU95664@mdounin.ru>
References: <20111121111732.GU95664@mdounin.ru>
Message-ID: <c7cc65882e66b03f979fc0d996f20ec4.NginxMailingListEnglish@forum.nginx.org>

Thanks Maxim

You wouldn't know off the top of your head if I could use auth_request
to authorize with shibboleth would you?

V

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218694,218696#msg-218696


From nginx-forum at nginx.us  Mon Nov 21 12:01:44 2011
From: nginx-forum at nginx.us (vedivad)
Date: Mon, 21 Nov 2011 07:01:44 -0500
Subject: 2 steps authentication!
Message-ID: <441d52ada2110e447a364a9aa57529ba.NginxMailingListEnglish@forum.nginx.org>

Google 2 step authentication has a pam module:

http://code.google.com/p/google-authenticator/wiki/PamModuleInstructions

So, theoretically, it could be implemented into nginx. What do you think
about it?

My goal is to have 2 step auth, somehow, on some dir of my web server
(some directory also has static content, so it isn't a php script i can
edit)

I think that it would be interesting. Many thanks!

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218638,218638#msg-218638


From appa at perusio.net  Mon Nov 21 12:24:52 2011
From: appa at perusio.net (=?UTF-8?B?QW50w7NuaW8=?= P. P. Almeida)
Date: Mon, 21 Nov 2011 12:24:52 +0000
Subject: 2 steps authentication!
In-Reply-To: <441d52ada2110e447a364a9aa57529ba.NginxMailingListEnglish@forum.nginx.org>
References: <441d52ada2110e447a364a9aa57529ba.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <874nxxr7cb.wl%appa@perusio.net>

On 21 Nov 2011 12h01 WET, nginx-forum at nginx.us wrote:

> Google 2 step authentication has a pam module:
>
> http://code.google.com/p/google-authenticator/wiki/PamModuleInstructions

You can implement 2 factor authentication on Nginx using the
auth_request module:

http://mdounin.ru/hg/ngx_http_auth_request_module/file/a29d74804ff1/README

--- appa


From rami.essaid at gmail.com  Mon Nov 21 12:25:39 2011
From: rami.essaid at gmail.com (Rami Essaid)
Date: Mon, 21 Nov 2011 07:25:39 -0500
Subject: nginx behind load balancer
Message-ID: <CAPMF7wjD8BYcac6NVjmOMSEVfwnKsT-_rrqh2sRn8YexPKZ4sA@mail.gmail.com>

Hi Guys,

This weekend for scalability we tried putting our nginx servers behind
amazon's elastic load balancers and came across a road block: it does not
transparently pass the user IP and header information to nginx. This caused
issues with several pieces of nginx we use including the IP allow / deny
rules, the limit_req module, and the limit_con module.  Has anyone
successfully put nginx behind a load balancer?  Any ideas on how to make
this work?

Thanks!
Rami
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20111121/8add0ff1/attachment.html>

From mdounin at mdounin.ru  Mon Nov 21 12:35:41 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Mon, 21 Nov 2011 16:35:41 +0400
Subject: shibboleth and authorizer mode
In-Reply-To: <c7cc65882e66b03f979fc0d996f20ec4.NginxMailingListEnglish@forum.nginx.org>
References: <20111121111732.GU95664@mdounin.ru>
	<c7cc65882e66b03f979fc0d996f20ec4.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20111121123541.GV95664@mdounin.ru>

Hello!

On Mon, Nov 21, 2011 at 06:22:26AM -0500, vanderkerkoff wrote:

> Thanks Maxim
> 
> You wouldn't know off the top of your head if I could use auth_request
> to authorize with shibboleth would you?

No idea, most likely the answer is "no" as long as it needs 
fastcgi authorizer role.

Auth_request is similar in the functionality it provides, but a 
bit different in details how it's implemented.

Maxim Dounin


From mdounin at mdounin.ru  Mon Nov 21 12:38:22 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Mon, 21 Nov 2011 16:38:22 +0400
Subject: nginx behind load balancer
In-Reply-To: <CAPMF7wjD8BYcac6NVjmOMSEVfwnKsT-_rrqh2sRn8YexPKZ4sA@mail.gmail.com>
References: <CAPMF7wjD8BYcac6NVjmOMSEVfwnKsT-_rrqh2sRn8YexPKZ4sA@mail.gmail.com>
Message-ID: <20111121123822.GW95664@mdounin.ru>

Hello!

On Mon, Nov 21, 2011 at 07:25:39AM -0500, Rami Essaid wrote:

> Hi Guys,
> 
> This weekend for scalability we tried putting our nginx servers behind
> amazon's elastic load balancers and came across a road block: it does not
> transparently pass the user IP and header information to nginx. This caused
> issues with several pieces of nginx we use including the IP allow / deny
> rules, the limit_req module, and the limit_con module.  Has anyone
> successfully put nginx behind a load balancer?  Any ideas on how to make
> this work?

http://wiki.nginx.org/HttpRealIpModule

Maxim Dounin


From rami.essaid at gmail.com  Mon Nov 21 12:40:49 2011
From: rami.essaid at gmail.com (Rami Essaid)
Date: Mon, 21 Nov 2011 07:40:49 -0500
Subject: nginx behind load balancer
In-Reply-To: <20111121123822.GW95664@mdounin.ru>
References: <CAPMF7wjD8BYcac6NVjmOMSEVfwnKsT-_rrqh2sRn8YexPKZ4sA@mail.gmail.com>
	<20111121123822.GW95664@mdounin.ru>
Message-ID: <CAPMF7wggdXCT=EQH57gGgM6SWP-o47pBmJWnsbCgSBCSk0OfQQ@mail.gmail.com>

Thanks Maxim,

This looks like exactly what we need.  In your experience does this resolve
most issues behind a load balancer?



On Mon, Nov 21, 2011 at 7:38 AM, Maxim Dounin <mdounin at mdounin.ru> wrote:

> Hello!
>
> On Mon, Nov 21, 2011 at 07:25:39AM -0500, Rami Essaid wrote:
>
> > Hi Guys,
> >
> > This weekend for scalability we tried putting our nginx servers behind
> > amazon's elastic load balancers and came across a road block: it does not
> > transparently pass the user IP and header information to nginx. This
> caused
> > issues with several pieces of nginx we use including the IP allow / deny
> > rules, the limit_req module, and the limit_con module.  Has anyone
> > successfully put nginx behind a load balancer?  Any ideas on how to make
> > this work?
>
> http://wiki.nginx.org/HttpRealIpModule
>
> 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/20111121/86abc937/attachment.html>

From mdounin at mdounin.ru  Mon Nov 21 12:46:02 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Mon, 21 Nov 2011 16:46:02 +0400
Subject: nginx behind load balancer
In-Reply-To: <CAPMF7wggdXCT=EQH57gGgM6SWP-o47pBmJWnsbCgSBCSk0OfQQ@mail.gmail.com>
References: <CAPMF7wjD8BYcac6NVjmOMSEVfwnKsT-_rrqh2sRn8YexPKZ4sA@mail.gmail.com>
	<20111121123822.GW95664@mdounin.ru>
	<CAPMF7wggdXCT=EQH57gGgM6SWP-o47pBmJWnsbCgSBCSk0OfQQ@mail.gmail.com>
Message-ID: <20111121124602.GX95664@mdounin.ru>

Hello!

On Mon, Nov 21, 2011 at 07:40:49AM -0500, Rami Essaid wrote:

> Thanks Maxim,
> 
> This looks like exactly what we need.  In your experience does this resolve
> most issues behind a load balancer?

Yes.

Maxim Dounin

> On Mon, Nov 21, 2011 at 7:38 AM, Maxim Dounin <mdounin at mdounin.ru> wrote:
> 
> > Hello!
> >
> > On Mon, Nov 21, 2011 at 07:25:39AM -0500, Rami Essaid wrote:
> >
> > > Hi Guys,
> > >
> > > This weekend for scalability we tried putting our nginx servers behind
> > > amazon's elastic load balancers and came across a road block: it does not
> > > transparently pass the user IP and header information to nginx. This
> > caused
> > > issues with several pieces of nginx we use including the IP allow / deny
> > > rules, the limit_req module, and the limit_con module.  Has anyone
> > > successfully put nginx behind a load balancer?  Any ideas on how to make
> > > this work?
> >
> > http://wiki.nginx.org/HttpRealIpModule
> >
> > 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  Mon Nov 21 12:49:21 2011
From: nginx-forum at nginx.us (vanderkerkoff)
Date: Mon, 21 Nov 2011 07:49:21 -0500
Subject: shibboleth and authorizer mode
In-Reply-To: <20111121123541.GV95664@mdounin.ru>
References: <20111121123541.GV95664@mdounin.ru>
Message-ID: <8c53ae3bd0dc13b71ea6fe4675520f08.NginxMailingListEnglish@forum.nginx.org>

Thanks Maxim

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218694,218705#msg-218705


From nginx-forum at nginx.us  Mon Nov 21 13:15:03 2011
From: nginx-forum at nginx.us (TECK)
Date: Mon, 21 Nov 2011 08:15:03 -0500
Subject: Enable SSL for a specific directory?
In-Reply-To: <87k46vrbc5.wl%appa@perusio.net>
References: <87k46vrbc5.wl%appa@perusio.net>
Message-ID: <5345cb4d51d7b11fa399a7c47badfebb.NginxMailingListEnglish@forum.nginx.org>

Actually, in my case the fix was to add:
fastcgi_param HTTPS on;

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218610,218706#msg-218706


From nginx-forum at nginx.us  Mon Nov 21 14:12:57 2011
From: nginx-forum at nginx.us (izrodix)
Date: Mon, 21 Nov 2011 09:12:57 -0500
Subject: mod_mysql_userdir ?
Message-ID: <68fe21150140c60c319091d9125052e2.NginxMailingListEnglish@forum.nginx.org>

Hello Nginx Team,
we would like to ask have you  got solution for following problem.
We have ftp users that are authorized in mysql they want to see their
homedirs
through the web by using the same user and password.

for example:
http://site.com/username1/ -> requires the username1 for username and
his password1
http://site.com/username2/ -> requires the username2 for username and
his password2

But when open http://site.com/username1/  and auth with username2
password  the reply from webserver must be  access denied or rewrites
the path to http://site.com/username2/ and show his homedir.

it is similar like UserDir in Apache or
(http://www.mgamble.ca/oss/mod_mysql_userdir/) but we want to use it in
nginx.

Regards,

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218708,218708#msg-218708


From nginx-forum at nginx.us  Mon Nov 21 17:03:27 2011
From: nginx-forum at nginx.us (scriptonian)
Date: Mon, 21 Nov 2011 12:03:27 -0500
Subject: Cant start Nginx on Mac OS X
In-Reply-To: <1321851413.2289.2.camel@sharl-desktop>
References: <1321851413.2289.2.camel@sharl-desktop>
Message-ID: <454ad2f2bbd87295230da54b2704f880.NginxMailingListEnglish@forum.nginx.org>

Thanks so much guys. I got all the errors to go away.... however, when i
start the server i get this

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()


looks like a port conflict to me. i dont know. but i dont have any other
servers running. when i hit http://localhost/ 404 Not Found (
nginx/1.0.8 )

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218663,218713#msg-218713


From nginx-forum at nginx.us  Mon Nov 21 17:05:09 2011
From: nginx-forum at nginx.us (scriptonian)
Date: Mon, 21 Nov 2011 12:05:09 -0500
Subject: Cant start Nginx on Mac OS X
In-Reply-To: <454ad2f2bbd87295230da54b2704f880.NginxMailingListEnglish@forum.nginx.org>
References: <1321851413.2289.2.camel@sharl-desktop>
	<454ad2f2bbd87295230da54b2704f880.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <9a15bb48cfedde2c17a59f5732f56c45.NginxMailingListEnglish@forum.nginx.org>

and when i do 

nginx -s stop

i can NOT even get a 404 error page. so something is working right at
least. but i still have a blocker.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218663,218714#msg-218714


From calin at presslabs.com  Mon Nov 21 18:41:15 2011
From: calin at presslabs.com (Calin Don)
Date: Mon, 21 Nov 2011 20:41:15 +0200
Subject: $request_time meaning
Message-ID: <CAEOe2JzF3si7pz3vi+4VBj7S_qAnX2HWcWT6Css=OCGr4dBtng@mail.gmail.com>

Hi,
How is the $request_time calculated in the case of a proxied request. I'm
interested especially in the case where the resource is stale in cache.

Thanks
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20111121/4a252f0e/attachment.html>

From mdounin at mdounin.ru  Mon Nov 21 18:55:59 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Mon, 21 Nov 2011 22:55:59 +0400
Subject: $request_time meaning
In-Reply-To: <CAEOe2JzF3si7pz3vi+4VBj7S_qAnX2HWcWT6Css=OCGr4dBtng@mail.gmail.com>
References: <CAEOe2JzF3si7pz3vi+4VBj7S_qAnX2HWcWT6Css=OCGr4dBtng@mail.gmail.com>
Message-ID: <20111121185559.GB95664@mdounin.ru>

Hello!

On Mon, Nov 21, 2011 at 08:41:15PM +0200, Calin Don wrote:

> Hi,
> How is the $request_time calculated in the case of a proxied request. I'm
> interested especially in the case where the resource is stale in cache.

$request_time is always time since start of the request (when 
first bytes are read from client) till end of the request (when 
last bytes are sent to client and logging happens).

Maxim Dounin


From nginx-forum at nginx.us  Mon Nov 21 19:21:03 2011
From: nginx-forum at nginx.us (bizshop)
Date: Mon, 21 Nov 2011 14:21:03 -0500
Subject: localhost path to images directory
Message-ID: <00590090b2e42628f5ac4ff4d8694419.NginxMailingListEnglish@forum.nginx.org>

I have nginx and php5-fpm set up and working to serve /var/www/index.php
as localhost/index.php
That works fine. The images on the page are not showing up however. The
images are in a folder /var/www/images
The  folder has same user/group permissions as the index page
I can see an image in my file manager at /var/www/images/key.gif and
gave it 777 permission
Page source shows image should be at relative path images/key.gif 
Pageinfo  (Firefox) says image should be at
http://localhost/images/key.gif
Typing directly in browser window http://localhost/images/key.gif gets a
404

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218718,218718#msg-218718


From francis at daoine.org  Mon Nov 21 19:55:25 2011
From: francis at daoine.org (Francis Daly)
Date: Mon, 21 Nov 2011 19:55:25 +0000
Subject: localhost path to images directory
In-Reply-To: <00590090b2e42628f5ac4ff4d8694419.NginxMailingListEnglish@forum.nginx.org>
References: <00590090b2e42628f5ac4ff4d8694419.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20111121195525.GA27443@craic.sysops.org>

On Mon, Nov 21, 2011 at 02:21:03PM -0500, bizshop wrote:

Hi there,

> That works fine. The images on the page are not showing up however. The
> images are in a folder /var/www/images

> Typing directly in browser window http://localhost/images/key.gif gets a
> 404

The error log probably tells you what file nginx failed to find.

Do you have "root /var/www;" in the server{} block, outside of all
location{}s?

	f
-- 
Francis Daly        francis at daoine.org


From nginx-forum at nginx.us  Mon Nov 21 20:10:48 2011
From: nginx-forum at nginx.us (bizshop)
Date: Mon, 21 Nov 2011 15:10:48 -0500
Subject: localhost path to images directory
In-Reply-To: <20111121195525.GA27443@craic.sysops.org>
References: <20111121195525.GA27443@craic.sysops.org>
Message-ID: <2b719b9e1204d81e5ac132512b05899a.NginxMailingListEnglish@forum.nginx.org>

Thanks - that got me started in the right direction. Had to set images
path in the default conf. Not something I had to do with apache as
server, still learning this nginx!

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218718,218720#msg-218720


From quintinpar at gmail.com  Tue Nov 22 03:23:52 2011
From: quintinpar at gmail.com (Quintin Par)
Date: Tue, 22 Nov 2011 08:53:52 +0530
Subject: Is proxy_cache_bypass recaching?
Message-ID: <CACW9FdOHmK2S-O83LWA7tA_krjANZ6+7DffbPWanUkZ=Fr6ABQ@mail.gmail.com>

Hi all,

I have this in my location directive

                    proxy_cache_bypass                     $http_soon;

now I do a curl call to invalidate



[user at ip-10-111 ~]$ curl "www.site.com/teaser" -I -H "soon:true"

HTTP/1.1 200 OK

Server: nginx

Date: Tue, 22 Nov 2011 03:07:20 GMT

Content-Type: text/html; charset=utf-8

Connection: keep-alive

Keep-Alive: timeout=60

Vary: Accept-Encoding

Vary: Cookie

X-Mod-Pagespeed: 0.9.17.7-716

Cache-Control: max-age=0, no-cache, no-store

X-Cache-Status: BYPASS



[user at ip-10-111 ~]$ curl "www.site.com/teaser" -I

HTTP/1.1 200 OK

Server: nginx

Date: Tue, 22 Nov 2011 03:07:37 GMT

Content-Type: text/html; charset=utf-8

Connection: keep-alive

Keep-Alive: timeout=60

Vary: Accept-Encoding

Vary: Cookie

X-Mod-Pagespeed: 0.9.17.7-716

Cache-Control: max-age=0, no-cache, no-store

X-Cache-Status: HIT



Does this mean that the curl call with the header is* recaching the page
and the subsequent hits are cache hits?*

-Quintin
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20111122/b0b98e3e/attachment.html>

From lists at ruby-forum.com  Tue Nov 22 05:51:56 2011
From: lists at ruby-forum.com (Sandesh K.)
Date: Tue, 22 Nov 2011 06:51:56 +0100
Subject: Nginx stuck processes
In-Reply-To: <7efb85fdbf5ec94ab17531ab41499073@ruby-forum.com>
References: <d04888efb006752e263fc5bca2c0d314@ruby-forum.com>
	<7efb85fdbf5ec94ab17531ab41499073@ruby-forum.com>
Message-ID: <31b70aa141cf1024aba7576099862a99@ruby-forum.com>

Sandesh K. wrote in post #1027339:
> Hello Igor,
>
> Do you have any update on this?

Igor,

We upgraded the OS as well as Nginx to 1.1.10 latest stable. But the 
problem still persists.


*2.6.18-164.el5PAE
*CentOS release 5.4 (Final)
*nginx: nginx version: nginx/1.0.10
*nginx: built by gcc 4.1.2 20080704 (Red Hat 4.1.2-46)
nginx: configure arguments: --prefix=/usr/local/nginx 
--sbin-path=/usr/local/nginx/sbin/nginx 
--conf-path=/etc/nginx/conf/nginx.conf 
--error-log-path=/var/log/nginx/logs/error.log 
--pid-path=/var/run/nginx.pid --lock-path=/var/log/nginx/logs/nginx.lock 
--with-mail --without-http --without-mail_imap_module 
--without-mail_smtp_module



GNU gdb Red Hat Linux (6.3.0.0-0.30.1rh)
Copyright 2004 Free Software Foundation, Inc.
GDB is free software, covered by the GNU General Public License, and you 
are
welcome to change it and/or distribute copies of it under certain 
conditions.
Type "show copying" to see the conditions.
There is absolutely no warranty for GDB. Type "show warranty" for 
details.
This GDB was configured as "i386-redhat-linux-gnu".
(gdb) attach 65233
Attaching to process 65233
Reading symbols from /usr/local/nginx/sbin/nginx...done.
Using host libthread_db library "/lib/tls/libthread_db.so.1".
Reading symbols from /lib/tls/libpthread.so.0...done.
[Thread debugging using libthread_db enabled]
[New Thread -1218550432 (LWP 65233)]
Loaded symbols for /lib/tls/libpthread.so.0
Reading symbols from /lib/libcrypto.so.4...done.
Loaded symbols for /lib/libcrypto.so.4
Reading symbols from /lib/tls/libc.so.6...done.
Loaded symbols for /lib/tls/libc.so.6
Reading symbols from /lib/ld-linux.so.2...done.
Loaded symbols for /lib/ld-linux.so.2
Reading symbols from /usr/kerberos/lib/libgssapi_krb5.so.2...done.
Loaded symbols for /usr/kerberos/lib/libgssapi_krb5.so.2
Reading symbols from /usr/kerberos/lib/libkrb5.so.3...done.
Loaded symbols for /usr/kerberos/lib/libkrb5.so.3
Reading symbols from /usr/kerberos/lib/libcom_err.so.3...done.
Loaded symbols for /usr/kerberos/lib/libcom_err.so.3
Reading symbols from /usr/kerberos/lib/libk5crypto.so.3...done.
Loaded symbols for /usr/kerberos/lib/libk5crypto.so.3
Reading symbols from /lib/libresolv.so.2...done.
Loaded symbols for /lib/libresolv.so.2
Reading symbols from /lib/libdl.so.2...done.
Loaded symbols for /lib/libdl.so.2
Reading symbols from /usr/lib/libz.so.1...done.
Loaded symbols for /usr/lib/libz.so.1
Reading symbols from /lib/libnss_files.so.2...done.
Loaded symbols for /lib/libnss_files.so.2
0x0031ebba in _int_malloc () from /lib/tls/libc.so.6
(gdb) backtrace
#0 0x0031ebba in _int_malloc () from /lib/tls/libc.so.6
#1 0x0031f89c in _int_memalign () from /lib/tls/libc.so.6
#2 0x0031e2c7 in memalign () from /lib/tls/libc.so.6
#3 0x00320367 in posix_memalign () from /lib/tls/libc.so.6
#4 0x0805da38 in ngx_memalign (alignment=16, size=4069528, log=0x108) at 
src/os/unix/ngx_alloc.c:56
#5 0x0804b917 in ngx_palloc_block (pool=0x8d0b9b0, size=52) at 
src/core/ngx_palloc.c:184
#6 0x0804b892 in ngx_palloc (pool=0x8d0b9b0, size=52) at 
src/core/ngx_palloc.c:138
#7 0x0804babf in ngx_pcalloc (pool=0x8d0b9b0, size=52) at 
src/core/ngx_palloc.c:304
#8 0x0804d0b7 in ngx_create_temp_buf (pool=0x8d0b9b0, size=4096) at 
src/core/ngx_buf.c:16
#9 0x08067508 in ngx_mail_proxy_init (s=0x8d0da20, peer=0x8d0c294) at 
src/mail/ngx_mail_proxy_module.c:167
#10 0x080666e2 in ngx_mail_auth_http_process_headers (s=0x8d0da20, 
ctx=0x8d12498) at src/mail/ngx_mail_auth_http_module.c:832
#11 0x0806606c in ngx_mail_auth_http_ignore_status_line (s=0x8d0da20, 
ctx=0x8d12498) at src/mail/ngx_mail_auth_http_module.c:448
#12 0x08065f4b in ngx_mail_auth_http_read_handler (rev=0x3e2180) at 
src/mail/ngx_mail_auth_http_module.c:331
#13 0x0806194e in ngx_rtsig_process_events (cycle=0x8cdb248, 
timer=147634972, flags=1) at src/event/modules/ngx_rtsig_module.c:400
#14 0x0805ad0c in ngx_process_events_and_timers (cycle=0x8ccb228) at 
src/event/ngx_event.c:245
#15 0x08060952 in ngx_worker_process_cycle (cycle=0x8ccb228, data=0x0) 
at src/os/unix/ngx_process_cycle.c:801
#16 0x0805ef60 in ngx_spawn_process (cycle=0x8ccb228, proc=0x806089d 
<ngx_worker_process_cycle>, data=0x0, name=0x806be29 "worker process", 
respawn=-3) at src/os/unix/ngx_process.c:196
#17 0x080601f6 in ngx_start_worker_processes (cycle=0x8ccb228, n=30, 
type=-3) at src/os/unix/ngx_process_cycle.c:360
#18 0x0805fbe7 in ngx_master_process_cycle (cycle=0x8ccb228) at 
src/os/unix/ngx_process_cycle.c:136
#19 0x0804a116 in main (argc=3, argv=0x8ccb9a4) at src/core/nginx.c:405
(gdb) quit

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


From nginx-forum at nginx.us  Tue Nov 22 06:43:53 2011
From: nginx-forum at nginx.us (sibsoft)
Date: Tue, 22 Nov 2011 01:43:53 -0500
Subject: MP4 - start time is out mp4 stts samples
In-Reply-To: <5e238ac1a229e14931b6ea72c107997c.NginxMailingListEnglish@forum.nginx.org>
References: <20111117050603.GA10385@nginx.com>
	<5e238ac1a229e14931b6ea72c107997c.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <1d125d44f314efdf2dc871c3544f0614.NginxMailingListEnglish@forum.nginx.org>

Here it goes
http://dl.dropbox.com/u/50505452/error.log.gz

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218451,218731#msg-218731


From mdounin at mdounin.ru  Tue Nov 22 08:58:57 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Tue, 22 Nov 2011 12:58:57 +0400
Subject: Nginx stuck processes
In-Reply-To: <31b70aa141cf1024aba7576099862a99@ruby-forum.com>
References: <d04888efb006752e263fc5bca2c0d314@ruby-forum.com>
	<7efb85fdbf5ec94ab17531ab41499073@ruby-forum.com>
	<31b70aa141cf1024aba7576099862a99@ruby-forum.com>
Message-ID: <20111122085856.GD95664@mdounin.ru>

Hello!

On Tue, Nov 22, 2011 at 06:51:56AM +0100, Sandesh K. wrote:

> Sandesh K. wrote in post #1027339:
> > Hello Igor,
> >
> > Do you have any update on this?
> 
> Igor,
> 
> We upgraded the OS as well as Nginx to 1.1.10 latest stable. But the 
> problem still persists.
> 
> 
> *2.6.18-164.el5PAE
> *CentOS release 5.4 (Final)
> *nginx: nginx version: nginx/1.0.10
> *nginx: built by gcc 4.1.2 20080704 (Red Hat 4.1.2-46)
> nginx: configure arguments: --prefix=/usr/local/nginx 
> --sbin-path=/usr/local/nginx/sbin/nginx 
> --conf-path=/etc/nginx/conf/nginx.conf 
> --error-log-path=/var/log/nginx/logs/error.log 
> --pid-path=/var/run/nginx.pid --lock-path=/var/log/nginx/logs/nginx.lock 
> --with-mail --without-http --without-mail_imap_module 
> --without-mail_smtp_module

[...]

> (gdb) backtrace
> #0 0x0031ebba in _int_malloc () from /lib/tls/libc.so.6
> #1 0x0031f89c in _int_memalign () from /lib/tls/libc.so.6
> #2 0x0031e2c7 in memalign () from /lib/tls/libc.so.6
> #3 0x00320367 in posix_memalign () from /lib/tls/libc.so.6
> #4 0x0805da38 in ngx_memalign (alignment=16, size=4069528, log=0x108) at 
> src/os/unix/ngx_alloc.c:56

The "size=4069528" looks really strange...

> #5 0x0804b917 in ngx_palloc_block (pool=0x8d0b9b0, size=52) at 
> src/core/ngx_palloc.c:184

Could you please show

fr 5
p *pool

output from gdb?

> #6 0x0804b892 in ngx_palloc (pool=0x8d0b9b0, size=52) at 
> src/core/ngx_palloc.c:138
> #7 0x0804babf in ngx_pcalloc (pool=0x8d0b9b0, size=52) at 
> src/core/ngx_palloc.c:304
> #8 0x0804d0b7 in ngx_create_temp_buf (pool=0x8d0b9b0, size=4096) at 
> src/core/ngx_buf.c:16
> #9 0x08067508 in ngx_mail_proxy_init (s=0x8d0da20, peer=0x8d0c294) at 
> src/mail/ngx_mail_proxy_module.c:167
> #10 0x080666e2 in ngx_mail_auth_http_process_headers (s=0x8d0da20, 
> ctx=0x8d12498) at src/mail/ngx_mail_auth_http_module.c:832
> #11 0x0806606c in ngx_mail_auth_http_ignore_status_line (s=0x8d0da20, 
> ctx=0x8d12498) at src/mail/ngx_mail_auth_http_module.c:448
> #12 0x08065f4b in ngx_mail_auth_http_read_handler (rev=0x3e2180) at 
> src/mail/ngx_mail_auth_http_module.c:331
> #13 0x0806194e in ngx_rtsig_process_events (cycle=0x8cdb248, 
> timer=147634972, flags=1) at src/event/modules/ngx_rtsig_module.c:400
> #14 0x0805ad0c in ngx_process_events_and_timers (cycle=0x8ccb228) at 
> src/event/ngx_event.c:245
> #15 0x08060952 in ngx_worker_process_cycle (cycle=0x8ccb228, data=0x0) 
> at src/os/unix/ngx_process_cycle.c:801
> #16 0x0805ef60 in ngx_spawn_process (cycle=0x8ccb228, proc=0x806089d 
> <ngx_worker_process_cycle>, data=0x0, name=0x806be29 "worker process", 
> respawn=-3) at src/os/unix/ngx_process.c:196
> #17 0x080601f6 in ngx_start_worker_processes (cycle=0x8ccb228, n=30, 
> type=-3) at src/os/unix/ngx_process_cycle.c:360
> #18 0x0805fbe7 in ngx_master_process_cycle (cycle=0x8ccb228) at 
> src/os/unix/ngx_process_cycle.c:136
> #19 0x0804a116 in main (argc=3, argv=0x8ccb9a4) at src/core/nginx.c:405
> (gdb) quit

Maxim Dounin


From igor at sysoev.ru  Tue Nov 22 10:30:46 2011
From: igor at sysoev.ru (Igor Sysoev)
Date: Tue, 22 Nov 2011 14:30:46 +0400
Subject: MP4 - start time is out mp4 stts samples
In-Reply-To: <1d125d44f314efdf2dc871c3544f0614.NginxMailingListEnglish@forum.nginx.org>
References: <20111117050603.GA10385@nginx.com>
	<5e238ac1a229e14931b6ea72c107997c.NginxMailingListEnglish@forum.nginx.org>
	<1d125d44f314efdf2dc871c3544f0614.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20111122103045.GE2383@nginx.com>

On Tue, Nov 22, 2011 at 01:43:53AM -0500, sibsoft wrote:
> Here it goes
> http://dl.dropbox.com/u/50505452/error.log.gz

The attached patch should help.


-- 
Igor Sysoev
-------------- next part --------------
Index: src/http/modules/ngx_http_mp4_module.c
===================================================================
--- src/http/modules/ngx_http_mp4_module.c	(revision 4303)
+++ src/http/modules/ngx_http_mp4_module.c	(working copy)
@@ -1899,7 +1899,7 @@
         ngx_log_debug2(NGX_LOG_DEBUG_HTTP, mp4->file.log, 0,
                        "count:%uD, duration:%uD", count, duration);
 
-        if (start_time < count * duration) {
+        if (start_time < (uint64_t) count * duration) {
             start_sample += (ngx_uint_t) (start_time / duration);
             count -= start_sample;
             ngx_mp4_set_32value(entry->count, count);

From nginx-forum at nginx.us  Tue Nov 22 10:38:11 2011
From: nginx-forum at nginx.us (wangbin579)
Date: Tue, 22 Nov 2011 05:38:11 -0500
Subject: nginx 100% cpu and it may be a nginx bug
Message-ID: <8aa1c8fc8687fcab280b76c5e85c55c4.NginxMailingListEnglish@forum.nginx.org>

The function ngx_http_upstream_get_round_robin_peer in nginx will loop
for a long period of times when pc->tries equals to 0.

Well then, under what conditions does pc->tries equal to zero?

When executing ngx_http_upstream_get_round_robin_peer for the first
time, in some cases it returns NGX_OK while pc->tries equals to 1. 

When in ngx_http_upstream_process_header, nginx process may enter the
following:
1524         if (n == NGX_ERROR || n == 0) {
1525             ngx_http_upstream_next(r, u,
NGX_HTTP_UPSTREAM_FT_ERROR);
1526             return;
1527         }

then enter ngx_http_upstream_next function

In the ngx_http_upstream_next,nginx process will call
ngx_http_upstream_free_round_robin_peer

2796     if (ft_type != NGX_HTTP_UPSTREAM_FT_NOLIVE) {
2797         u->peer.free(&u->peer, u->peer.data, state);
2798     }


we now enter ngx_http_upstream_free_round_robin_peer.

Pay attention to the following codes in the function:
if (pc->tries) {
   pc->tries--;
}

After executing the above, pc->tries changes from 1 to 0.

then the process returns to ngx_http_upstream_next

2892     ngx_http_upstream_connect(r, u);
2893 }

we enter ngx_http_upstream_connect here:

this function will call 

1081     rc = ngx_event_connect_peer(&u->peer);


...


It finally enter ngx_http_upstream_get_round_robin_peer again. And then
the process will loop for a long time which is unpredictable.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218736,218736#msg-218736


From nginx-forum at nginx.us  Tue Nov 22 10:41:13 2011
From: nginx-forum at nginx.us (wangbin579)
Date: Tue, 22 Nov 2011 05:41:13 -0500
Subject: nginx 100% cpu and it may be a nginx bug
In-Reply-To: <8aa1c8fc8687fcab280b76c5e85c55c4.NginxMailingListEnglish@forum.nginx.org>
References: <8aa1c8fc8687fcab280b76c5e85c55c4.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <0bebbbf3c03d18b641887ca445da92ec.NginxMailingListEnglish@forum.nginx.org>

we find all processes of nginx are all nearly cpu 100%.

we used tcpcopy to find this problem.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218736,218737#msg-218737


From nginx-forum at nginx.us  Tue Nov 22 10:44:07 2011
From: nginx-forum at nginx.us (wangbin579)
Date: Tue, 22 Nov 2011 05:44:07 -0500
Subject: nginx 100% cpu and it may be a nginx bug
In-Reply-To: <0bebbbf3c03d18b641887ca445da92ec.NginxMailingListEnglish@forum.nginx.org>
References: <8aa1c8fc8687fcab280b76c5e85c55c4.NginxMailingListEnglish@forum.nginx.org>
	<0bebbbf3c03d18b641887ca445da92ec.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <dff09c08fd5c70aead4f5f69cd5f2335.NginxMailingListEnglish@forum.nginx.org>

we use gcore to generate a core file.

the backtrace is as the following:


#0  0x000000000046ae0b in ngx_http_upstream_get_round_robin_peer
(pc=0x1b822678, data=0x1b826e10)
    at src/http/ngx_http_upstream_round_robin.c:482
#1  0x00000000004a6a60 in ngx_http_upstream_get_keepalive_peer
(pc=0x1b822678, data=0x1b826de0)
    at
../ngx_http_upstream_keepalive-c6396fef9295/ngx_http_upstream_keepalive_module.c:227
#2  0x0000000000429fab in ngx_event_connect_peer (pc=0x1b822678) at
src/event/ngx_event_connect.c:24
#3  0x000000000046219c in ngx_http_upstream_connect (r=0x1b821230,
u=0x1b822668) at src/http/ngx_http_upstream.c:1103
#4  0x0000000000465f61 in ngx_http_upstream_next (r=0x1b821230,
u=0x1b822668, ft_type=2)
    at src/http/ngx_http_upstream.c:2900
#5  0x00000000004631c0 in ngx_http_upstream_process_header
(r=0x1b821230, u=0x1b822668) at src/http/ngx_http_upstream.c:1547
#6  0x0000000000462d8c in ngx_http_upstream_send_request (r=0x1b821230,
u=0x1b822668) at src/http/ngx_http_upstream.c:1414
#7  0x00000000004625de in ngx_http_upstream_connect (r=0x1b821230,
u=0x1b822668) at src/http/ngx_http_upstream.c:1204
#8  0x000000000046129d in ngx_http_upstream_init_request (r=0x1b821230)
at src/http/ngx_http_upstream.c:631
#9  0x0000000000460a6d in ngx_http_upstream_init (r=0x1b821230) at
src/http/ngx_http_upstream.c:432
#10 0x000000000045782b in ngx_http_read_client_request_body
(r=0x1b821230, post_handler=0x460989 <ngx_http_upstream_init>)
    at src/http/ngx_http_request_body.c:154
#11 0x00000000004a0875 in ngx_hmux_handler (r=0x1b821230) at
../hmux/ngx_hmux_module.c:673
#12 0x000000000043ffd1 in ngx_http_core_content_phase (r=0x1b821230,
ph=0x1b5d4f88) at src/http/ngx_http_core_module.c:1365
#13 0x000000000043ebf5 in ngx_http_core_run_phases (r=0x1b821230) at
src/http/ngx_http_core_module.c:861
#14 0x000000000043eb71 in ngx_http_handler (r=0x1b821230) at
src/http/ngx_http_core_module.c:844
#15 0x000000000044c9a1 in ngx_http_process_request (r=0x1b821230) at
src/http/ngx_http_request.c:1665
#16 0x000000000044b34d in ngx_http_process_request_headers
(rev=0x2b3f10a3f4f0) at src/http/ngx_http_request.c:1111
#17 0x000000000044ab26 in ngx_http_process_request_line
(rev=0x2b3f10a3f4f0) at src/http/ngx_http_request.c:911
#18 0x0000000000449d7d in ngx_http_init_request (rev=0x2b3f10a3f4f0) at
src/http/ngx_http_request.c:518
#19 0x0000000000434ea7 in ngx_epoll_process_events (cycle=0x1b5ae090,
timer=1352, flags=1)
    at src/event/modules/ngx_epoll_module.c:678
#20 0x0000000000426480 in ngx_process_events_and_timers
(cycle=0x1b5ae090) at src/event/ngx_event.c:245
---Type <return> to continue, or q <return> to quit---
#21 0x0000000000432d0c in ngx_worker_process_cycle (cycle=0x1b5ae090,
data=0x0) at src/os/unix/ngx_process_cycle.c:801
#22 0x000000000042faad in ngx_spawn_process (cycle=0x1b5ae090,
proc=0x432b60 <ngx_worker_process_cycle>, data=0x0,
    name=0x57450e "worker process", respawn=-3) at
src/os/unix/ngx_process.c:196
#23 0x0000000000431c12 in ngx_start_worker_processes (cycle=0x1b5ae090,
n=8, type=-3) at src/os/unix/ngx_process_cycle.c:360
#24 0x00000000004312fe in ngx_master_process_cycle (cycle=0x1b5ae090) at
src/os/unix/ngx_process_cycle.c:136
#25 0x0000000000403c8c in main (argc=1, argv=0x7fffba823048) at
src/core/nginx.c:405
(gdb) info lo
rrp = 0x1b826e10
now = 1321860194
m = 8
rc = 461502464
i = 18446744061305903636  

n = 0
c = 0x40
peer = 0x1b5cf2c0
peers = 0x1

(gdb) p *rrp
$5 = {peers = 0x1b5cf230, current = 3, tried = 0x1b826e28, data = 4}
(gdb) p rrp->peers->peer[0]
$6 = {sockaddr = 0x1b5c63b8, socklen = 16, name = {len = 19, data =
0x1b5c63c8 "xxx.xxx.xxx.xxx:36901server"},
  current_weight = 0, weight = 1, fails = 1, accessed = 1321860193,
max_fails = 1, fail_timeout = 10, down = 0,
  ssl_session = 0x0}
(gdb) p rrp->peers->peer[1]
$7 = {sockaddr = 0x1b5c6418, socklen = 16, name = {len = 19, data =
0x1b5c6428 "xxx.xxx.xxx.xxx:36902server"},
  current_weight = 0, weight = 1, fails = 1, accessed = 1321860194,
max_fails = 1, fail_timeout = 10, down = 0,
  ssl_session = 0x0}
(gdb) p rrp->peers->peer[2]
$8 = {sockaddr = 0x1b5c6478, socklen = 16, name = {len = 19, data =
0x1b5c6488 "xxx.xxx.xxx.xxx:36903server"},
  current_weight = 0, weight = 1, fails = 1, accessed = 1321860194,
max_fails = 1, fail_timeout = 10, down = 0,
  ssl_session = 0x0}
(gdb) p rrp->peers->peer[3]
$9 = {sockaddr = 0x1b5c64d8, socklen = 16, name = {len = 19, data =
0x1b5c64e8 "xxx.xxx.xxx.xxx:36904keepalive"},
  current_weight = 0, weight = 1, fails = 1, accessed = 1321860193,
max_fails = 1, fail_timeout = 10, down = 0,
  ssl_session = 0x0}

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218736,218738#msg-218738


From nginx-forum at nginx.us  Tue Nov 22 10:44:53 2011
From: nginx-forum at nginx.us (wangbin579)
Date: Tue, 22 Nov 2011 05:44:53 -0500
Subject: nginx 100% cpu and it may be a nginx bug
In-Reply-To: <dff09c08fd5c70aead4f5f69cd5f2335.NginxMailingListEnglish@forum.nginx.org>
References: <8aa1c8fc8687fcab280b76c5e85c55c4.NginxMailingListEnglish@forum.nginx.org>
	<0bebbbf3c03d18b641887ca445da92ec.NginxMailingListEnglish@forum.nginx.org>
	<dff09c08fd5c70aead4f5f69cd5f2335.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <4a1251180ea80f6248addbf3c4383962.NginxMailingListEnglish@forum.nginx.org>

[root at 208 sbin]# /home/nginx/sbin/nginx -V
nginx: nginx version: nginx/1.0.10
nginx: built by gcc 4.1.2 20080704 (Red Hat 4.1.2-48)
nginx: TLS SNI support enabled
nginx: configure arguments: --prefix=/home/nginx --user=nginx
--group=nginx --without-select_module --without-poll_module
--with-http_ssl_module --with-http_realip_module
--with-http_gzip_static_module --with-http_stub_status_module
--without-http_fastcgi_module --without-http_memcached_module
--without-http_empty_gif_module --with-pcre=../pcre-7.9
--with-openssl=../openssl-0.9.8r --add-module=../hmux
--add-module=../ngx_http_upstream_keepalive-c6396fef9295 --with-debug

part of nginx conf file? 

events {
    worker_connections  8192;
    use epoll;
    epoll_events 4096;
    accept_mutex off;
}

...
    upstream yyy  {
        server xxx.xxx.xxx.xxx:36901;
        server xxx.xxx.xxx.xxx:36902;
        server xxx.xxx.xxx.xxx:36903;
        server xxx.xxx.xxx.xxx:36904;

        keepalive 1024;
    }

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218736,218739#msg-218739


From nginx-forum at nginx.us  Tue Nov 22 10:45:25 2011
From: nginx-forum at nginx.us (wangbin579)
Date: Tue, 22 Nov 2011 05:45:25 -0500
Subject: nginx 100% cpu and it may be a nginx bug
In-Reply-To: <4a1251180ea80f6248addbf3c4383962.NginxMailingListEnglish@forum.nginx.org>
References: <8aa1c8fc8687fcab280b76c5e85c55c4.NginxMailingListEnglish@forum.nginx.org>
	<0bebbbf3c03d18b641887ca445da92ec.NginxMailingListEnglish@forum.nginx.org>
	<dff09c08fd5c70aead4f5f69cd5f2335.NginxMailingListEnglish@forum.nginx.org>
	<4a1251180ea80f6248addbf3c4383962.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <c4f0ff3e27deb893737f407486f029b9.NginxMailingListEnglish@forum.nginx.org>

nginx-0.8.53 has the same problem

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218736,218740#msg-218740


From mdounin at mdounin.ru  Tue Nov 22 11:01:55 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Tue, 22 Nov 2011 15:01:55 +0400
Subject: nginx 100% cpu and it may be a nginx bug
In-Reply-To: <8aa1c8fc8687fcab280b76c5e85c55c4.NginxMailingListEnglish@forum.nginx.org>
References: <8aa1c8fc8687fcab280b76c5e85c55c4.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20111122110155.GF95664@mdounin.ru>

Hello!

On Tue, Nov 22, 2011 at 05:38:11AM -0500, wangbin579 wrote:

> The function ngx_http_upstream_get_round_robin_peer in nginx will loop
> for a long period of times when pc->tries equals to 0.
> 
> Well then, under what conditions does pc->tries equal to zero?
> 
> When executing ngx_http_upstream_get_round_robin_peer for the first
> time, in some cases it returns NGX_OK while pc->tries equals to 1. 
> 
> When in ngx_http_upstream_process_header, nginx process may enter the
> following:
> 1524         if (n == NGX_ERROR || n == 0) {
> 1525             ngx_http_upstream_next(r, u,
> NGX_HTTP_UPSTREAM_FT_ERROR);
> 1526             return;
> 1527         }
> 
> then enter ngx_http_upstream_next function
> 
> In the ngx_http_upstream_next,nginx process will call
> ngx_http_upstream_free_round_robin_peer
> 
> 2796     if (ft_type != NGX_HTTP_UPSTREAM_FT_NOLIVE) {
> 2797         u->peer.free(&u->peer, u->peer.data, state);
> 2798     }
> 
> 
> we now enter ngx_http_upstream_free_round_robin_peer.
> 
> Pay attention to the following codes in the function:
> if (pc->tries) {
>    pc->tries--;
> }
> 
> After executing the above, pc->tries changes from 1 to 0.
> 
> then the process returns to ngx_http_upstream_next
> 
> 2892     ngx_http_upstream_connect(r, u);
> 2893 }

The ngx_http_upstream_next() will finalize request if tries == 0.

You are probably facing the problem with keepalive backend 
connections in 1.0.x (where they aren't officially supported).  It 
is fixed with this changeset while introducing official keepalive 
backend connections support in 1.1.4:

http://trac.nginx.org/nginx/changeset/4116/nginx

Maxim Dounin


From nginx-forum at nginx.us  Tue Nov 22 13:55:59 2011
From: nginx-forum at nginx.us (janedenone)
Date: Tue, 22 Nov 2011 08:55:59 -0500
Subject: Conflict: index and try_files
In-Reply-To: <12b16bcca767bd5986409e54ad6bad9c.NginxMailingListEnglish@forum.nginx.org>
References: <12b16bcca767bd5986409e54ad6bad9c.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <63a6e0b21656160610ccec1a3818385e.NginxMailingListEnglish@forum.nginx.org>

Thanks again for the helpful hint. I noticed that it seems to be
possible to remove the $uri part ?

try_files $uri/ $uri.html =404;

? and URLs like /image.jpg, /style.css etc are still found. I would
expect that these requests return 404, because nginx tries image.jpg/
and then image.jpg.html, but not the actual $uri. Am I missing
something?

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218562,218744#msg-218744


From nginx-forum at nginx.us  Tue Nov 22 17:10:49 2011
From: nginx-forum at nginx.us (kung_fu_mike)
Date: Tue, 22 Nov 2011 12:10:49 -0500
Subject: URL decoding a variable and/or using rewite effectively
Message-ID: <8a1bc9ef7b75d9a0160f4920943d0e3e.NginxMailingListEnglish@forum.nginx.org>

Thank you in advance to anyone who can point me in the right direction.

We currently use nginx as an asset server. Requests are made to the
asset server in the form of:
http://downloads.ourdomain.com?unique_key=MD5hashOFstuff&name=the_nice_file_name(that
might have quotes or spaces)

The unique key is the actual name of the file, however we use $arg_NAME
in the headers to control the downloaded name. eg:
add_header    Content-Disposition "attachment;filename=$arg_NAME.xslx";
(yes they are excel files).

Our current issue is that space, parenthesis and other characters become
URL encoded in the variable. As such, the file-name (as it appears in
download) is not quite what we would like. My research has led me to
find that there is currently no access to a 'url-decode' style function.
I was curious, though, if there might be another method someone could
think to employ. Perhaps I have access to regular expression replace
style functionality that I am unaware of. Something that simple, would
more then likely solve all of the issues for this particular case.

Thank you for taking the time to read this,

-Mike
(kung_fu_mike)

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218754,218754#msg-218754


From reallfqq-nginx at yahoo.fr  Tue Nov 22 19:32:52 2011
From: reallfqq-nginx at yahoo.fr (B.R.)
Date: Tue, 22 Nov 2011 14:32:52 -0500
Subject: PH Fast-CGI security pitfall
Message-ID: <CALqce=1u1Zwj3k6MiDuiuapFeQxhfVkdtoUnhsj52vCU5hHp7A@mail.gmail.com>

Hello,

I juste read this
article<https://nealpoole.com/blog/2011/04/setting-up-php-fastcgi-and-nginx-dont-trust-the-tutorials-check-your-configuration/>which
highlight a common security pitfall to serve PHP files.
I don't see any similar advice in your PHP on Fast-CGI
tutorial<http://wiki.nginx.org/PHPFcgiExample>nor your
pitfalls page <http://wiki.nginx.org/Pitfalls>.

On the last page, you tell about the problem in the *Pass Non-PHP Requests
to PHP* section, you seem to point in the right direction in the *Proxy
everything* section, but not for the right reasons.
You tell people to use an 'if' to check for file existence, but the use of
'try' is much better, a you know it since you redirect to the IfIsEvil page.

The article I gave you reference to offers 5 different wys to secure the
server. The 'try_files $uri =404;' seems to be a nice way of preventing
non-PHP script from being executed, isn't it?
Thanks,
---
*B. R.*
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20111122/4f84c353/attachment.html>

From max at mxcrypt.com  Tue Nov 22 19:58:07 2011
From: max at mxcrypt.com (Maxim Khitrov)
Date: Tue, 22 Nov 2011 14:58:07 -0500
Subject: PH Fast-CGI security pitfall
In-Reply-To: <CALqce=1u1Zwj3k6MiDuiuapFeQxhfVkdtoUnhsj52vCU5hHp7A@mail.gmail.com>
References: <CALqce=1u1Zwj3k6MiDuiuapFeQxhfVkdtoUnhsj52vCU5hHp7A@mail.gmail.com>
Message-ID: <CAJcQMWf+ntFyu9UVprqm8ODvJqNLjpLSGwPR6v97mT5BS4p9MA@mail.gmail.com>

On Tue, Nov 22, 2011 at 2:32 PM, B.R. <reallfqq-nginx at yahoo.fr> wrote:
> Hello,
>
> I juste read this article which highlight a common security pitfall to serve
> PHP files.
> I don't see any similar advice in your PHP on Fast-CGI tutorial nor your
> pitfalls page.
>
> On the last page, you tell about the problem in the Pass Non-PHP Requests to
> PHP section, you seem to point in the right direction in the Proxy
> everything section, but not for the right reasons.
> You tell people to use an 'if' to check for file existence, but the use of
> 'try' is much better, a you know it since you redirect to the IfIsEvil page.
>
> The article I gave you reference to offers 5 different wys to secure the
> server. The 'try_files $uri =404;' seems to be a nice way of preventing
> non-PHP script from being executed, isn't it?

I generally use the following template for serving PHP via FastCGI:

location ~ \.php$ {
    if (!-f $request_filename) { return 404; }

    fastcgi_pass 127.0.0.1:8000;
    fastcgi_param SCRIPT_FILENAME $request_filename;
    ...
}

The 'if' statement causes 404 to be returned unless the requested file
actually exists. Making sure that people can't upload files ending in
'.php' is a separate mater, but I believe that this configuration
takes care of the security issue described in your first link.

For the given example, nginx detects that
http://www.bambookites.com/uploads/random.gif/somefilename.php doesn't
refer to an actual php file, so nothing is passed to the interpreter.

- Max


From reallfqq-nginx at yahoo.fr  Tue Nov 22 20:04:50 2011
From: reallfqq-nginx at yahoo.fr (B.R.)
Date: Tue, 22 Nov 2011 15:04:50 -0500
Subject: PH Fast-CGI security pitfall
In-Reply-To: <CAJcQMWf+ntFyu9UVprqm8ODvJqNLjpLSGwPR6v97mT5BS4p9MA@mail.gmail.com>
References: <CALqce=1u1Zwj3k6MiDuiuapFeQxhfVkdtoUnhsj52vCU5hHp7A@mail.gmail.com>
	<CAJcQMWf+ntFyu9UVprqm8ODvJqNLjpLSGwPR6v97mT5BS4p9MA@mail.gmail.com>
Message-ID: <CALqce=0YOLyz-RodfJEDz8EYVVz4PSNRnPTJtn700WdV1et1aw@mail.gmail.com>

Your use of the if switch is enlisted in the pitfalls page of
nginx<http://wiki.nginx.org/Pitfalls>(see
*Check IF File Exists* section) :o)
It is said you should use the try one!

My message was: is using the try switch with the '=404' the right way? If
yes can someone add that to the nginx wiki?
---
*B. R.*


On Tue, Nov 22, 2011 at 14:58, Maxim Khitrov <max at mxcrypt.com> wrote:

> On Tue, Nov 22, 2011 at 2:32 PM, B.R. <reallfqq-nginx at yahoo.fr> wrote:
> > Hello,
> >
> > I juste read this article which highlight a common security pitfall to
> serve
> > PHP files.
> > I don't see any similar advice in your PHP on Fast-CGI tutorial nor your
> > pitfalls page.
> >
> > On the last page, you tell about the problem in the Pass Non-PHP
> Requests to
> > PHP section, you seem to point in the right direction in the Proxy
> > everything section, but not for the right reasons.
> > You tell people to use an 'if' to check for file existence, but the use
> of
> > 'try' is much better, a you know it since you redirect to the IfIsEvil
> page.
> >
> > The article I gave you reference to offers 5 different wys to secure the
> > server. The 'try_files $uri =404;' seems to be a nice way of preventing
> > non-PHP script from being executed, isn't it?
>
> I generally use the following template for serving PHP via FastCGI:
>
> location ~ \.php$ {
>    if (!-f $request_filename) { return 404; }
>
>    fastcgi_pass 127.0.0.1:8000;
>    fastcgi_param SCRIPT_FILENAME $request_filename;
>    ...
> }
>
> The 'if' statement causes 404 to be returned unless the requested file
> actually exists. Making sure that people can't upload files ending in
> '.php' is a separate mater, but I believe that this configuration
> takes care of the security issue described in your first link.
>
> For the given example, nginx detects that
> http://www.bambookites.com/uploads/random.gif/somefilename.php doesn't
> refer to an actual php file, so nothing is passed to the interpreter.
>
> - Max
>
> _______________________________________________
> 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/20111122/c00c09b3/attachment.html>

From roman.vasilyev at yousendit.com  Tue Nov 22 20:06:36 2011
From: roman.vasilyev at yousendit.com (Roman Vasilyev)
Date: Tue, 22 Nov 2011 12:06:36 -0800
Subject: PH Fast-CGI security pitfall
In-Reply-To: <CAJcQMWf+ntFyu9UVprqm8ODvJqNLjpLSGwPR6v97mT5BS4p9MA@mail.gmail.com>
References: <CALqce=1u1Zwj3k6MiDuiuapFeQxhfVkdtoUnhsj52vCU5hHp7A@mail.gmail.com>
	<CAJcQMWf+ntFyu9UVprqm8ODvJqNLjpLSGwPR6v97mT5BS4p9MA@mail.gmail.com>
Message-ID: <4ECC00CC.10203@yousendit.com>

According previous article 
http://wiki.nginx.org/Pitfalls#Front_Controller_Pattern_based_packages

in your template better to write:

location ~ \.php$ {
     try_files $uri =404;

     fastcgi_pass 127.0.0.1:8000;
     fastcgi_param SCRIPT_FILENAME $request_filename;
     ...
}



On 11/22/2011 11:58 AM, Maxim Khitrov wrote:
> On Tue, Nov 22, 2011 at 2:32 PM, B.R.<reallfqq-nginx at yahoo.fr>  wrote:
>    
>> Hello,
>>
>> I juste read this article which highlight a common security pitfall to serve
>> PHP files.
>> I don't see any similar advice in your PHP on Fast-CGI tutorial nor your
>> pitfalls page.
>>
>> On the last page, you tell about the problem in the Pass Non-PHP Requests to
>> PHP section, you seem to point in the right direction in the Proxy
>> everything section, but not for the right reasons.
>> You tell people to use an 'if' to check for file existence, but the use of
>> 'try' is much better, a you know it since you redirect to the IfIsEvil page.
>>
>> The article I gave you reference to offers 5 different wys to secure the
>> server. The 'try_files $uri =404;' seems to be a nice way of preventing
>> non-PHP script from being executed, isn't it?
>>      
> I generally use the following template for serving PHP via FastCGI:
>
> location ~ \.php$ {
>      if (!-f $request_filename) { return 404; }
>
>      fastcgi_pass 127.0.0.1:8000;
>      fastcgi_param SCRIPT_FILENAME $request_filename;
>      ...
> }
>
> The 'if' statement causes 404 to be returned unless the requested file
> actually exists. Making sure that people can't upload files ending in
> '.php' is a separate mater, but I believe that this configuration
> takes care of the security issue described in your first link.
>
> For the given example, nginx detects that
> http://www.bambookites.com/uploads/random.gif/somefilename.php doesn't
> refer to an actual php file, so nothing is passed to the interpreter.
>
> - Max
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx
>    


From reallfqq-nginx at yahoo.fr  Tue Nov 22 20:24:03 2011
From: reallfqq-nginx at yahoo.fr (B.R.)
Date: Tue, 22 Nov 2011 15:24:03 -0500
Subject: Spawning PHP as FastCGI
Message-ID: <CALqce=1dKuxC9JyPrNnMcw=0P=2n+kLnWDgOtK2omdtV9Avjgg@mail.gmail.com>

I used the bash script given on this
page<http://wiki.nginx.org/PHPFcgiExample#Spawning_a_FastCGI_Process>,
after having changed the binding to a UNIX socket (and not a network one).

The update-rc configuration tool gives me this output:
update-rc.d: using dependency based boot sequencing
insserv: warning: script 'php-cgi' missing LSB tags and overrides
insserv: There is a loop between service nginx and php-cgi if stopped
insserv:  loop involving service php-cgi at depth 2
insserv:  loop involving service nginx at depth 1
insserv: Stopping php-cgi depends on nginx and therefore on system facility
`$all' which can not be true!
insserv: exiting now without changing boot order!
update-rc.d: error: insserv rejected the script header

I'm not used to the services configuration, but dependencies between nginx
and php seems to be somehow buggy.
How to fix that in a cleaner way?

Many thanks,
---
*B. R.*
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20111122/7f0bbf62/attachment-0001.html>

From nginx-forum at nginx.us  Tue Nov 22 20:28:46 2011
From: nginx-forum at nginx.us (Ensiferous)
Date: Tue, 22 Nov 2011 15:28:46 -0500
Subject: PH Fast-CGI security pitfall
In-Reply-To: <CALqce=1u1Zwj3k6MiDuiuapFeQxhfVkdtoUnhsj52vCU5hHp7A@mail.gmail.com>
References: <CALqce=1u1Zwj3k6MiDuiuapFeQxhfVkdtoUnhsj52vCU5hHp7A@mail.gmail.com>
Message-ID: <dc11dd4c9f177bf67da039d23784ba52.NginxMailingListEnglish@forum.nginx.org>

It doesn't matter whether you use if or try_files. If people would
actually read the if is evil page then it specifically states that "It
is important to note that the behaviour of if is not inconsistent, given
two identical requests it will not randomly fail on one and work on the
other, with proper testing and understanding ifs can be used."

If you prefer if then use an if, if you prefer try_files then use a
try_files, there's really no reason to agonize over this issue.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218755,218761#msg-218761


From ne at vbart.ru  Tue Nov 22 20:40:01 2011
From: ne at vbart.ru (Valentin V. Bartenev)
Date: Wed, 23 Nov 2011 00:40:01 +0400
Subject: Spawning PHP as FastCGI
In-Reply-To: <CALqce=1dKuxC9JyPrNnMcw=0P=2n+kLnWDgOtK2omdtV9Avjgg@mail.gmail.com>
References: <CALqce=1dKuxC9JyPrNnMcw=0P=2n+kLnWDgOtK2omdtV9Avjgg@mail.gmail.com>
Message-ID: <201111230040.01919.ne@vbart.ru>

On Wednesday 23 November 2011 00:24:03 B.R. wrote:
> I used the bash script given on this
> page<http://wiki.nginx.org/PHPFcgiExample#Spawning_a_FastCGI_Process>,
> after having changed the binding to a UNIX socket (and not a network one).
[...]

Sorry for offtopic, but is there any reason to use it instead of php-fpm?

wbr, Valentin V. Bartenev


From calin.don at gmail.com  Tue Nov 22 21:52:29 2011
From: calin.don at gmail.com (Calin Don)
Date: Tue, 22 Nov 2011 23:52:29 +0200
Subject: $request_time meaning
In-Reply-To: <20111121185559.GB95664@mdounin.ru>
References: <CAEOe2JzF3si7pz3vi+4VBj7S_qAnX2HWcWT6Css=OCGr4dBtng@mail.gmail.com>
	<20111121185559.GB95664@mdounin.ru>
Message-ID: <CAEOe2JyooV5pS4+Qk0RBu_WiQzQW_7PREOxZKOHUz+MPYF6Zuw@mail.gmail.com>

So the time spent doing a post_action is added to $request_time or not?

On Mon, Nov 21, 2011 at 20:55, Maxim Dounin <mdounin at mdounin.ru> wrote:

> Hello!
>
> On Mon, Nov 21, 2011 at 08:41:15PM +0200, Calin Don wrote:
>
> > Hi,
> > How is the $request_time calculated in the case of a proxied request. I'm
> > interested especially in the case where the resource is stale in cache.
>
> $request_time is always time since start of the request (when
> first bytes are read from client) till end of the request (when
> last bytes are sent to client and logging happens).
>
> 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/20111122/a0a40d8f/attachment.html>

From francis at daoine.org  Tue Nov 22 22:44:41 2011
From: francis at daoine.org (Francis Daly)
Date: Tue, 22 Nov 2011 22:44:41 +0000
Subject: $request_time meaning
In-Reply-To: <CAEOe2JyooV5pS4+Qk0RBu_WiQzQW_7PREOxZKOHUz+MPYF6Zuw@mail.gmail.com>
References: <CAEOe2JzF3si7pz3vi+4VBj7S_qAnX2HWcWT6Css=OCGr4dBtng@mail.gmail.com>
	<20111121185559.GB95664@mdounin.ru>
	<CAEOe2JyooV5pS4+Qk0RBu_WiQzQW_7PREOxZKOHUz+MPYF6Zuw@mail.gmail.com>
Message-ID: <20111122224441.GD27443@craic.sysops.org>

> > On Mon, Nov 21, 2011 at 08:41:15PM +0200, Calin Don wrote:
> On Mon, Nov 21, 2011 at 20:55, Maxim Dounin <mdounin at mdounin.ru> wrote:
On Tue, Nov 22, 2011 at 11:52:29PM +0200, Calin Don wrote:

Hi there,

> > > How is the $request_time calculated in the case of a proxied request. I'm
> > > interested especially in the case where the resource is stale in cache.
> >
> > $request_time is always time since start of the request (when
> > first bytes are read from client) till end of the request (when
> > last bytes are sent to client and logging happens).

> So the time spent doing a post_action is added to $request_time or not?

It seems straightforward enough to test.

Define a post_action in a location of (for example) a php script which
does sleep(3), then see if $request_time is 3 seconds bigger in that
location than elsewhere.

When I try it in a location which just serves static files, I do see
$request_time = 3.002 in access.log.

You can do the same in your stale-in-cache location.

Good luck,

	f
-- 
Francis Daly        francis at daoine.org


From quintinpar at gmail.com  Wed Nov 23 03:31:18 2011
From: quintinpar at gmail.com (Quintin Par)
Date: Wed, 23 Nov 2011 09:01:18 +0530
Subject: Is proxy_cache_bypass recaching?
In-Reply-To: <CACW9FdOHmK2S-O83LWA7tA_krjANZ6+7DffbPWanUkZ=Fr6ABQ@mail.gmail.com>
References: <CACW9FdOHmK2S-O83LWA7tA_krjANZ6+7DffbPWanUkZ=Fr6ABQ@mail.gmail.com>
Message-ID: <CACW9FdMEV4x=oJLoV0PeV0dSOW7KEKLJ0WRZLxsmPygfstquVg@mail.gmail.com>

Can someone help me if this please?

On Tue, Nov 22, 2011 at 8:53 AM, Quintin Par <quintinpar at gmail.com> wrote:

> Hi all,
>
> I have this in my location directive
>
>                     proxy_cache_bypass                     $http_soon;
>
> now I do a curl call to invalidate
>
>
>
> [user at ip-10-111 ~]$ curl "www.site.com/teaser" -I -H "soon:true"
>
> HTTP/1.1 200 OK
>
> Server: nginx
>
> Date: Tue, 22 Nov 2011 03:07:20 GMT
>
> Content-Type: text/html; charset=utf-8
>
> Connection: keep-alive
>
> Keep-Alive: timeout=60
>
> Vary: Accept-Encoding
>
> Vary: Cookie
>
> X-Mod-Pagespeed: 0.9.17.7-716
>
> Cache-Control: max-age=0, no-cache, no-store
>
> X-Cache-Status: BYPASS
>
>
>
> [user at ip-10-111 ~]$ curl "www.site.com/teaser" -I
>
> HTTP/1.1 200 OK
>
> Server: nginx
>
> Date: Tue, 22 Nov 2011 03:07:37 GMT
>
> Content-Type: text/html; charset=utf-8
>
> Connection: keep-alive
>
> Keep-Alive: timeout=60
>
> Vary: Accept-Encoding
>
> Vary: Cookie
>
> X-Mod-Pagespeed: 0.9.17.7-716
>
> Cache-Control: max-age=0, no-cache, no-store
>
> X-Cache-Status: HIT
>
>
>
> Does this mean that the curl call with the header is* recaching the page
> and the subsequent hits are cache hits?*
>
> -Quintin
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20111123/5c92d717/attachment.html>

From nginx-forum at nginx.us  Wed Nov 23 04:31:07 2011
From: nginx-forum at nginx.us (Salem)
Date: Tue, 22 Nov 2011 23:31:07 -0500
Subject: No nice Permalinks with Wordpress
Message-ID: <ce5ab2dcdb90596969bc7fd613a92aaa.NginxMailingListEnglish@forum.nginx.org>

Hello,

after days of testing i don't find a way to use short permalinks with
Wordpress, i can only configure URLS with /index.php/xxx in it, not sure
why.

The rewrite-rules are there, but wordpress ignores them.

I tried
    try_files  $uri $uri/ /index.php?$args;
and
  try_files $uri $uri/ /index.php?q=$request_uri;
(both inside location / {}

If i try /xxx/ instead of /index.php/xxx/ i only see an "Nothing found
for xxx" error from wordpress.

Is there any way to fix this? I found many faqs but i don't find my
problem :-(

Thanks

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218771,218771#msg-218771


From edho at myconan.net  Wed Nov 23 05:07:29 2011
From: edho at myconan.net (Edho Arief)
Date: Wed, 23 Nov 2011 12:07:29 +0700
Subject: No nice Permalinks with Wordpress
In-Reply-To: <ce5ab2dcdb90596969bc7fd613a92aaa.NginxMailingListEnglish@forum.nginx.org>
References: <ce5ab2dcdb90596969bc7fd613a92aaa.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <CABTjkK=begitZ1S9c=orHOdjeqB-4dtqc9-Yzv3X7cRinkSZrw@mail.gmail.com>

On Wed, Nov 23, 2011 at 11:31 AM, Salem <nginx-forum at nginx.us> wrote:
> Hello,
>
> after days of testing i don't find a way to use short permalinks with
> Wordpress, i can only configure URLS with /index.php/xxx in it, not sure
> why.
>

Use "Custom Structure" and remove the index.php/ part.

For the try_files, use this:

try_files $uri $uri/ /index.php?q=$uri&$args;



-- 
O< ascii ribbon campaign - stop html mail - www.asciiribbon.org


From nginx-forum at nginx.us  Wed Nov 23 07:19:19 2011
From: nginx-forum at nginx.us (sibsoft)
Date: Wed, 23 Nov 2011 02:19:19 -0500
Subject: MP4 - start time is out mp4 stts samples
In-Reply-To: <5e238ac1a229e14931b6ea72c107997c.NginxMailingListEnglish@forum.nginx.org>
References: <20111117050603.GA10385@nginx.com>
	<5e238ac1a229e14931b6ea72c107997c.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <78357bec940d9e785d7bb42f2eb5e229.NginxMailingListEnglish@forum.nginx.org>

Thanks a lot. It's working fine now.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218451,218773#msg-218773


From lists at ruby-forum.com  Wed Nov 23 07:38:26 2011
From: lists at ruby-forum.com (Sandesh K.)
Date: Wed, 23 Nov 2011 08:38:26 +0100
Subject: Nginx stuck processes
In-Reply-To: <20111122085856.GD95664@mdounin.ru>
References: <d04888efb006752e263fc5bca2c0d314@ruby-forum.com>
	<7efb85fdbf5ec94ab17531ab41499073@ruby-forum.com>
	<31b70aa141cf1024aba7576099862a99@ruby-forum.com>
	<20111122085856.GD95664@mdounin.ru>
Message-ID: <78913b8db44bcdfd04b7bca6ee737233@ruby-forum.com>

Maxim Dounin wrote in post #1033042:
> Hello!
>
> On Tue, Nov 22, 2011 at 06:51:56AM +0100, Sandesh K. wrote:
>
>>
>> --without-mail_smtp_module
> [...]
>
>> (gdb) backtrace
>> #0 0x0031ebba in _int_malloc () from /lib/tls/libc.so.6
>> #1 0x0031f89c in _int_memalign () from /lib/tls/libc.so.6
>> #2 0x0031e2c7 in memalign () from /lib/tls/libc.so.6
>> #3 0x00320367 in posix_memalign () from /lib/tls/libc.so.6
>> #4 0x0805da38 in ngx_memalign (alignment=16, size=4069528, log=0x108) at
>> src/os/unix/ngx_alloc.c:56
>
> The "size=4069528" looks really strange...
>
>> #5 0x0804b917 in ngx_palloc_block (pool=0x8d0b9b0, size=52) at
>> src/core/ngx_palloc.c:184
>
> Could you please show
>
> fr 5
> p *pool
>
> output from gdb?
>
>> #11 0x0806606c in ngx_mail_auth_http_ignore_status_line (s=0x8d0da20,
>> <ngx_worker_process_cycle>, data=0x0, name=0x806be29 "worker process",
>> respawn=-3) at src/os/unix/ngx_process.c:196
>> #17 0x080601f6 in ngx_start_worker_processes (cycle=0x8ccb228, n=30,
>> type=-3) at src/os/unix/ngx_process_cycle.c:360
>> #18 0x0805fbe7 in ngx_master_process_cycle (cycle=0x8ccb228) at
>> src/os/unix/ngx_process_cycle.c:136
>> #19 0x0804a116 in main (argc=3, argv=0x8ccb9a4) at src/core/nginx.c:405
>> (gdb) quit
>
> Maxim Dounin

Hello Maxim,

Please find the output.

43121 ?        S      0:00 nginx: master process
/usr/local/nginx/sbin/nginx -c /etc/nginx/conf/nginx.conf
43123 ?        S      1:58  \_ nginx: worker process
43124 ?        S      1:46  \_ nginx: worker process
43127 ?        S      2:02  \_ nginx: worker process
43128 ?        S      1:37  \_ nginx: worker process
43130 ?        S      1:50  \_ nginx: worker process
43131 ?        S      2:23  \_ nginx: worker process
43132 ?        S      2:44  \_ nginx: worker process
43133 ?        S      2:09  \_ nginx: worker process
43134 ?        S      1:47  \_ nginx: worker process
43135 ?        S      2:07  \_ nginx: worker process
43129 ?        R    866:11  \_ nginx: worker process
43137 ?        S      1:35  \_ nginx: worker process
43138 ?        S      1:28  \_ nginx: worker process
43139 ?        S      3:04  \_ nginx: worker process
43140 ?        S      1:42  \_ nginx: worker process
43142 ?        S      1:23  \_ nginx: worker process
43143 ?        S      2:24  \_ nginx: worker process
 7328 ?        S      0:59  \_ nginx: worker process
 9444 ?        S      0:46  \_ nginx: worker process
57217 ?        S      0:00  \_ nginx: worker process


GNU gdb Red Hat Linux (6.3.0.0-1.132.EL3rh)
Copyright 2004 Free Software Foundation, Inc.
GDB is free software, covered by the GNU General Public License, and you
are
welcome to change it and/or distribute copies of it under certain
conditions.
Type "show copying" to see the conditions.
There is absolutely no warranty for GDB.  Type "show warranty" for
details.
This GDB was configured as "i386-redhat-linux-gnu".
(gdb) attach 43129
Attaching to process 43129
Reading symbols from /usr/local/nginx/sbin/nginx...done.
Using host libthread_db library "/lib/tls/libthread_db.so.1".
Reading symbols from /lib/tls/libpthread.so.0...done.
[Thread debugging using libthread_db enabled]
[New Thread -1218529952 (LWP 43129)]
Loaded symbols for /lib/tls/libpthread.so.0
Reading symbols from /lib/libcrypto.so.4...done.
Loaded symbols for /lib/libcrypto.so.4
Reading symbols from /lib/tls/libc.so.6...done.
Loaded symbols for /lib/tls/libc.so.6
Reading symbols from /lib/ld-linux.so.2...done.
Loaded symbols for /lib/ld-linux.so.2
Reading symbols from /usr/kerberos/lib/libgssapi_krb5.so.2...done.
Loaded symbols for /usr/kerberos/lib/libgssapi_krb5.so.2
Reading symbols from /usr/kerberos/lib/libkrb5.so.3...done.
Loaded symbols for /usr/kerberos/lib/libkrb5.so.3
Reading symbols from /usr/kerberos/lib/libcom_err.so.3...done.
Loaded symbols for /usr/kerberos/lib/libcom_err.so.3
Reading symbols from /usr/kerberos/lib/libk5crypto.so.3...done.
Loaded symbols for /usr/kerberos/lib/libk5crypto.so.3
Reading symbols from /lib/libresolv.so.2...done.
Loaded symbols for /lib/libresolv.so.2
Reading symbols from /lib/libdl.so.2...done.
Loaded symbols for /lib/libdl.so.2
Reading symbols from /usr/lib/libz.so.1...done.
Loaded symbols for /usr/lib/libz.so.1
Reading symbols from /lib/libnss_files.so.2...done.
Loaded symbols for /lib/libnss_files.so.2
0x00f36130 in _int_malloc () from /lib/tls/libc.so.6
(gdb) fr 5
#5  0x0804b917 in ngx_palloc_block (pool=0x90f58e0, size=128) at
src/core/ngx_palloc.c:184
184         m = ngx_memalign(NGX_POOL_ALIGNMENT, psize, pool->log);
(gdb) p *pool
$1 = {d = {last = 0x90f599c
"xxxx at fx.xxx.1xx.2xx.8x.8x.mail.xx.xx.com\r\t\b\001", end = 0x90f59e0
"\b\001", next = 0x910cb20, failed = 0}, max = 216,
  current = 0x90f58e0, chain = 0x0, large = 0x0, cleanup = 0x0, log =
0x90f5918}
(gdb) backtrace
#0  0x00f36130 in _int_malloc () from /lib/tls/libc.so.6
#1  0x00f36b3c in _int_memalign () from /lib/tls/libc.so.6
#2  0x00f354f7 in memalign () from /lib/tls/libc.so.6
#3  0x00f37607 in posix_memalign () from /lib/tls/libc.so.6
#4  0x0805da3c in ngx_memalign (alignment=16, size=16751288, log=0x108)
at src/os/unix/ngx_alloc.c:56
#5  0x0804b917 in ngx_palloc_block (pool=0x90f58e0, size=128) at
src/core/ngx_palloc.c:184
#6  0x0804b892 in ngx_palloc (pool=0x90f58e0, size=128) at
src/core/ngx_palloc.c:138
#7  0x0804d0cf in ngx_create_temp_buf (pool=0x90f58e0, size=128) at
src/core/ngx_buf.c:21
#8  0x0806566a in ngx_mail_pop3_init_protocol (rev=0x8ebd4b4) at
src/mail/ngx_mail_pop3_handler.c:104
#9  0x0805bc4d in ngx_event_process_posted (cycle=0x8e951d8,
posted=0x8071ea0) at src/event/ngx_event_posted.c:39
#10 0x0805ad89 in ngx_process_events_and_timers (cycle=0x8e951d8) at
src/event/ngx_event.c:272
#11 0x08060956 in ngx_worker_process_cycle (cycle=0x8e951d8, data=0x0)
at src/os/unix/ngx_process_cycle.c:801
#12 0x0805ef64 in ngx_spawn_process (cycle=0x8e951d8, proc=0x80608a1
<ngx_worker_process_cycle>, data=0x0, name=0x806be29 "worker process",
respawn=-3) at src/os/unix/ngx_process.c:196
#13 0x080601fa in ngx_start_worker_processes (cycle=0x8e951d8, n=20,
type=-3) at src/os/unix/ngx_process_cycle.c:360
#14 0x0805fbeb in ngx_master_process_cycle (cycle=0x8e951d8) at
src/os/unix/ngx_process_cycle.c:136
#15 0x0804a116 in main (argc=3, argv=0x8e95954) at src/core/nginx.c:405

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


From mdounin at mdounin.ru  Wed Nov 23 09:57:37 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Wed, 23 Nov 2011 13:57:37 +0400
Subject: $request_time meaning
In-Reply-To: <CAEOe2JyooV5pS4+Qk0RBu_WiQzQW_7PREOxZKOHUz+MPYF6Zuw@mail.gmail.com>
References: <CAEOe2JzF3si7pz3vi+4VBj7S_qAnX2HWcWT6Css=OCGr4dBtng@mail.gmail.com>
	<20111121185559.GB95664@mdounin.ru>
	<CAEOe2JyooV5pS4+Qk0RBu_WiQzQW_7PREOxZKOHUz+MPYF6Zuw@mail.gmail.com>
Message-ID: <20111123095737.GO95664@mdounin.ru>

Hello!

On Tue, Nov 22, 2011 at 11:52:29PM +0200, Calin Don wrote:

> So the time spent doing a post_action is added to $request_time or not?

Added.

In general, post_action is a dirty hack which works in context of 
a request.  It delays all processing of the request (i.e. logging) 
as well as processing of furhter requests in the connection in 
question (if any).  This is one of the reasons why post_action 
isn't documented, BTW.

Maxim Dounin

> 
> On Mon, Nov 21, 2011 at 20:55, Maxim Dounin <mdounin at mdounin.ru> wrote:
> 
> > Hello!
> >
> > On Mon, Nov 21, 2011 at 08:41:15PM +0200, Calin Don wrote:
> >
> > > Hi,
> > > How is the $request_time calculated in the case of a proxied request. I'm
> > > interested especially in the case where the resource is stale in cache.
> >
> > $request_time is always time since start of the request (when
> > first bytes are read from client) till end of the request (when
> > last bytes are sent to client and logging happens).
> >
> > 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  Wed Nov 23 10:30:46 2011
From: nginx-forum at nginx.us (token)
Date: Wed, 23 Nov 2011 05:30:46 -0500
Subject: server_name host order
Message-ID: <9a2e056ad6a61db90e7ecb9a88e24dbf.NginxMailingListEnglish@forum.nginx.org>

Hello, I would like to know if this is correct

Will NGINX allways return the first of the provided hostname name in
from the `server_name` as beign the "SERVER_NAME",

eg server_name hostname1 hostname2;

I have two server blocks in this example 

server {
	listen 80;       
        server_name www.example.co.uk *.example.co.uk;
}

the URI = http://something.example.co.uk/

// PHP APPLICATION -------------- 
I see as expected   `SERVER_NAME` and `HTTP_HOST`

SERVER_PORT =>  80                  
SERVER_NAME =>  www.example.co.uk
HTTP_HOST   =>  something.example.co.uk
--------------

server {
      listen  443;
      server_name  www.example.co.uk secure.example.co.uk;
}

// PHP APPLICATION -------------- 
again the `SERVER_NAME` and `HTTP_HOST` as i would expect

the URI = https://secure.example.co.uk/

//  SERVER_PORT =>   443
//  SERVER_NAME =>   www.example.co.uk
//  HTTP_HOST   =>   secure.example.co.uk

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218786,218786#msg-218786


From nginx-forum at nginx.us  Wed Nov 23 10:43:15 2011
From: nginx-forum at nginx.us (janedenone)
Date: Wed, 23 Nov 2011 05:43:15 -0500
Subject: Using ab to benchmark nginx: Connection reset by peer (54)
In-Reply-To: <CAB4Tn6NxfRkfoKX24i6uetmspHA6ROdbLM9KQUeUAFEa3WwwvQ@mail.gmail.com>
References: <CAB4Tn6NxfRkfoKX24i6uetmspHA6ROdbLM9KQUeUAFEa3WwwvQ@mail.gmail.com>
Message-ID: <4f0ce25d856b3091d02fd8c742b94374.NginxMailingListEnglish@forum.nginx.org>

agentzh Wrote:
-------------------------------------------------------

> On Mac OS X 10.6, I have to use "127.0.0.1"
> instead of "localhost"
> while running the ab command, or I'll get exactly
> the same error
> message.
> 
> And for 10.7 (Lion), people say they have to apply
> a patch to ab to get it work.
> 
> Hope this helps,
> -agentzh

Thanks a lot ? I now use the -r flag, which makes ab proceed in spite
of failed requests.

- jan

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218554,218789#msg-218789


From ikaraszi at adverticum.com  Wed Nov 23 10:56:21 2011
From: ikaraszi at adverticum.com (KARASZI Istvan)
Date: Wed, 23 Nov 2011 11:56:21 +0100
Subject: log request only when they reached a size
Message-ID: <51089950-5F45-4455-B16B-E579EA2FB91E@adverticum.com>

Dear list!

I'm new here, so I just hope I'm following the rules.

I would like to send a patch for http_log_module which I made for version nginx-1.0.10.

My company is in adserving business and it is important for us to log the big requests with the access_log feature. But we were unable to do that with nginx, because logging all the requests would produce too much I/O load on our servers.

So I made a patch for http_log_module to accept a `size` parameter which is optional (the last parameter after the `buffer`). If this `size` is provided nginx logs only the big enough requests with access_log (checks for `body_bytes_sent`).

I hope you find it useful and you will include this in the upstream.

-------------- next part --------------
A non-text attachment was scrubbed...
Name: http_log_module-size.patch
Type: application/octet-stream
Size: 2868 bytes
Desc: not available
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20111123/d5ec5e83/attachment.obj>
-------------- next part --------------


Thanks,
-- 
KARASZI Istvan
http://adverticum.com/


From nginx-forum at nginx.us  Wed Nov 23 11:22:24 2011
From: nginx-forum at nginx.us (nicksoft)
Date: Wed, 23 Nov 2011 06:22:24 -0500
Subject: configuration problem with subdomain and proxy_pass
Message-ID: <aa157f0fae3e5fb1c29175c7ba4f5c58.NginxMailingListEnglish@forum.nginx.org>

Hi,

I have a problem with configuration subdomains for nginx. I want to
configure domains:

mydomain.com (port 80) --> http://localhost:8080/
www.mydomain.com (port 80) --> http://localhost:8080/
subdomain.mydomain.com (port 80) --> http://localhost:8080/subdomain


Tomcat listening at 8080 port on localhost. I have application
'subdomain' in tomcat webapps folder.



This is my configuration file:
------------------------------------------------------
server
{
       listen       80;
       server_name  mydomain.com;

       # Main location
       location /
       {
           proxy_pass         http://127.0.0.1:8080/;
           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       52m;
           client_body_buffer_size    128k;

           proxy_connect_timeout      3600;
           proxy_send_timeout         3600;
           proxy_read_timeout         90;

           proxy_buffer_size          4k;
           proxy_buffers              4 32k;
           proxy_busy_buffers_size    64k;
           proxy_temp_file_write_size 64k;
           proxy_cache_valid 5m;
       }
}

server
{
       listen       80;
       server_name  www.mydomain.com;

       # Main location
       location /
       {
           proxy_pass         http://127.0.0.1:8080/;
           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       52m;
           client_body_buffer_size    128k;

           proxy_connect_timeout      3600;
           proxy_send_timeout         3600;
           proxy_read_timeout         90;

           proxy_buffer_size          4k;
           proxy_buffers              4 32k;
           proxy_busy_buffers_size    64k;
           proxy_temp_file_write_size 64k;
           proxy_cache_valid 5m;
       }
}

server
{
       listen       80;
       server_name subdomain.mydomain.com;

       # Main location
       location /
       {
           proxy_pass         http://127.0.0.1:8080/subdomain/;
           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       52m;
           client_body_buffer_size    128k;

           proxy_connect_timeout      3600;
           proxy_send_timeout         3600;
           proxy_read_timeout         90;

           proxy_buffer_size          4k;
           proxy_buffers              4 32k;
           proxy_busy_buffers_size    64k;
           proxy_temp_file_write_size 64k;
           proxy_cache_valid 5m;
       }
}
------------------------------------------------------


For this configuration http:/www.mydomain.com  and http:/mydomain.com
working very well, but subdomain.mydomain.com doesn't work (there is no
response from server).

Oh, and http://mydomain.com/subdomain in browser works well too.


I will be grateful for your help.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218793,218793#msg-218793


From calin.don at gmail.com  Wed Nov 23 11:25:46 2011
From: calin.don at gmail.com (Calin Don)
Date: Wed, 23 Nov 2011 13:25:46 +0200
Subject: $request_time meaning
In-Reply-To: <20111123095737.GO95664@mdounin.ru>
References: <CAEOe2JzF3si7pz3vi+4VBj7S_qAnX2HWcWT6Css=OCGr4dBtng@mail.gmail.com>
	<20111121185559.GB95664@mdounin.ru>
	<CAEOe2JyooV5pS4+Qk0RBu_WiQzQW_7PREOxZKOHUz+MPYF6Zuw@mail.gmail.com>
	<20111123095737.GO95664@mdounin.ru>
Message-ID: <CAEOe2JxmA3iRqJh+tDQCwZkU=pK_gKe2_Pt4mpgOsADwsJrRFQ@mail.gmail.com>

So it is safe to use post_action then?
I'm relying on it to serve cached response for users and update cache entry
in background if needed, so that users get most of the time a cache hit.

Is there a cleaner way to do this?

On Wed, Nov 23, 2011 at 11:57, Maxim Dounin <mdounin at mdounin.ru> wrote:

> Hello!
>
> On Tue, Nov 22, 2011 at 11:52:29PM +0200, Calin Don wrote:
>
> > So the time spent doing a post_action is added to $request_time or not?
>
> Added.
>
> In general, post_action is a dirty hack which works in context of
> a request.  It delays all processing of the request (i.e. logging)
> as well as processing of furhter requests in the connection in
> question (if any).  This is one of the reasons why post_action
> isn't documented, BTW.
>
> Maxim Dounin
>
> >
> > On Mon, Nov 21, 2011 at 20:55, Maxim Dounin <mdounin at mdounin.ru> wrote:
> >
> > > Hello!
> > >
> > > On Mon, Nov 21, 2011 at 08:41:15PM +0200, Calin Don wrote:
> > >
> > > > Hi,
> > > > How is the $request_time calculated in the case of a proxied
> request. I'm
> > > > interested especially in the case where the resource is stale in
> cache.
> > >
> > > $request_time is always time since start of the request (when
> > > first bytes are read from client) till end of the request (when
> > > last bytes are sent to client and logging happens).
> > >
> > > 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/20111123/618fb5a5/attachment.html>

From calin.don at gmail.com  Wed Nov 23 11:30:58 2011
From: calin.don at gmail.com (Calin Don)
Date: Wed, 23 Nov 2011 13:30:58 +0200
Subject: Is proxy_cache_bypass recaching?
In-Reply-To: <CACW9FdMEV4x=oJLoV0PeV0dSOW7KEKLJ0WRZLxsmPygfstquVg@mail.gmail.com>
References: <CACW9FdOHmK2S-O83LWA7tA_krjANZ6+7DffbPWanUkZ=Fr6ABQ@mail.gmail.com>
	<CACW9FdMEV4x=oJLoV0PeV0dSOW7KEKLJ0WRZLxsmPygfstquVg@mail.gmail.com>
Message-ID: <CAEOe2Jw4WBY3WKR74=4O7UwnfiA7wBNgTuW8PtkDwFuAD0e4=g@mail.gmail.com>

Yes. But be aware of Cache-Control header sent by your backend. (
http://wiki.nginx.org/HttpProxyModule#proxy_cache : The cache honors
backend's "Expires", "Cache-Control: no-cache", and "Cache-Control:
max-age=XXX" headers since version 0.7.48. Since version 7.66, "private"
and "no-store" are also honored.)

On Wed, Nov 23, 2011 at 05:31, Quintin Par <quintinpar at gmail.com> wrote:

> Can someone help me if this please?
>
> On Tue, Nov 22, 2011 at 8:53 AM, Quintin Par <quintinpar at gmail.com> wrote:
>
>> Hi all,
>>
>> I have this in my location directive
>>
>>                     proxy_cache_bypass                     $http_soon;
>>
>> now I do a curl call to invalidate
>>
>>
>>
>> [user at ip-10-111 ~]$ curl "www.site.com/teaser" -I -H "soon:true"
>>
>> HTTP/1.1 200 OK
>>
>> Server: nginx
>>
>> Date: Tue, 22 Nov 2011 03:07:20 GMT
>>
>> Content-Type: text/html; charset=utf-8
>>
>> Connection: keep-alive
>>
>> Keep-Alive: timeout=60
>>
>> Vary: Accept-Encoding
>>
>> Vary: Cookie
>>
>> X-Mod-Pagespeed: 0.9.17.7-716
>>
>> Cache-Control: max-age=0, no-cache, no-store
>>
>> X-Cache-Status: BYPASS
>>
>>
>>
>> [user at ip-10-111 ~]$ curl "www.site.com/teaser" -I
>>
>> HTTP/1.1 200 OK
>>
>> Server: nginx
>>
>> Date: Tue, 22 Nov 2011 03:07:37 GMT
>>
>> Content-Type: text/html; charset=utf-8
>>
>> Connection: keep-alive
>>
>> Keep-Alive: timeout=60
>>
>> Vary: Accept-Encoding
>>
>> Vary: Cookie
>>
>> X-Mod-Pagespeed: 0.9.17.7-716
>>
>> Cache-Control: max-age=0, no-cache, no-store
>>
>> X-Cache-Status: HIT
>>
>>
>>
>> Does this mean that the curl call with the header is* recaching the page
>> and the subsequent hits are cache hits?*
>>
>> -Quintin
>>
>
>
> _______________________________________________
> 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/20111123/da55441a/attachment-0001.html>

From faskiri.devel at gmail.com  Wed Nov 23 12:31:48 2011
From: faskiri.devel at gmail.com (Fasih)
Date: Wed, 23 Nov 2011 18:01:48 +0530
Subject: configuration problem with subdomain and proxy_pass
In-Reply-To: <aa157f0fae3e5fb1c29175c7ba4f5c58.NginxMailingListEnglish@forum.nginx.org>
References: <aa157f0fae3e5fb1c29175c7ba4f5c58.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <CAGBooBYi8dU70_RCL=EqD6C_0=nHrPT+UbrekXCK-9cXh+4udQ@mail.gmail.com>

Hi,

Small observation, from your configuration

http://subdomain.mydomain.com

Will initiate a request to

http://localhost:8080/subdomain/

and not

http://localhost:8080/subdomain

[notice the /] which you said works in the browser. You might want to check
that out.

Also, you can check out the tomcat's access logs to see what request you
got.

On Wed, Nov 23, 2011 at 4:52 PM, nicksoft <nginx-forum at nginx.us> wrote:

> Hi,
>
> I have a problem with configuration subdomains for nginx. I want to
> configure domains:
>
> mydomain.com (port 80) --> http://localhost:8080/
> www.mydomain.com (port 80) --> http://localhost:8080/
> subdomain.mydomain.com (port 80) --> http://localhost:8080/subdomain
>
>
> Tomcat listening at 8080 port on localhost. I have application
> 'subdomain' in tomcat webapps folder.
>
>
>
> This is my configuration file:
> ------------------------------------------------------
> server
> {
>       listen       80;
>       server_name  mydomain.com;
>
>       # Main location
>       location /
>       {
>           proxy_pass         http://127.0.0.1:8080/;
>           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       52m;
>           client_body_buffer_size    128k;
>
>           proxy_connect_timeout      3600;
>           proxy_send_timeout         3600;
>           proxy_read_timeout         90;
>
>           proxy_buffer_size          4k;
>           proxy_buffers              4 32k;
>           proxy_busy_buffers_size    64k;
>           proxy_temp_file_write_size 64k;
>           proxy_cache_valid 5m;
>       }
> }
>
> server
> {
>       listen       80;
>       server_name  www.mydomain.com;
>
>       # Main location
>       location /
>       {
>           proxy_pass         http://127.0.0.1:8080/;
>           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       52m;
>           client_body_buffer_size    128k;
>
>           proxy_connect_timeout      3600;
>           proxy_send_timeout         3600;
>           proxy_read_timeout         90;
>
>           proxy_buffer_size          4k;
>           proxy_buffers              4 32k;
>           proxy_busy_buffers_size    64k;
>           proxy_temp_file_write_size 64k;
>           proxy_cache_valid 5m;
>       }
> }
>
> server
> {
>       listen       80;
>       server_name subdomain.mydomain.com;
>
>       # Main location
>       location /
>       {
>           proxy_pass         http://127.0.0.1:8080/subdomain/;
>           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       52m;
>           client_body_buffer_size    128k;
>
>           proxy_connect_timeout      3600;
>           proxy_send_timeout         3600;
>           proxy_read_timeout         90;
>
>           proxy_buffer_size          4k;
>           proxy_buffers              4 32k;
>           proxy_busy_buffers_size    64k;
>           proxy_temp_file_write_size 64k;
>           proxy_cache_valid 5m;
>       }
> }
> ------------------------------------------------------
>
>
> For this configuration http:/www.mydomain.com  and http:/mydomain.com
> working very well, but subdomain.mydomain.com doesn't work (there is no
> response from server).
>
> Oh, and http://mydomain.com/subdomain in browser works well too.
>
>
> I will be grateful for your help.
>
> Posted at Nginx Forum:
> http://forum.nginx.org/read.php?2,218793,218793#msg-218793
>
> _______________________________________________
> 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/20111123/801c7670/attachment.html>

From mdounin at mdounin.ru  Wed Nov 23 12:39:51 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Wed, 23 Nov 2011 16:39:51 +0400
Subject: $request_time meaning
In-Reply-To: <CAEOe2JxmA3iRqJh+tDQCwZkU=pK_gKe2_Pt4mpgOsADwsJrRFQ@mail.gmail.com>
References: <CAEOe2JzF3si7pz3vi+4VBj7S_qAnX2HWcWT6Css=OCGr4dBtng@mail.gmail.com>
	<20111121185559.GB95664@mdounin.ru>
	<CAEOe2JyooV5pS4+Qk0RBu_WiQzQW_7PREOxZKOHUz+MPYF6Zuw@mail.gmail.com>
	<20111123095737.GO95664@mdounin.ru>
	<CAEOe2JxmA3iRqJh+tDQCwZkU=pK_gKe2_Pt4mpgOsADwsJrRFQ@mail.gmail.com>
Message-ID: <20111123123951.GQ95664@mdounin.ru>

Hello!

On Wed, Nov 23, 2011 at 01:25:46PM +0200, Calin Don wrote:

> So it is safe to use post_action then?
> I'm relying on it to serve cached response for users and update cache entry
> in background if needed, so that users get most of the time a cache hit.

It's not expected to cause crashes and/or protocol violations.  It 
may have unexpected side effects though (e.g. it delays processing 
of further requests in the connection, as I alredy outlined 
below).

> Is there a cleaner way to do this?

The way I would recommend is to use "tail -F /path/to/log | 
<script which does HEAD requests to update needed cache entries>".

Maxim Dounin

> 
> On Wed, Nov 23, 2011 at 11:57, Maxim Dounin <mdounin at mdounin.ru> wrote:
> 
> > Hello!
> >
> > On Tue, Nov 22, 2011 at 11:52:29PM +0200, Calin Don wrote:
> >
> > > So the time spent doing a post_action is added to $request_time or not?
> >
> > Added.
> >
> > In general, post_action is a dirty hack which works in context of
> > a request.  It delays all processing of the request (i.e. logging)
> > as well as processing of furhter requests in the connection in
> > question (if any).  This is one of the reasons why post_action
> > isn't documented, BTW.
> >
> > Maxim Dounin
> >
> > >
> > > On Mon, Nov 21, 2011 at 20:55, Maxim Dounin <mdounin at mdounin.ru> wrote:
> > >
> > > > Hello!
> > > >
> > > > On Mon, Nov 21, 2011 at 08:41:15PM +0200, Calin Don wrote:
> > > >
> > > > > Hi,
> > > > > How is the $request_time calculated in the case of a proxied
> > request. I'm
> > > > > interested especially in the case where the resource is stale in
> > cache.
> > > >
> > > > $request_time is always time since start of the request (when
> > > > first bytes are read from client) till end of the request (when
> > > > last bytes are sent to client and logging happens).
> > > >
> > > > 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  Wed Nov 23 13:12:04 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Wed, 23 Nov 2011 17:12:04 +0400
Subject: Nginx stuck processes
In-Reply-To: <78913b8db44bcdfd04b7bca6ee737233@ruby-forum.com>
References: <d04888efb006752e263fc5bca2c0d314@ruby-forum.com>
	<7efb85fdbf5ec94ab17531ab41499073@ruby-forum.com>
	<31b70aa141cf1024aba7576099862a99@ruby-forum.com>
	<20111122085856.GD95664@mdounin.ru>
	<78913b8db44bcdfd04b7bca6ee737233@ruby-forum.com>
Message-ID: <20111123131204.GS95664@mdounin.ru>

Hello!

On Wed, Nov 23, 2011 at 08:38:26AM +0100, Sandesh K. wrote:

> Maxim Dounin wrote in post #1033042:
> > Hello!
> >
> > On Tue, Nov 22, 2011 at 06:51:56AM +0100, Sandesh K. wrote:
> >
> >>
> >> --without-mail_smtp_module
> > [...]
> >
> >> (gdb) backtrace
> >> #0 0x0031ebba in _int_malloc () from /lib/tls/libc.so.6
> >> #1 0x0031f89c in _int_memalign () from /lib/tls/libc.so.6
> >> #2 0x0031e2c7 in memalign () from /lib/tls/libc.so.6
> >> #3 0x00320367 in posix_memalign () from /lib/tls/libc.so.6
> >> #4 0x0805da38 in ngx_memalign (alignment=16, size=4069528, log=0x108) at
> >> src/os/unix/ngx_alloc.c:56
> >
> > The "size=4069528" looks really strange...
> >
> >> #5 0x0804b917 in ngx_palloc_block (pool=0x8d0b9b0, size=52) at
> >> src/core/ngx_palloc.c:184
> >
> > Could you please show
> >
> > fr 5
> > p *pool
> >
> > output from gdb?

[...]

> (gdb) fr 5
> #5  0x0804b917 in ngx_palloc_block (pool=0x90f58e0, size=128) at
> src/core/ngx_palloc.c:184
> 184         m = ngx_memalign(NGX_POOL_ALIGNMENT, psize, pool->log);
> (gdb) p *pool
> $1 = {d = {last = 0x90f599c
> "xxxx at fx.xxx.1xx.2xx.8x.8x.mail.xx.xx.com\r\t\b\001", end = 0x90f59e0
> "\b\001", next = 0x910cb20, failed = 0}, max = 216,
>   current = 0x90f58e0, chain = 0x0, large = 0x0, cleanup = 0x0, log =
> 0x90f5918}

Ok, it looks like pool is ok (pool->d.end - pool == 0x100).  
Probably the cause of strange arguments is either corrupted stack 
or some compiler optimizations.

> (gdb) backtrace
> #0  0x00f36130 in _int_malloc () from /lib/tls/libc.so.6
> #1  0x00f36b3c in _int_memalign () from /lib/tls/libc.so.6
> #2  0x00f354f7 in memalign () from /lib/tls/libc.so.6
> #3  0x00f37607 in posix_memalign () from /lib/tls/libc.so.6
> #4  0x0805da3c in ngx_memalign (alignment=16, size=16751288, log=0x108)
> at src/os/unix/ngx_alloc.c:56
> #5  0x0804b917 in ngx_palloc_block (pool=0x90f58e0, size=128) at
> src/core/ngx_palloc.c:184

For now I tend to think the problem is somewhere in (g)libc.  You may 
want to update it, and if it doesn't help - recompile with debug 
and dig into it further.

Maxim Dounin


From nginx-forum at nginx.us  Wed Nov 23 13:18:44 2011
From: nginx-forum at nginx.us (Sylvia)
Date: Wed, 23 Nov 2011 08:18:44 -0500
Subject: No nice Permalinks with Wordpress
In-Reply-To: <ce5ab2dcdb90596969bc7fd613a92aaa.NginxMailingListEnglish@forum.nginx.org>
References: <ce5ab2dcdb90596969bc7fd613a92aaa.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <91e910b507df79c2804ecd0465fb3c30.NginxMailingListEnglish@forum.nginx.org>

http://wordpress.org/extend/plugins/nginx-compatibility/
You need this plugin, as WP disables permalinks when there is no
Apache's mod_rewrite.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218771,218817#msg-218817


From josh at joshparker.us  Wed Nov 23 13:21:10 2011
From: josh at joshparker.us (Parker, Joshua)
Date: Wed, 23 Nov 2011 08:21:10 -0500
Subject: php.ini & Virtual Hosts
Message-ID: <619A6AD9-BBF2-46F5-A857-0332B887A0AC@joshparker.us>

Is it possible to define a php.ini file per virtual host. If so, how can it be done? Thanks.


--
Joshua Parker
http://www.joshparker.us/

From nginx-forum at nginx.us  Wed Nov 23 13:27:55 2011
From: nginx-forum at nginx.us (mennanov)
Date: Wed, 23 Nov 2011 08:27:55 -0500
Subject: Match any location except particular one
Message-ID: <a9a54f553a370c9cf35b10f5c58fd40f.NginxMailingListEnglish@forum.nginx.org>

Hello, i have a website with the following files structure:

engine/
__index.php
webroot/
__images/
__css/
__cms/
______engine/
_________index.php
______webroot/
_________images/
_________css/
others/

So the folder "cms" is placed in a webroot of the main site. The point
is that all queries like "/js/jquery.js" must be treated as
"/webroot/js/jquery.js" and all cms queries like "/cms/js/jquery.js"
must be treated as "/webroot/cms/webroot/js/jquery.js". I spent 2 days
writing a proper config, the main problem is that all queries to /cms/
location MUST NOT be parsed in "location / {...}" and all queries in
"location /cms/ {...}" must not be parsed in "location / {...}"

Here is my config:

server {
    listen       8080;
    server_name  glinka.fm;
    root /home/renat/www/glinka;
    index index.php index.html;

    # rules for all queries except /cms/ (how to do that??) "location ~
/(?!cms)" makes a cycle and 500 error
    location / {
	try_files	/webroot/$uri /engine/index.php;
    }

    # rules for /cms/ queries only
    location /cms/ {
	rewrite /cms/(.*) /webroot/cms/webroot/$1;
	if (!-f $request_filename) {
	    rewrite (.*) /webroot/cms/engine/index.php;
	}
    }

    location ~ \.php$ {
      include fastcgi.conf;
    }
}

Please help me

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218819,218819#msg-218819


From r at roze.lv  Wed Nov 23 13:38:21 2011
From: r at roze.lv (Reinis Rozitis)
Date: Wed, 23 Nov 2011 15:38:21 +0200
Subject: php.ini & Virtual Hosts
In-Reply-To: <619A6AD9-BBF2-46F5-A857-0332B887A0AC@joshparker.us>
References: <619A6AD9-BBF2-46F5-A857-0332B887A0AC@joshparker.us>
Message-ID: <10F5FB89836F40EF8FCA27530252CDF1@DD21>

> Is it possible to define a php.ini file per virtual host. If so, how can it be done? Thanks.

To actually use different php.ini files you will have to spawn seperate php processes but you can achieve different settings for 
each virtualhost using a single php.ini file:

1. You can use the [HOST=yourvirtualhost] ( http://php.net/manual/en/ini.sections.php ) approach and define the settings directly in 
php.ini
2. You can make multiple pools in the FPM configuration  ( http://php.net/manual/en/install.fpm.configuration.php ) and change the 
settings via php_flag / php_admin_value etc directives
3. You can pass the config changes also from nginx using fastcgi_param PHP_VALUE ...
4. Can also use the .user.ini approach http://www.php.net/manual/en/configuration.file.per-user.php

rr 


From mdounin at mdounin.ru  Wed Nov 23 13:43:34 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Wed, 23 Nov 2011 17:43:34 +0400
Subject: server_name host order
In-Reply-To: <9a2e056ad6a61db90e7ecb9a88e24dbf.NginxMailingListEnglish@forum.nginx.org>
References: <9a2e056ad6a61db90e7ecb9a88e24dbf.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20111123134334.GT95664@mdounin.ru>

Hello!

On Wed, Nov 23, 2011 at 05:30:46AM -0500, token wrote:

> Hello, I would like to know if this is correct
> 
> Will NGINX allways return the first of the provided hostname name in
> from the `server_name` as beign the "SERVER_NAME",
> 
> eg server_name hostname1 hostname2;

The $server_name variable is always first name as defined by the 
"server_name" directive in the server{} in question.

With default fastcgi_params as shipped with nginx, SERVER_NAME as 
passed to php via fastcgi will be set to $server_name, i.e. first 
name defined with the "server_name" directive.

> I have two server blocks in this example 
> 
> server {
> 	listen 80;       
>         server_name www.example.co.uk *.example.co.uk;
> }
> 
> the URI = http://something.example.co.uk/
> 
> // PHP APPLICATION -------------- 
> I see as expected   `SERVER_NAME` and `HTTP_HOST`
> 
> SERVER_PORT =>  80                  
> SERVER_NAME =>  www.example.co.uk
> HTTP_HOST   =>  something.example.co.uk
> --------------
> 
> server {
>       listen  443;
>       server_name  www.example.co.uk secure.example.co.uk;
> }
> 
> // PHP APPLICATION -------------- 
> again the `SERVER_NAME` and `HTTP_HOST` as i would expect
> 
> the URI = https://secure.example.co.uk/
> 
> //  SERVER_PORT =>   443
> //  SERVER_NAME =>   www.example.co.uk
> //  HTTP_HOST   =>   secure.example.co.uk

Yes, this is correct and expected behaviour.

Maxim Dounin


From edho at myconan.net  Wed Nov 23 13:49:04 2011
From: edho at myconan.net (Edho Arief)
Date: Wed, 23 Nov 2011 20:49:04 +0700
Subject: No nice Permalinks with Wordpress
In-Reply-To: <91e910b507df79c2804ecd0465fb3c30.NginxMailingListEnglish@forum.nginx.org>
References: <ce5ab2dcdb90596969bc7fd613a92aaa.NginxMailingListEnglish@forum.nginx.org>
	<91e910b507df79c2804ecd0465fb3c30.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <CABTjkKn0C4kpB884hnB+crBMgawVw8+GVmA6B8e0G3iOdU6R9Q@mail.gmail.com>

On Wed, Nov 23, 2011 at 8:18 PM, Sylvia <nginx-forum at nginx.us> wrote:
> http://wordpress.org/extend/plugins/nginx-compatibility/
> You need this plugin, as WP disables permalinks when there is no
> Apache's mod_rewrite.
>

It is not disabled, just not given as default choices. It's still
possible using custom structure and removing the index.php/ part.

-- 
O< ascii ribbon campaign - stop html mail - www.asciiribbon.org


From nginx-forum at nginx.us  Wed Nov 23 13:55:28 2011
From: nginx-forum at nginx.us (token)
Date: Wed, 23 Nov 2011 08:55:28 -0500
Subject: server_name host order
In-Reply-To: <9a2e056ad6a61db90e7ecb9a88e24dbf.NginxMailingListEnglish@forum.nginx.org>
References: <9a2e056ad6a61db90e7ecb9a88e24dbf.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <7b79c3dda419963c9911f039d0e6d7d9.NginxMailingListEnglish@forum.nginx.org>

Thank you for clearing this up. Thanks Again Maxim!

> Hello!

> On Wed, Nov 23, 2011 at 05:30:46AM -0500, token wrote:

> Hello, I would like to know if this is correct
>
> Will NGINX allways return the first of the provided hostname name in
> from the `server_name` as beign the "SERVER_NAME",
>
> eg server_name hostname1 hostname2;

> The $server_name variable is always first name as defined by the
> "server_name" directive in the server{} in question.

> With default fastcgi_params as shipped with nginx, SERVER_NAME as
> passed to php via fastcgi will be set to $server_name, i.e. first
> name defined with the "server_name" directive.

> I have two server blocks in this example
>
> server {
> listen 80;
> server_name www.example.co.uk *.example.co.uk;
> }
>
> the URI = http://something.example.co.uk/
>
> // PHP APPLICATION --------------
> I see as expected `SERVER_NAME` and `HTTP_HOST`
>
> SERVER_PORT => 80
> SERVER_NAME => www.example.co.uk
> HTTP_HOST => something.example.co.uk
> --------------
>
> server {
> listen 443;
> server_name www.example.co.uk secure.example.co.uk;
> }
>
> // PHP APPLICATION --------------
> again the `SERVER_NAME` and `HTTP_HOST` as i would expect
>
> the URI = https://secure.example.co.uk/
>
> // SERVER_PORT => 443
> // SERVER_NAME => www.example.co.uk
> // HTTP_HOST => secure.example.co.uk

> Yes, this is correct and expected behaviour.

> Maxim Dounin

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218786,218825#msg-218825


From josh at joshparker.us  Wed Nov 23 13:59:00 2011
From: josh at joshparker.us (Parker, Joshua)
Date: Wed, 23 Nov 2011 08:59:00 -0500
Subject: php.ini & Virtual Hosts
In-Reply-To: <10F5FB89836F40EF8FCA27530252CDF1@DD21>
References: <619A6AD9-BBF2-46F5-A857-0332B887A0AC@joshparker.us>
	<10F5FB89836F40EF8FCA27530252CDF1@DD21>
Message-ID: <CAOa7tqYAmjgD9dcvudUFBot8aqJqZ6vzgW3TdfRkpDruZUe6bQ@mail.gmail.com>

Thank you very much for this info.



On Wednesday, November 23, 2011, Reinis Rozitis <r at roze.lv> wrote:
>> Is it possible to define a php.ini file per virtual host. If so, how can
it be done? Thanks.
>
> To actually use different php.ini files you will have to spawn seperate
php processes but you can achieve different settings for each virtualhost
using a single php.ini file:
>
> 1. You can use the [HOST=yourvirtualhost] (
http://php.net/manual/en/ini.sections.php ) approach and define the
settings directly in php.ini
> 2. You can make multiple pools in the FPM configuration  (
http://php.net/manual/en/install.fpm.configuration.php ) and change the
settings via php_flag / php_admin_value etc directives
> 3. You can pass the config changes also from nginx using fastcgi_param
PHP_VALUE ...
> 4. Can also use the .user.ini approach
http://www.php.net/manual/en/configuration.file.per-user.php
>
> rr
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx
>

-- 
[image: Josh Parker WordPress Consultant] <http://www.joshparker.us/>
Joshua Parker
WordPress Consultant
888.255.1798 x701

Consulting Site
www.joshparker.us
[image: Twitter] <http://twitter.com/#!/joshmac>[image:
Linkedin]<http://www.linkedin.com/in/joshmac>[image:
Josh Parker :: WordPress Consultant] <http://www.joshparker.us/feed/>[image:
Skype][image: Google+] <http://gplus.to/joshuaparker>[image: WordPress
Profile] <http://wordpress.org/extend/plugins/profile/parkerj>[image:
About.me] <http://about.me/joshuaparker>[image:
Gravatar]<http://en.gravatar.com/joshmac3>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20111123/4cd4acc9/attachment.html>

From nginx-forum at nginx.us  Wed Nov 23 13:59:08 2011
From: nginx-forum at nginx.us (nicksoft)
Date: Wed, 23 Nov 2011 08:59:08 -0500
Subject: configuration problem with subdomain and proxy_pass
In-Reply-To: <aa157f0fae3e5fb1c29175c7ba4f5c58.NginxMailingListEnglish@forum.nginx.org>
References: <aa157f0fae3e5fb1c29175c7ba4f5c58.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <ed2b2b3f990ce63d3f6f28463ea32ebf.NginxMailingListEnglish@forum.nginx.org>

@faskiri.devel


http://localhost:8080/subdomain/
     is equals to
http://localhost:8080/subdomain

Both work the same.



In DNS control Panel of my domain I have a 'A record' :
subdomain.mydomain.com with IP adress of my VPS server (with nginx and
tomcat). It should works.... I don't know what is wrong in my
configuration.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218793,218827#msg-218827


From nginx-forum at nginx.us  Wed Nov 23 14:26:42 2011
From: nginx-forum at nginx.us (nicksoft)
Date: Wed, 23 Nov 2011 09:26:42 -0500
Subject: configuration problem with subdomain and proxy_pass
In-Reply-To: <ed2b2b3f990ce63d3f6f28463ea32ebf.NginxMailingListEnglish@forum.nginx.org>
References: <aa157f0fae3e5fb1c29175c7ba4f5c58.NginxMailingListEnglish@forum.nginx.org>
	<ed2b2b3f990ce63d3f6f28463ea32ebf.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <f83d55285be9b4392cdde176f9637bde.NginxMailingListEnglish@forum.nginx.org>

I found error in nginx log:

2011/11/22 16:26:12 [error] 1423#0: *1 connect() failed (111: Connection
refused) while connecting to upstream, client: xxx.yyy.zzz.vvv, server:
subdomain.mydomain.com, request: "GET / HTTP/1.1", upstream:
"http://127.0.0.1:8080/subdomain/", host: "subdomain.mydomain.com"

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218793,218839#msg-218839


From francis at daoine.org  Wed Nov 23 14:45:56 2011
From: francis at daoine.org (Francis Daly)
Date: Wed, 23 Nov 2011 14:45:56 +0000
Subject: Match any location except particular one
In-Reply-To: <a9a54f553a370c9cf35b10f5c58fd40f.NginxMailingListEnglish@forum.nginx.org>
References: <a9a54f553a370c9cf35b10f5c58fd40f.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20111123144556.GE27443@craic.sysops.org>

On Wed, Nov 23, 2011 at 08:27:55AM -0500, mennanov wrote:

Hi there,

> I spent 2 days
> writing a proper config, the main problem is that all queries to /cms/
> location MUST NOT be parsed in "location / {...}" and all queries in
> "location /cms/ {...}" must not be parsed in "location / {...}"
> 
> Here is my config:

You have three top-level location{} blocks here.

You probably only want two.

>     location / {

This will match all requests that don't match another location.

>     # rules for /cms/ queries only
>     location /cms/ {

This will match all requests that begin "/cms/", unless they match
another location.

>     location ~ \.php$ {

This will match all requests that end in ".php". Specifically including
both "/index.php" and "/cms/index.php". Which is likely not what you want.

In nginx, every request is handled by one location{} block.

See, for example, http://wiki.nginx.org/HttpCoreModule#location for
details of how the one is chosen.

So you probably want something like

  location / {
   # all "location /" config
   location ~ \.php$ {
    # all "php inside location /" config
   }
  }

  location /cms/ {
   # all "location /cms/" config
   location ~ \.php$ {
    # all "php inside location /cms/" config
   }
  }

Good luck with it,

	f
-- 
Francis Daly        francis at daoine.org


From nginx-forum at nginx.us  Wed Nov 23 15:11:08 2011
From: nginx-forum at nginx.us (mennanov)
Date: Wed, 23 Nov 2011 10:11:08 -0500
Subject: Match any location except particular one
In-Reply-To: <a9a54f553a370c9cf35b10f5c58fd40f.NginxMailingListEnglish@forum.nginx.org>
References: <a9a54f553a370c9cf35b10f5c58fd40f.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <d4fcef6adc07cb9ff434449864bb3ebf.NginxMailingListEnglish@forum.nginx.org>

Thanks but i've already found a solution in the russian nginx forum.

location ~ ^/cms/(.*)$ {
try_files /webroot/cms/webroot/$1 /webroot/cms/engine/index.php;
}

Is what i needed.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218819,218849#msg-218849


From fasihullah.askiri at gmail.com  Wed Nov 23 15:07:56 2011
From: fasihullah.askiri at gmail.com (Fasihullah Askiri)
Date: Wed, 23 Nov 2011 20:37:56 +0530
Subject: configuration problem with subdomain and proxy_pass
In-Reply-To: <f83d55285be9b4392cdde176f9637bde.NginxMailingListEnglish@forum.nginx.org>
References: <aa157f0fae3e5fb1c29175c7ba4f5c58.NginxMailingListEnglish@forum.nginx.org>
	<ed2b2b3f990ce63d3f6f28463ea32ebf.NginxMailingListEnglish@forum.nginx.org>
	<f83d55285be9b4392cdde176f9637bde.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <CAGBooBb3+4iMa13=e1OpYVA9PLf7KZ5GNN1gNY=2C77eroBo-w@mail.gmail.com>

Can you try:

curl http://127.0.0.1:8080/subdomain/ -H "host: subdomain.mydomain.com" -v

or

wget  http://127.0.0.1:8080/subdomain/ --header "host:
subdomain.mydomain.com" --debug

from the nginx machine and send the output, looks like there is some error
in the network configuration

+Fasih

On Wed, Nov 23, 2011 at 7:56 PM, nicksoft <nginx-forum at nginx.us> wrote:

> I found error in nginx log:
>
> 2011/11/22 16:26:12 [error] 1423#0: *1 connect() failed (111: Connection
> refused) while connecting to upstream, client: xxx.yyy.zzz.vvv, server:
> subdomain.mydomain.com, request: "GET / HTTP/1.1", upstream:
> "http://127.0.0.1:8080/subdomain/", host: "subdomain.mydomain.com"
>
> Posted at Nginx Forum:
> http://forum.nginx.org/read.php?2,218793,218839#msg-218839
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx
>



-- 
+Fasih

*Wisdom doesn't automatically come with old age. Nothing does - except
wrinkles. It's true, some wines improve with age. But only if the grapes
were good in the first place*
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20111123/03e0aa32/attachment-0001.html>

From igor at sysoev.ru  Wed Nov 23 16:16:18 2011
From: igor at sysoev.ru (Igor Sysoev)
Date: Wed, 23 Nov 2011 20:16:18 +0400
Subject: Match any location except particular one
In-Reply-To: <d4fcef6adc07cb9ff434449864bb3ebf.NginxMailingListEnglish@forum.nginx.org>
References: <a9a54f553a370c9cf35b10f5c58fd40f.NginxMailingListEnglish@forum.nginx.org>
	<d4fcef6adc07cb9ff434449864bb3ebf.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20111123161618.GB49193@nginx.com>

On Wed, Nov 23, 2011 at 10:11:08AM -0500, mennanov wrote:
> Thanks but i've already found a solution in the russian nginx forum.
> 
> location ~ ^/cms/(.*)$ {
> try_files /webroot/cms/webroot/$1 /webroot/cms/engine/index.php;
> }
> 
> Is what i needed.

location /cms/ {
    alias /home/renat/www/glinka/webroot/cms/webroot/;
    try_files  $uri   /webroot/cms/engine/index.php;
}


-- 
Igor Sysoev


From nginx-forum at nginx.us  Wed Nov 23 16:26:02 2011
From: nginx-forum at nginx.us (mennanov)
Date: Wed, 23 Nov 2011 11:26:02 -0500
Subject: Match any location except particular one
In-Reply-To: <20111123161618.GB49193@nginx.com>
References: <20111123161618.GB49193@nginx.com>
Message-ID: <e58f7d258698705912da7cffbbad37b7.NginxMailingListEnglish@forum.nginx.org>

Igor Sysoev Wrote:
-------------------------------------------------------
> On Wed, Nov 23, 2011 at 10:11:08AM -0500, mennanov
> wrote:
> > Thanks but i've already found a solution in the
> russian nginx forum.
> > 
> > location ~ ^/cms/(.*)$ {
> > try_files /webroot/cms/webroot/$1
> /webroot/cms/engine/index.php;
> > }
> > 
> > Is what i needed.
> 
> location /cms/ {
>     alias
> /home/renat/www/glinka/webroot/cms/webroot/;
>     try_files  $uri  
> /webroot/cms/engine/index.php;
> }
> 
> 
> -- 
> Igor Sysoev
> 
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx

Is there any way to use relative path in alias?
It does not look comfortable to write the whole absolute path, it would
be rather better if it is current root + path

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218819,218856#msg-218856


From simone.fumagalli at contactlab.com  Wed Nov 23 16:26:54 2011
From: simone.fumagalli at contactlab.com (Simone Fumagalli)
Date: Wed, 23 Nov 2011 17:26:54 +0100
Subject: Nginx Cache and WP canonical URL create infinite loop (bug ?)
Message-ID: <4ECD1ECE.4060701@contactlab.com>

Hello everybody.

Just want to report this to all people who have an NGINX that cache request in front a Wordpress installation

Scenario

 - Nginx is configured as reverse proxy in front of an Apache that run Wordpress. 
 - proxy_cache_key is not defined so is $scheme$proxy_host$request_uri;

Let's say a client require the page http://www.mysite.com/my-page// (yes double slash at the end) that is not cached

 1) Nginx pass the request to Apache/Wordpress
 2) Apache/Wordpress process the request and reply with a 301 and sends the client to http://www.mysite.com/my-page/ because is the canonical URL
 3) Nginx save in the cache the response with these data 
        KEY: http://www.mysite.com/my-page/
        HTTP/1.0 301 Moved Permanently
	[..]
        Location: http://www.mysite.com/my-page/
	[..]
    As you can see the KEY is equal to Location
 4) The user/browser send a request for http://www.mysite.com/my-page/   (just one slash)
 5) Nginx find the key in the cache and so redirect the browser to http://www.mysite.com/my-page/ causing an infinite loop
 
Conclusions

 - All the users that require the page http://www.mysite.com/my-page/ while is cached are send in the loop
 - This happen of course also for requests with a number of slashes at the end > 2
 - I can't call this bug but is just an "unexpected" behaviour, for sure a malicious user can overload your site or make pages un-available with these kind of requests.
 - There are probably other type of URL that create this situation. Have found any until now, can someone point one out ?
 - Is there a way to avoid this ?

Thanks

--
Simone
   



From igor at sysoev.ru  Wed Nov 23 16:29:07 2011
From: igor at sysoev.ru (Igor Sysoev)
Date: Wed, 23 Nov 2011 20:29:07 +0400
Subject: Match any location except particular one
In-Reply-To: <e58f7d258698705912da7cffbbad37b7.NginxMailingListEnglish@forum.nginx.org>
References: <20111123161618.GB49193@nginx.com>
	<e58f7d258698705912da7cffbbad37b7.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20111123162907.GC49193@nginx.com>

On Wed, Nov 23, 2011 at 11:26:02AM -0500, mennanov wrote:
> Igor Sysoev Wrote:
> -------------------------------------------------------
> > On Wed, Nov 23, 2011 at 10:11:08AM -0500, mennanov
> > wrote:
> > > Thanks but i've already found a solution in the
> > russian nginx forum.
> > > 
> > > location ~ ^/cms/(.*)$ {
> > > try_files /webroot/cms/webroot/$1
> > /webroot/cms/engine/index.php;
> > > }
> > > 
> > > Is what i needed.
> > 
> > location /cms/ {
> >     alias
> > /home/renat/www/glinka/webroot/cms/webroot/;
> >     try_files  $uri  
> > /webroot/cms/engine/index.php;
> > }
> > 
> > 
> > -- 
> > Igor Sysoev
> > 
> > _______________________________________________
> > nginx mailing list
> > nginx at nginx.org
> > http://mailman.nginx.org/mailman/listinfo/nginx
> 
> Is there any way to use relative path in alias?
> It does not look comfortable to write the whole absolute path, it would
> be rather better if it is current root + path

alias is more flexible root.


-- 
Igor Sysoev


From mdounin at mdounin.ru  Wed Nov 23 17:04:24 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Wed, 23 Nov 2011 21:04:24 +0400
Subject: Nginx Cache and WP canonical URL create infinite loop (bug ?)
In-Reply-To: <4ECD1ECE.4060701@contactlab.com>
References: <4ECD1ECE.4060701@contactlab.com>
Message-ID: <20111123170424.GX95664@mdounin.ru>

Hello!

On Wed, Nov 23, 2011 at 05:26:54PM +0100, Simone Fumagalli wrote:

> Hello everybody.
> 
> Just want to report this to all people who have an NGINX that cache request in front a Wordpress installation
> 
> Scenario
> 
>  - Nginx is configured as reverse proxy in front of an Apache that run Wordpress. 
>  - proxy_cache_key is not defined so is $scheme$proxy_host$request_uri;
> 
> Let's say a client require the page http://www.mysite.com/my-page// (yes double slash at the end) that is not cached
> 
>  1) Nginx pass the request to Apache/Wordpress
>  2) Apache/Wordpress process the request and reply with a 301 and sends the client to http://www.mysite.com/my-page/ because is the canonical URL
>  3) Nginx save in the cache the response with these data 
>         KEY: http://www.mysite.com/my-page/
>         HTTP/1.0 301 Moved Permanently
> 	[..]
>         Location: http://www.mysite.com/my-page/
> 	[..]
>     As you can see the KEY is equal to Location
>  4) The user/browser send a request for http://www.mysite.com/my-page/   (just one slash)
>  5) Nginx find the key in the cache and so redirect the browser to http://www.mysite.com/my-page/ causing an infinite loop
>  
> Conclusions
> 
>  - All the users that require the page http://www.mysite.com/my-page/ while is cached are send in the loop
>  - This happen of course also for requests with a number of slashes at the end > 2
>  - I can't call this bug but is just an "unexpected" behaviour, for sure a malicious user can overload your site or make pages un-available with these kind of requests.
>  - There are probably other type of URL that create this situation. Have found any until now, can someone point one out ?
>  - Is there a way to avoid this ?

Could you please show your config?  By default nginx will use the 
same uri in cache as it will use in request to upstream, and this 
shouldn't happen.

There are some nuances though, at least when using proxy_pass with 
variables (see http://trac.nginx.org/nginx/ticket/34, patch is 
awaiting Igor's review).

Maxim Dounin


From nginx-forum at nginx.us  Wed Nov 23 17:08:44 2011
From: nginx-forum at nginx.us (nicksoft)
Date: Wed, 23 Nov 2011 12:08:44 -0500
Subject: configuration problem with subdomain and proxy_pass
In-Reply-To: <f83d55285be9b4392cdde176f9637bde.NginxMailingListEnglish@forum.nginx.org>
References: <aa157f0fae3e5fb1c29175c7ba4f5c58.NginxMailingListEnglish@forum.nginx.org>
	<ed2b2b3f990ce63d3f6f28463ea32ebf.NginxMailingListEnglish@forum.nginx.org>
	<f83d55285be9b4392cdde176f9637bde.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <bb1837d8d9836db8a39d7c2582355c26.NginxMailingListEnglish@forum.nginx.org>

This is otput from console (Debian 6 min.):

wget http://127.0.0.1:8080/subdomain/ --header "host:
subdomain.mydomain.com" --debug


DEBUG output created by Wget 1.12 on linux-gnu.

--2011-11-23 17:04:15--  http://127.0.0.1:8080/subdomain/
Connecting to 127.0.0.1:8080... connected.
Created socket 3.
Releasing 0x08dfc180 (new refcount 0).
Deleting unused 0x08dfc180.

---request begin---
GET /jaspresso/ HTTP/1.0
User-Agent: Wget/1.12 (linux-gnu)
Accept: */*
host: subdomain.mydomain.com
Connection: Keep-Alive

---request end---
HTTP request sent, awaiting response...
---response begin---
HTTP/1.1 400 Bad Request
Server: Apache-Coyote/1.1
Date: Wed, 23 Nov 2011 17:04:15 GMT
Connection: close

---response end---
400 Bad Request
Closed fd 3
2011-11-23 17:04:15 ERROR 400: Bad Request.



Strange... What could it be?

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218793,218867#msg-218867


From nginx-forum at nginx.us  Wed Nov 23 17:10:20 2011
From: nginx-forum at nginx.us (nicksoft)
Date: Wed, 23 Nov 2011 12:10:20 -0500
Subject: configuration problem with subdomain and proxy_pass
In-Reply-To: <bb1837d8d9836db8a39d7c2582355c26.NginxMailingListEnglish@forum.nginx.org>
References: <aa157f0fae3e5fb1c29175c7ba4f5c58.NginxMailingListEnglish@forum.nginx.org>
	<ed2b2b3f990ce63d3f6f28463ea32ebf.NginxMailingListEnglish@forum.nginx.org>
	<f83d55285be9b4392cdde176f9637bde.NginxMailingListEnglish@forum.nginx.org>
	<bb1837d8d9836db8a39d7c2582355c26.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <082866d24a5fc5f0298e7f9a146cb306.NginxMailingListEnglish@forum.nginx.org>

For: wget http://127.0.0.1:8080/subdomain/  otuput:


--2011-11-23 17:09:10--  http://127.0.0.1:8080/subdomain/
Connecting to 127.0.0.1:8080... connected.
HTTP request sent, awaiting response... 200 OK
Length: 4761 (4.6K) [text/html]
Saving to: `index.html'

100%[===========================================================================================>]
4,761       --.-K/s   in 0s

2011-11-23 17:09:10 (413 MB/s) - `index.html' saved [4761/4761]

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218793,218869#msg-218869


From simone.fumagalli at contactlab.com  Wed Nov 23 17:11:41 2011
From: simone.fumagalli at contactlab.com (Simone Fumagalli)
Date: Wed, 23 Nov 2011 18:11:41 +0100
Subject: Nginx Cache and WP canonical URL create infinite loop (bug ?)
In-Reply-To: <20111123170424.GX95664@mdounin.ru>
References: <4ECD1ECE.4060701@contactlab.com>
	<20111123170424.GX95664@mdounin.ru>
Message-ID: <4ECD294D.90108@contactlab.com>

> Could you please show your config?  

Nothing special, here we go (only important parts)

server {

	server_name www.mysite.com
	listen 1.2.3.4:80

	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;
	proxy_store_access  user:rw  group:rw  all:r;

	location / {
	
		# If logged in, don't cache.
		if ($http_cookie ~* "comment_author_|wordpress_(?!test_cookie)|wp-postpass_" ) {
			set $do_not_cache 1;
		}
		
		# Response in not cached	
		proxy_no_cache $do_not_cache;
		
		# Response is not taken from the cache
		proxy_cache_bypass $do_not_cache;
		
		proxy_cache MY_CACHE;
		proxy_pass http://ALL_backend;

	}
	
	location ~ /purge(/.*) {
		proxy_cache_purge MY_CACHE "$scheme://$host$1$is_args$args";
	}

}

> By default nginx will use the same uri in cache as it will use in request to upstream, and this shouldn't happen

AFAIK when I require http://www.mysite.com/my-page// the key for the cache is set to http://www.mysite.com/my-page/

Ciao

--
Simone


From faskiri.devel at gmail.com  Wed Nov 23 17:17:39 2011
From: faskiri.devel at gmail.com (Fasih)
Date: Wed, 23 Nov 2011 22:47:39 +0530
Subject: configuration problem with subdomain and proxy_pass
In-Reply-To: <082866d24a5fc5f0298e7f9a146cb306.NginxMailingListEnglish@forum.nginx.org>
References: <aa157f0fae3e5fb1c29175c7ba4f5c58.NginxMailingListEnglish@forum.nginx.org>
	<ed2b2b3f990ce63d3f6f28463ea32ebf.NginxMailingListEnglish@forum.nginx.org>
	<f83d55285be9b4392cdde176f9637bde.NginxMailingListEnglish@forum.nginx.org>
	<bb1837d8d9836db8a39d7c2582355c26.NginxMailingListEnglish@forum.nginx.org>
	<082866d24a5fc5f0298e7f9a146cb306.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <CAGBooBYffA8FOmDjTqw_xakqp4ofLYDzzkYP8hxUoLj=egnYuw@mail.gmail.com>

Your Apache is not set correctly. Please make sure that the VirtualHost is
setup correctly.
Basically normal wget is working but if you set the host: header apache
rejects the request.

I think you have explicity put a VirtualHost domain:8080

changing that to VirtualHost *:8080 should fix it. In any case, since plain
wget also gives you errors, you are better off looking at Apache
logs/manual than in nginx.

On Wed, Nov 23, 2011 at 10:40 PM, nicksoft <nginx-forum at nginx.us> wrote:

> For: wget http://127.0.0.1:8080/subdomain/  otuput:
>
>
> --2011-11-23 17:09:10--  http://127.0.0.1:8080/subdomain/
> Connecting to 127.0.0.1:8080... connected.
> HTTP request sent, awaiting response... 200 OK
> Length: 4761 (4.6K) [text/html]
> Saving to: `index.html'
>
>
> 100%[===========================================================================================>]
> 4,761       --.-K/s   in 0s
>
> 2011-11-23 17:09:10 (413 MB/s) - `index.html' saved [4761/4761]
>
> Posted at Nginx Forum:
> http://forum.nginx.org/read.php?2,218793,218869#msg-218869
>
> _______________________________________________
> 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/20111123/01b0247d/attachment.html>

From mdounin at mdounin.ru  Wed Nov 23 17:41:23 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Wed, 23 Nov 2011 21:41:23 +0400
Subject: Nginx Cache and WP canonical URL create infinite loop (bug ?)
In-Reply-To: <4ECD294D.90108@contactlab.com>
References: <4ECD1ECE.4060701@contactlab.com>
	<20111123170424.GX95664@mdounin.ru> <4ECD294D.90108@contactlab.com>
Message-ID: <20111123174123.GY95664@mdounin.ru>

Hello!

On Wed, Nov 23, 2011 at 06:11:41PM +0100, Simone Fumagalli wrote:

> > Could you please show your config?  
> 
> Nothing special, here we go (only important parts)
> 
> server {
> 
> 	server_name www.mysite.com
> 	listen 1.2.3.4:80
> 
> 	proxy_redirect     off;
> 	proxy_set_header   Host             $host;
> 	proxy_set_header   X-Real-IP        $remote_addr;
> 	proxy_set_header   X-Forwarded-For  $proxy_add_x_forwarded_for;
> 	proxy_store_access  user:rw  group:rw  all:r;
> 
> 	location / {
> 	
> 		# If logged in, don't cache.
> 		if ($http_cookie ~* "comment_author_|wordpress_(?!test_cookie)|wp-postpass_" ) {
> 			set $do_not_cache 1;
> 		}
> 		
> 		# Response in not cached	
> 		proxy_no_cache $do_not_cache;
> 		
> 		# Response is not taken from the cache
> 		proxy_cache_bypass $do_not_cache;
> 		
> 		proxy_cache MY_CACHE;
> 		proxy_pass http://ALL_backend;
> 
> 	}
> 	
> 	location ~ /purge(/.*) {
> 		proxy_cache_purge MY_CACHE "$scheme://$host$1$is_args$args";
> 	}
> 
> }
> 
> > By default nginx will use the same uri in cache as it will use in request to upstream, and this shouldn't happen
> 
> AFAIK when I require http://www.mysite.com/my-page// the key for the cache is set to http://www.mysite.com/my-page/

With this config, the cache key for the request to 
"http://www.mysite.com/my-page//" will be 
"http://ALL_backend/my-page//".

Maxim Dounin


From nginx-forum at nginx.us  Wed Nov 23 18:18:04 2011
From: nginx-forum at nginx.us (nicksoft)
Date: Wed, 23 Nov 2011 13:18:04 -0500
Subject: configuration problem with subdomain and proxy_pass
In-Reply-To: <082866d24a5fc5f0298e7f9a146cb306.NginxMailingListEnglish@forum.nginx.org>
References: <aa157f0fae3e5fb1c29175c7ba4f5c58.NginxMailingListEnglish@forum.nginx.org>
	<ed2b2b3f990ce63d3f6f28463ea32ebf.NginxMailingListEnglish@forum.nginx.org>
	<f83d55285be9b4392cdde176f9637bde.NginxMailingListEnglish@forum.nginx.org>
	<bb1837d8d9836db8a39d7c2582355c26.NginxMailingListEnglish@forum.nginx.org>
	<082866d24a5fc5f0298e7f9a146cb306.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <042f7bd7a1f48b36d6f13b53cec8f4ae.NginxMailingListEnglish@forum.nginx.org>

Thank you. I will check tomcat configuration.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218793,218875#msg-218875


From nginx-forum at nginx.us  Wed Nov 23 18:31:15 2011
From: nginx-forum at nginx.us (rateitstar)
Date: Wed, 23 Nov 2011 13:31:15 -0500
Subject: Hi, i have a question,
	Im currently got a caching system in place but it use php to utilize
	it...
Message-ID: <79395dd25a806e2d1701f88bdd3c399d.NginxMailingListEnglish@forum.nginx.org>

Im currently got a caching system in place but it use php to utilize it,


is their some way to direct a user to html cache file ie

www.domain.com/view.php?id={id}

would hidden redirect to:

www.domain.com/cache/{id}.html

not using any 304 directs **totally invisible..if you know what i mean
due to search engines being tricky about those things

*now this is the bit im thinking about::

if the {id}.html file does  not exist

carry the request out as normal with the view.php script

for example with this script:

www.domain.com/view.php?id={id}



Thank you before hand.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218877,218877#msg-218877


From pasik at iki.fi  Wed Nov 23 19:11:03 2011
From: pasik at iki.fi (Pasi =?iso-8859-1?Q?K=E4rkk=E4inen?=)
Date: Wed, 23 Nov 2011 21:11:03 +0200
Subject: Disable buffering/caching of HTTP PUT upload requests in nginx	proxy
Message-ID: <20111123191103.GW12984@reaktio.net>

Hello,

During the upcoming days I'm going to try to implement a feature in nginx proxy
that allows disabling local disk buffering/caching of HTTP PUT upload requests,
e.g. allow directly stream the uploads to the backend server.

Right now I'm having problems because nginx proxy stores all the uploads temporarily
to the nginx server local disk, and thus fills up the disk when big files are being uploaded.

If someone has comments/ideas about this, feel free to write to me.
I'm currently in the process of getting familiar with nginx code
and the request 'flow' inside nginx.

I'll let you know how it goes :)

Thanks,

-- Pasi


From zzz at zzz.org.ua  Wed Nov 23 19:22:39 2011
From: zzz at zzz.org.ua (Alexandr Gomoliako)
Date: Wed, 23 Nov 2011 21:22:39 +0200
Subject: Disable buffering/caching of HTTP PUT upload requests in nginx
	proxy
In-Reply-To: <20111123191103.GW12984@reaktio.net>
References: <20111123191103.GW12984@reaktio.net>
Message-ID: <CAKb9G9hztkXxmRL+YL35BvLyHWV2S6wkpyVC-8O2P7T2AT2Fbw@mail.gmail.com>

On Wed, Nov 23, 2011 at 9:11 PM, Pasi K?rkk?inen <pasik at iki.fi> wrote:
> Hello,
>
> During the upcoming days I'm going to try to implement a feature in nginx proxy
> that allows disabling local disk buffering/caching of HTTP PUT upload requests,
> e.g. allow directly stream the uploads to the backend server.
>
> Right now I'm having problems because nginx proxy stores all the uploads temporarily
> to the nginx server local disk, and thus fills up the disk when big files are being uploaded.
>
> If someone has comments/ideas about this, feel free to write to me.
> I'm currently in the process of getting familiar with nginx code
> and the request 'flow' inside nginx.

It depends how are you going to approach this. I already did something
similar by simply hijacking connection of the request.

Consider subscribing to nginx-devel@ for further questions.


From pasik at iki.fi  Wed Nov 23 19:34:23 2011
From: pasik at iki.fi (Pasi =?iso-8859-1?Q?K=E4rkk=E4inen?=)
Date: Wed, 23 Nov 2011 21:34:23 +0200
Subject: Disable buffering/caching of HTTP PUT upload requests in
	nginx	proxy
In-Reply-To: <CAKb9G9hztkXxmRL+YL35BvLyHWV2S6wkpyVC-8O2P7T2AT2Fbw@mail.gmail.com>
References: <20111123191103.GW12984@reaktio.net>
	<CAKb9G9hztkXxmRL+YL35BvLyHWV2S6wkpyVC-8O2P7T2AT2Fbw@mail.gmail.com>
Message-ID: <20111123193423.GX12984@reaktio.net>

On Wed, Nov 23, 2011 at 09:22:39PM +0200, Alexandr Gomoliako wrote:
> On Wed, Nov 23, 2011 at 9:11 PM, Pasi K?rkk?inen <pasik at iki.fi> wrote:
> > Hello,
> >
> > During the upcoming days I'm going to try to implement a feature in nginx proxy
> > that allows disabling local disk buffering/caching of HTTP PUT upload requests,
> > e.g. allow directly stream the uploads to the backend server.
> >
> > Right now I'm having problems because nginx proxy stores all the uploads temporarily
> > to the nginx server local disk, and thus fills up the disk when big files are being uploaded.
> >
> > If someone has comments/ideas about this, feel free to write to me.
> > I'm currently in the process of getting familiar with nginx code
> > and the request 'flow' inside nginx.
> 
> It depends how are you going to approach this. I already did something
> similar by simply hijacking connection of the request.
> 

Ok, good to hear.

> Consider subscribing to nginx-devel@ for further questions.
> 

Good idea. I'll do that. 

Thanks,

-- Pasi


From nginx-forum at nginx.us  Wed Nov 23 19:46:13 2011
From: nginx-forum at nginx.us (EricTheRed03)
Date: Wed, 23 Nov 2011 14:46:13 -0500
Subject: Simple rewrite (redirection)
Message-ID: <e94c00ba41258efee577167cad3d92ee.NginxMailingListEnglish@forum.nginx.org>

Hi Guys,

In my HTTP config section in the config file, I would like to implement
a redirection based on this scenario:

If the URL is not "http://sitea.com/exactmatch" then redirect to the
"https" home. else go to our hidden url (our testing boxes)

#logic would be somthing like this..
If (The URL is anything but "/ahiddenurl") 
{
        go to "https://sitea.com"
}
else 
{
      go to "http://hiddenurl"
}

Is this possible in the nginx.conf file?? Any help would be so very much
appreciated!

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218893,218893#msg-218893


From nginx-forum at nginx.us  Wed Nov 23 19:54:06 2011
From: nginx-forum at nginx.us (EricTheRed03)
Date: Wed, 23 Nov 2011 14:54:06 -0500
Subject: Simple rewrite (redirection)
In-Reply-To: <e94c00ba41258efee577167cad3d92ee.NginxMailingListEnglish@forum.nginx.org>
References: <e94c00ba41258efee577167cad3d92ee.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <cffdb7d206c70dd854655837507d00ea.NginxMailingListEnglish@forum.nginx.org>

This is the block of code I've been working on for hours ... :(

--------------- nginx.conf ------------
   server {
            listen   80;
            server_name sitea.com;
            rewrite ^(/hiddendevpath)(.*)$ http://sitea.com/ ;
            rewrite ^(.*) https://sitea.com
               }

------------- testing below ----------

sudo /usr/sbin/nginx -t
Error: nginx: [emerg] invalid number of arguments in "rewrite" directive
in /etc/nginx/nginx.conf:44

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218893,218894#msg-218894


From ne at vbart.ru  Wed Nov 23 20:17:28 2011
From: ne at vbart.ru (Valentin V. Bartenev)
Date: Thu, 24 Nov 2011 00:17:28 +0400
Subject: Simple rewrite (redirection)
In-Reply-To: <cffdb7d206c70dd854655837507d00ea.NginxMailingListEnglish@forum.nginx.org>
References: <e94c00ba41258efee577167cad3d92ee.NginxMailingListEnglish@forum.nginx.org>
	<cffdb7d206c70dd854655837507d00ea.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <201111240017.28573.ne@vbart.ru>

On Wednesday 23 November 2011 23:54:06 EricTheRed03 wrote:
> This is the block of code I've been working on for hours ... :(
> 
> --------------- nginx.conf ------------
>    server {
>             listen   80;
>             server_name sitea.com;
>             rewrite ^(/hiddendevpath)(.*)$ http://sitea.com/ ;
>             rewrite ^(.*) https://sitea.com
>                }
> 
> ------------- testing below ----------
> 
> sudo /usr/sbin/nginx -t
> Error: nginx: [emerg] invalid number of arguments in "rewrite" directive
> in /etc/nginx/nginx.conf:44

You forgot a ; at the end of this directive:
rewrite ^(.*) https://sitea.com


Try this:

  server {

    listen  80;
    server_name sitea.com;

    location /hiddendevpath {
      rewrite  ^  https://sitea.com/;
    }

  }

wbr, Valentin V. Bartenev


From gelonida at gmail.com  Wed Nov 23 22:23:50 2011
From: gelonida at gmail.com (Gelonida N)
Date: Wed, 23 Nov 2011 23:23:50 +0100
Subject: same config for two virtual hosts (except hostname / certificate)
Message-ID: <jajrpm$cn$1@dough.gmane.org>

Hi,

I'm still a little shaky with how to best structure my nginx
configurations and where one can / should use includes and where one
couldn't.

somehow I like the idea of a directory structure like

/etc/nginx/  # for basic files / certificates / . . .
/etc/nginx/sites-available # the available configs
/etc/nginx/sites-enabled  # the sites, that shall really be started


Now I have two sites (ssl which should share exactly the same config
(except the host name and the server certificate)
the first one would be for example server_1.mysite.net
the other server_2.mysite.net:4433


Of course both would need their own certificate and host name.

What would be the recommended way for such a setup?

thanks in advance for any feedback


From appa at perusio.net  Wed Nov 23 22:51:10 2011
From: appa at perusio.net (=?UTF-8?B?QW50w7NuaW8=?= P. P. Almeida)
Date: Wed, 23 Nov 2011 22:51:10 +0000
Subject: same config for two virtual hosts (except hostname / certificate)
In-Reply-To: <jajrpm$cn$1@dough.gmane.org>
References: <jajrpm$cn$1@dough.gmane.org>
Message-ID: <87k46qo3kx.wl%appa@perusio.net>

On 23 Nov 2011 22h23 WET, gelonida at gmail.com wrote:

> Hi,
>
> I'm still a little shaky with how to best structure my nginx
> configurations and where one can / should use includes and where one
> couldn't.
>
> somehow I like the idea of a directory structure like
>
> /etc/nginx/  # for basic files / certificates / . . .
> /etc/nginx/sites-available # the available configs
> /etc/nginx/sites-enabled  # the sites, that shall really be started
>
>
> Now I have two sites (ssl which should share exactly the same config
> (except the host name and the server certificate)
> the first one would be for example server_1.mysite.net
> the other server_2.mysite.net:4433
>
>
> Of course both would need their own certificate and host name.
>
> What would be the recommended way for such a setup?
>

My suggestion is for you to place the common things in both vhosts in
a file, e.g., my_common_config.conf and do an include in each vhost:

server_1.mysite.net.conf
server_2.mysite.net.conf

You can try this script for enabling each site and disabling using the
sites_enabled/sites-available layout.

https://github.com/perusio/nginx_ensite

--- appa


From gelonida at gmail.com  Wed Nov 23 23:21:09 2011
From: gelonida at gmail.com (Gelonida N)
Date: Thu, 24 Nov 2011 00:21:09 +0100
Subject: same config for two virtual hosts (except hostname / certificate)
In-Reply-To: <87k46qo3kx.wl%appa@perusio.net>
References: <jajrpm$cn$1@dough.gmane.org> <87k46qo3kx.wl%appa@perusio.net>
Message-ID: <jajv55$n97$1@dough.gmane.org>

Thanks for your reply,

On 11/23/2011 11:51 PM, Ant?nio P. P. Almeida wrote:
> On 23 Nov 2011 22h23 WET, gelonida at gmail.com wrote:
>> somehow I like the idea of a directory structure like
. . .
>> /etc/nginx/  # for basic files / certificates / . . .
>> /etc/nginx/sites-available # the available configs
>> /etc/nginx/sites-enabled  # the sites, that shall really be started
>>
>>
>> Now I have two sites (ssl which should share exactly the same config
>> (except the host name and the server certificate)
>> the first one would be for example server_1.mysite.net
>> the other server_2.mysite.net:4433
>>
. . .
>>
>> What would be the recommended way for such a setup?
>>
> 
> My suggestion is for you to place the common things in both vhosts in
> a file, e.g., my_common_config.conf and do an include in each vhost:
> 
> server_1.mysite.net.conf
> server_2.mysite.net.conf

Where would you place my_common_config.conf?
would you clutter /etc/nginx with it or create a subdir like
/etf/nginx/sites-baseconfig
would you then just do an include with a relative path like
include ../my_common_config.conf ?


> 
> You can try this script for enabling each site and disabling using the
> sites_enabled/sites-available layout.
> 
> https://github.com/perusio/nginx_ensite

That looks nice :-) WIll try it.




From appa at perusio.net  Wed Nov 23 23:30:03 2011
From: appa at perusio.net (=?UTF-8?B?QW50w7NuaW8=?= P. P. Almeida)
Date: Wed, 23 Nov 2011 23:30:03 +0000
Subject: same config for two virtual hosts (except hostname / certificate)
In-Reply-To: <jajv55$n97$1@dough.gmane.org>
References: <jajrpm$cn$1@dough.gmane.org> <87k46qo3kx.wl%appa@perusio.net>
	<jajv55$n97$1@dough.gmane.org>
Message-ID: <87hb1uo1s4.wl%appa@perusio.net>

On 23 Nov 2011 23h21 WET, gelonida at gmail.com wrote:

> Thanks for your reply,
>
> On 11/23/2011 11:51 PM, Ant?nio P. P. Almeida wrote:
>> On 23 Nov 2011 22h23 WET, gelonida at gmail.com wrote:
>>> somehow I like the idea of a directory structure like
> . . .
>>> /etc/nginx/ # for basic files / certificates / . . .
>>> /etc/nginx/sites-available # the available configs
>>> /etc/nginx/sites-enabled # the sites, that shall really be started
>>>
>>>
>>> Now I have two sites (ssl which should share exactly the same
>>> config (except the host name and the server certificate) the first
>>> one would be for example server_1.mysite.net the other
>>> server_2.mysite.net:4433
>>>
> . . .
>>>
>>> What would be the recommended way for such a setup?
>>>
>>
>> My suggestion is for you to place the common things in both vhosts
>> in a file, e.g., my_common_config.conf and do an include in each
>> vhost:
>>
>> server_1.mysite.net.conf
>> server_2.mysite.net.conf
>
> Where would you place my_common_config.conf?
> would you clutter /etc/nginx with it or create a subdir like
> /etf/nginx/sites-baseconfig
> would you then just do an include with a relative path like
> include ../my_common_config.conf ?

I prefer to have it under the sites-available directory. Because is
something specific to one or more of the enabled sites.

There's in Debian a /etc/nginx/conf.d dir but I never use it. I put
all generic stuff in /etc/nginx and site specific stuff under
/etc/nginx/sites-available.

I always name the vhost config files with names that identify it
explicitly, .e.g, www.example.com.conf. This way there's no risk in
confusing vhost config with other files that have directives that can
be used in several vhosts.

--- appa


From gelonida at gmail.com  Wed Nov 23 23:35:42 2011
From: gelonida at gmail.com (Gelonida N)
Date: Thu, 24 Nov 2011 00:35:42 +0100
Subject: same config for two virtual hosts (except hostname / certificate)
In-Reply-To: <87hb1uo1s4.wl%appa@perusio.net>
References: <jajrpm$cn$1@dough.gmane.org>
	<87k46qo3kx.wl%appa@perusio.net>	<jajv55$n97$1@dough.gmane.org>
	<87hb1uo1s4.wl%appa@perusio.net>
Message-ID: <jak00f$trb$1@dough.gmane.org>

On 11/24/2011 12:30 AM, Ant?nio P. P. Almeida wrote:
> On 23 Nov 2011 23h21 WET, gelonida at gmail.com wrote:
>>>
>>> My suggestion is for you to place the common things in both vhosts
>>> in a file, e.g., my_common_config.conf and do an include in each
>>> vhost:
>>>
>>> server_1.mysite.net.conf
>>> server_2.mysite.net.conf
>>
>> Where would you place my_common_config.conf?
>> would you clutter /etc/nginx with it or create a subdir like
>> /etf/nginx/sites-baseconfig
>> would you then just do an include with a relative path like
>> include ../my_common_config.conf ?
> 
> I prefer to have it under the sites-available directory. Because is
> something specific to one or more of the enabled sites.
> 
> There's in Debian a /etc/nginx/conf.d dir but I never use it. I put
> all generic stuff in /etc/nginx and site specific stuff under
> /etc/nginx/sites-available.
> 
> I always name the vhost config files with names that identify it
> explicitly, .e.g, www.example.com.conf. This way there's no risk in
> confusing vhost config with other files that have directives that can
> be used in several vhosts.
> 
Yes combined with your naming style this should be clear enough.
I assume you will just symlink the vhosts file to sites available.
and the include will thus be something like
include ../sites-available/common_settings.conf, right?



From appa at perusio.net  Wed Nov 23 23:43:01 2011
From: appa at perusio.net (=?UTF-8?B?QW50w7NuaW8=?= P. P. Almeida)
Date: Wed, 23 Nov 2011 23:43:01 +0000
Subject: same config for two virtual hosts (except hostname / certificate)
In-Reply-To: <jak00f$trb$1@dough.gmane.org>
References: <jajrpm$cn$1@dough.gmane.org> <87k46qo3kx.wl%appa@perusio.net>
	<jajv55$n97$1@dough.gmane.org> <87hb1uo1s4.wl%appa@perusio.net>
	<jak00f$trb$1@dough.gmane.org>
Message-ID: <87fwheo16i.wl%appa@perusio.net>

On 23 Nov 2011 23h35 WET, gelonida at gmail.com wrote:

> On 11/24/2011 12:30 AM, Ant?nio P. P. Almeida wrote:
>> On 23 Nov 2011 23h21 WET, gelonida at gmail.com wrote:
>>>>
>>>> My suggestion is for you to place the common things in both
>>>> vhosts in a file, e.g., my_common_config.conf and do an include
>>>> in each vhost:
>>>>
>>>> server_1.mysite.net.conf
>>>> server_2.mysite.net.conf
>>>
>>> Where would you place my_common_config.conf?
>>> would you clutter /etc/nginx with it or create a subdir like
>>> /etf/nginx/sites-baseconfig
>>> would you then just do an include with a relative path like
>>> include ../my_common_config.conf ?
>>
>> I prefer to have it under the sites-available directory. Because is
>> something specific to one or more of the enabled sites.
>>
>> There's in Debian a /etc/nginx/conf.d dir but I never use it. I put
>> all generic stuff in /etc/nginx and site specific stuff under
>> /etc/nginx/sites-available.
>>
>> I always name the vhost config files with names that identify it
>> explicitly, .e.g, www.example.com.conf. This way there's no risk in
>> confusing vhost config with other files that have directives that
>> can be used in several vhosts.
>>
> Yes combined with your naming style this should be clear enough.
> I assume you will just symlink the vhosts file to sites available.
> and the include will thus be something like
> include ../sites-available/common_settings.conf, right?

Nope. I do nginx_ensite www.example.com.conf 

This creates a symlink:

ls -la /etc/sites-enabled

www.example.com.conf -->
/etc/nginx/sites-available/www.example.com.conf

Of course you must have a line like:

include /etc/nginx/sites-enabled/*; 

in your nginx.conf

The common_settings.conf is included in the www.example.com.conf file:

include sites-available/common_settings.conf;

It only makes sense to symlink vhost configs.

--- appa


From nginx-forum at nginx.us  Thu Nov 24 01:54:56 2011
From: nginx-forum at nginx.us (balrog2000)
Date: Wed, 23 Nov 2011 20:54:56 -0500
Subject: Nginx 0.7.67 + Apache how to forbid responses with Set-Cookie being
	stored
Message-ID: <bfed6b8d5eb3d39115e7adb6aff4fe6b.NginxMailingListEnglish@forum.nginx.org>

Hi,

I am caching apache with nginx. Application is developed by several
people and consists of many modules.
I would like to be sure that if by any case application sets Set-Cookie
in the response and sets X-Accel-Expires > 0 it would not get cached
anyway.

Can i any how use proxy_no_cache with $upstream_http_ variables?
Can I use apache mod_headers to append or modify specific header only in
case set-cookie occurs in the response?


It is easy by looking at the request headers, but have no luck finding
solution for response headers. 
My general idea is to have 100% certainty that responses with Set-Cookie
does not get cached.

0.7.67 is not negotiable as it is debian stable in production.
Apache version 2.2.x

Thanks in advance, Tomasz Kraus
Regards

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218910,218910#msg-218910


From bangzhong at gmail.com  Thu Nov 24 05:04:18 2011
From: bangzhong at gmail.com (Chen Bangzhong)
Date: Thu, 24 Nov 2011 13:04:18 +0800
Subject: how to configure the expires header if my static file is served by
	resin?
Message-ID: <CAK6WCizrS6biJ5hYZNdA3-Pa8BLkNO6Y8C0maksRoCP3TyBnQw@mail.gmail.com>

Hello, all

I use nginx as a reverse proxy and all traffic goes to resin. The
css/js files are also hosted on resin.

Now I want to add the expires directives to css/js, is the following
configuration correct?

location ~ .*\.(css|js)$
      {
          include proxy_params;
          proxy_pass http://back_end;
          #access_log        off;
          expires           6h;
      }

      location / {
          include proxy_params;
          proxy_pass http://back_end;
      }

thanks.


From simone.fumagalli at contactlab.com  Thu Nov 24 08:50:10 2011
From: simone.fumagalli at contactlab.com (Simone Fumagalli)
Date: Thu, 24 Nov 2011 09:50:10 +0100
Subject: Nginx Cache and WP canonical URL create infinite loop (bug ?)
In-Reply-To: <20111123174123.GY95664@mdounin.ru>
References: <4ECD1ECE.4060701@contactlab.com>	<20111123170424.GX95664@mdounin.ru>
	<4ECD294D.90108@contactlab.com> <20111123174123.GY95664@mdounin.ru>
Message-ID: <4ECE0542.1000202@contactlab.com>

On 11/23/2011 06:41 PM, Maxim Dounin wrote:
> With this config, the cache key for the request to
> "http://www.mysite.com/my-page//" will be
> "http://ALL_backend/my-page//".

Well, I was wrong. In another file, on main level, I've

	proxy_cache_key "$scheme://$host$request_uri";

--
Simone


From igor at sysoev.ru  Thu Nov 24 10:07:33 2011
From: igor at sysoev.ru (Igor Sysoev)
Date: Thu, 24 Nov 2011 14:07:33 +0400
Subject: Simple rewrite (redirection)
In-Reply-To: <e94c00ba41258efee577167cad3d92ee.NginxMailingListEnglish@forum.nginx.org>
References: <e94c00ba41258efee577167cad3d92ee.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20111124100733.GG70072@nginx.com>

On Wed, Nov 23, 2011 at 02:46:13PM -0500, EricTheRed03 wrote:
> Hi Guys,
> 
> In my HTTP config section in the config file, I would like to implement
> a redirection based on this scenario:
> 
> If the URL is not "http://sitea.com/exactmatch" then redirect to the
> "https" home. else go to our hidden url (our testing boxes)
> 
> #logic would be somthing like this..
> If (The URL is anything but "/ahiddenurl") 
> {
>         go to "https://sitea.com"
> }
> else 
> {
>       go to "http://hiddenurl"
> }
> 
> Is this possible in the nginx.conf file?? Any help would be so very much
> appreciated!

server {
    listen       80;
    server_name  sitea.com;

    location / {
        return https://sitea.com;
    }

    location /hiddenurl {
        ...
    }
}


-- 
Igor Sysoev


From nginx-forum at nginx.us  Thu Nov 24 10:16:04 2011
From: nginx-forum at nginx.us (higkoo)
Date: Thu, 24 Nov 2011 05:16:04 -0500
Subject: Nginx Can't get the request port when doing proxy
Message-ID: <cb92dadb23d2455581baaa7016545ade.NginxMailingListEnglish@forum.nginx.org>

such as :
    Client(192.168.1.101) want visit http://192.168.1.201:9000/myfile
example,and need visit other ports.
    but client must use proxy to visit. Nginx is the proxyer!

    Nginx proxy running on 192.168.1.202 and listen at  port 8080 .
        I use '''proxy_pass $scheme://$host$request_uri;''' to make
proxy,but I get http-502 when I use this proxy-server.
    Becase nginx use the default port 80 to proxy:
        | request = http://192.168.1.201:9000/myfile
        | proxy-server = 192.168.1.202 8080
        | proxy-to  = http://192.168.1.201:80/myfile       
    There are only $remote_port and $server_port on nginx-variables:
http://wiki.nginx.org/HttpCoreModule#Variables

    If I just want visit port 9000 on 192.168.1.201,I can write by
'''proxy_pass $scheme://$host:9000$request_uri;'''
    But it's not a good idea ,and there are some other port to use.

    How can I do ?

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218934,218934#msg-218934


From mdounin at mdounin.ru  Thu Nov 24 10:17:44 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Thu, 24 Nov 2011 14:17:44 +0400
Subject: Nginx Cache and WP canonical URL create infinite loop (bug ?)
In-Reply-To: <4ECE0542.1000202@contactlab.com>
References: <4ECD1ECE.4060701@contactlab.com>
	<20111123170424.GX95664@mdounin.ru> <4ECD294D.90108@contactlab.com>
	<20111123174123.GY95664@mdounin.ru>
	<4ECE0542.1000202@contactlab.com>
Message-ID: <20111124101744.GA95664@mdounin.ru>

Hello!

On Thu, Nov 24, 2011 at 09:50:10AM +0100, Simone Fumagalli wrote:

> On 11/23/2011 06:41 PM, Maxim Dounin wrote:
> >With this config, the cache key for the request to
> >"http://www.mysite.com/my-page//" will be
> >"http://ALL_backend/my-page//".
> 
> Well, I was wrong. In another file, on main level, I've
> 
> 	proxy_cache_key "$scheme://$host$request_uri";

With this proxy_cache_key the key (for the same request) will be 
"http://www.mysite.com/my-page//".

Maxim Dounin


From mdounin at mdounin.ru  Thu Nov 24 10:27:34 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Thu, 24 Nov 2011 14:27:34 +0400
Subject: Nginx Can't get the request port when doing proxy
In-Reply-To: <cb92dadb23d2455581baaa7016545ade.NginxMailingListEnglish@forum.nginx.org>
References: <cb92dadb23d2455581baaa7016545ade.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20111124102734.GC95664@mdounin.ru>

Hello!

On Thu, Nov 24, 2011 at 05:16:04AM -0500, higkoo wrote:

> such as :
>     Client(192.168.1.101) want visit http://192.168.1.201:9000/myfile
> example,and need visit other ports.
>     but client must use proxy to visit. Nginx is the proxyer!
> 
>     Nginx proxy running on 192.168.1.202 and listen at  port 8080 .
>         I use '''proxy_pass $scheme://$host$request_uri;''' to make
> proxy,but I get http-502 when I use this proxy-server.
>     Becase nginx use the default port 80 to proxy:
>         | request = http://192.168.1.201:9000/myfile
>         | proxy-server = 192.168.1.202 8080
>         | proxy-to  = http://192.168.1.201:80/myfile       
>     There are only $remote_port and $server_port on nginx-variables:
> http://wiki.nginx.org/HttpCoreModule#Variables
> 
>     If I just want visit port 9000 on 192.168.1.201,I can write by
> '''proxy_pass $scheme://$host:9000$request_uri;'''
>     But it's not a good idea ,and there are some other port to use.
> 
>     How can I do ?

Just a side note:

nginx isn't forward proxy, it's reverse proxy.  By using nginx as 
a forward proxy you are looking for problems, including security 
ones.

Maxim Dounin


From simone.fumagalli at contactlab.com  Thu Nov 24 13:34:59 2011
From: simone.fumagalli at contactlab.com (Simone Fumagalli)
Date: Thu, 24 Nov 2011 14:34:59 +0100
Subject: Nginx Cache and WP canonical URL create infinite loop (bug ?)
In-Reply-To: <20111124101744.GA95664@mdounin.ru>
References: <4ECD1ECE.4060701@contactlab.com>	<20111123170424.GX95664@mdounin.ru>
	<4ECD294D.90108@contactlab.com>	<20111123174123.GY95664@mdounin.ru>	<4ECE0542.1000202@contactlab.com>
	<20111124101744.GA95664@mdounin.ru>
Message-ID: <4ECE4803.4090101@contactlab.com>

On 11/24/2011 11:17 AM, Maxim Dounin wrote:
> With this proxy_cache_key the key (for the same request) will be
> "http://www.mysite.com/my-page//".

Well, actually when I require http://www.mysite.com/my-page// the key I've in the cache is http://www.mysite.com/my-page/
I'm on NGINX 0.8.54 and I've installed "Cache purge" module http://labs.frickle.com/nginx_ngx_cache_purge/

Also when I require a URL like http://www.mysite.com/my-page/###### in the cache the KEY is http://www.mysite.com/my-page/

If I've an entry in the cache with the key http://www.mysite.com/my-page/ and I require http://www.mysite.com/my-page/###### the request hit the cache 

Regards

--
Simone


From delta.yeh at gmail.com  Fri Nov 25 03:03:39 2011
From: delta.yeh at gmail.com (Delta Yeh)
Date: Fri, 25 Nov 2011 11:03:39 +0800
Subject: nginx-1.1.8
In-Reply-To: <20111114154403.GF33940@nginx.com>
References: <20111114154403.GF33940@nginx.com>
Message-ID: <CADKMG50iPH+H3S5Kefi3rTEkFL5pDTT6H0NxqB1GLJcJQq596Q@mail.gmail.com>

2011/11/14 Igor Sysoev <igor at sysoev.ru>:
> Changes with nginx 1.1.8 ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? 14 Nov 2011
>
> ? ?*) Change: the ngx_http_limit_zone_module was renamed to the
> ? ? ? ngx_http_limit_conn_module.
>
> ? ?*) Change: the "limit_zone" directive was superseded by the
> ? ? ? "limit_conn_zone" directive with a new syntax.
>
> ? ?*) Feature: support for multiple "limit_conn" limits on the same level.
>

 Usually we use   limit_zone module like:
http{

 limit_zone   one  $binary_remote_addr  10m;
server
{
   location /mypage{
      limit_conn   one  number_of connections;
  }
}

What is the  scenario that we should use multiple limit_conn ?

Is there a way to limit total connections to one location ?


From tseveendorj at gmail.com  Fri Nov 25 05:19:41 2011
From: tseveendorj at gmail.com (=?UTF-8?B?0KbRjdCy0Y3RjdC90LTQvtGA0LY=?=)
Date: Fri, 25 Nov 2011 13:19:41 +0800
Subject: Rewrite translation apache to nginx
Message-ID: <4ECF256D.5090604@gmail.com>


Hello,

I have always difficulty with rewrite translating apache to nginx.

<IfModule mod_rewrite.c>
         RewriteEngine On
         #
         #  Full path to your site
         #
         RewriteBase /
         # Rename Photo Names
         RewriteRule 
^file/pic/photo/([0-9]+)/([0-9]+)/([A-Za-z0-9]{32}+)\-(.*?)([_0-9]*?)\.(.*)$     
file/pic/photo/$1/$2/$3$5.$6
         #
         #  Rules
         #
         RewriteCond %{REQUEST_FILENAME} !-f
         RewriteCond %{REQUEST_FILENAME} !-d
         RewriteRule ^(.*) index.php?do=/$1
         # Image Access Protection
         RewriteRule ^file/pic/photo/(.*)\.(.*)$  
static/image.php?file=$1&ext=$2
</IfModule>


It is my translation

location / {
rewrite 
^file/pic/photo/([0-9]+)/([0-9]+)/([A-Za-z0-9]{32}+)\-(.*?)([_0-9]*?)\.(.*)$       
file/pic/photo/$1/$2/$3$5.$6 break;

if (!-f $request_filename ) {
break;
}

if (!-d $request_filename) {
break;
}

rewrite ^(.*)$ index.php?do=/$1
}


How to convert it properly.

Thanks
Tseveen.

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20111125/dc232826/attachment.html>

From nginx-forum at nginx.us  Fri Nov 25 05:53:11 2011
From: nginx-forum at nginx.us (higkoo)
Date: Fri, 25 Nov 2011 00:53:11 -0500
Subject: Nginx Can't get the request port when doing proxy
In-Reply-To: <20111124102734.GC95664@mdounin.ru>
References: <20111124102734.GC95664@mdounin.ru>
Message-ID: <d59acec3f80f776c95839e189e3df739.NginxMailingListEnglish@forum.nginx.org>

Thanks , Maxim Dounin.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218934,218999#msg-218999


From edho at myconan.net  Fri Nov 25 06:17:24 2011
From: edho at myconan.net (Edho Arief)
Date: Fri, 25 Nov 2011 13:17:24 +0700
Subject: Rewrite translation apache to nginx
In-Reply-To: <4ECF256D.5090604@gmail.com>
References: <4ECF256D.5090604@gmail.com>
Message-ID: <CABTjkKm=MpVE3gAjFvc2gy9=0xJ9n-_rQkzqtv7wqVRkZzfYYQ@mail.gmail.com>

2011/11/25 ?????????? <tseveendorj at gmail.com>:
>
> Hello,
>
> I have always difficulty with rewrite translating apache to nginx.
>
> <IfModule mod_rewrite.c>
> ??????? RewriteEngine On
> ??????? #
> ??????? #? Full path to your site
> ??????? #
> ??????? RewriteBase /
> ??????? # Rename Photo Names
> ??????? RewriteRule
> ^file/pic/photo/([0-9]+)/([0-9]+)/([A-Za-z0-9]{32}+)\-(.*?)([_0-9]*?)\.(.*)$
> file/pic/photo/$1/$2/$3$5.$6

rewrite ^/file/pic/photo/([0-9]+)/([0-9]+)/([A-Za-z0-9]{32}+)\-(.*?)([_0-9]*?)\.(.*)$
/file/pic/photo/$1/$2/$3$5.$6;


> ??????? #
> ??????? #? Rules
> ??????? #
> ??????? RewriteCond %{REQUEST_FILENAME} !-f
> ??????? RewriteCond %{REQUEST_FILENAME} !-d
> ??????? RewriteRule ^(.*) index.php?do=/$1

try_files $uri $uri/ /index.php?do=$uri;

> ??????? # Image Access Protection
> ??????? RewriteRule ^file/pic/photo/(.*)\.(.*)$
> static/image.php?file=$1&ext=$2

rewrite ^/file/pic/photo/(.*)\.(.*)$ /static/image.php?file=$1&ext=$2




-- 
O< ascii ribbon campaign - stop html mail - www.asciiribbon.org

From davide.damico at contactlab.com  Fri Nov 25 09:22:05 2011
From: davide.damico at contactlab.com (Davide D'Amico)
Date: Fri, 25 Nov 2011 10:22:05 +0100
Subject: Nginx Cache and WP canonical URL create infinite loop
In-Reply-To: <mailman.1840.1322130455.3032.nginx@nginx.org>
References: <mailman.1840.1322130455.3032.nginx@nginx.org>
Message-ID: <2D1AD7DF-8284-437E-8CF3-649EB92A17CD@contactlab.com>

> 
> ------------------------------
> 
> Message: 9
> Date: Thu, 24 Nov 2011 14:17:44 +0400
> From: Maxim Dounin <mdounin at mdounin.ru>
> To: nginx at nginx.org
> Subject: Re: Nginx Cache and WP canonical URL create infinite loop
> 	(bug ?)
> Message-ID: <20111124101744.GA95664 at mdounin.ru>
> Content-Type: text/plain; charset=us-ascii
> 
> Hello!
> 
> On Thu, Nov 24, 2011 at 09:50:10AM +0100, Simone Fumagalli wrote:
> 
>> On 11/23/2011 06:41 PM, Maxim Dounin wrote:
>>> With this config, the cache key for the request to
>>> "http://www.mysite.com/my-page//" will be
>>> "http://ALL_backend/my-page//".
>> 
>> Well, I was wrong. In another file, on main level, I've
>> 
>> 	proxy_cache_key "$scheme://$host$request_uri";
> 
> With this proxy_cache_key the key (for the same request) will be 
> "http://www.mysite.com/my-page//".
> 
> Maxim Dounin
> 

Hi, this is the situation:
dave at therock:~> uname -a
FreeBSD therock.office.lan 8.2-STABLE FreeBSD 8.2-STABLE #0: Sat Aug 27 08:47:51 CEST 2011     root at therock.office.lan:/usr/obj/usr/src/sys/THEROCK  amd64
dave at therock:~> pkg_info | grep -i nginx
nginx-1.0.10_1,1    Robust and small WWW server
dave at therock:/usr/ports/www/nginx> sudo make showconfig | grep '=on'
     HTTP_MODULE=on "Enable HTTP module"
     HTTP_CACHE_MODULE=on "Enable http_cache module"
     HTTP_GEOIP_MODULE=on "Enable http_geoip module"
     HTTP_REWRITE_MODULE=on "Enable http_rewrite module"
     HTTP_SSL_MODULE=on "Enable http_ssl module"
     HTTP_STATUS_MODULE=on "Enable http_stub_status module"
     CACHE_PURGE_MODULE=on "3rd party cache_purge module"
     HEADERS_MORE_MODULE=on "3rd party headers_more module"
dave at therock:/usr/ports/www/nginx> 

proxy_cache_path /usr/local/www/cache/myhost levels=1:2 keys_zone=MYHOST:125m inactive=24h max_size=2g;

server {
  listen 80;
  access_log    /var/log/nginx/myhost-test-access.log upstream;
  error_log     /var/log/nginx/myhost-test-error.log;
  server_name   myhost.foo.bar;

  proxy_redirect     off;
  proxy_set_header   Host             $host;
  proxy_set_header   X-Real-IP        $remote_addr;
  proxy_set_header   X-Forwarded-For  $proxy_add_x_forwarded_for;
  proxy_store_access   user:rw  group:rw  all:r;
  proxy_max_temp_file_size   0;
  proxy_cache_key "$scheme://$host$request_uri";
  proxy_intercept_errors on;
  
  proxy_ignore_headers "Expires" "X-Accel-Expires" "Cache-Control";
  proxy_connect_timeout      10;
 
  location ~ /purge(/.*) {
    allow              127.0.0.1;
    allow              172.16.16.189;
    deny               all;
    proxy_cache_purge  MYHOST "$scheme://$host$1$is_args$args";
  }


  location / {
    proxy_cache MYHOST;
    proxy_cache_valid 200 1h;
    proxy_cache_valid 404 30m;
    proxy_cache_use_stale error timeout invalid_header updating http_500 http_502 http_503 http_504;
    proxy_pass http://172.16.16.227:80;
  }
}

then I'm requesting the following url (2 times):
[dave at ext ~]$ wget "http://myhost.foo.bar/tomato//#"
--2011-11-25 10:18:20--  http://myhost.foo.bar/tomato//
[?]

100%[======================================>] 10          --.-K/s   in 0s      

2011-11-25 10:18:20 (778 KB/s) - `index.html' saved [10/10]

[dave at ext ~]$ 

In the cache folder I see:
root at therock:/usr/local/www/cache/myhost# more 9/75/2d047ea530c6f984d115c0b9fd23c759 
Bk<CF>N^@^@^@^@<FF><FF><FF><FF><FF><FF><FF><FF>2]<CF>N^@^@^@^@M<A0>A<91>^@^@Z^@i^A^@^@^@^@^@^@
KEY: http://myhost.foo.bar/tomato//
HTTP/1.1 200 OK
Date: Fri, 25 Nov 2011 09:17:38 GMT
Server: Apache/2.2.21 (FreeBSD) mod_ssl/2.2.21 OpenSSL/0.9.8q DAV/2 PHP/5.3.8 with Suhosin-Patch mod_perl/2.0.5 Perl/v5.12.4
X-Powered-By: PHP/5.3.8
Content-Length: 10
Connection: close
Content-Type: text/html

1322212658
root at therock:/usr/local/www/cache/myhost# 

And it seems that the key has a wrong value, isn't?

Thanks in advance,
d.

From mdounin at mdounin.ru  Fri Nov 25 10:00:24 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Fri, 25 Nov 2011 14:00:24 +0400
Subject: nginx-1.1.8
In-Reply-To: <CADKMG50iPH+H3S5Kefi3rTEkFL5pDTT6H0NxqB1GLJcJQq596Q@mail.gmail.com>
References: <20111114154403.GF33940@nginx.com>
	<CADKMG50iPH+H3S5Kefi3rTEkFL5pDTT6H0NxqB1GLJcJQq596Q@mail.gmail.com>
Message-ID: <20111125100024.GH95664@mdounin.ru>

Hello!

On Fri, Nov 25, 2011 at 11:03:39AM +0800, Delta Yeh wrote:

> 2011/11/14 Igor Sysoev <igor at sysoev.ru>:
> > Changes with nginx 1.1.8 ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? 14 Nov 2011
> >
> > ? ?*) Change: the ngx_http_limit_zone_module was renamed to the
> > ? ? ? ngx_http_limit_conn_module.
> >
> > ? ?*) Change: the "limit_zone" directive was superseded by the
> > ? ? ? "limit_conn_zone" directive with a new syntax.
> >
> > ? ?*) Feature: support for multiple "limit_conn" limits on the same level.
> >
> 
>  Usually we use   limit_zone module like:
> http{
> 
>  limit_zone   one  $binary_remote_addr  10m;
> server
> {
>    location /mypage{
>       limit_conn   one  number_of connections;
>   }
> }
> 
> What is the  scenario that we should use multiple limit_conn ?

E.g. if you want to limit number of connections per ip *and* total 
number of connections per virtual host.

    limit_conn_zone $binary_remote_addr zone=perip:10m;
    limit_conn_zone $server_name zone=perserver:10m;

    server {
        ...
        limit_conn perip 10;
        limit_conn perserver 100;
    }

> Is there a way to limit total connections to one location ?

Yes.

Maxim Dounin


From delta.yeh at gmail.com  Fri Nov 25 10:50:25 2011
From: delta.yeh at gmail.com (Delta Yeh)
Date: Fri, 25 Nov 2011 18:50:25 +0800
Subject: nginx-1.1.8
In-Reply-To: <20111125100024.GH95664@mdounin.ru>
References: <20111114154403.GF33940@nginx.com>
	<CADKMG50iPH+H3S5Kefi3rTEkFL5pDTT6H0NxqB1GLJcJQq596Q@mail.gmail.com>
	<20111125100024.GH95664@mdounin.ru>
Message-ID: <CADKMG52M+4kJ-h1nO3mno-VozH8PpBdnh5xruKL7XvOuQOxVXA@mail.gmail.com>

So if I want to limit total connection to one uri, I can do it by:


limit_conn_zone $host$uri zone=peruri:10m;


    server {
        ...
        location / {
            limit_conn peruri 10;
      }

    }


but because uri is

2011/11/25 Maxim Dounin <mdounin at mdounin.ru>:
> Hello!
>
> On Fri, Nov 25, 2011 at 11:03:39AM +0800, Delta Yeh wrote:
>
>> 2011/11/14 Igor Sysoev <igor at sysoev.ru>:
>> > Changes with nginx 1.1.8 ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? 14 Nov 2011
>> >
>> > ? ?*) Change: the ngx_http_limit_zone_module was renamed to the
>> > ? ? ? ngx_http_limit_conn_module.
>> >
>> > ? ?*) Change: the "limit_zone" directive was superseded by the
>> > ? ? ? "limit_conn_zone" directive with a new syntax.
>> >
>> > ? ?*) Feature: support for multiple "limit_conn" limits on the same level.
>> >
>>
>> ?Usually we use ? limit_zone module like:
>> http{
>>
>> ?limit_zone ? one ?$binary_remote_addr ?10m;
>> server
>> {
>> ? ?location /mypage{
>> ? ? ? limit_conn ? one ?number_of connections;
>> ? }
>> }
>>
>> What is the ?scenario that we should use multiple limit_conn ?
>
> E.g. if you want to limit number of connections per ip *and* total
> number of connections per virtual host.
>
> ? ?limit_conn_zone $binary_remote_addr zone=perip:10m;
> ? ?limit_conn_zone $server_name zone=perserver:10m;
>
> ? ?server {
> ? ? ? ?...
> ? ? ? ?limit_conn perip 10;
> ? ? ? ?limit_conn perserver 100;
> ? ?}
>
>> Is there a way to limit total connections to one location ?
>
> Yes.
>
> Maxim Dounin
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx


From rainer at ultra-secure.de  Fri Nov 25 12:03:44 2011
From: rainer at ultra-secure.de (Rainer Duffner)
Date: Fri, 25 Nov 2011 13:03:44 +0100
Subject: Fw: [rt-users] New extension: advanced setup of RT with Nginx
Message-ID: <20111125130344.7f78c760@suse2.iptech.internal>

FYI:

Great news, IMO.





Beginn der weitergeleiteten Nachricht:

Datum: Fri, 25 Nov 2011 15:53:57 +0400
Von: Ruslan Zakirov <ruz at bestpractical.com>
An: RT users <RT-Users at lists.bestpractical.com>
Betreff: [rt-users] New extension: advanced setup of RT with Nginx


Hi,
Recently I spent some time on configuring nginx for RT to serve as
much aspossible by web server, lower memory footprint, make RT more
responsive andincrease number of concurrent requests. At the end this
effort resulted in a newextension RT-Extension-Nginx that comes with
configuration generator andscript to start/stop/restart nginx and fcgi
servers.
http://search.cpan.org/dist/RT-Extension-Nginx/lib/RT/Extension/Nginx.pm
Try, enjoy and send feedback.
-- 
Best regards, Ruslan.
--------
RT Training Sessions (http://bestpractical.com/services/training.html)
*  Barcelona, Spain  November 28 & 29, 2011


-- 
-- 
Everyware AG
Rainer Duffner
CISSP, LPI, MCSE
Zurlindenstrasse 52a
CH-8003 Zurich
Telefon +41 44 466 60 25
Telefax +41 44 466 60 10


From nginx-forum at nginx.us  Fri Nov 25 12:49:38 2011
From: nginx-forum at nginx.us (Fleshgrinder)
Date: Fri, 25 Nov 2011 07:49:38 -0500
Subject: nginx 1.x.x segfault at e8 ip xxx sp xxx error 4 in libssl.so.1.0.0
Message-ID: <74e368a93cf35d43c926660b4ddf733a.NginxMailingListEnglish@forum.nginx.org>

nginx -V output:

nginx version: nginx/1.1.8
built by gcc 4.6.2 (Debian 4.6.2-5)
TLS SNI support enabled
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=/build/buildd-nginx_1.1.8-2-i386-6PKxRV/nginx-1.1.8/debian/modules/nginx-echo
--add-module=/build/buildd-nginx_1.1.8-2-i386-6PKxRV/nginx-1.1.8/debian/modules/nginx-upstream-fair

This is the nginx-full package from the Debian unstable branch. I also
had the same problem with nginx-full from dotdeb (1.0.09 and 1.0.10).
The same problem also occurs with libssl.0.9.8.

My certificate is from StartCom http://www.startssl.com/

Any help is much appreciated, if you need further information to help me
on this issue please let me know.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,219030,219030#msg-219030


From mdounin at mdounin.ru  Fri Nov 25 12:54:21 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Fri, 25 Nov 2011 16:54:21 +0400
Subject: nginx 1.x.x segfault at e8 ip xxx sp xxx error 4 in
	libssl.so.1.0.0
In-Reply-To: <74e368a93cf35d43c926660b4ddf733a.NginxMailingListEnglish@forum.nginx.org>
References: <74e368a93cf35d43c926660b4ddf733a.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20111125125421.GP95664@mdounin.ru>

Hello!

On Fri, Nov 25, 2011 at 07:49:38AM -0500, Fleshgrinder wrote:

> nginx -V output:
> 
> nginx version: nginx/1.1.8
> built by gcc 4.6.2 (Debian 4.6.2-5)
> TLS SNI support enabled
> 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=/build/buildd-nginx_1.1.8-2-i386-6PKxRV/nginx-1.1.8/debian/modules/nginx-echo
> --add-module=/build/buildd-nginx_1.1.8-2-i386-6PKxRV/nginx-1.1.8/debian/modules/nginx-upstream-fair
> 
> This is the nginx-full package from the Debian unstable branch. I also
> had the same problem with nginx-full from dotdeb (1.0.09 and 1.0.10).
> The same problem also occurs with libssl.0.9.8.
> 
> My certificate is from StartCom http://www.startssl.com/
> 
> Any help is much appreciated, if you need further information to help me
> on this issue please let me know.

Please try the patch or the workaround from here:

http://trac.nginx.org/nginx/ticket/54

If it doesn't help, please provide more information, as outlined 
at http://wiki.nginx.org/Debugging.

Maxim Dounin


From nginx-forum at nginx.us  Fri Nov 25 13:12:43 2011
From: nginx-forum at nginx.us (Fleshgrinder)
Date: Fri, 25 Nov 2011 08:12:43 -0500
Subject: nginx 1.x.x segfault at e8 ip xxx sp xxx error 4 in
	libssl.so.1.0.0
In-Reply-To: <20111125125421.GP95664@mdounin.ru>
References: <20111125125421.GP95664@mdounin.ru>
Message-ID: <792027e2e42c53ac69e368e23f779e16.NginxMailingListEnglish@forum.nginx.org>

Thanks a lot, I'm going to give the workaround a try, because I really
have to blocks that are listening to 443 and one of both blocks has no
ssl_certificate defined (it only redirects from www to non-www). I just
copied the whole ssl block from the other server context and restarted.
I'll let you know if it worked (but I'm very confident).

I don't want to apply the patch because I want to stick to the Debian
packages (integrate perfectly in the system).

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,219030,219033#msg-219033


From nginx-forum at nginx.us  Fri Nov 25 13:44:06 2011
From: nginx-forum at nginx.us (demon326)
Date: Fri, 25 Nov 2011 08:44:06 -0500
Subject: Nginx 404 not found/503 error in subdir.
Message-ID: <0dbed7942c64699987eb6207e43f60da.NginxMailingListEnglish@forum.nginx.org>

Okay, i setup a debian install with ispconfig3 ,php5-fpm and nginx, the
main site, drupal is working fine, even perfect. But i also have a
modded phpbb3 board, in a subdir, named 'forum'.

Now for some reason i get the following error from nginx when accesing
the dir:

'ERROR 500 - Internal Server Error!'

This is the content of the nginx directiv field(for vhost):

http://pastebin.com/rTacVDSa


I checked the php and nginx logs, nothing in it that suggest an error
with php...

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,219034,219034#msg-219034


From grails at jmsd.co.uk  Fri Nov 25 14:51:24 2011
From: grails at jmsd.co.uk (John Moore)
Date: Fri, 25 Nov 2011 14:51:24 +0000
Subject: Can nginx override encoding of Tomcat back end?
Message-ID: <4ECFAB6C.8020401@jmsd.co.uk>

I have been trying for some time now to get Tomcat to serve everything 
with UTF-8 encoding, but some pages keep coming back with ISO-8859-1 
encoding. As Tomcat is sitting behind an nginx remote proxy server, I 
was wondering whether I could configure nginx to override the encoding 
automatically? I have added 'charset utf-8' in the server block but this 
by itself obviously doesn't do the trick. Can I set some response 
headers, perhaps?


From rami.essaid at gmail.com  Fri Nov 25 14:54:14 2011
From: rami.essaid at gmail.com (Rami Essaid)
Date: Fri, 25 Nov 2011 09:54:14 -0500
Subject: nginx behind load balancer
In-Reply-To: <20111121124602.GX95664@mdounin.ru>
References: <CAPMF7wjD8BYcac6NVjmOMSEVfwnKsT-_rrqh2sRn8YexPKZ4sA@mail.gmail.com>
	<20111121123822.GW95664@mdounin.ru>
	<CAPMF7wggdXCT=EQH57gGgM6SWP-o47pBmJWnsbCgSBCSk0OfQQ@mail.gmail.com>
	<20111121124602.GX95664@mdounin.ru>
Message-ID: <CAPMF7wgk00ZX4jWo2EAgL2Fr6etD7pr6uHJGVSvxdC9Sm_k00A@mail.gmail.com>

Hi Maxim,

We implemented to module and still had some trouble.  A lot of the
connections would return " 503 Service Temporarily Unavailable".  Our
configuration works fine without the load balancer but then gives these 503
errors behind the load balancer.

Looking into the error logs I notice a lot of these errors both with and
without the load balancer "connect() failed (111: Connection refused) while
connecting to upstream".  Could this be the reason that we are having
issues?

Thanks!
Rami


On Mon, Nov 21, 2011 at 7:46 AM, Maxim Dounin <mdounin at mdounin.ru> wrote:

> Hello!
>
> On Mon, Nov 21, 2011 at 07:40:49AM -0500, Rami Essaid wrote:
>
> > Thanks Maxim,
> >
> > This looks like exactly what we need.  In your experience does this
> resolve
> > most issues behind a load balancer?
>
> Yes.
>
> Maxim Dounin
>
> > On Mon, Nov 21, 2011 at 7:38 AM, Maxim Dounin <mdounin at mdounin.ru>
> wrote:
> >
> > > Hello!
> > >
> > > On Mon, Nov 21, 2011 at 07:25:39AM -0500, Rami Essaid wrote:
> > >
> > > > Hi Guys,
> > > >
> > > > This weekend for scalability we tried putting our nginx servers
> behind
> > > > amazon's elastic load balancers and came across a road block: it
> does not
> > > > transparently pass the user IP and header information to nginx. This
> > > caused
> > > > issues with several pieces of nginx we use including the IP allow /
> deny
> > > > rules, the limit_req module, and the limit_con module.  Has anyone
> > > > successfully put nginx behind a load balancer?  Any ideas on how to
> make
> > > > this work?
> > >
> > > http://wiki.nginx.org/HttpRealIpModule
> > >
> > > 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/20111125/ee8cd974/attachment.html>

From igor at sysoev.ru  Fri Nov 25 15:16:19 2011
From: igor at sysoev.ru (Igor Sysoev)
Date: Fri, 25 Nov 2011 19:16:19 +0400
Subject: Can nginx override encoding of Tomcat back end?
In-Reply-To: <4ECFAB6C.8020401@jmsd.co.uk>
References: <4ECFAB6C.8020401@jmsd.co.uk>
Message-ID: <20111125151619.GE6241@nginx.com>

On Fri, Nov 25, 2011 at 02:51:24PM +0000, John Moore wrote:
> I have been trying for some time now to get Tomcat to serve everything 
> with UTF-8 encoding, but some pages keep coming back with ISO-8859-1 
> encoding. As Tomcat is sitting behind an nginx remote proxy server, I 
> was wondering whether I could configure nginx to override the encoding 
> automatically? I have added 'charset utf-8' in the server block but this 
> by itself obviously doesn't do the trick. Can I set some response 
> headers, perhaps?

You can try this, but nginx will scan response body with ISO-8859-1
trying to convert it to utf-8:

http {
    charset_map  iso-8859-1 utf-8 {}

    server {
        charset           utf-8;
        override_charset  on;


-- 
Igor Sysoev


From grails at jmsd.co.uk  Fri Nov 25 15:24:18 2011
From: grails at jmsd.co.uk (John Moore)
Date: Fri, 25 Nov 2011 15:24:18 +0000
Subject: Can nginx override encoding of Tomcat back end?
In-Reply-To: <20111125151619.GE6241@nginx.com>
References: <4ECFAB6C.8020401@jmsd.co.uk> <20111125151619.GE6241@nginx.com>
Message-ID: <4ECFB322.3000903@jmsd.co.uk>

On 25/11/11 15:16, Igor Sysoev wrote:
>
> You can try this, but nginx will scan response body with ISO-8859-1
> trying to convert it to utf-8:
>
> http {
>      charset_map  iso-8859-1 utf-8 {}
>
>      server {
>          charset           utf-8;
>          override_charset  on;
>
>    
Thanks Igor. I tried this, but it's still leaving the encoding as 
ISO-8859-1 (as reported under Page Info in Firefox). This is 0.6.32 on 
Windows - I don't know whether a more recent version would help?

JM


From igor at sysoev.ru  Fri Nov 25 15:34:18 2011
From: igor at sysoev.ru (Igor Sysoev)
Date: Fri, 25 Nov 2011 19:34:18 +0400
Subject: Can nginx override encoding of Tomcat back end?
In-Reply-To: <4ECFB322.3000903@jmsd.co.uk>
References: <4ECFAB6C.8020401@jmsd.co.uk> <20111125151619.GE6241@nginx.com>
	<4ECFB322.3000903@jmsd.co.uk>
Message-ID: <20111125153418.GA10606@nginx.com>

On Fri, Nov 25, 2011 at 03:24:18PM +0000, John Moore wrote:
> On 25/11/11 15:16, Igor Sysoev wrote:
> >
> > You can try this, but nginx will scan response body with ISO-8859-1
> > trying to convert it to utf-8:
> >
> > http {
> >      charset_map  iso-8859-1 utf-8 {}
> >
> >      server {
> >          charset           utf-8;
> >          override_charset  on;
> >
> >    
> Thanks Igor. I tried this, but it's still leaving the encoding as 
> ISO-8859-1 (as reported under Page Info in Firefox).

Are you sure that Firefox did not get it from cache ?

> This is 0.6.32 on 
> Windows - I don't know whether a more recent version would help?

Where have you found such version ?
The first nginx/Windows version was 0.7.52.
There were some change in charset module in 0.8.0, so you can try
more modern version.


-- 
Igor Sysoev


From grails at jmsd.co.uk  Fri Nov 25 15:41:30 2011
From: grails at jmsd.co.uk (John Moore)
Date: Fri, 25 Nov 2011 15:41:30 +0000
Subject: Can nginx override encoding of Tomcat back end?
In-Reply-To: <20111125153418.GA10606@nginx.com>
References: <4ECFAB6C.8020401@jmsd.co.uk>
	<20111125151619.GE6241@nginx.com>	<4ECFB322.3000903@jmsd.co.uk>
	<20111125153418.GA10606@nginx.com>
Message-ID: <4ECFB72A.3020608@jmsd.co.uk>

On 25/11/11 15:34, Igor Sysoev wrote:
> Are you sure that Firefox did not get it from cache ?
>    

Pretty sure. Just doing some checking as there are some strange things 
going on...

>> This is 0.6.32 on
>> Windows - I don't know whether a more recent version would help?
>>      
> Where have you found such version ?
> The first nginx/Windows version was 0.7.52.
> There were some change in charset module in 0.8.0, so you can try
> more modern version.
>
>    

It was one of the Kevin Worthington versions. To be honest I've not 
really looked at it in a long while. I'll install a newer version and 
report back.

JM


From grails at jmsd.co.uk  Fri Nov 25 16:09:50 2011
From: grails at jmsd.co.uk (John Moore)
Date: Fri, 25 Nov 2011 16:09:50 +0000
Subject: Can nginx override encoding of Tomcat back end?
In-Reply-To: <20111125153418.GA10606@nginx.com>
References: <4ECFAB6C.8020401@jmsd.co.uk>
	<20111125151619.GE6241@nginx.com>	<4ECFB322.3000903@jmsd.co.uk>
	<20111125153418.GA10606@nginx.com>
Message-ID: <4ECFBDCE.5090802@jmsd.co.uk>

Igor, thanks for the suggestion. This is now working as it should!


From mdounin at mdounin.ru  Fri Nov 25 17:14:40 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Fri, 25 Nov 2011 21:14:40 +0400
Subject: nginx behind load balancer
In-Reply-To: <CAPMF7wgk00ZX4jWo2EAgL2Fr6etD7pr6uHJGVSvxdC9Sm_k00A@mail.gmail.com>
References: <CAPMF7wjD8BYcac6NVjmOMSEVfwnKsT-_rrqh2sRn8YexPKZ4sA@mail.gmail.com>
	<20111121123822.GW95664@mdounin.ru>
	<CAPMF7wggdXCT=EQH57gGgM6SWP-o47pBmJWnsbCgSBCSk0OfQQ@mail.gmail.com>
	<20111121124602.GX95664@mdounin.ru>
	<CAPMF7wgk00ZX4jWo2EAgL2Fr6etD7pr6uHJGVSvxdC9Sm_k00A@mail.gmail.com>
Message-ID: <20111125171440.GX95664@mdounin.ru>

Hello!

On Fri, Nov 25, 2011 at 09:54:14AM -0500, Rami Essaid wrote:

> Hi Maxim,
> 
> We implemented to module and still had some trouble.  A lot of the
> connections would return " 503 Service Temporarily Unavailable".  Our
> configuration works fine without the load balancer but then gives these 503
> errors behind the load balancer.

nginx itself will only return 503 if it hits either limit_conn or 
limit_req.

If you see this returned by nginx, and it only happens with load 
balancer, this may indicate you've not configured realip module 
properly (or your load balancer doesn't provide appropriate 
headers) and you are hitting per-ip limits configured due to all 
requests appear to be from load balancer.

Check if client's ip logged is really client's one, not an ip of 
your load balancer.

> Looking into the error logs I notice a lot of these errors both with and
> without the load balancer "connect() failed (111: Connection refused) while
> connecting to upstream".  Could this be the reason that we are having
> issues?

Unlikely, but It's a good idea to track and fix this in any case.

Maxim Dounin

> 
> Thanks!
> Rami
> 
> 
> On Mon, Nov 21, 2011 at 7:46 AM, Maxim Dounin <mdounin at mdounin.ru> wrote:
> 
> > Hello!
> >
> > On Mon, Nov 21, 2011 at 07:40:49AM -0500, Rami Essaid wrote:
> >
> > > Thanks Maxim,
> > >
> > > This looks like exactly what we need.  In your experience does this
> > resolve
> > > most issues behind a load balancer?
> >
> > Yes.
> >
> > Maxim Dounin
> >
> > > On Mon, Nov 21, 2011 at 7:38 AM, Maxim Dounin <mdounin at mdounin.ru>
> > wrote:
> > >
> > > > Hello!
> > > >
> > > > On Mon, Nov 21, 2011 at 07:25:39AM -0500, Rami Essaid wrote:
> > > >
> > > > > Hi Guys,
> > > > >
> > > > > This weekend for scalability we tried putting our nginx servers
> > behind
> > > > > amazon's elastic load balancers and came across a road block: it
> > does not
> > > > > transparently pass the user IP and header information to nginx. This
> > > > caused
> > > > > issues with several pieces of nginx we use including the IP allow /
> > deny
> > > > > rules, the limit_req module, and the limit_con module.  Has anyone
> > > > > successfully put nginx behind a load balancer?  Any ideas on how to
> > make
> > > > > this work?
> > > >
> > > > http://wiki.nginx.org/HttpRealIpModule
> > > >
> > > > 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 rami.essaid at gmail.com  Fri Nov 25 17:37:53 2011
From: rami.essaid at gmail.com (Rami Essaid)
Date: Fri, 25 Nov 2011 12:37:53 -0500
Subject: nginx behind load balancer
In-Reply-To: <20111125171440.GX95664@mdounin.ru>
References: <CAPMF7wjD8BYcac6NVjmOMSEVfwnKsT-_rrqh2sRn8YexPKZ4sA@mail.gmail.com>
	<20111121123822.GW95664@mdounin.ru>
	<CAPMF7wggdXCT=EQH57gGgM6SWP-o47pBmJWnsbCgSBCSk0OfQQ@mail.gmail.com>
	<20111121124602.GX95664@mdounin.ru>
	<CAPMF7wgk00ZX4jWo2EAgL2Fr6etD7pr6uHJGVSvxdC9Sm_k00A@mail.gmail.com>
	<20111125171440.GX95664@mdounin.ru>
Message-ID: <CAPMF7wiggTJZUYz-tdBvw_frbfvNYYggznVXY-+bfwGU6sepqA@mail.gmail.com>

Thanks for the help Maxim!  We disabled our limit_req and that seemed to
have fixed the problem.  Looking at the logs it seems that only 1/3 of the
requests are correctly getting the new IP assigned via the realIP module,
the remainder are still logging the load balancer IP.  This probably is
more of an issue with the amazon load balancer but do you have any idea on
what may be going on?

Also, where would you recommend as a place to start tracking and fixing the
other issue?





On Fri, Nov 25, 2011 at 12:14 PM, Maxim Dounin <mdounin at mdounin.ru> wrote:

> Hello!
>
> On Fri, Nov 25, 2011 at 09:54:14AM -0500, Rami Essaid wrote:
>
> > Hi Maxim,
> >
> > We implemented to module and still had some trouble.  A lot of the
> > connections would return " 503 Service Temporarily Unavailable".  Our
> > configuration works fine without the load balancer but then gives these
> 503
> > errors behind the load balancer.
>
> nginx itself will only return 503 if it hits either limit_conn or
> limit_req.
>
> If you see this returned by nginx, and it only happens with load
> balancer, this may indicate you've not configured realip module
> properly (or your load balancer doesn't provide appropriate
> headers) and you are hitting per-ip limits configured due to all
> requests appear to be from load balancer.
>
> Check if client's ip logged is really client's one, not an ip of
> your load balancer.
>
> > Looking into the error logs I notice a lot of these errors both with and
> > without the load balancer "connect() failed (111: Connection refused)
> while
> > connecting to upstream".  Could this be the reason that we are having
> > issues?
>
> Unlikely, but It's a good idea to track and fix this in any case.
>
> Maxim Dounin
>
> >
> > Thanks!
> > Rami
> >
> >
> > On Mon, Nov 21, 2011 at 7:46 AM, Maxim Dounin <mdounin at mdounin.ru>
> wrote:
> >
> > > Hello!
> > >
> > > On Mon, Nov 21, 2011 at 07:40:49AM -0500, Rami Essaid wrote:
> > >
> > > > Thanks Maxim,
> > > >
> > > > This looks like exactly what we need.  In your experience does this
> > > resolve
> > > > most issues behind a load balancer?
> > >
> > > Yes.
> > >
> > > Maxim Dounin
> > >
> > > > On Mon, Nov 21, 2011 at 7:38 AM, Maxim Dounin <mdounin at mdounin.ru>
> > > wrote:
> > > >
> > > > > Hello!
> > > > >
> > > > > On Mon, Nov 21, 2011 at 07:25:39AM -0500, Rami Essaid wrote:
> > > > >
> > > > > > Hi Guys,
> > > > > >
> > > > > > This weekend for scalability we tried putting our nginx servers
> > > behind
> > > > > > amazon's elastic load balancers and came across a road block: it
> > > does not
> > > > > > transparently pass the user IP and header information to nginx.
> This
> > > > > caused
> > > > > > issues with several pieces of nginx we use including the IP
> allow /
> > > deny
> > > > > > rules, the limit_req module, and the limit_con module.  Has
> anyone
> > > > > > successfully put nginx behind a load balancer?  Any ideas on how
> to
> > > make
> > > > > > this work?
> > > > >
> > > > > http://wiki.nginx.org/HttpRealIpModule
> > > > >
> > > > > 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
>
> _______________________________________________
> 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/20111125/1c9e0e2c/attachment.html>

From fla_torres at yahoo.com.br  Fri Nov 25 20:39:58 2011
From: fla_torres at yahoo.com.br (Flavio Torres)
Date: Fri, 25 Nov 2011 18:39:58 -0200
Subject: nginx mail proxy - dovecot ssl backend
Message-ID: <4ECFFD1E.2030002@yahoo.com.br>


-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hello!

Is SSL backends (dovecot) supported by nginx mail proxy ?

Thank you!
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.11 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/

iEYEARECAAYFAk7P/RwACgkQNRQApncg294jrgCfW/lRpMywRk2tq++wRE5u3sxM
X1kAn31+gvYibdHLYM3Mv2HDaW/TjaHa
=DK/D
-----END PGP SIGNATURE-----


From stef at caunter.ca  Fri Nov 25 21:19:33 2011
From: stef at caunter.ca (Stefan Caunter)
Date: Fri, 25 Nov 2011 16:19:33 -0500
Subject: nginx behind load balancer
In-Reply-To: <CAPMF7wiggTJZUYz-tdBvw_frbfvNYYggznVXY-+bfwGU6sepqA@mail.gmail.com>
References: <CAPMF7wjD8BYcac6NVjmOMSEVfwnKsT-_rrqh2sRn8YexPKZ4sA@mail.gmail.com>
	<20111121123822.GW95664@mdounin.ru>
	<CAPMF7wggdXCT=EQH57gGgM6SWP-o47pBmJWnsbCgSBCSk0OfQQ@mail.gmail.com>
	<20111121124602.GX95664@mdounin.ru>
	<CAPMF7wgk00ZX4jWo2EAgL2Fr6etD7pr6uHJGVSvxdC9Sm_k00A@mail.gmail.com>
	<20111125171440.GX95664@mdounin.ru>
	<CAPMF7wiggTJZUYz-tdBvw_frbfvNYYggznVXY-+bfwGU6sepqA@mail.gmail.com>
Message-ID: <CALdpnrzFdfxvP0dvGKHg4cLSzvUZ6HHBr53FRgzzPVKLvRKGmw@mail.gmail.com>

On Fri, Nov 25, 2011 at 12:37 PM, Rami Essaid <rami.essaid at gmail.com> wrote:
> Thanks for the help Maxim! ?We disabled our limit_req and that seemed to
> have fixed the problem. ?Looking at the logs it seems that only 1/3 of the
> requests are correctly getting the new IP assigned via the realIP module,
> the remainder are still logging the load balancer IP. ?This probably is more
> of an issue with the amazon load balancer but do you have any idea on what
> may be going on?
> Also, where would you recommend as a?place?to start tracking and fixing the
> other issue?
>

You have not established that load balancer is setting X-Forwarded-For
or some other header to pass Real IP.

Does the amazon pseudo device have enough sockets available to connect to nginx?

What do you expect from adding this layer? Were you hitting
performance limits? Has it improved or degraded performance?

>
>
> On Fri, Nov 25, 2011 at 12:14 PM, Maxim Dounin <mdounin at mdounin.ru> wrote:
>>
>> Hello!
>>
>> On Fri, Nov 25, 2011 at 09:54:14AM -0500, Rami Essaid wrote:
>>
>> > Hi Maxim,
>> >
>> > We implemented to module and still had some trouble. ?A lot of the
>> > connections would return " 503 Service Temporarily Unavailable". ?Our
>> > configuration works fine without the load balancer but then gives these
>> > 503
>> > errors behind the load balancer.
>>
>> nginx itself will only return 503 if it hits either limit_conn or
>> limit_req.
>>
>> If you see this returned by nginx, and it only happens with load
>> balancer, this may indicate you've not configured realip module
>> properly (or your load balancer doesn't provide appropriate
>> headers) and you are hitting per-ip limits configured due to all
>> requests appear to be from load balancer.
>>
>> Check if client's ip logged is really client's one, not an ip of
>> your load balancer.
>>
>> > Looking into the error logs I notice a lot of these errors both with and
>> > without the load balancer "connect() failed (111: Connection refused)
>> > while
>> > connecting to upstream". ?Could this be the reason that we are having
>> > issues?
>>
>> Unlikely, but It's a good idea to track and fix this in any case.
>>
>> Maxim Dounin
>>
>> >
>> > Thanks!
>> > Rami
>> >
>> >
>> > On Mon, Nov 21, 2011 at 7:46 AM, Maxim Dounin <mdounin at mdounin.ru>
>> > wrote:
>> >
>> > > Hello!
>> > >
>> > > On Mon, Nov 21, 2011 at 07:40:49AM -0500, Rami Essaid wrote:
>> > >
>> > > > Thanks Maxim,
>> > > >
>> > > > This looks like exactly what we need. ?In your experience does this
>> > > resolve
>> > > > most issues behind a load balancer?
>> > >
>> > > Yes.
>> > >
>> > > Maxim Dounin
>> > >
>> > > > On Mon, Nov 21, 2011 at 7:38 AM, Maxim Dounin <mdounin at mdounin.ru>
>> > > wrote:
>> > > >
>> > > > > Hello!
>> > > > >
>> > > > > On Mon, Nov 21, 2011 at 07:25:39AM -0500, Rami Essaid wrote:
>> > > > >
>> > > > > > Hi Guys,
>> > > > > >
>> > > > > > This weekend for scalability we tried putting our nginx servers
>> > > behind
>> > > > > > amazon's elastic load balancers and came across a road block: it
>> > > does not
>> > > > > > transparently pass the user IP and header information to nginx.
>> > > > > > This
>> > > > > caused
>> > > > > > issues with several pieces of nginx we use including the IP
>> > > > > > allow /
>> > > deny
>> > > > > > rules, the limit_req module, and the limit_con module. ?Has
>> > > > > > anyone
>> > > > > > successfully put nginx behind a load balancer? ?Any ideas on how
>> > > > > > to
>> > > make
>> > > > > > this work?
>> > > > >
>> > > > > http://wiki.nginx.org/HttpRealIpModule
>> > > > >
>> > > > > 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
>>
>> _______________________________________________
>> 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 Nov 25 21:30:05 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Sat, 26 Nov 2011 01:30:05 +0400
Subject: nginx mail proxy - dovecot ssl backend
In-Reply-To: <4ECFFD1E.2030002@yahoo.com.br>
References: <4ECFFD1E.2030002@yahoo.com.br>
Message-ID: <20111125213005.GC95664@mdounin.ru>

Hello!

On Fri, Nov 25, 2011 at 06:39:58PM -0200, Flavio Torres wrote:

> Is SSL backends (dovecot) supported by nginx mail proxy ?

No, mail proxy doesn't support ssl backends.

Maxim Dounin


From rami.essaid at gmail.com  Fri Nov 25 21:54:27 2011
From: rami.essaid at gmail.com (Rami Essaid)
Date: Fri, 25 Nov 2011 16:54:27 -0500
Subject: nginx behind load balancer
In-Reply-To: <CALdpnrzFdfxvP0dvGKHg4cLSzvUZ6HHBr53FRgzzPVKLvRKGmw@mail.gmail.com>
References: <CAPMF7wjD8BYcac6NVjmOMSEVfwnKsT-_rrqh2sRn8YexPKZ4sA@mail.gmail.com>
	<20111121123822.GW95664@mdounin.ru>
	<CAPMF7wggdXCT=EQH57gGgM6SWP-o47pBmJWnsbCgSBCSk0OfQQ@mail.gmail.com>
	<20111121124602.GX95664@mdounin.ru>
	<CAPMF7wgk00ZX4jWo2EAgL2Fr6etD7pr6uHJGVSvxdC9Sm_k00A@mail.gmail.com>
	<20111125171440.GX95664@mdounin.ru>
	<CAPMF7wiggTJZUYz-tdBvw_frbfvNYYggznVXY-+bfwGU6sepqA@mail.gmail.com>
	<CALdpnrzFdfxvP0dvGKHg4cLSzvUZ6HHBr53FRgzzPVKLvRKGmw@mail.gmail.com>
Message-ID: <CAPMF7wiwWKqu43FnUJY4R36ubSWO7Wh3TEkfkR2MOq_3JSUJ=A@mail.gmail.com>

In looking at the $proxy_add_x_forwarded_for variable I believe that the
Loadbalancer is in fact passing the variables but nginx is taking the wrong
value?

Here is what i get from the variables.
$proxy_add_x_forwarded_for: 217.27.244.18, 10.160.43.200, 10.160.43.200
$remote_addr: 10.160.43.200

Does this mean that nginx is taking the last value in the x-forwarded-for?

To answer your other question, yes, the Amazon plenty of sockets available
to connect to nginx.

The reason we added this was to be able to utilize Amazon's auto scaling
feature for redundancy and scalability.  We have not been able to test
performance yet since we are still trying to make it work but as soon as we
do I will be sure to report back.


On Fri, Nov 25, 2011 at 4:19 PM, Stefan Caunter <stef at caunter.ca> wrote:

> On Fri, Nov 25, 2011 at 12:37 PM, Rami Essaid <rami.essaid at gmail.com>
> wrote:
> > Thanks for the help Maxim!  We disabled our limit_req and that seemed to
> > have fixed the problem.  Looking at the logs it seems that only 1/3 of
> the
> > requests are correctly getting the new IP assigned via the realIP module,
> > the remainder are still logging the load balancer IP.  This probably is
> more
> > of an issue with the amazon load balancer but do you have any idea on
> what
> > may be going on?
> > Also, where would you recommend as a place to start tracking and fixing
> the
> > other issue?
> >
>
> You have not established that load balancer is setting X-Forwarded-For
> or some other header to pass Real IP.
>
> Does the amazon pseudo device have enough sockets available to connect to
> nginx?
>
> What do you expect from adding this layer? Were you hitting
> performance limits? Has it improved or degraded performance?
>
> >
> >
> > On Fri, Nov 25, 2011 at 12:14 PM, Maxim Dounin <mdounin at mdounin.ru>
> wrote:
> >>
> >> Hello!
> >>
> >> On Fri, Nov 25, 2011 at 09:54:14AM -0500, Rami Essaid wrote:
> >>
> >> > Hi Maxim,
> >> >
> >> > We implemented to module and still had some trouble.  A lot of the
> >> > connections would return " 503 Service Temporarily Unavailable".  Our
> >> > configuration works fine without the load balancer but then gives
> these
> >> > 503
> >> > errors behind the load balancer.
> >>
> >> nginx itself will only return 503 if it hits either limit_conn or
> >> limit_req.
> >>
> >> If you see this returned by nginx, and it only happens with load
> >> balancer, this may indicate you've not configured realip module
> >> properly (or your load balancer doesn't provide appropriate
> >> headers) and you are hitting per-ip limits configured due to all
> >> requests appear to be from load balancer.
> >>
> >> Check if client's ip logged is really client's one, not an ip of
> >> your load balancer.
> >>
> >> > Looking into the error logs I notice a lot of these errors both with
> and
> >> > without the load balancer "connect() failed (111: Connection refused)
> >> > while
> >> > connecting to upstream".  Could this be the reason that we are having
> >> > issues?
> >>
> >> Unlikely, but It's a good idea to track and fix this in any case.
> >>
> >> Maxim Dounin
> >>
> >> >
> >> > Thanks!
> >> > Rami
> >> >
> >> >
> >> > On Mon, Nov 21, 2011 at 7:46 AM, Maxim Dounin <mdounin at mdounin.ru>
> >> > wrote:
> >> >
> >> > > Hello!
> >> > >
> >> > > On Mon, Nov 21, 2011 at 07:40:49AM -0500, Rami Essaid wrote:
> >> > >
> >> > > > Thanks Maxim,
> >> > > >
> >> > > > This looks like exactly what we need.  In your experience does
> this
> >> > > resolve
> >> > > > most issues behind a load balancer?
> >> > >
> >> > > Yes.
> >> > >
> >> > > Maxim Dounin
> >> > >
> >> > > > On Mon, Nov 21, 2011 at 7:38 AM, Maxim Dounin <mdounin at mdounin.ru
> >
> >> > > wrote:
> >> > > >
> >> > > > > Hello!
> >> > > > >
> >> > > > > On Mon, Nov 21, 2011 at 07:25:39AM -0500, Rami Essaid wrote:
> >> > > > >
> >> > > > > > Hi Guys,
> >> > > > > >
> >> > > > > > This weekend for scalability we tried putting our nginx
> servers
> >> > > behind
> >> > > > > > amazon's elastic load balancers and came across a road block:
> it
> >> > > does not
> >> > > > > > transparently pass the user IP and header information to
> nginx.
> >> > > > > > This
> >> > > > > caused
> >> > > > > > issues with several pieces of nginx we use including the IP
> >> > > > > > allow /
> >> > > deny
> >> > > > > > rules, the limit_req module, and the limit_con module.  Has
> >> > > > > > anyone
> >> > > > > > successfully put nginx behind a load balancer?  Any ideas on
> how
> >> > > > > > to
> >> > > make
> >> > > > > > this work?
> >> > > > >
> >> > > > > http://wiki.nginx.org/HttpRealIpModule
> >> > > > >
> >> > > > > 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
> >>
> >> _______________________________________________
> >> 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/20111125/bec15e1c/attachment.html>

From rami.essaid at gmail.com  Fri Nov 25 22:25:25 2011
From: rami.essaid at gmail.com (Rami Essaid)
Date: Fri, 25 Nov 2011 17:25:25 -0500
Subject: nginx behind load balancer
In-Reply-To: <CAPMF7wiwWKqu43FnUJY4R36ubSWO7Wh3TEkfkR2MOq_3JSUJ=A@mail.gmail.com>
References: <CAPMF7wjD8BYcac6NVjmOMSEVfwnKsT-_rrqh2sRn8YexPKZ4sA@mail.gmail.com>
	<20111121123822.GW95664@mdounin.ru>
	<CAPMF7wggdXCT=EQH57gGgM6SWP-o47pBmJWnsbCgSBCSk0OfQQ@mail.gmail.com>
	<20111121124602.GX95664@mdounin.ru>
	<CAPMF7wgk00ZX4jWo2EAgL2Fr6etD7pr6uHJGVSvxdC9Sm_k00A@mail.gmail.com>
	<20111125171440.GX95664@mdounin.ru>
	<CAPMF7wiggTJZUYz-tdBvw_frbfvNYYggznVXY-+bfwGU6sepqA@mail.gmail.com>
	<CALdpnrzFdfxvP0dvGKHg4cLSzvUZ6HHBr53FRgzzPVKLvRKGmw@mail.gmail.com>
	<CAPMF7wiwWKqu43FnUJY4R36ubSWO7Wh3TEkfkR2MOq_3JSUJ=A@mail.gmail.com>
Message-ID: <CAPMF7wgD20dhJftsc_hQr63-30m+nVLpWQ9NxiuY7UBmcadTzQ@mail.gmail.com>

As a quick update, it looks like this has happened before. The load
balancer bounces off of several internal IP's sometimes and nginx picks the
last one only. Does anyone know of a workaround to remove the last two
trusted IPs from the x-forwarded-for header?

http://forum.nginx.org/read.php?11,26102,214069

Rami

On Fri, Nov 25, 2011 at 4:54 PM, Rami Essaid <rami.essaid at gmail.com> wrote:

> In looking at the $proxy_add_x_forwarded_for variable I believe that the
> Loadbalancer is in fact passing the variables but nginx is taking the wrong
> value?
>
> Here is what i get from the variables.
> $proxy_add_x_forwarded_for: 217.27.244.18, 10.160.43.200, 10.160.43.200
> $remote_addr: 10.160.43.200
>
> Does this mean that nginx is taking the last value in the x-forwarded-for?
>
>
> To answer your other question, yes, the Amazon plenty of sockets
> available to connect to nginx.
>
> The reason we added this was to be able to utilize Amazon's auto scaling
> feature for redundancy and scalability.  We have not been able to test
> performance yet since we are still trying to make it work but as soon as we
> do I will be sure to report back.
>
>
> On Fri, Nov 25, 2011 at 4:19 PM, Stefan Caunter <stef at caunter.ca> wrote:
>
>> On Fri, Nov 25, 2011 at 12:37 PM, Rami Essaid <rami.essaid at gmail.com>
>> wrote:
>> > Thanks for the help Maxim!  We disabled our limit_req and that seemed to
>> > have fixed the problem.  Looking at the logs it seems that only 1/3 of
>> the
>> > requests are correctly getting the new IP assigned via the realIP
>> module,
>> > the remainder are still logging the load balancer IP.  This probably is
>> more
>> > of an issue with the amazon load balancer but do you have any idea on
>> what
>> > may be going on?
>> > Also, where would you recommend as a place to start tracking and fixing
>> the
>> > other issue?
>> >
>>
>> You have not established that load balancer is setting X-Forwarded-For
>> or some other header to pass Real IP.
>>
>> Does the amazon pseudo device have enough sockets available to connect to
>> nginx?
>>
>> What do you expect from adding this layer? Were you hitting
>> performance limits? Has it improved or degraded performance?
>>
>> >
>> >
>> > On Fri, Nov 25, 2011 at 12:14 PM, Maxim Dounin <mdounin at mdounin.ru>
>> wrote:
>> >>
>> >> Hello!
>> >>
>> >> On Fri, Nov 25, 2011 at 09:54:14AM -0500, Rami Essaid wrote:
>> >>
>> >> > Hi Maxim,
>> >> >
>> >> > We implemented to module and still had some trouble.  A lot of the
>> >> > connections would return " 503 Service Temporarily Unavailable".  Our
>> >> > configuration works fine without the load balancer but then gives
>> these
>> >> > 503
>> >> > errors behind the load balancer.
>> >>
>> >> nginx itself will only return 503 if it hits either limit_conn or
>> >> limit_req.
>> >>
>> >> If you see this returned by nginx, and it only happens with load
>> >> balancer, this may indicate you've not configured realip module
>> >> properly (or your load balancer doesn't provide appropriate
>> >> headers) and you are hitting per-ip limits configured due to all
>> >> requests appear to be from load balancer.
>> >>
>> >> Check if client's ip logged is really client's one, not an ip of
>> >> your load balancer.
>> >>
>> >> > Looking into the error logs I notice a lot of these errors both with
>> and
>> >> > without the load balancer "connect() failed (111: Connection refused)
>> >> > while
>> >> > connecting to upstream".  Could this be the reason that we are having
>> >> > issues?
>> >>
>> >> Unlikely, but It's a good idea to track and fix this in any case.
>> >>
>> >> Maxim Dounin
>> >>
>> >> >
>> >> > Thanks!
>> >> > Rami
>> >> >
>> >> >
>> >> > On Mon, Nov 21, 2011 at 7:46 AM, Maxim Dounin <mdounin at mdounin.ru>
>> >> > wrote:
>> >> >
>> >> > > Hello!
>> >> > >
>> >> > > On Mon, Nov 21, 2011 at 07:40:49AM -0500, Rami Essaid wrote:
>> >> > >
>> >> > > > Thanks Maxim,
>> >> > > >
>> >> > > > This looks like exactly what we need.  In your experience does
>> this
>> >> > > resolve
>> >> > > > most issues behind a load balancer?
>> >> > >
>> >> > > Yes.
>> >> > >
>> >> > > Maxim Dounin
>> >> > >
>> >> > > > On Mon, Nov 21, 2011 at 7:38 AM, Maxim Dounin <
>> mdounin at mdounin.ru>
>> >> > > wrote:
>> >> > > >
>> >> > > > > Hello!
>> >> > > > >
>> >> > > > > On Mon, Nov 21, 2011 at 07:25:39AM -0500, Rami Essaid wrote:
>> >> > > > >
>> >> > > > > > Hi Guys,
>> >> > > > > >
>> >> > > > > > This weekend for scalability we tried putting our nginx
>> servers
>> >> > > behind
>> >> > > > > > amazon's elastic load balancers and came across a road
>> block: it
>> >> > > does not
>> >> > > > > > transparently pass the user IP and header information to
>> nginx.
>> >> > > > > > This
>> >> > > > > caused
>> >> > > > > > issues with several pieces of nginx we use including the IP
>> >> > > > > > allow /
>> >> > > deny
>> >> > > > > > rules, the limit_req module, and the limit_con module.  Has
>> >> > > > > > anyone
>> >> > > > > > successfully put nginx behind a load balancer?  Any ideas on
>> how
>> >> > > > > > to
>> >> > > make
>> >> > > > > > this work?
>> >> > > > >
>> >> > > > > http://wiki.nginx.org/HttpRealIpModule
>> >> > > > >
>> >> > > > > 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
>> >>
>> >> _______________________________________________
>> >> 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/20111125/b8e44921/attachment-0001.html>

From Richard.Kearsley at m247.com  Fri Nov 25 22:36:21 2011
From: Richard.Kearsley at m247.com (Richard Kearsley)
Date: Fri, 25 Nov 2011 22:36:21 +0000
Subject: add_header for 206?
Message-ID: <EFC5A19DBF3A2E489904733B467D990B38AC71@office-server2.m247.local>

Hi!

Quote from wiki:

Add_header
"Adds headers to the HTTP response when the response code is equal to 200, 204, 301, 302 or 304."

Any reason we missing 206?


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20111125/7c4cfdd5/attachment.html>

From nginx-forum at nginx.us  Sat Nov 26 07:25:24 2011
From: nginx-forum at nginx.us (anagio)
Date: Sat, 26 Nov 2011 02:25:24 -0500
Subject: List modules compiled with nginx in ec2?
Message-ID: <6c5e5433b01e5d55a11eb39aff1e55a9.NginxMailingListEnglish@forum.nginx.org>

I setup an amazon ec2 instance that had nginx pre-compiled is there a
way to see what optional modules have been configured?

Thanks

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,219082,219082#msg-219082


From edho at myconan.net  Sat Nov 26 07:31:32 2011
From: edho at myconan.net (Edho Arief)
Date: Sat, 26 Nov 2011 14:31:32 +0700
Subject: List modules compiled with nginx in ec2?
In-Reply-To: <6c5e5433b01e5d55a11eb39aff1e55a9.NginxMailingListEnglish@forum.nginx.org>
References: <6c5e5433b01e5d55a11eb39aff1e55a9.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <CABTjkKnSbvJF7C5ehKAyhUZUerwUY9r_4UA=kEPEs0LX31cMVw@mail.gmail.com>

On Sat, Nov 26, 2011 at 2:25 PM, anagio <nginx-forum at nginx.us> wrote:
> I setup an amazon ec2 instance that had nginx pre-compiled is there a
> way to see what optional modules have been configured?
>

nginx -V

-- 
O< ascii ribbon campaign - stop html mail - www.asciiribbon.org


From nginx-forum at nginx.us  Sat Nov 26 07:35:46 2011
From: nginx-forum at nginx.us (anagio)
Date: Sat, 26 Nov 2011 02:35:46 -0500
Subject: List modules compiled with nginx in ec2?
In-Reply-To: <CABTjkKnSbvJF7C5ehKAyhUZUerwUY9r_4UA=kEPEs0LX31cMVw@mail.gmail.com>
References: <CABTjkKnSbvJF7C5ehKAyhUZUerwUY9r_4UA=kEPEs0LX31cMVw@mail.gmail.com>
Message-ID: <e9a9e0d11ef788b6670301e1eb35236a.NginxMailingListEnglish@forum.nginx.org>

Thanks, is it possible to add a module after it's been compiled?

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,219082,219084#msg-219084


From nginx-forum at nginx.us  Sat Nov 26 07:36:43 2011
From: nginx-forum at nginx.us (strawman)
Date: Sat, 26 Nov 2011 02:36:43 -0500
Subject: nginx-lua and nginx upload module
Message-ID: <fbab13973b47e4b42b5be26affae2d08.NginxMailingListEnglish@forum.nginx.org>

Hi all,

I'm hoping to write an nginx-lua script which handles file uploads from
the nginx upload module.  I can do this fine using an external backend
and the upload_pass directive, but the lua module has no functionality
for parsing the multipart form data that this generates.  I've also
tried accessing the $upload_file_name, $upload_tmp_path etc. variables
generated by the upload module directly, but I don't seem to be able to
access these outside of the upload_set_form_field directive.

Is there any way to get the contents of those variables into a lua
handler in a form that can be easily parsed?

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,219085,219085#msg-219085


From edho at myconan.net  Sat Nov 26 07:43:49 2011
From: edho at myconan.net (Edho Arief)
Date: Sat, 26 Nov 2011 14:43:49 +0700
Subject: List modules compiled with nginx in ec2?
In-Reply-To: <e9a9e0d11ef788b6670301e1eb35236a.NginxMailingListEnglish@forum.nginx.org>
References: <CABTjkKnSbvJF7C5ehKAyhUZUerwUY9r_4UA=kEPEs0LX31cMVw@mail.gmail.com>
	<e9a9e0d11ef788b6670301e1eb35236a.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <CABTjkKkB5P1FvKSA9NdDzgkeq+aeijbbScbymKf-PCNQL3fVHw@mail.gmail.com>

On Sat, Nov 26, 2011 at 2:35 PM, anagio <nginx-forum at nginx.us> wrote:
> Thanks, is it possible to add a module after it's been compiled?
>

Nope. nginx must be recompiled to add/remove modules.


-- 
O< ascii ribbon campaign - stop html mail - www.asciiribbon.org


From agentzh at gmail.com  Sat Nov 26 08:08:05 2011
From: agentzh at gmail.com (agentzh)
Date: Sat, 26 Nov 2011 16:08:05 +0800
Subject: nginx-lua and nginx upload module
In-Reply-To: <fbab13973b47e4b42b5be26affae2d08.NginxMailingListEnglish@forum.nginx.org>
References: <fbab13973b47e4b42b5be26affae2d08.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <CAB4Tn6MHTU5YYi9f=c4Qb+h_dnf=o4vVxRC1Mo=CqDHgWTfSbQ@mail.gmail.com>

On Sat, Nov 26, 2011 at 3:36 PM, strawman <nginx-forum at nginx.us> wrote:
> Hi all,
>
> I'm hoping to write an nginx-lua script which handles file uploads from
> the nginx upload module. ?I can do this fine using an external backend
> and the upload_pass directive, but the lua module has no functionality
> for parsing the multipart form data that this generates. ?I've also
> tried accessing the $upload_file_name, $upload_tmp_path etc. variables
> generated by the upload module directly, but I don't seem to be able to
> access these outside of the upload_set_form_field directive.
>

Well, there will be a streaming request body reading API for ngx_lua
soon. For instance:

    local data, err = ngx.req.recv_body_data()

Atop this, we can easily build a multipart streaming parser in pure
Lua and no longer rely on the ngx_upload module to read and process
the request body. (The UI of ngx_upload is not so friendly for Lua
scripting and it'll be technically difficult to make them work
together without lots of line noises.)

Someone is already willing to sponsor this work but I'm too busy to
implement this lately :P

Regards,
-agentzh


From nginx at nginxuser.net  Sat Nov 26 09:16:47 2011
From: nginx at nginxuser.net (Nginx User)
Date: Sat, 26 Nov 2011 12:16:47 +0300
Subject: nginx-lua and nginx upload module
In-Reply-To: <fbab13973b47e4b42b5be26affae2d08.NginxMailingListEnglish@forum.nginx.org>
References: <fbab13973b47e4b42b5be26affae2d08.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <CAN=WzV0qTq3vNcAaNNe+w65KZZkNTCOaTu2Y2xJwh_4v_1gJqQ@mail.gmail.com>

On 26 November 2011 10:36, strawman <nginx-forum at nginx.us> wrote:
> Hi all,
>
> I'm hoping to write an nginx-lua script which handles file uploads from
> the nginx upload module. ?I can do this fine using an external backend
> and the upload_pass directive, but the lua module has no functionality
> for parsing the multipart form data that this generates. ?I've also
> tried accessing the $upload_file_name, $upload_tmp_path etc. variables
> generated by the upload module directly, but I don't seem to be able to
> access these outside of the upload_set_form_field directive.
>
> Is there any way to get the contents of those variables into a lua
> handler in a form that can be easily parsed?

Hi There,

I hit the multipart form limitation of the ngx_lua module recently and
wrote a lua module to do the job.

Basically I combined a number of functions from various cgilua modules
and added a few of my own to create a single module.

I have attached a copy and the instructions for calling it are added
as comments at the top of the module.

It will handle both url encoded and multipart encoded forms and return
the post data as a lua key/value pairs
-------------- next part --------------
A non-text attachment was scrubbed...
Name: form_parser_module.lua
Type: application/octet-stream
Size: 19610 bytes
Desc: not available
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20111126/725278c7/attachment-0001.obj>

From agentzh at gmail.com  Sat Nov 26 09:46:34 2011
From: agentzh at gmail.com (agentzh)
Date: Sat, 26 Nov 2011 17:46:34 +0800
Subject: nginx-lua and nginx upload module
In-Reply-To: <CAN=WzV0qTq3vNcAaNNe+w65KZZkNTCOaTu2Y2xJwh_4v_1gJqQ@mail.gmail.com>
References: <fbab13973b47e4b42b5be26affae2d08.NginxMailingListEnglish@forum.nginx.org>
	<CAN=WzV0qTq3vNcAaNNe+w65KZZkNTCOaTu2Y2xJwh_4v_1gJqQ@mail.gmail.com>
Message-ID: <CAB4Tn6P28wHFpNGYD8Y8XmoYcW6by3JHvJPwUb_9sLfNShZa9Q@mail.gmail.com>

On Sat, Nov 26, 2011 at 5:16 PM, Nginx User <nginx at nginxuser.net> wrote:
> Hi There,
>
> I hit the multipart form limitation of the ngx_lua module recently and
> wrote a lua module to do the job.
>
> Basically I combined a number of functions from various cgilua modules
> and added a few of my own to create a single module.
>
> I have attached a copy and the instructions for calling it are added
> as comments at the top of the module.
>
> It will handle both url encoded and multipart encoded forms and return
> the post data as a lua key/value pairs
>

This approach certainly works but it requires the request body to be
hold completely in memory, which may be unacceptable for big file
uploading.

Regards,
-agentzh


From nginx-forum at nginx.us  Sat Nov 26 10:03:24 2011
From: nginx-forum at nginx.us (talisto)
Date: Sat, 26 Nov 2011 05:03:24 -0500
Subject: limit issue, i think it's a bug
In-Reply-To: <ecb6bc9f6bbd03f1556f098eb801cf88.NginxMailingListEnglish@forum.nginx.org>
References: <ecb6bc9f6bbd03f1556f098eb801cf88.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <b696210fa7ce8c204aca0ce4b99b405e.NginxMailingListEnglish@forum.nginx.org>

I realize this is a two-year-old topic, but I'm having the same issue. 
It doesn't matter what I have the rate set to, it only allows about 2
requests per second.  This is my config:

in http:
limit_req_zone $binary_remote_addr zone=flood:10m rate=15r/s;

in server location:
limit_req zone=flood;

That should really be all that is necessary, right?  I'm using Nginx
1.1.6.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,25729,219093#msg-219093


From nginx at nginxuser.net  Sat Nov 26 10:07:36 2011
From: nginx at nginxuser.net (Nginx User)
Date: Sat, 26 Nov 2011 13:07:36 +0300
Subject: nginx-lua and nginx upload module
In-Reply-To: <CAB4Tn6P28wHFpNGYD8Y8XmoYcW6by3JHvJPwUb_9sLfNShZa9Q@mail.gmail.com>
References: <fbab13973b47e4b42b5be26affae2d08.NginxMailingListEnglish@forum.nginx.org>
	<CAN=WzV0qTq3vNcAaNNe+w65KZZkNTCOaTu2Y2xJwh_4v_1gJqQ@mail.gmail.com>
	<CAB4Tn6P28wHFpNGYD8Y8XmoYcW6by3JHvJPwUb_9sLfNShZa9Q@mail.gmail.com>
Message-ID: <CAN=WzV3SZ1ThQoNq8gM5tV0uZbbL96srf-PutXig7-wT3xZw6A@mail.gmail.com>

On 26 November 2011 12:46, agentzh <agentzh at gmail.com> wrote:
> This approach certainly works but it requires the request body to be
> hold completely in memory, which may be unacceptable for big file
> uploading.

Correct. This is why there is a default 2mb "maxinput" flag.

Hopefully a directive to handle multipart encoded forms will be added
to ngx_lua by the developer soon :)


From nginx at nginxuser.net  Sat Nov 26 10:17:30 2011
From: nginx at nginxuser.net (Nginx User)
Date: Sat, 26 Nov 2011 13:17:30 +0300
Subject: nginx-lua and nginx upload module
In-Reply-To: <CAN=WzV0qTq3vNcAaNNe+w65KZZkNTCOaTu2Y2xJwh_4v_1gJqQ@mail.gmail.com>
References: <fbab13973b47e4b42b5be26affae2d08.NginxMailingListEnglish@forum.nginx.org>
	<CAN=WzV0qTq3vNcAaNNe+w65KZZkNTCOaTu2Y2xJwh_4v_1gJqQ@mail.gmail.com>
Message-ID: <CAN=WzV1c8iF2n-AmtLiL0xgYAA_+r33Z1icdj7SFPrYmtCJ8cA@mail.gmail.com>

On 26 November 2011 12:16, Nginx User <nginx at nginxuser.net> wrote:
> I have attached a copy and the instructions for calling it are added
> as comments at the top of the module.
>
> It will handle both url encoded and multipart encoded forms and return
> the post data as a lua key/value pairs

Use this version instead
-------------- next part --------------
A non-text attachment was scrubbed...
Name: form_parser_module.lua
Type: application/octet-stream
Size: 19612 bytes
Desc: not available
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20111126/79214c7d/attachment-0001.obj>

From mdounin at mdounin.ru  Sat Nov 26 10:39:32 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Sat, 26 Nov 2011 14:39:32 +0400
Subject: limit issue, i think it's a bug
In-Reply-To: <b696210fa7ce8c204aca0ce4b99b405e.NginxMailingListEnglish@forum.nginx.org>
References: <ecb6bc9f6bbd03f1556f098eb801cf88.NginxMailingListEnglish@forum.nginx.org>
	<b696210fa7ce8c204aca0ce4b99b405e.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20111126103931.GA67687@mdounin.ru>

Hello!

On Sat, Nov 26, 2011 at 05:03:24AM -0500, talisto wrote:

> I realize this is a two-year-old topic, but I'm having the same issue. 
> It doesn't matter what I have the rate set to, it only allows about 2
> requests per second.  This is my config:
> 
> in http:
> limit_req_zone $binary_remote_addr zone=flood:10m rate=15r/s;
> 
> in server location:
> limit_req zone=flood;
> 
> That should really be all that is necessary, right?  I'm using Nginx
> 1.1.6.

How do you test it?

Maxim Dounin


From grails at jmsd.co.uk  Sat Nov 26 11:29:47 2011
From: grails at jmsd.co.uk (John Moore)
Date: Sat, 26 Nov 2011 11:29:47 +0000
Subject: nginx 1.1.0 not working on Windows 2000 Server
Message-ID: <4ED0CDAB.2070301@jmsd.co.uk>

A customer of mine is running an ancient Windows 2000 Server 
installation, on which I have been running an old version of nginx. I 
want to upgrade to a more recent version, but nginx 1.1.0 is not working 
for me. I get the following in the error log:

2011/11/26 11:30:56 [notice] 25552#26824: 
WSAIoctl(SIO_GET_EXTENSION_FUNCTION_POINTER, WSAID_TRANSMITPACKETS) 
failed (10022: An invalid argument was supplied)
2011/11/26 11:30:56 [notice] 25552#26824: 
WSAIoctl(SIO_GET_EXTENSION_FUNCTION_POINTER, WSAID_CONNECTEX) failed 
(10022: An invalid argument was supplied)
2011/11/26 11:30:56 [notice] 25552#26824: 
WSAIoctl(SIO_GET_EXTENSION_FUNCTION_POINTER, WSAID_DISCONNECTEX) failed 
(10022: An invalid argument was supplied)

No idea what this signifies. Is this fixable, or does anyone know of a 
recent nginx version which does run OK on such an old OS?


From grails at jmsd.co.uk  Sat Nov 26 11:46:40 2011
From: grails at jmsd.co.uk (John Moore)
Date: Sat, 26 Nov 2011 11:46:40 +0000
Subject: nginx 1.1.0 not working on Windows 2000 Server
In-Reply-To: <4ED0CDAB.2070301@jmsd.co.uk>
References: <4ED0CDAB.2070301@jmsd.co.uk>
Message-ID: <4ED0D1A0.6050202@jmsd.co.uk>

On 26/11/11 11:29, John Moore wrote:
> A customer of mine is running an ancient Windows 2000 Server 
> installation, on which I have been running an old version of nginx. I 
> want to upgrade to a more recent version, but nginx 1.1.0 is not 
> working for me. I get the following in the error log:
>
> 2011/11/26 11:30:56 [notice] 25552#26824: 
> WSAIoctl(SIO_GET_EXTENSION_FUNCTION_POINTER, WSAID_TRANSMITPACKETS) 
> failed (10022: An invalid argument was supplied)
> 2011/11/26 11:30:56 [notice] 25552#26824: 
> WSAIoctl(SIO_GET_EXTENSION_FUNCTION_POINTER, WSAID_CONNECTEX) failed 
> (10022: An invalid argument was supplied)
> 2011/11/26 11:30:56 [notice] 25552#26824: 
> WSAIoctl(SIO_GET_EXTENSION_FUNCTION_POINTER, WSAID_DISCONNECTEX) 
> failed (10022: An invalid argument was supplied)
>
> No idea what this signifies. Is this fixable, or does anyone know of a 
> recent nginx version which does run OK on such an old OS?
>

I installed the Kevn Worthington Cygwin-based build instead and this 
time it worked fine, so problem solved.


From nginx-forum at nginx.us  Sat Nov 26 12:22:51 2011
From: nginx-forum at nginx.us (talisto)
Date: Sat, 26 Nov 2011 07:22:51 -0500
Subject: limit issue, i think it's a bug
In-Reply-To: <20111126103931.GA67687@mdounin.ru>
References: <20111126103931.GA67687@mdounin.ru>
Message-ID: <b05605db79aefd7ba3ebafb7b4fcda30.NginxMailingListEnglish@forum.nginx.org>

Maxim Dounin Wrote:
> How do you test it?

I've been testing it with Siege, but I can actually generate error 503
responses just by refreshing my web browser fast enough.  My results
with Siege are somewhat random; sometimes the very first request fails,
then a couple will pass, then another will fail, all within a couple
seconds.  That's with my rate set to 15r/s, which should be more than
enough.. in Siege I'm only using 5 concurrent users with 1 request per
second, yet it still fails, often on the first or second hit.  In my
browser, it's a bit more reliable;  I have a simple page which makes 2
ajax requests; the 2nd ajax request will always fail.

As soon as I remove the limit_req line, I can flood my server with as
many requests as Siege can handle and it never generates a 503, so I
know that the errors aren't being caused by something else.  I'm not
sure why the limit_req isn't working properly though.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,25729,219104#msg-219104


From nginx-forum at nginx.us  Sat Nov 26 12:41:24 2011
From: nginx-forum at nginx.us (nviennot)
Date: Sat, 26 Nov 2011 07:41:24 -0500
Subject: Is there a bug in ngx_http_upstream_module's broken connection
	check code?
In-Reply-To: <0488794304f40cfbb44db6ebff747a42.NginxMailingListEnglish@forum.nginx.org>
References: <266e18e392bdbf122d245d5b240a6747.NginxMailingListEnglish@forum.nginx.org>
	<0488794304f40cfbb44db6ebff747a42.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <6f2ffe47d712aa859b938125ebf28b0f.NginxMailingListEnglish@forum.nginx.org>

I am experiencing this bug in a production system. (SSL closes are not
detected,  thus sockets stay in CLOSE_WAIT state forever -- nice DoS).

I was looking at nginx sources, and it seems that this bug has not been
fixed.
The alternative is to use stunnel with the X-Forwarded-For patch, but
that's way too messy.

In ngx_http_upstream_check_broken_connection(), there seems to be a
different path for kqueue. What about modifying the poll/epoll behavior
to detect disconnections for other event modules ? In
ngx_epoll_add_connection(), we can add the EPOLLHUP event, and mark the
connection as disconnected when processing HUP events instead of using
the buggy MSG_PEEK hack

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,209276,219105#msg-219105


From nginx-forum at nginx.us  Sat Nov 26 13:10:58 2011
From: nginx-forum at nginx.us (strawman)
Date: Sat, 26 Nov 2011 08:10:58 -0500
Subject: nginx-lua and nginx upload module
In-Reply-To: <fbab13973b47e4b42b5be26affae2d08.NginxMailingListEnglish@forum.nginx.org>
References: <fbab13973b47e4b42b5be26affae2d08.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <1e594ab2d21ee357596dca18bef215bd.NginxMailingListEnglish@forum.nginx.org>

Thanks for the swift replies, unfortunately I'm still having trouble
getting it to play nicely with the upload module.

Here's the config I'm using: http://pastebin.com/GsrYHwj0

It seems as though the request body isn't being read correctly;
ngx.req.get_body_data() works as expected if I run it from inside
/upload and make a multipart post to it, but it only returns the first
line of the body data when run from inside @handler.  echo_request_body
works fine either way.

Here's what I get when posting to it with the above config:

$ curl -F file=@TDXBN.gif http://localhost/upload
* post defs:
content_type: multipart/form-data;
boundary=----------------------------ff78acc43147
maxinput: 2097152
content_length: 358
maxfilesize: 1048576
args: 

* post args:

* request body:
------------------------------ff78acc43147

On a slightly different tack, I noticed one of the release announcements
for ngx_echo showed this directive being used:

echo_subrequest PUT /some_upstream/$upload_file_md5 -f $upload_tmp_path

Is there something in particular you need to do to get access to
$upload_tmp_path etc. outside of the upload_set_form_field directive? 
I've tried hacking this into working with echo_subrequest, but if i try
to use any of the $upload_* variables anywhere outside of that directive
then it doesn't work at all.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,219085,219107#msg-219107


From agentzh at gmail.com  Sat Nov 26 13:16:16 2011
From: agentzh at gmail.com (agentzh)
Date: Sat, 26 Nov 2011 21:16:16 +0800
Subject: nginx-lua and nginx upload module
In-Reply-To: <1e594ab2d21ee357596dca18bef215bd.NginxMailingListEnglish@forum.nginx.org>
References: <fbab13973b47e4b42b5be26affae2d08.NginxMailingListEnglish@forum.nginx.org>
	<1e594ab2d21ee357596dca18bef215bd.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <CAB4Tn6MNJBrwa42k+WkTQFf1MvC2PVt6qMx9YhAKTv04QqYHjw@mail.gmail.com>

On Sat, Nov 26, 2011 at 9:10 PM, strawman <nginx-forum at nginx.us> wrote:
> Thanks for the swift replies, unfortunately I'm still having trouble
> getting it to play nicely with the upload module.
>
> Here's the config I'm using: http://pastebin.com/GsrYHwj0
>
> It seems as though the request body isn't being read correctly;
> ngx.req.get_body_data() works as expected if I run it from inside
> /upload and make a multipart post to it, but it only returns the first
> line of the body data when run from inside @handler. ?echo_request_body
> works fine either way.
>

Sigh. I really hope that ngx_upload could be implemented as a rewrite
or access phase handler rather than a content handler, then we can
trivially combine with content_by_lua in the same location, for
example, and no longer need to do any kind of internal redirections
that clear everything including nginx variables and module contexts.

Maybe Valery can work on an alternative UI for his ngx_upload module? ;)

Regards,
-agentzh


From stef at caunter.ca  Sat Nov 26 13:21:16 2011
From: stef at caunter.ca (Stefan Caunter)
Date: Sat, 26 Nov 2011 08:21:16 -0500
Subject: limit issue, i think it's a bug
In-Reply-To: <b696210fa7ce8c204aca0ce4b99b405e.NginxMailingListEnglish@forum.nginx.org>
References: <ecb6bc9f6bbd03f1556f098eb801cf88.NginxMailingListEnglish@forum.nginx.org>
	<b696210fa7ce8c204aca0ce4b99b405e.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <CALdpnrwH=TnsPrGsurHHpyVUCHG7TYT65negM95Xwkiy-VWBpA@mail.gmail.com>

On Sat, Nov 26, 2011 at 5:03 AM, talisto <nginx-forum at nginx.us> wrote:
> I realize this is a two-year-old topic, but I'm having the same issue.
> It doesn't matter what I have the rate set to, it only allows about 2
> requests per second. ?This is my config:
>
> in http:
> limit_req_zone $binary_remote_addr zone=flood:10m rate=15r/s;
>
> in server location:
> limit_req zone=flood;
>
> That should really be all that is necessary, right? ?I'm using Nginx
> 1.1.6.
>

> Maxim Dounin Wrote:
> How do you test it?

> I've been testing it with Siege, but I can actually generate error 503
> responses just by refreshing my web browser fast enough.  My results
> with Siege are somewhat random; sometimes the very first request fails,
> then a couple will pass, then another will fail, all within a couple
> seconds.  That's with my rate set to 15r/s, which should be more than
> enough.. in Siege I'm only using 5 concurrent users with 1 request per
> second, yet it still fails, often on the first or second hit.  In my
> browser, it's a bit more reliable;  I have a simple page which makes 2
> ajax requests; the 2nd ajax request will always fail.

> As soon as I remove the limit_req line, I can flood my server with as
> many requests as Siege can handle and it never generates a 503, so I
> know that the errors aren't being caused by something else.  I'm not
> sure why the limit_req isn't working properly though.

You are truncating your response. Don't top post or do that, it makes
it difficult to track things or for us to help you.

>From a review of the documentation this is by design.
Siege with 5 concurrent users at 1r/s could easily exceed 15r/s if
there is other traffic or if it is not behaving predictably.
Are you sure there are no other requests to the site, and that the
browser test only makes 2 requests, and that siege only sends 10
requests? We need logs.
>From what you have sent, you will get 503 because you are limiting the
requests and have not defined burst. Set burst to nodelay and see if
it allows 15/s.
And you cannot claim bug without evidence; please see the DEBUG
README, show relevant configuration, define limit_req_log_level
correctly, and send the output.


Stefan Caunter
http://scaleengine.com/contact
"People who enjoy having meetings should never be allowed to be in
charge of anything."


From nginx at nginxuser.net  Sat Nov 26 13:24:57 2011
From: nginx at nginxuser.net (Nginx User)
Date: Sat, 26 Nov 2011 16:24:57 +0300
Subject: nginx-lua and nginx upload module
In-Reply-To: <CAB4Tn6MNJBrwa42k+WkTQFf1MvC2PVt6qMx9YhAKTv04QqYHjw@mail.gmail.com>
References: <fbab13973b47e4b42b5be26affae2d08.NginxMailingListEnglish@forum.nginx.org>
	<1e594ab2d21ee357596dca18bef215bd.NginxMailingListEnglish@forum.nginx.org>
	<CAB4Tn6MNJBrwa42k+WkTQFf1MvC2PVt6qMx9YhAKTv04QqYHjw@mail.gmail.com>
Message-ID: <CAN=WzV3yYhekZR493ZQR8p5ZABCgGaEJQsbRV6ZTWLWhN-SXNQ@mail.gmail.com>

On 26 November 2011 16:16, agentzh <agentzh at gmail.com> wrote:
> Sigh. I really hope that ngx_upload could be implemented as a rewrite
> or access phase handler rather than a content handler, then we can
> trivially combine with content_by_lua in the same location, for
> example, and no longer need to do any kind of internal redirections
> that clear everything including nginx variables and module contexts.

For your info, I believe agentzh may be refering, obliquely, to the
restriction in the last line here:
http://wiki.nginx.org/HttpLuaModule#content_by_lua (which also applies
to content_by_lua_file).


From nginx-forum at nginx.us  Sat Nov 26 13:55:46 2011
From: nginx-forum at nginx.us (WGH)
Date: Sat, 26 Nov 2011 08:55:46 -0500
Subject: Feature request: WebSocket proxying :-)
In-Reply-To: <4C8F37F6.6010704@gmail.com>
References: <4C8F37F6.6010704@gmail.com>
Message-ID: <61b7c5e5c8c146685357b592ac7096ba.NginxMailingListEnglish@forum.nginx.org>

Bump! HTTP/1.1 proxying is now being implemented in devel version of
nginx, for unfortunately important headers like "Upgrade" and
"Connection" are still stripped/replaced. 

>How about use my nginx_tcp_proxy_module?
>You can't use the same listening port with HTTP modules.
No.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,129658,219111#msg-219111


From nginx-forum at nginx.us  Sat Nov 26 15:21:47 2011
From: nginx-forum at nginx.us (utkung)
Date: Sat, 26 Nov 2011 10:21:47 -0500
Subject: Data synchronization (load balancing)
Message-ID: <77bb72e3fca0b63879cc31728d962644.NginxMailingListEnglish@forum.nginx.org>

Hi, I have a problem:
upstream backend {
			server 172.16.0.1:80;
			server 172.16.0.2:80;
}	
server {
			listen 80;
			server_name		www.domain.com;
			location / {
				proxy_pass	http://backend;
			}
		}
Now I have the model as above, but I don't know can configure to sync
data between 2 servers? please, help me ...

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,219113,219113#msg-219113


From stef at caunter.ca  Sat Nov 26 15:41:04 2011
From: stef at caunter.ca (Stefan Caunter)
Date: Sat, 26 Nov 2011 10:41:04 -0500
Subject: Data synchronization (load balancing)
In-Reply-To: <77bb72e3fca0b63879cc31728d962644.NginxMailingListEnglish@forum.nginx.org>
References: <77bb72e3fca0b63879cc31728d962644.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <CALdpnrxHsmrAeVzftWSrKxj+s04LjV2eXQU0Q8t-7WTVXJD4Zg@mail.gmail.com>

On Sat, Nov 26, 2011 at 10:21 AM, utkung <nginx-forum at nginx.us> wrote:
> Hi, I have a problem:
> upstream backend {
> ? ? ? ? ? ? ? ? ? ? ? ?server 172.16.0.1:80;
> ? ? ? ? ? ? ? ? ? ? ? ?server 172.16.0.2:80;
> }
> server {
> ? ? ? ? ? ? ? ? ? ? ? ?listen 80;
> ? ? ? ? ? ? ? ? ? ? ? ?server_name ? ? ? ? ? ? www.domain.com;
> ? ? ? ? ? ? ? ? ? ? ? ?location / {
> ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?proxy_pass ? ? ?http://backend;
> ? ? ? ? ? ? ? ? ? ? ? ?}
> ? ? ? ? ? ? ? ?}
> Now I have the model as above, but I don't know can configure to sync
> data between 2 servers? please, help me ...

Hardly an nginx problem, but whatever the backends are, use nfs for
the web space, or push your site code to every backend's web space.
Database should be accessible on an internal network.

---

Stefan Caunter
http://scaleengine.com/contact
"People who enjoy having meetings should never be allowed to be in
charge of anything."





>
> Posted at Nginx Forum: http://forum.nginx.org/read.php?2,219113,219113#msg-219113
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx
>


From nginx-forum at nginx.us  Sat Nov 26 15:59:37 2011
From: nginx-forum at nginx.us (utkung)
Date: Sat, 26 Nov 2011 10:59:37 -0500
Subject: Data synchronization (load balancing)
In-Reply-To: <77bb72e3fca0b63879cc31728d962644.NginxMailingListEnglish@forum.nginx.org>
References: <77bb72e3fca0b63879cc31728d962644.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <bc731e961c256cfb8220a461e05ada45.NginxMailingListEnglish@forum.nginx.org>

I am a student, and are learning more about nginx.
you can explain about this? sorry my knowledge is too little

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,219113,219116#msg-219116


From valery+nginxen at grid.net.ru  Sat Nov 26 20:14:44 2011
From: valery+nginxen at grid.net.ru (Valery Kholodkov)
Date: Sat, 26 Nov 2011 21:14:44 +0100
Subject: nginx-lua and nginx upload module
In-Reply-To: <CAB4Tn6MNJBrwa42k+WkTQFf1MvC2PVt6qMx9YhAKTv04QqYHjw@mail.gmail.com>
References: <fbab13973b47e4b42b5be26affae2d08.NginxMailingListEnglish@forum.nginx.org>
	<1e594ab2d21ee357596dca18bef215bd.NginxMailingListEnglish@forum.nginx.org>
	<CAB4Tn6MNJBrwa42k+WkTQFf1MvC2PVt6qMx9YhAKTv04QqYHjw@mail.gmail.com>
Message-ID: <1450E6A6-8B57-433E-A1BD-51748DD8794E@grid.net.ru>

The limitation on the phase that is used to process uploads at the moment comes from nginx itself and not from upload module.

If you describe what sort of API you need from upload module, I will be able to implement it.

-- 
Best regards,
Valery Kholodkov

On 26 Nov 2011, at 14:16, agentzh <agentzh at gmail.com> wrote:

> On Sat, Nov 26, 2011 at 9:10 PM, strawman <nginx-forum at nginx.us> wrote:
>> Thanks for the swift replies, unfortunately I'm still having trouble
>> getting it to play nicely with the upload module.
>> 
>> Here's the config I'm using: http://pastebin.com/GsrYHwj0
>> 
>> It seems as though the request body isn't being read correctly;
>> ngx.req.get_body_data() works as expected if I run it from inside
>> /upload and make a multipart post to it, but it only returns the first
>> line of the body data when run from inside @handler.  echo_request_body
>> works fine either way.
>> 
> 
> Sigh. I really hope that ngx_upload could be implemented as a rewrite
> or access phase handler rather than a content handler, then we can
> trivially combine with content_by_lua in the same location, for
> example, and no longer need to do any kind of internal redirections
> that clear everything including nginx variables and module contexts.
> 
> Maybe Valery can work on an alternative UI for his ngx_upload module? ;)
> 
> Regards,
> -agentzh
> 
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx


From mdounin at mdounin.ru  Sat Nov 26 20:28:31 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Sun, 27 Nov 2011 00:28:31 +0400
Subject: limit issue, i think it's a bug
In-Reply-To: <b05605db79aefd7ba3ebafb7b4fcda30.NginxMailingListEnglish@forum.nginx.org>
References: <20111126103931.GA67687@mdounin.ru>
	<b05605db79aefd7ba3ebafb7b4fcda30.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20111126202831.GC67687@mdounin.ru>

Hello!

On Sat, Nov 26, 2011 at 07:22:51AM -0500, talisto wrote:

> Maxim Dounin Wrote:
> > How do you test it?
> 
> I've been testing it with Siege, but I can actually generate error 503
> responses just by refreshing my web browser fast enough.  My results
> with Siege are somewhat random; sometimes the very first request fails,
> then a couple will pass, then another will fail, all within a couple
> seconds.  That's with my rate set to 15r/s, which should be more than
> enough.. in Siege I'm only using 5 concurrent users with 1 request per
> second, yet it still fails, often on the first or second hit.  In my
> browser, it's a bit more reliable;  I have a simple page which makes 2
> ajax requests; the 2nd ajax request will always fail.
> 
> As soon as I remove the limit_req line, I can flood my server with as
> many requests as Siege can handle and it never generates a 503, so I
> know that the errors aren't being caused by something else.  I'm not
> sure why the limit_req isn't working properly though.

The limit_req is expected to generate 503 as soon it sees more 
requests than burst set.  As you have no burst set, it will 
generate 503 as soon as it sees requests coming with less than 
1/15 second interval.  This is expected behaviour.

While flooding your your server with requests (with limit_req 
set), you should see about 15 successfull requests per second, 
others will return 503.

Maxim Dounin


From 911csj at gmail.com  Sat Nov 26 20:44:46 2011
From: 911csj at gmail.com (=?GB2312?B?vrTVxQ==?=)
Date: Sun, 27 Nov 2011 04:44:46 +0800
Subject: nginx behind load balancer
In-Reply-To: <CAPMF7wgD20dhJftsc_hQr63-30m+nVLpWQ9NxiuY7UBmcadTzQ@mail.gmail.com>
References: <CAPMF7wjD8BYcac6NVjmOMSEVfwnKsT-_rrqh2sRn8YexPKZ4sA@mail.gmail.com>
	<20111121123822.GW95664@mdounin.ru>
	<CAPMF7wggdXCT=EQH57gGgM6SWP-o47pBmJWnsbCgSBCSk0OfQQ@mail.gmail.com>
	<20111121124602.GX95664@mdounin.ru>
	<CAPMF7wgk00ZX4jWo2EAgL2Fr6etD7pr6uHJGVSvxdC9Sm_k00A@mail.gmail.com>
	<20111125171440.GX95664@mdounin.ru>
	<CAPMF7wiggTJZUYz-tdBvw_frbfvNYYggznVXY-+bfwGU6sepqA@mail.gmail.com>
	<CALdpnrzFdfxvP0dvGKHg4cLSzvUZ6HHBr53FRgzzPVKLvRKGmw@mail.gmail.com>
	<CAPMF7wiwWKqu43FnUJY4R36ubSWO7Wh3TEkfkR2MOq_3JSUJ=A@mail.gmail.com>
	<CAPMF7wgD20dhJftsc_hQr63-30m+nVLpWQ9NxiuY7UBmcadTzQ@mail.gmail.com>
Message-ID: <CAFipNWzCzh4Zczj=VeDcRPLbL3MRs-rKJdEZcSbJ6C9s1X+GyA@mail.gmail.com>

$remote_addr(OSI 3layer) and $http_x_forwarded_for(OSI 7layer) of these two
variables is different. behind LB server may be you can use config like
this.
http {
    map $proxy_add_x_forwarded_for $xff_pass {
        default             "xff403";
        ~^61\.152\.90\.8    "xffpass";
        ~^61\.172\.241\.    "xffpass";
    limit_req_zone          $http_x_forwarded_for zone=one:10m rate=200r/s;
    ...
    server {
        if ($xff_pass !~* xffpass) {
            return             403;
        }
        limit_req zone=one;
        ...
    }
}

2011/11/26 Rami Essaid <rami.essaid at gmail.com>

> As a quick update, it looks like this has happened before. The load
> balancer bounces off of several internal IP's sometimes and nginx picks the
> last one only. Does anyone know of a workaround to remove the last two
> trusted IPs from the x-forwarded-for header?
>
> http://forum.nginx.org/read.php?11,26102,214069
>
> Rami
>
> On Fri, Nov 25, 2011 at 4:54 PM, Rami Essaid <rami.essaid at gmail.com>wrote:
>
>> In looking at the $proxy_add_x_forwarded_for variable I believe that the
>> Loadbalancer is in fact passing the variables but nginx is taking the wrong
>> value?
>>
>> Here is what i get from the variables.
>> $proxy_add_x_forwarded_for: 217.27.244.18, 10.160.43.200, 10.160.43.200
>> $remote_addr: 10.160.43.200
>>
>> Does this mean that nginx is taking the last value in the
>> x-forwarded-for?
>>
>> To answer your other question, yes, the Amazon plenty of sockets
>> available to connect to nginx.
>>
>> The reason we added this was to be able to utilize Amazon's auto scaling
>> feature for redundancy and scalability.  We have not been able to test
>> performance yet since we are still trying to make it work but as soon as we
>> do I will be sure to report back.
>>
>>
>> On Fri, Nov 25, 2011 at 4:19 PM, Stefan Caunter <stef at caunter.ca> wrote:
>>
>>> On Fri, Nov 25, 2011 at 12:37 PM, Rami Essaid <rami.essaid at gmail.com>
>>> wrote:
>>> > Thanks for the help Maxim!  We disabled our limit_req and that seemed
>>> to
>>> > have fixed the problem.  Looking at the logs it seems that only 1/3 of
>>> the
>>> > requests are correctly getting the new IP assigned via the realIP
>>> module,
>>> > the remainder are still logging the load balancer IP.  This probably
>>> is more
>>> > of an issue with the amazon load balancer but do you have any idea on
>>> what
>>> > may be going on?
>>> > Also, where would you recommend as a place to start tracking and
>>> fixing the
>>> > other issue?
>>> >
>>>
>>> You have not established that load balancer is setting X-Forwarded-For
>>> or some other header to pass Real IP.
>>>
>>> Does the amazon pseudo device have enough sockets available to connect
>>> to nginx?
>>>
>>> What do you expect from adding this layer? Were you hitting
>>> performance limits? Has it improved or degraded performance?
>>>
>>> >
>>> >
>>> > On Fri, Nov 25, 2011 at 12:14 PM, Maxim Dounin <mdounin at mdounin.ru>
>>> wrote:
>>> >>
>>> >> Hello!
>>> >>
>>> >> On Fri, Nov 25, 2011 at 09:54:14AM -0500, Rami Essaid wrote:
>>> >>
>>> >> > Hi Maxim,
>>> >> >
>>> >> > We implemented to module and still had some trouble.  A lot of the
>>> >> > connections would return " 503 Service Temporarily Unavailable".
>>>  Our
>>> >> > configuration works fine without the load balancer but then gives
>>> these
>>> >> > 503
>>> >> > errors behind the load balancer.
>>> >>
>>> >> nginx itself will only return 503 if it hits either limit_conn or
>>> >> limit_req.
>>> >>
>>> >> If you see this returned by nginx, and it only happens with load
>>> >> balancer, this may indicate you've not configured realip module
>>> >> properly (or your load balancer doesn't provide appropriate
>>> >> headers) and you are hitting per-ip limits configured due to all
>>> >> requests appear to be from load balancer.
>>> >>
>>> >> Check if client's ip logged is really client's one, not an ip of
>>> >> your load balancer.
>>> >>
>>> >> > Looking into the error logs I notice a lot of these errors both
>>> with and
>>> >> > without the load balancer "connect() failed (111: Connection
>>> refused)
>>> >> > while
>>> >> > connecting to upstream".  Could this be the reason that we are
>>> having
>>> >> > issues?
>>> >>
>>> >> Unlikely, but It's a good idea to track and fix this in any case.
>>> >>
>>> >> Maxim Dounin
>>> >>
>>> >> >
>>> >> > Thanks!
>>> >> > Rami
>>> >> >
>>> >> >
>>> >> > On Mon, Nov 21, 2011 at 7:46 AM, Maxim Dounin <mdounin at mdounin.ru>
>>> >> > wrote:
>>> >> >
>>> >> > > Hello!
>>> >> > >
>>> >> > > On Mon, Nov 21, 2011 at 07:40:49AM -0500, Rami Essaid wrote:
>>> >> > >
>>> >> > > > Thanks Maxim,
>>> >> > > >
>>> >> > > > This looks like exactly what we need.  In your experience does
>>> this
>>> >> > > resolve
>>> >> > > > most issues behind a load balancer?
>>> >> > >
>>> >> > > Yes.
>>> >> > >
>>> >> > > Maxim Dounin
>>> >> > >
>>> >> > > > On Mon, Nov 21, 2011 at 7:38 AM, Maxim Dounin <
>>> mdounin at mdounin.ru>
>>> >> > > wrote:
>>> >> > > >
>>> >> > > > > Hello!
>>> >> > > > >
>>> >> > > > > On Mon, Nov 21, 2011 at 07:25:39AM -0500, Rami Essaid wrote:
>>> >> > > > >
>>> >> > > > > > Hi Guys,
>>> >> > > > > >
>>> >> > > > > > This weekend for scalability we tried putting our nginx
>>> servers
>>> >> > > behind
>>> >> > > > > > amazon's elastic load balancers and came across a road
>>> block: it
>>> >> > > does not
>>> >> > > > > > transparently pass the user IP and header information to
>>> nginx.
>>> >> > > > > > This
>>> >> > > > > caused
>>> >> > > > > > issues with several pieces of nginx we use including the IP
>>> >> > > > > > allow /
>>> >> > > deny
>>> >> > > > > > rules, the limit_req module, and the limit_con module.  Has
>>> >> > > > > > anyone
>>> >> > > > > > successfully put nginx behind a load balancer?  Any ideas
>>> on how
>>> >> > > > > > to
>>> >> > > make
>>> >> > > > > > this work?
>>> >> > > > >
>>> >> > > > > http://wiki.nginx.org/HttpRealIpModule
>>> >> > > > >
>>> >> > > > > 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
>>> >>
>>> >> _______________________________________________
>>> >> 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/20111127/d7abbaec/attachment.html>

From mdounin at mdounin.ru  Sat Nov 26 20:58:02 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Sun, 27 Nov 2011 00:58:02 +0400
Subject: nginx 1.1.0 not working on Windows 2000 Server
In-Reply-To: <4ED0CDAB.2070301@jmsd.co.uk>
References: <4ED0CDAB.2070301@jmsd.co.uk>
Message-ID: <20111126205802.GD67687@mdounin.ru>

Hello!

On Sat, Nov 26, 2011 at 11:29:47AM +0000, John Moore wrote:

> A customer of mine is running an ancient Windows 2000 Server
> installation, on which I have been running an old version of nginx.
> I want to upgrade to a more recent version, but nginx 1.1.0 is not
> working for me. I get the following in the error log:
> 
> 2011/11/26 11:30:56 [notice] 25552#26824:
> WSAIoctl(SIO_GET_EXTENSION_FUNCTION_POINTER, WSAID_TRANSMITPACKETS)
> failed (10022: An invalid argument was supplied)
> 2011/11/26 11:30:56 [notice] 25552#26824:
> WSAIoctl(SIO_GET_EXTENSION_FUNCTION_POINTER, WSAID_CONNECTEX) failed
> (10022: An invalid argument was supplied)
> 2011/11/26 11:30:56 [notice] 25552#26824:
> WSAIoctl(SIO_GET_EXTENSION_FUNCTION_POINTER, WSAID_DISCONNECTEX)
> failed (10022: An invalid argument was supplied)
> 
> No idea what this signifies. Is this fixable, or does anyone know of
> a recent nginx version which does run OK on such an old OS?

These messages are just say that TransmitPackets(), ConnectEx(), 
and DisconnectEx() functions aren't available on your OS.  It's 
harmless, nginx should work ok regardless of these messages 
(actually, functions in question aren't used as of now).

Maxim Dounin


From nginx at nginxuser.net  Sat Nov 26 21:17:33 2011
From: nginx at nginxuser.net (Nginx User)
Date: Sun, 27 Nov 2011 00:17:33 +0300
Subject: nginx-lua and nginx upload module
In-Reply-To: <1e594ab2d21ee357596dca18bef215bd.NginxMailingListEnglish@forum.nginx.org>
References: <fbab13973b47e4b42b5be26affae2d08.NginxMailingListEnglish@forum.nginx.org>
	<1e594ab2d21ee357596dca18bef215bd.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <CAN=WzV2DjQ4FAXpcSv2eaR4QZ6e9pv2Jti9eLcVS0bDyoB5B+g@mail.gmail.com>

On 26 November 2011 16:10, strawman <nginx-forum at nginx.us> wrote:
> echo_request_body works fine either way.

If it was possible to access echo_request_body from ngx_lua, then
could possibly should solve your issue as the form parser should be
able to use this with some modifications.


From mdounin at mdounin.ru  Sat Nov 26 21:57:18 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Sun, 27 Nov 2011 01:57:18 +0400
Subject: Feature request: WebSocket proxying :-)
In-Reply-To: <61b7c5e5c8c146685357b592ac7096ba.NginxMailingListEnglish@forum.nginx.org>
References: <4C8F37F6.6010704@gmail.com>
	<61b7c5e5c8c146685357b592ac7096ba.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20111126215718.GE67687@mdounin.ru>

Hello!

On Sat, Nov 26, 2011 at 08:55:46AM -0500, WGH wrote:

> Bump! HTTP/1.1 proxying is now being implemented in devel version of
> nginx, for unfortunately important headers like "Upgrade" and
> "Connection" are still stripped/replaced. 

It's not enough to talk HTTP/1.1 to proxy WebSockets.  It requires 
either talking WebSockets protocol or being able to establish 
opaque pipe between client and backend.

Maxim Dounin


From agentzh at gmail.com  Sat Nov 26 22:08:53 2011
From: agentzh at gmail.com (agentzh)
Date: Sun, 27 Nov 2011 06:08:53 +0800
Subject: nginx-lua and nginx upload module
In-Reply-To: <1450E6A6-8B57-433E-A1BD-51748DD8794E@grid.net.ru>
References: <fbab13973b47e4b42b5be26affae2d08.NginxMailingListEnglish@forum.nginx.org>
	<1e594ab2d21ee357596dca18bef215bd.NginxMailingListEnglish@forum.nginx.org>
	<CAB4Tn6MNJBrwa42k+WkTQFf1MvC2PVt6qMx9YhAKTv04QqYHjw@mail.gmail.com>
	<1450E6A6-8B57-433E-A1BD-51748DD8794E@grid.net.ru>
Message-ID: <CAB4Tn6M0mhXEDQsfEhdjOZScjiozjFjNK+tsktBTOjwO=B9d8Q@mail.gmail.com>

Hi, Valery!

On Sun, Nov 27, 2011 at 4:14 AM, Valery Kholodkov
<valery+nginxen at grid.net.ru> wrote:
> The limitation on the phase that is used to process uploads at the moment comes from nginx itself and not from upload module.
>

Happily we no longer have such limitations for nginx 0.8.54+ :)

> If you describe what sort of API you need from upload module, I will be able to implement it.
>

Great! Just a quick example from my head:

    location /upload {
        upload_in_access_phase;
        proxy_pass http://backend;
    }

That is, no longer introducing an internal location here for internal
redirections ;)

Thanks!
-agentzh


From mdounin at mdounin.ru  Sat Nov 26 23:38:58 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Sun, 27 Nov 2011 03:38:58 +0400
Subject: Is there a bug in ngx_http_upstream_module's broken connection
	check code?
In-Reply-To: <6f2ffe47d712aa859b938125ebf28b0f.NginxMailingListEnglish@forum.nginx.org>
References: <266e18e392bdbf122d245d5b240a6747.NginxMailingListEnglish@forum.nginx.org>
	<0488794304f40cfbb44db6ebff747a42.NginxMailingListEnglish@forum.nginx.org>
	<6f2ffe47d712aa859b938125ebf28b0f.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20111126233858.GF67687@mdounin.ru>

Hello!

On Sat, Nov 26, 2011 at 07:41:24AM -0500, nviennot wrote:

> I am experiencing this bug in a production system. (SSL closes are not
> detected,  thus sockets stay in CLOSE_WAIT state forever -- nice DoS).

I believe I've already stated this several times, but let me 
repeat: this is not a bug, it's rather lack of optimization.

And if this makes DoS possible in you setup, this means your setup 
is broken (i.e. there is a bug in your setup).  Even if nginx will 
be able to detect connection close in simple SSL case, this 
doesn't mean all connection closes will be detectable.  

Consider connection with a pipelined request not yet read from 
client: it won't be possible to detect pending connection close 
with standard interfaces (without reading the request first, which 
isn't an option).  You must use some timeouts/keepalives to handle 
this.

> I was looking at nginx sources, and it seems that this bug has not been
> fixed.
> The alternative is to use stunnel with the X-Forwarded-For patch, but
> that's way too messy.
> 
> In ngx_http_upstream_check_broken_connection(), there seems to be a
> different path for kqueue. What about modifying the poll/epoll behavior
> to detect disconnections for other event modules ? In
> ngx_epoll_add_connection(), we can add the EPOLLHUP event, and mark the
> connection as disconnected when processing HUP events instead of using
> the buggy MSG_PEEK hack

Likely, EPOLLRDHUP is needed here.  Patches are welcome.

Maxim Dounin


From mdounin at mdounin.ru  Sat Nov 26 23:53:07 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Sun, 27 Nov 2011 03:53:07 +0400
Subject: add_header for 206?
In-Reply-To: <EFC5A19DBF3A2E489904733B467D990B38AC71@office-server2.m247.local>
References: <EFC5A19DBF3A2E489904733B467D990B38AC71@office-server2.m247.local>
Message-ID: <20111126235307.GG67687@mdounin.ru>

Hello!

On Fri, Nov 25, 2011 at 10:36:21PM +0000, Richard Kearsley wrote:

> Quote from wiki:
> 
> Add_header
> "Adds headers to the HTTP response when the response code is equal to 200, 204, 301, 302 or 304."
> 
> Any reason we missing 206?

I tend to think it was just forgotten, probably because nginx's 
own range filter (which will generate 206 from 200) only appears 
later in headers filter chain.  Therefore nginx's own 206 will 
have headers added via add_header, while proxied 206 responses 
won't.

Trivial patch:

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
@@ -145,6 +145,7 @@ ngx_http_headers_filter(ngx_http_request
         || r != r->main
         || (r->headers_out.status != NGX_HTTP_OK
             && r->headers_out.status != NGX_HTTP_NO_CONTENT
+            && r->headers_out.status != NGX_HTTP_PARTIAL_CONTENT
             && r->headers_out.status != NGX_HTTP_MOVED_PERMANENTLY
             && r->headers_out.status != NGX_HTTP_MOVED_TEMPORARILY
             && r->headers_out.status != NGX_HTTP_NOT_MODIFIED))

Maxim Dounin


From xb.zhou at qq.com  Sun Nov 27 00:05:45 2011
From: xb.zhou at qq.com (=?ISO-8859-1?B?WkhPVSBYaWFvYm8=?=)
Date: Sun, 27 Nov 2011 08:05:45 +0800
Subject: positioning of nginx
Message-ID: <tencent_425C3AEE7066DA510AB9014C@qq.com>

We can deploy a web system like below:

1)   HAProxy <---->[optinal Varnish] <----->nginx+uWSGI(communicate locally)
2)   nginx <------> uWSGI(or cluster)

as nginx can 'cache' and 'proxy',  the positioning of nginx is confusing me.

my questions are:
1, is it wrong to that nginx talks to uWSGI locally?
2, is it proper of using nginx to take over haproxy+varnish?

thx

------------------
Sincerely yours
                         ZHOU Xiaobo

From nginx-forum at nginx.us  Sun Nov 27 06:07:16 2011
From: nginx-forum at nginx.us (pk899)
Date: Sun, 27 Nov 2011 01:07:16 -0500
Subject: Really, truly controlling Nginx error pages
Message-ID: <9d2dbc33b499e78f82ebebaf65b0aae5.NginxMailingListEnglish@forum.nginx.org>

I have googled and searched through this forum. 

But this syntax in the config file does NOT help: 


    # recursive_error_pages on;
    error_page 403 404 500 502 503 504  /error.htm;


It works sometimes, but often it just shows me the stock Nginx page. 

How do I avoid this once and for all, or is this impossible?

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,219133,219133#msg-219133


From varia at e-healthexpert.org  Sun Nov 27 08:56:24 2011
From: varia at e-healthexpert.org (Mark Alan)
Date: Sun, 27 Nov 2011 08:56:24 +0000
Subject: Integrating Nginx with polarssl instead of openssl
Message-ID: <20111127085624.70bfacc4@e-healthexpert.org>

In the domain of SSL libraries, while OpenSSL is like Apache, PolarSSL
is much like Nginx: more compact and more secure.

PolarSSL http://polarssl.org/ has compact modular code, small memory
footprint and is being favored in high security applications (the
integration in the OpenVPN project being an example of that).

I wonder, has anybody tried to compile Nginx linked to PolarSSL,
instead of OpenSSL ?


Regards,

Mark


From nginx at nginxuser.net  Sun Nov 27 14:01:39 2011
From: nginx at nginxuser.net (Nginx User)
Date: Sun, 27 Nov 2011 17:01:39 +0300
Subject: nginx-lua and nginx upload module
In-Reply-To: <fbab13973b47e4b42b5be26affae2d08.NginxMailingListEnglish@forum.nginx.org>
References: <fbab13973b47e4b42b5be26affae2d08.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <CAN=WzV2WEBML96r8CWqW1guokX3kcCh2TMD3tpOBh9wqU9PnOQ@mail.gmail.com>

On 26 November 2011 10:36, strawman <nginx-forum at nginx.us> wrote:
> Hi all,
>
> I'm hoping to write an nginx-lua script which handles file uploads from
> the nginx upload module. ?I can do this fine using an external backend
> and the upload_pass directive, but the lua module has no functionality
> for parsing the multipart form data that this generates. ?I've also
> tried accessing the $upload_file_name, $upload_tmp_path etc. variables
> generated by the upload module directly, but I don't seem to be able to
> access these outside of the upload_set_form_field directive.
>
> Is there any way to get the contents of those variables into a lua
> handler in a form that can be easily parsed?

Hi,

I have been intrigued by your question and been looking at various
possible solutions but then it struck me as odd as to why you would
want to do this in the first place.

A bit of trying to step out of a potential XY Problem situation. What
is it exactly that you are trying to achieve?


From nginx-forum at nginx.us  Sun Nov 27 16:20:17 2011
From: nginx-forum at nginx.us (jstalin)
Date: Sun, 27 Nov 2011 11:20:17 -0500
Subject: 404 error when %20 is in file name
Message-ID: <1556399b019620cce345b10469fd3c81.NginxMailingListEnglish@forum.nginx.org>

I'm running the latest version of nginx. When I try to serve a file with
a "%20" in the name (example funny%20picture.jpg) I get a 404 error. The
filename itself has the "%20" in it, not a space. But nginx refuses to
serve the file. Any ideas on a configuration change I can make so that
this works?

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,219147,219147#msg-219147


From igor at sysoev.ru  Sun Nov 27 17:21:46 2011
From: igor at sysoev.ru (Igor Sysoev)
Date: Sun, 27 Nov 2011 21:21:46 +0400
Subject: 404 error when %20 is in file name
In-Reply-To: <1556399b019620cce345b10469fd3c81.NginxMailingListEnglish@forum.nginx.org>
References: <1556399b019620cce345b10469fd3c81.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20111127172146.GA71537@nginx.com>

On Sun, Nov 27, 2011 at 11:20:17AM -0500, jstalin wrote:
> I'm running the latest version of nginx. When I try to serve a file with
> a "%20" in the name (example funny%20picture.jpg) I get a 404 error. The
> filename itself has the "%20" in it, not a space. But nginx refuses to
> serve the file. Any ideas on a configuration change I can make so that
> this works?

If you want to get "funny%20picture.jpg", you should encode "%" in request
as "%25": "funny%2520picture.jpg".


-- 
Igor Sysoev


From nginx-forum at nginx.us  Sun Nov 27 18:59:40 2011
From: nginx-forum at nginx.us (Veerle)
Date: Sun, 27 Nov 2011 13:59:40 -0500
Subject: Perl - awstats.pl config
Message-ID: <2a5d87b4603b5c07f2f1d55d51c719a6.NginxMailingListEnglish@forum.nginx.org>

Hello,

i try to implement a awstats.pl for every vhost on my nginx
installation.

I tried it the following:

location ~ ^/awstats/ {
  gzip off;
  include /etc/nginx/fastcgi_params;
  fastcgi_pass unix:/var/run/fcgiwrap.socket;
  fastcgi_param SCRIPT_FILENAME /usr/lib/cgi-bin/awstats.pl;
}

I think i'm to stupid for the location part, it works fine with \.pl$
but not with a path, it would be nice to could configure a "directoy"
like /awstats/ in every vhost and then runs the awstats.pl through the
socket (socket is working), but not the SCRIPT_FILENAME-part.

I only found documentions with a php-application who runs the
perl-file.

Thanks :/

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,219151,219151#msg-219151


From appa at perusio.net  Sun Nov 27 19:07:06 2011
From: appa at perusio.net (=?UTF-8?B?QW50w7NuaW8=?= P. P. Almeida)
Date: Sun, 27 Nov 2011 19:07:06 +0000
Subject: Perl - awstats.pl config
In-Reply-To: <2a5d87b4603b5c07f2f1d55d51c719a6.NginxMailingListEnglish@forum.nginx.org>
References: <2a5d87b4603b5c07f2f1d55d51c719a6.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <87ipm5mlk5.wl%appa@perusio.net>

On 27 Nov 2011 18h59 WET, nginx-forum at nginx.us wrote:

> Hello,
>
> i try to implement a awstats.pl for every vhost on my nginx
> installation.
>
> I tried it the following:
>
> location ~ ^/awstats/ {
> gzip off;
> include /etc/nginx/fastcgi_params;
> fastcgi_pass unix:/var/run/fcgiwrap.socket;
> fastcgi_param SCRIPT_FILENAME /usr/lib/cgi-bin/awstats.pl;
> }
>
> I think i'm to stupid for the location part, it works fine with
> \.pl$ but not with a path, it would be nice to could configure a
> "directoy" like /awstats/ in every vhost and then runs the
> awstats.pl through the socket (socket is working), but not the
> SCRIPT_FILENAME-part.
>
> I only found documentions with a php-application who runs the
> perl-file.
>
> Thanks :/
>

Try:

location = /awstats.pl {
    root /usr/lib/cgi-bin;     
    gzip off;
    include /etc/nginx/fastcgi_params;
    fastcgi_pass unix:/var/run/fcgiwrap.socket;
    # Not needed if already in fastcgi_params.
    fastcgi_param  SCRIPT_FILENAME $document_root$fastcgi_script_name;
}

--- appa


From nginx-forum at nginx.us  Sun Nov 27 21:14:17 2011
From: nginx-forum at nginx.us (nviennot)
Date: Sun, 27 Nov 2011 16:14:17 -0500
Subject: Is there a bug in ngx_http_upstream_module's broken connection
	check code?
In-Reply-To: <20111126233858.GF67687@mdounin.ru>
References: <20111126233858.GF67687@mdounin.ru>
Message-ID: <0f73acc956f113dc26c3d19da6381a5d.NginxMailingListEnglish@forum.nginx.org>

Consider the case where nginx is used as a reverse proxy, and upstream
is doing some server push (the http request never ends).
You are saying that timeouts/keepalive should be used. I would agree
with you, and I actually made a module for nginx+tcp_keepalive:
https://github.com/nviennot/nginx-tcp-keepalive

But that's not enough. The closed SSL connection stays open from nginx
point of view, even though it doesn't appear anymore in netstat.
On the other hand, when upstream sends something, nginx write() to the
closed socket will fail with EPIPE.
Maybe you were referring using keepalive by having upstream periodically
sending data to the client.
But that's adding a requirement on the upstream logic, which is why it's
not just an optimization problem, but an actual bug, when upstream rely
on nginx to close the connection when the client does to detect its
presence.

Bottom line: I'll send a patch with EPOLLRDHUP. Also feel free to
comment on the tcp-keepalive module.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,209276,219158#msg-219158


From nginx-forum at nginx.us  Sun Nov 27 22:25:58 2011
From: nginx-forum at nginx.us (Veerle)
Date: Sun, 27 Nov 2011 17:25:58 -0500
Subject: Perl - awstats.pl config
In-Reply-To: <87ipm5mlk5.wl%appa@perusio.net>
References: <87ipm5mlk5.wl%appa@perusio.net>
Message-ID: <42c01e90c7fc886123141b9eb30cd1fa.NginxMailingListEnglish@forum.nginx.org>

Thanks, works fine this way.

One question about the root-part.

Atm i have the root part inside the server {} like:
server {
 ...
  root /xxx/xxx/xxx/
  lcoation / {}
}

But now i need to set the root inside the location, but it seems i can't
override it. Do i really need to remove the root-line from the main
server-part and add it to every location?

Thanks

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,219151,219160#msg-219160


From nginx-forum at nginx.us  Sun Nov 27 22:37:25 2011
From: nginx-forum at nginx.us (Veerle)
Date: Sun, 27 Nov 2011 17:37:25 -0500
Subject: Perl - awstats.pl config
In-Reply-To: <42c01e90c7fc886123141b9eb30cd1fa.NginxMailingListEnglish@forum.nginx.org>
References: <87ipm5mlk5.wl%appa@perusio.net>
	<42c01e90c7fc886123141b9eb30cd1fa.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <da79e836617bc00b2f81d1e47e668b5d.NginxMailingListEnglish@forum.nginx.org>

Hello again, just tried something like that:

  location = /cgi-bin/awstats.pl {
    root /usr/lib/cgi-bin;
    gzip off;
    include /etc/nginx/fastcgi_params;
    fastcgi_pass unix:/var/run/fcgiwrap.socket;
    fastcgi_param SCRIPT_FILENAME /usr/lib/cgi-bin/awstats.pl;       
  }

But i only get a 403 as error.

Error message is:
 *415 FastCGI sent in stderr: "Cannot get script name, is DOCUMENT_ROOT
and SCRIPT_NAME set and is the script executable?" while reading
response header from upstream, client: 80.24.x.x, server: test, request:
"GET /cgi-bin/awstats.pl HTTP/1.1", upstream:
"fastcgi://unix:/var/run/fcgiwrap.socket:",

And why is the root-part needed? SCRIPT_FILENAME is not enough?

Thanks

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,219151,219161#msg-219161


From cyril.lavier at davromaniak.eu  Sun Nov 27 22:48:16 2011
From: cyril.lavier at davromaniak.eu (Cyril Lavier)
Date: Sun, 27 Nov 2011 23:48:16 +0100
Subject: Perl - awstats.pl config
In-Reply-To: <da79e836617bc00b2f81d1e47e668b5d.NginxMailingListEnglish@forum.nginx.org>
References: <87ipm5mlk5.wl%appa@perusio.net>
	<42c01e90c7fc886123141b9eb30cd1fa.NginxMailingListEnglish@forum.nginx.org>
	<da79e836617bc00b2f81d1e47e668b5d.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <4ED2BE30.9060102@davromaniak.eu>

Hi Veerle.

I don't know if this will help you, but I also run an awstats with 
nginx, and here is how I managed to make it work.

First, I made a symlink from /var/www/stats/awstats to /usr/lib/cgi-bin

Then, here is the configuration file (it's a virtual host) :

server {

     listen   80;
     listen   [::]:80;

     server_name  stats.yourwebsite.tld;

     access_log  /var/log/nginx/stats.yourwebsite.tld.access.log combined;
     error_log /var/log/nginx/stats.yourwebsite.tld.error.log;
     root   /var/www/stats;

     location ^~ /awstats-icon {
             alias /usr/share/awstats/icon/;
             access_log off;
     }

     location ^~ /awstatscss {
             alias /usr/share/doc/awstats/examples/css/;
             access_log off;
     }

     location ^~ /awstatsclasses {
             alias /usr/share/doc/awstats/examples/classes/;
                access_log off;
     }

     location ~ ^/.*\.pl$ {
             gzip off;
         fastcgi_pass  unix:/tmp/cgi.sock;
             fastcgi_index index.cgi;
             fastcgi_param SCRIPT_FILENAME  
/var/www/stats$fastcgi_script_name;
             fastcgi_param SCRIPT_NAME    $fastcgi_script_name;
             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_param GATEWAY_INTERFACE  CGI/1.1;
             fastcgi_param SERVER_SOFTWARE    nginx;
             fastcgi_param REQUEST_URI        $request_uri;
             fastcgi_param DOCUMENT_URI       $document_uri;
             fastcgi_param DOCUMENT_ROOT      $document_root;
             fastcgi_param SERVER_PROTOCOL    $server_protocol;
             fastcgi_param REMOTE_ADDR        $remote_addr;
             fastcgi_param REMOTE_PORT        $remote_port;
             fastcgi_param SERVER_ADDR        $server_addr;
                fastcgi_param SERVER_PORT        $server_port;
             fastcgi_param SERVER_NAME        $server_name;
         fastcgi_param REMOTE_USER        $remote_user;
     }

}

Then, you will be able to access your awstats by 
http://stats.yourwebsite.tld/awstats/awstats.pl

And don't hesitate to add authentication.

Thanks.

On 11/27/2011 11:37 PM, Veerle wrote:
> Hello again, just tried something like that:
>
>    location = /cgi-bin/awstats.pl {
>      root /usr/lib/cgi-bin;
>      gzip off;
>      include /etc/nginx/fastcgi_params;
>      fastcgi_pass unix:/var/run/fcgiwrap.socket;
>      fastcgi_param SCRIPT_FILENAME /usr/lib/cgi-bin/awstats.pl;
>    }
>
> But i only get a 403 as error.
>
> Error message is:
>   *415 FastCGI sent in stderr: "Cannot get script name, is DOCUMENT_ROOT
> and SCRIPT_NAME set and is the script executable?" while reading
> response header from upstream, client: 80.24.x.x, server: test, request:
> "GET /cgi-bin/awstats.pl HTTP/1.1", upstream:
> "fastcgi://unix:/var/run/fcgiwrap.socket:",
>
> And why is the root-part needed? SCRIPT_FILENAME is not enough?
>
> Thanks
>
> Posted at Nginx Forum: http://forum.nginx.org/read.php?2,219151,219161#msg-219161
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx


-- 
Cyril "Davromaniak" Lavier


From ru at nginx.com  Mon Nov 28 07:27:12 2011
From: ru at nginx.com (Ruslan Ermilov)
Date: Mon, 28 Nov 2011 11:27:12 +0400
Subject: nginx-1.1.8
In-Reply-To: <20111125100024.GH95664@mdounin.ru>
References: <20111114154403.GF33940@nginx.com>
	<CADKMG50iPH+H3S5Kefi3rTEkFL5pDTT6H0NxqB1GLJcJQq596Q@mail.gmail.com>
	<20111125100024.GH95664@mdounin.ru>
Message-ID: <20111128072712.GB46362@lo0.su>

On Fri, Nov 25, 2011 at 02:00:24PM +0400, Maxim Dounin wrote:
> Hello!
> 
> On Fri, Nov 25, 2011 at 11:03:39AM +0800, Delta Yeh wrote:
> 
> > 2011/11/14 Igor Sysoev <igor at sysoev.ru>:
> > > Changes with nginx 1.1.8 ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? 14 Nov 2011
> > >
> > > ? ?*) Change: the ngx_http_limit_zone_module was renamed to the
> > > ? ? ? ngx_http_limit_conn_module.
> > >
> > > ? ?*) Change: the "limit_zone" directive was superseded by the
> > > ? ? ? "limit_conn_zone" directive with a new syntax.
> > >
> > > ? ?*) Feature: support for multiple "limit_conn" limits on the same level.
> > >
> > 
> >  Usually we use   limit_zone module like:
> > http{
> > 
> >  limit_zone   one  $binary_remote_addr  10m;
> > server
> > {
> >    location /mypage{
> >       limit_conn   one  number_of connections;
> >   }
> > }
> > 
> > What is the  scenario that we should use multiple limit_conn ?
> 
> E.g. if you want to limit number of connections per ip *and* total 
> number of connections per virtual host.
> 
>     limit_conn_zone $binary_remote_addr zone=perip:10m;
>     limit_conn_zone $server_name zone=perserver:10m;
> 
>     server {
>         ...
>         limit_conn perip 10;
>         limit_conn perserver 100;
>     }
> 
> > Is there a way to limit total connections to one location ?
> 
> Yes.

I've just updated documentation with this particular example:

http://nginx.org/en/docs/http/ngx_http_limit_conn_module.html#limit_conn


From nginx-forum at nginx.us  Mon Nov 28 08:11:46 2011
From: nginx-forum at nginx.us (mennanov)
Date: Mon, 28 Nov 2011 03:11:46 -0500
Subject: Simple arithmetics in nginx config
Message-ID: <28ada9c6edc92f8b605674bff350b4b7.NginxMailingListEnglish@forum.nginx.org>

Hello! I have an image folder called "i" in site's root folder, it has
many subfolders like 0, 100, 200, 300 ... 600000, etc..
So in folder "200" there are images from 200.jpg to 299.jpg, in folder
"300" there are 300.jpg to 399.jpg ans so on. As you can see i create a
new folder for every 100th image because i don't want to have all the
images in one folder (there are too much).

The problem is that i can't calculate a proper folder name, i tried like
this:

location ~ /i/(.*)\.jpg {
	    set $folder $1-($1%100);
	    try_files /i/$folder/$1.jpg =400;
        }

but of course it does not work due to syntax error.

Help me please

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,219172,219172#msg-219172


From agentzh at gmail.com  Mon Nov 28 08:32:09 2011
From: agentzh at gmail.com (agentzh)
Date: Mon, 28 Nov 2011 16:32:09 +0800
Subject: Simple arithmetics in nginx config
In-Reply-To: <28ada9c6edc92f8b605674bff350b4b7.NginxMailingListEnglish@forum.nginx.org>
References: <28ada9c6edc92f8b605674bff350b4b7.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <CAB4Tn6OrczzYKfdBL+Pzbd3Q26Oit5uzN6hV8Kkr8nSOy7Qc0w@mail.gmail.com>

On Mon, Nov 28, 2011 at 4:11 PM, mennanov <nginx-forum at nginx.us> wrote:
>
> location ~ /i/(.*)\.jpg {
> ? ? ? ? ? ?set $folder $1-($1%100);
> ? ? ? ? ? ?try_files /i/$folder/$1.jpg =400;
> ? ? ? ?}
>
> but of course it does not work due to syntax error.
>

The standard "set" directive does not support arithmetic operations at
all. But it's an ideal use case for our ngx_lua module:

    location ~ /i/(.*)\.jpg {
       set_by_lua $folder 'return ngx.var[1] - (ngx.var[1] % 100)';
       try_files /i/$folder/$1.jpg =400;
    }

See http://wiki.nginx.org/HttpLuaModule for details.

Regards,
-agentzh


From andrew at nginx.com  Mon Nov 28 09:14:16 2011
From: andrew at nginx.com (Andrew Alexeev)
Date: Mon, 28 Nov 2011 13:14:16 +0400
Subject: List modules compiled with nginx in ec2?
In-Reply-To: <6c5e5433b01e5d55a11eb39aff1e55a9.NginxMailingListEnglish@forum.nginx.org>
References: <6c5e5433b01e5d55a11eb39aff1e55a9.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <A259A679-4A32-45AB-A8FF-5456C58C140C@nginx.com>

On Nov 26, 2011, at 11:25 AM, anagio wrote:

> I setup an amazon ec2 instance that had nginx pre-compiled is there a
> way to see what optional modules have been configured?

Do you mean an AMI with precompiled nginx? Which one is that?

> Thanks
> 
> Posted at Nginx Forum: http://forum.nginx.org/read.php?2,219082,219082#msg-219082
> 
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx


From valery+nginxen at grid.net.ru  Mon Nov 28 09:45:49 2011
From: valery+nginxen at grid.net.ru (Valery Kholodkov)
Date: Mon, 28 Nov 2011 09:45:49 +0000 (GMT)
Subject: nginx-lua and nginx upload module
In-Reply-To: <CAB4Tn6M0mhXEDQsfEhdjOZScjiozjFjNK+tsktBTOjwO=B9d8Q@mail.gmail.com>
Message-ID: <9183678.4921.1322473549620.JavaMail.root@zone.mtgsy.net>


----- agentzh <agentzh at gmail.com> wrote:
> Hi, Valery!
> 
> On Sun, Nov 27, 2011 at 4:14 AM, Valery Kholodkov
> <valery+nginxen at grid.net.ru> wrote:
> > The limitation on the phase that is used to process uploads at the moment comes from nginx itself and not from upload module.
> >
> 
> Happily we no longer have such limitations for nginx 0.8.54+ :)

Explain?

> 
> > If you describe what sort of API you need from upload module, I will be able to implement it.
> >
> 
> Great! Just a quick example from my head:
> 
>     location /upload {
>         upload_in_access_phase;
>         proxy_pass http://backend;
>     }
> 
> That is, no longer introducing an internal location here for internal
> redirections ;)

Doesn't look like API. We are talking about being able to use accelerated uploads feature from lua, aren't we?

-- 
Regards,
Valery Kholodkov


From agentzh at gmail.com  Mon Nov 28 09:51:41 2011
From: agentzh at gmail.com (agentzh)
Date: Mon, 28 Nov 2011 17:51:41 +0800
Subject: nginx-lua and nginx upload module
In-Reply-To: <9183678.4921.1322473549620.JavaMail.root@zone.mtgsy.net>
References: <CAB4Tn6M0mhXEDQsfEhdjOZScjiozjFjNK+tsktBTOjwO=B9d8Q@mail.gmail.com>
	<9183678.4921.1322473549620.JavaMail.root@zone.mtgsy.net>
Message-ID: <CAB4Tn6MBC5Xn1oRKJezEX4hhwNtsL5QiVEAVukMbBvsDtOwD6w@mail.gmail.com>

On Mon, Nov 28, 2011 at 5:45 PM, Valery Kholodkov
<valery+nginxen at grid.net.ru> wrote:
>>
>> Happily we no longer have such limitations for nginx 0.8.54+ :)
>
> Explain?
>

Now we can happily read and process the request body in both rewrite
and access phase handlers in nginx 0.8.54+. And ngx_lua is already
doing that for both the lua_need_request_body config directive and the
"ngx.req.read_body()" Lua API.

>> Great! Just a quick example from my head:
>>
>> ? ? location /upload {
>> ? ? ? ? upload_in_access_phase;
>> ? ? ? ? proxy_pass http://backend;
>> ? ? }
>>
>> That is, no longer introducing an internal location here for internal
>> redirections ;)
>
> Doesn't look like API. We are talking about being able to use accelerated uploads feature from lua, aren't we?
>

No, I think we're talking about making ngx_upload easier to work with
other nginx modules, especially with those registering a content
handler like ngx_lua, ngx_echo, ngx_proxy, and ngx_fastcgi.

The whole point here is to make ngx_upload read and process the
request body in an earlier phase like "rewrite" and "access" phases,
such that we can preserve the content handler for other modules and
can eliminate internal redirects altogether to reduce runtime cost.

Sorry for the confusions in my previous emails :)

Thanks!
-agentzh


From valery+nginxen at grid.net.ru  Mon Nov 28 10:22:25 2011
From: valery+nginxen at grid.net.ru (Valery Kholodkov)
Date: Mon, 28 Nov 2011 10:22:25 +0000 (GMT)
Subject: nginx-lua and nginx upload module
In-Reply-To: <CAB4Tn6MBC5Xn1oRKJezEX4hhwNtsL5QiVEAVukMbBvsDtOwD6w@mail.gmail.com>
Message-ID: <5133784.4937.1322475745829.JavaMail.root@zone.mtgsy.net>


----- agentzh <agentzh at gmail.com> wrote:
> On Mon, Nov 28, 2011 at 5:45 PM, Valery Kholodkov
> <valery+nginxen at grid.net.ru> wrote:
> >>
> >> Happily we no longer have such limitations for nginx 0.8.54+ :)
> >
> > Explain?
> >
> 
> Now we can happily read and process the request body in both rewrite
> and access phase handlers in nginx 0.8.54+. And ngx_lua is already
> doing that for both the lua_need_request_body config directive and the
> "ngx.req.read_body()" Lua API.

I think we've been able to do that even before. But that's not point. The problem is not in what particular phase upload module will run, but that it must run whenever some module tries to read the request body. This kind of customisation is not supported by nginx yet.

> 
> >> Great! Just a quick example from my head:
> >>
> >> ? ? location /upload {
> >> ? ? ? ? upload_in_access_phase;
> >> ? ? ? ? proxy_pass http://backend;
> >> ? ? }
> >>
> >> That is, no longer introducing an internal location here for internal
> >> redirections ;)
> >
> > Doesn't look like API. We are talking about being able to use accelerated uploads feature from lua, aren't we?
> >
> 
> No, I think we're talking about making ngx_upload easier to work with
> other nginx modules, especially with those registering a content
> handler like ngx_lua, ngx_echo, ngx_proxy, and ngx_fastcgi.
> 
> The whole point here is to make ngx_upload read and process the
> request body in an earlier phase like "rewrite" and "access" phases,
> such that we can preserve the content handler for other modules and
> can eliminate internal redirects altogether to reduce runtime cost.
> 
> Sorry for the confusions in my previous emails :)

Again, the point is not in what particular phase it will run... And by the way, the costs of internal redirects are negligible.

-- 
Regards,
Valery Kholodkov


From nginx-forum at nginx.us  Mon Nov 28 10:44:32 2011
From: nginx-forum at nginx.us (anagio)
Date: Mon, 28 Nov 2011 05:44:32 -0500
Subject: Is it possible to use a remote file to store servernames? server_name
	$somevariable-file?
Message-ID: <b24debaac687d74fec26c070859aacdc.NginxMailingListEnglish@forum.nginx.org>

I need to add a lot of sub-domains to a configuration file as users
register on our website our script needs to add the sub-domain. Is it
possible to use a variable or function in the config file so it loads a
remote file which has sub-domains listed one per line? It would be
easier for our developers to add a sub-domain one per line into a file.
Then trying to add them into the configuration file directly.

server_name  $some-function;

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,219193,219193#msg-219193


From ru at nginx.com  Mon Nov 28 11:51:09 2011
From: ru at nginx.com (Ruslan Ermilov)
Date: Mon, 28 Nov 2011 15:51:09 +0400
Subject: Is it possible to use a remote file to store servernames?
	server_name $somevariable-file?
In-Reply-To: <b24debaac687d74fec26c070859aacdc.NginxMailingListEnglish@forum.nginx.org>
References: <b24debaac687d74fec26c070859aacdc.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20111128115109.GB1529@lo0.su>

On Mon, Nov 28, 2011 at 05:44:32AM -0500, anagio wrote:
> I need to add a lot of sub-domains to a configuration file as users
> register on our website our script needs to add the sub-domain. Is it
> possible to use a variable or function in the config file so it loads a
> remote file which has sub-domains listed one per line? It would be
> easier for our developers to add a sub-domain one per line into a file.
> Then trying to add them into the configuration file directly.
> 
> server_name  $some-function;

http://nginx.org/en/docs/ngx_core_module.html#include


From nginx-forum at nginx.us  Mon Nov 28 11:52:59 2011
From: nginx-forum at nginx.us (anagio)
Date: Mon, 28 Nov 2011 06:52:59 -0500
Subject: Is it possible to use a remote file to store servernames?
	server_name $somevariable-file?
In-Reply-To: <20111128115109.GB1529@lo0.su>
References: <20111128115109.GB1529@lo0.su>
Message-ID: <c7358b0db0cb59d4e88768df32fc32bb.NginxMailingListEnglish@forum.nginx.org>

Isn't include only for nginx.conf file? How would I add this to the
sites-available/server file to include a remote file with list of
sub-domains?

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,219193,219202#msg-219202


From ru at nginx.com  Mon Nov 28 12:08:33 2011
From: ru at nginx.com (Ruslan Ermilov)
Date: Mon, 28 Nov 2011 16:08:33 +0400
Subject: Is it possible to use a remote file to store servernames?
	server_name $somevariable-file?
In-Reply-To: <c7358b0db0cb59d4e88768df32fc32bb.NginxMailingListEnglish@forum.nginx.org>
References: <20111128115109.GB1529@lo0.su>
	<c7358b0db0cb59d4e88768df32fc32bb.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20111128120833.GA476@lo0.su>

On Mon, Nov 28, 2011 at 06:52:59AM -0500, anagio wrote:
> Isn't include only for nginx.conf file? How would I add this to the
> sites-available/server file to include a remote file with list of
> sub-domains?

You can rsync/scp/curl/whatever the file in your preferable
format (with server names one per line), then generate an
nginx.conf syntax-compliant file with "server_name XXX;"
lines, then send the HUP signal (for details, please see
http://nginx.org/en/docs/control.html) to the master nginx
process, and have it update the configuration on the fly.
It's all pretty easy to achieve, and doesn't require adding
the "fetching" functionality to nginx.


From nginx-forum at nginx.us  Mon Nov 28 15:17:07 2011
From: nginx-forum at nginx.us (freshsun)
Date: Mon, 28 Nov 2011 10:17:07 -0500
Subject: client wrongly closed keepalive connection problem
Message-ID: <8da9c98d8593eb52ffa0d3d7ccad264a.NginxMailingListEnglish@forum.nginx.org>

Hi all, we are some college students maintaining a linux distribution
mirror. We are not experienced but now meet with
this problem.
There is a pxe system in our college to help every computer boot a linux
system or install one on Ethernet. Now most users 
cannot install, cause most of the packages will time out when being
downloaded.
We used lftp to mirror the directories, trying to find out the problem.
On this directory (on our mirror with same content)
http://ftp.debian.org/debian/pool/main/a/abiword/
lftp hints:

<--- Accept-Ranges: bytes
<--- 
---> HEAD /debian/pool/main/a/abiword/abiword_2.8.2.orig.tar.gz
HTTP/1.1
---> Host: mirrors.ustc.edu.cn
---> User-Agent: lftp/4.0.6
---> Accept: */*
---> Connection: keep-alive
---> 
<--- HTTP/1.1 200 OK
**** recv: <input-socket>: Connection reset by peer
---- Closing HTTP connection

sometimes might be 
"send:<output-socket>: Connection reset by peer"

and the nginx error.log shows"[info] 8864#0: *1661715 client xxxx closed
keepalive connection"

After trying mirror with lftp many times, we found that after request
around 100 files in this directory,
the server starts to send data back. And around receiving each 15-17
files, the reset error will show once,
until the server start to send out bigger files in this directory.

At first we suspect that it might because there are too many small files
at the start in this directory,
but when we move this directory to another host and mirror it, it is
just ok.

Following is our nginx conf without some not that import configuration:

user www-data;
worker_processes  16;
worker_cpu_affinity  xxxxx(just skip it);
worker_rlimit_nofile 65535;
events {
    use epoll;
    worker_connections  10000;
    #multi_accept off;
}
http {
    include       /etc/nginx/mime.types;
    access_log    /var/log/nginx/access.log;
    error_log  /var/log/nginx/error.log info;

    sendfile        on;
    tcp_nopush     on;

    keepalive_timeout  30;
    tcp_nodelay        off;(we also tried nopush off with nodelay on)
    limit_zone conn_per_ip $binary_remote_addr 10m;
}

server {
    root /srv/www/;
    autoindex on;
    limit_conn conn_per_ip 20;

    ## limit speed for MSIE UAs
    if ($http_user_agent ~ "MSIE" ) { 
        set $limit_rate 10k;
    } 
}

netstat -ant |wc -l :   4000-5000

sysctl.conf:
net.ipv4.tcp_max_syn_backlog = 65536
net.core.netdev_max_backlog = 32768
net.core.somaxconn = 32768
net.core.wmem_default = 8388608
net.core.rmem_default = 8388608
net.core.rmem_max = 16777216
net.core.wmem_max = 16777216
net.ipv4.tcp_timestamps = 0
net.ipv4.tcp_synack_retries = 2
net.ipv4.tcp_syn_retries = 2
net.ipv4.tcp_tw_recycle = 1
net.ipv4.tcp_tw_reuse = 1
net.ipv4.tcp_mem = 94500000 915000000 927000000
net.ipv4.tcp_max_orphans = 3276800

The nginx version we have tried: 1.0.5, 1.0.10, 0.7.67
Server system: debian squeeze.

Currently we found that limit_req might solve the problem, but is this
the right way? Because it
will restrict the performance.

Thank you for your help.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,219220,219220#msg-219220


From mdounin at mdounin.ru  Mon Nov 28 16:07:28 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Mon, 28 Nov 2011 20:07:28 +0400
Subject: nginx-1.1.9
Message-ID: <20111128160728.GQ67687@mdounin.ru>

Changes with nginx 1.1.9                                         28 Nov 2011

    *) Change: now double quotes are encoded in an "echo" SSI-command
       output.
       Thanks to Zaur Abasmirzoev.

    *) Feature: the "valid" parameter of the "resolver" directive. By
       default TTL returned by a DNS server is used.
       Thanks to Kirill A. Korinskiy.

    *) Bugfix: nginx might hang after a worker process abnormal termination.

    *) Bugfix: a segmentation fault might occur in a worker process if SNI
       was used; the bug had appeared in 1.1.2.

    *) Bugfix: in the "keepalive_disable" directive; the bug had appeared in
       1.1.8.
       Thanks to Alexander Usov.

    *) Bugfix: SIGWINCH signal did not work after first binary upgrade; the
       bug had appeared in 1.1.1.

    *) Bugfix: backend responses with length not matching "Content-Length"
       header line are no longer cached.

    *) Bugfix: in the "scgi_param" directive, if complex parameters were
       used.

    *) Bugfix: in the "epoll" event method.
       Thanks to Yichun Zhang.

    *) Bugfix: in the ngx_http_flv_module.
       Thanks to Piotr Sikora.

    *) Bugfix: in the ngx_http_mp4_module.

    *) Bugfix: IPv6 addresses are now handled properly in a request line and
       in a "Host" request header line.

    *) Bugfix: "add_header" and "expires" directives did not work if a
       request was proxied and response status code was 206.

    *) Bugfix: nginx could not be built on FreeBSD 10.

    *) Bugfix: nginx could not be built on AIX.


Maxim Dounin


From usirsiwal at verivue.com  Mon Nov 28 16:52:38 2011
From: usirsiwal at verivue.com (Umesh Sirsiwal)
Date: Mon, 28 Nov 2011 11:52:38 -0500
Subject: Dealing with buffered data with upstream generated response
Message-ID: <4ED3BC56.4000200@verivue.com>


I am developing a body filter which transforms outgoing stream. As part 
of the function, the incoming stream is copied in new buffer and sent. 
This in some cases results in a condition where my filter has busy 
buffers but the upper layers don't have any busy buffer. In our case the 
data is generated by upstream module. Since upstream module does not pay 
attention to connection->buffered, the output_filter is never called 
again to flush my busy buffers and the transfer just hangs.

Adding r->connection->buffered to the or condition solves the hang.

            if (u->out_bufs || u->busy_bufs) {
                 rc = ngx_http_output_filter(r, u->out_bufs);

                 if (rc == NGX_ERROR) {
                     ngx_http_upstream_finalize_request(r, u, 0);
                     return;
                 }

                 ngx_chain_update_chains(&u->free_bufs, &u->busy_bufs,
&u->out_bufs, u->output.tag);
             }

Is this nginx bug or am I missing something?

-Umesh


From mdounin at mdounin.ru  Mon Nov 28 18:13:49 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Mon, 28 Nov 2011 22:13:49 +0400
Subject: client wrongly closed keepalive connection problem
In-Reply-To: <8da9c98d8593eb52ffa0d3d7ccad264a.NginxMailingListEnglish@forum.nginx.org>
References: <8da9c98d8593eb52ffa0d3d7ccad264a.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20111128181349.GU67687@mdounin.ru>

Hello!

On Mon, Nov 28, 2011 at 10:17:07AM -0500, freshsun wrote:

> Hi all, we are some college students maintaining a linux distribution
> mirror. We are not experienced but now meet with
> this problem.
> There is a pxe system in our college to help every computer boot a linux
> system or install one on Ethernet. Now most users 
> cannot install, cause most of the packages will time out when being
> downloaded.
> We used lftp to mirror the directories, trying to find out the problem.
> On this directory (on our mirror with same content)
> http://ftp.debian.org/debian/pool/main/a/abiword/
> lftp hints:
> 
> <--- Accept-Ranges: bytes
> <--- 
> ---> HEAD /debian/pool/main/a/abiword/abiword_2.8.2.orig.tar.gz
> HTTP/1.1
> ---> Host: mirrors.ustc.edu.cn
> ---> User-Agent: lftp/4.0.6
> ---> Accept: */*
> ---> Connection: keep-alive
> ---> 
> <--- HTTP/1.1 200 OK
> **** recv: <input-socket>: Connection reset by peer
> ---- Closing HTTP connection
> 
> sometimes might be 
> "send:<output-socket>: Connection reset by peer"
> 
> and the nginx error.log shows"[info] 8864#0: *1661715 client xxxx closed
> keepalive connection"
> 
> After trying mirror with lftp many times, we found that after request
> around 100 files in this directory,
> the server starts to send data back. And around receiving each 15-17
> files, the reset error will show once,
> until the server start to send out bigger files in this directory.
> 
> At first we suspect that it might because there are too many small files
> at the start in this directory,
> but when we move this directory to another host and mirror it, it is
> just ok.

It looks like lftp uses pipelining excessively, you may try 
setting

    lingering_close always;

to see if it helps (available in nginx 1.1.0+, 1.0.6+).

Maxim Dounin


From matthieu.tourne at gmail.com  Mon Nov 28 18:57:59 2011
From: matthieu.tourne at gmail.com (Matthieu Tourne)
Date: Mon, 28 Nov 2011 10:57:59 -0800
Subject: nginx-1.1.9
In-Reply-To: <20111128160728.GQ67687@mdounin.ru>
References: <20111128160728.GQ67687@mdounin.ru>
Message-ID: <CAAGYmFdxvi+b6qEqohGwNPU_UEKjW4VJNMxRkHwgQEKfP+AMFg@mail.gmail.com>

Thanks, it looks like this new nginx has a lot of good fixes.

However it appears to be signed with a different key (
http://nginx.org/download/nginx-1.1.9.tar.gz.asc), than the previous
releases.


On Mon, Nov 28, 2011 at 8:07 AM, Maxim Dounin <mdounin at mdounin.ru> wrote:

> Changes with nginx 1.1.9                                         28 Nov
> 2011
>
>    *) Change: now double quotes are encoded in an "echo" SSI-command
>       output.
>       Thanks to Zaur Abasmirzoev.
>
>    *) Feature: the "valid" parameter of the "resolver" directive. By
>       default TTL returned by a DNS server is used.
>       Thanks to Kirill A. Korinskiy.
>
>    *) Bugfix: nginx might hang after a worker process abnormal termination.
>
>    *) Bugfix: a segmentation fault might occur in a worker process if SNI
>       was used; the bug had appeared in 1.1.2.
>
>    *) Bugfix: in the "keepalive_disable" directive; the bug had appeared in
>       1.1.8.
>       Thanks to Alexander Usov.
>
>    *) Bugfix: SIGWINCH signal did not work after first binary upgrade; the
>       bug had appeared in 1.1.1.
>
>    *) Bugfix: backend responses with length not matching "Content-Length"
>       header line are no longer cached.
>
>    *) Bugfix: in the "scgi_param" directive, if complex parameters were
>       used.
>
>    *) Bugfix: in the "epoll" event method.
>       Thanks to Yichun Zhang.
>
>    *) Bugfix: in the ngx_http_flv_module.
>       Thanks to Piotr Sikora.
>
>    *) Bugfix: in the ngx_http_mp4_module.
>
>    *) Bugfix: IPv6 addresses are now handled properly in a request line and
>       in a "Host" request header line.
>
>    *) Bugfix: "add_header" and "expires" directives did not work if a
>       request was proxied and response status code was 206.
>
>    *) Bugfix: nginx could not be built on FreeBSD 10.
>
>    *) Bugfix: nginx could not be built on AIX.
>
>
> 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/20111128/159d8c03/attachment.html>

From maxim at nginx.com  Mon Nov 28 19:03:09 2011
From: maxim at nginx.com (Maxim Konovalov)
Date: Mon, 28 Nov 2011 23:03:09 +0400
Subject: nginx-1.1.9
In-Reply-To: <CAAGYmFdxvi+b6qEqohGwNPU_UEKjW4VJNMxRkHwgQEKfP+AMFg@mail.gmail.com>
References: <20111128160728.GQ67687@mdounin.ru>
	<CAAGYmFdxvi+b6qEqohGwNPU_UEKjW4VJNMxRkHwgQEKfP+AMFg@mail.gmail.com>
Message-ID: <0AA03041-29D1-4136-AE0C-264ABFB94F20@nginx.com>

Right, correct. It is signed by Maxim Dounin, who handled the release process.

-- maxim


Nov 28, 2011, ? 10:57 PM, Matthieu Tourne <matthieu.tourne at gmail.com> ???????(?):

> Thanks, it looks like this new nginx has a lot of good fixes.
> 
> However it appears to be signed with a different key (http://nginx.org/download/nginx-1.1.9.tar.gz.asc), than the previous releases.
> 
> 
> On Mon, Nov 28, 2011 at 8:07 AM, Maxim Dounin <mdounin at mdounin.ru> wrote:
> Changes with nginx 1.1.9                                         28 Nov 2011
> 
>    *) Change: now double quotes are encoded in an "echo" SSI-command
>       output.
>       Thanks to Zaur Abasmirzoev.
> 
>    *) Feature: the "valid" parameter of the "resolver" directive. By
>       default TTL returned by a DNS server is used.
>       Thanks to Kirill A. Korinskiy.
> 
>    *) Bugfix: nginx might hang after a worker process abnormal termination.
> 
>    *) Bugfix: a segmentation fault might occur in a worker process if SNI
>       was used; the bug had appeared in 1.1.2.
> 
>    *) Bugfix: in the "keepalive_disable" directive; the bug had appeared in
>       1.1.8.
>       Thanks to Alexander Usov.
> 
>    *) Bugfix: SIGWINCH signal did not work after first binary upgrade; the
>       bug had appeared in 1.1.1.
> 
>    *) Bugfix: backend responses with length not matching "Content-Length"
>       header line are no longer cached.
> 
>    *) Bugfix: in the "scgi_param" directive, if complex parameters were
>       used.
> 
>    *) Bugfix: in the "epoll" event method.
>       Thanks to Yichun Zhang.
> 
>    *) Bugfix: in the ngx_http_flv_module.
>       Thanks to Piotr Sikora.
> 
>    *) Bugfix: in the ngx_http_mp4_module.
> 
>    *) Bugfix: IPv6 addresses are now handled properly in a request line and
>       in a "Host" request header line.
> 
>    *) Bugfix: "add_header" and "expires" directives did not work if a
>       request was proxied and response status code was 206.
> 
>    *) Bugfix: nginx could not be built on FreeBSD 10.
> 
>    *) Bugfix: nginx could not be built on AIX.
> 
> 
> 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 --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20111128/0c20c68d/attachment-0001.html>

From appa at perusio.net  Mon Nov 28 19:05:32 2011
From: appa at perusio.net (=?UTF-8?B?QW50w7NuaW8=?= P. P. Almeida)
Date: Mon, 28 Nov 2011 19:05:32 +0000
Subject: nginx-1.1.9
In-Reply-To: <CAAGYmFdxvi+b6qEqohGwNPU_UEKjW4VJNMxRkHwgQEKfP+AMFg@mail.gmail.com>
References: <20111128160728.GQ67687@mdounin.ru>
	<CAAGYmFdxvi+b6qEqohGwNPU_UEKjW4VJNMxRkHwgQEKfP+AMFg@mail.gmail.com>
Message-ID: <8762i4m5j7.wl%appa@perusio.net>

On 28 Nov 2011 18h57 WET, matthieu.tourne at gmail.com wrote:

> Thanks, it looks like this new nginx has a lot of good fixes.
>
> However it appears to be signed with a different key (
> http://nginx.org/download/nginx-1.1.9.tar.gz.asc), than the previous
> releases.

Yep my script bails out.

gpg: Signature made Seg 28 Nov 2011 15:21:49 WET using RSA key ID A1C052F8
gpg: Imposs?vel verificar assinatura: chave p?blica n?o encontrada

This release is signed by Maxim :)

gpg --recv-keys --keyserver pgp.mit.edu A1C052F8
gpg: requesting key A1C052F8 from hkp server pgp.mit.edu
gpg: key A1C052F8: public key "Maxim Dounin <mdounin at mdounin.ru>" imported
gpg: 3 marginal(s) needed, 1 complete(s) needed, PGP trust model
gpg: depth: 0  valid:   1  signed:   0  trust: 0-, 0q, 0n, 0m, 0f, 1u
gpg: proxima verifica??o da base de dados de confian?a a 2020-07-30
gpg: N?mero total processado: 1
gpg:               importados: 1  (RSA: 1

Now it works.

--- appa


From kworthington at gmail.com  Tue Nov 29 01:23:17 2011
From: kworthington at gmail.com (Kevin Worthington)
Date: Mon, 28 Nov 2011 20:23:17 -0500
Subject: [nginx-announce] nginx-1.1.9
In-Reply-To: <20111128160244.GO67687@mdounin.ru>
References: <20111128160244.GO67687@mdounin.ru>
Message-ID: <CAGo79UVVqxNzeRGUgqOPSS5U7Ao0yZXwhADEcJp743_P=qK3Vw@mail.gmail.com>

Hello Nginx Users,

Just released: Nginx 1.1.9 For Windows http://goo.gl/ytQhC (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

Thank you,
Kevin
--
Kevin Worthington
kworthington ~at] gmail [dot} .C0M~
http://www.kevinworthington.com/


On Mon, Nov 28, 2011 at 11:02 AM, Maxim Dounin <mdounin at mdounin.ru> wrote:
> Changes with nginx 1.1.9 ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? 28 Nov 2011
>
> ? ?*) Change: now double quotes are encoded in an "echo" SSI-command
> ? ? ? output.
> ? ? ? Thanks to Zaur Abasmirzoev.
>
> ? ?*) Feature: the "valid" parameter of the "resolver" directive. By
> ? ? ? default TTL returned by a DNS server is used.
> ? ? ? Thanks to Kirill A. Korinskiy.
>
> ? ?*) Bugfix: nginx might hang after a worker process abnormal termination.
>
> ? ?*) Bugfix: a segmentation fault might occur in a worker process if SNI
> ? ? ? was used; the bug had appeared in 1.1.2.
>
> ? ?*) Bugfix: in the "keepalive_disable" directive; the bug had appeared in
> ? ? ? 1.1.8.
> ? ? ? Thanks to Alexander Usov.
>
> ? ?*) Bugfix: SIGWINCH signal did not work after first binary upgrade; the
> ? ? ? bug had appeared in 1.1.1.
>
> ? ?*) Bugfix: backend responses with length not matching "Content-Length"
> ? ? ? header line are no longer cached.
>
> ? ?*) Bugfix: in the "scgi_param" directive, if complex parameters were
> ? ? ? used.
>
> ? ?*) Bugfix: in the "epoll" event method.
> ? ? ? Thanks to Yichun Zhang.
>
> ? ?*) Bugfix: in the ngx_http_flv_module.
> ? ? ? Thanks to Piotr Sikora.
>
> ? ?*) Bugfix: in the ngx_http_mp4_module.
>
> ? ?*) Bugfix: IPv6 addresses are now handled properly in a request line and
> ? ? ? in a "Host" request header line.
>
> ? ?*) Bugfix: "add_header" and "expires" directives did not work if a
> ? ? ? request was proxied and response status code was 206.
>
> ? ?*) Bugfix: nginx could not be built on FreeBSD 10.
>
> ? ?*) Bugfix: nginx could not be built on AIX.
>
>
> Maxim Dounin
>
> _______________________________________________
> nginx-announce mailing list
> nginx-announce at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx-announce
>


From agentzh at gmail.com  Tue Nov 29 06:01:23 2011
From: agentzh at gmail.com (agentzh)
Date: Tue, 29 Nov 2011 14:01:23 +0800
Subject: nginx-lua and nginx upload module
In-Reply-To: <5133784.4937.1322475745829.JavaMail.root@zone.mtgsy.net>
References: <CAB4Tn6MBC5Xn1oRKJezEX4hhwNtsL5QiVEAVukMbBvsDtOwD6w@mail.gmail.com>
	<5133784.4937.1322475745829.JavaMail.root@zone.mtgsy.net>
Message-ID: <CAB4Tn6PEtGMTH=vRrJmEOW95Pnz928CX7L4Zw=MP3DwF=KbqWA@mail.gmail.com>

On Mon, Nov 28, 2011 at 6:22 PM, Valery Kholodkov
<valery+nginxen at grid.net.ru> wrote:
>
> I think we've been able to do that even before.

Sadly for nginx 0.8.41 ~ 0.8.53, "IO interruptions" are explicitly
prohibited in rewrite phase handlers and for exactly the same reason
your ngx_eval module does not work at all for these versions of Nginx
:)

> But that's not point. The problem is not in what particular phase upload module will run, but that it must run whenever some module tries to read the request body. This kind of customisation is not supported by nginx yet.
>

True. We definitely need an input filter mechanism for Nginx. I've
cc'd Andrew Alexeev. Maybe he can help make this happen in the Nginx
core ;)

>
> Again, the point is not in what particular phase it will run... And by the way, the costs of internal redirects are negligible.
>

Not really if the user has a lot of regex style locations defined in
his nginx.conf. For now, the nginx core does not (yet) utilize the
PCRE JIT feature for its pattern matching, so it does consume quite a
few CPU cycles for complicated location patterns ;) And...if we *can*
eliminate the cost of internal redirects altogether, why shouldn't we
make that happen? ;)

Also, I believe it can simplify the user config file greatly and does
not cause variable scope problems (like $uri and $args' scope). I know
there is a upload_pass_args directive to forward the original URI
query args, but if the $args scope issue does not exist in the first
place, we do not need this config directive at all.

BTW, I notice that if ngx_upload's special variables like
$upload_file_md5 are used in a wrong context, it will simply crash the
nginx worker (segmentation faults) due to null module context pointer
access (at least for the latest 2.2.0 release). Maybe it's more
reasonable to check the ctx pointer and print out a helpful error
message in functions like ngx_http_upload_md5_variable?

Best,
-agentzh


From nginx-forum at nginx.us  Tue Nov 29 07:41:27 2011
From: nginx-forum at nginx.us (unai)
Date: Tue, 29 Nov 2011 02:41:27 -0500
Subject: NGINX server_name issues
Message-ID: <bd9cc3c456a239ad1f1bf1a433fd9feb.NginxMailingListEnglish@forum.nginx.org>

I have the following simple server block on NGINX:

 server {
        listen 80;
        listen 8090;
        server_name st.rdsx.net;
        autoindex on;

        root /home/users/saifbechan/st-rdsx-net;
    }

After I include the relevant settings on my hosts file I get the
following (unexpected) behavior:

(1) http://st.rdsx.net/ and http://st.rdsx.net:8090/ work fine;
(2) http://st.rdsx.net:8090/future-cell-phone-technology-01-150x150.jpg
works;
(3) http://st.rdsx.net/future-cell-phone-technology-01-150x150.jpg ->
ERROR! "The connection was reset"

I've been troubleshooting (3) for a couple hours and I'm unable to
identify the culprit. I'm running NGINX 1.0.10 (latest stable) on Debian
6.0.2 32 bits.

This NGINX instance runs another 40 or 50 sites with no problems.

Could anyone enlighten me? Any thoughts?

Thank you so much!
unai

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,219257,219257#msg-219257


From igor at sysoev.ru  Tue Nov 29 07:57:34 2011
From: igor at sysoev.ru (Igor Sysoev)
Date: Tue, 29 Nov 2011 11:57:34 +0400
Subject: NGINX server_name issues
In-Reply-To: <bd9cc3c456a239ad1f1bf1a433fd9feb.NginxMailingListEnglish@forum.nginx.org>
References: <bd9cc3c456a239ad1f1bf1a433fd9feb.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20111129075734.GE22752@nginx.com>

On Tue, Nov 29, 2011 at 02:41:27AM -0500, unai wrote:
> I have the following simple server block on NGINX:
> 
>  server {
>         listen 80;
>         listen 8090;
>         server_name st.rdsx.net;
>         autoindex on;
> 
>         root /home/users/saifbechan/st-rdsx-net;
>     }
> 
> After I include the relevant settings on my hosts file I get the
> following (unexpected) behavior:
> 
> (1) http://st.rdsx.net/ and http://st.rdsx.net:8090/ work fine;
> (2) http://st.rdsx.net:8090/future-cell-phone-technology-01-150x150.jpg
> works;
> (3) http://st.rdsx.net/future-cell-phone-technology-01-150x150.jpg ->
> ERROR! "The connection was reset"
> 
> I've been troubleshooting (3) for a couple hours and I'm unable to
> identify the culprit. I'm running NGINX 1.0.10 (latest stable) on Debian
> 6.0.2 32 bits.
> 
> This NGINX instance runs another 40 or 50 sites with no problems.
> 
> Could anyone enlighten me? Any thoughts?

Could you try debug log of the request ?

http://nginx.org/en/docs/debugging_log.html


-- 
Igor Sysoev


From nbubingo at gmail.com  Tue Nov 29 08:35:48 2011
From: nbubingo at gmail.com (=?GB2312?B?0qbOsLHz?=)
Date: Tue, 29 Nov 2011 16:35:48 +0800
Subject: [ANNOUNCE] nginx_cross_origin_module - support cross origin resource
	sharing protocol in Nginx
Message-ID: <CAGxuLK2UCHQU=DMXuu0RFOcY3fbir9S4eUXYBm1gMEngSfgs2g@mail.gmail.com>

Hi, folks,

I'm pleased to release the nginx_cross_origin_module.

This module can process the cross-origin resource sharing Javascript
request with this protocol (<http://www.w3.org/TR/cors/>). This module
 follows the protocol version of 20100727.

The project site: https://github.com/yaoweibin/nginx_cross_origin_module

Synopsis
  a simple example:
    http {

        cors on;
        cors_max_age     3600;
        cors_origin_list unbounded;
        cors_method_list GET HEAD PUT POST;
        cors_header_list unbounded;

        server {
            listen       80;
            server_name  localhost;

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


From nginx-forum at nginx.us  Tue Nov 29 08:35:52 2011
From: nginx-forum at nginx.us (mennanov)
Date: Tue, 29 Nov 2011 03:35:52 -0500
Subject: Simple arithmetics in nginx config
In-Reply-To: <CAB4Tn6OrczzYKfdBL+Pzbd3Q26Oit5uzN6hV8Kkr8nSOy7Qc0w@mail.gmail.com>
References: <CAB4Tn6OrczzYKfdBL+Pzbd3Q26Oit5uzN6hV8Kkr8nSOy7Qc0w@mail.gmail.com>
Message-ID: <0de8164f2dedd16bc1787a0028a3ba8b.NginxMailingListEnglish@forum.nginx.org>

agentzh Wrote:
-------------------------------------------------------
> On Mon, Nov 28, 2011 at 4:11 PM, mennanov
> <nginx-forum at nginx.us> wrote:
> >
> > location ~ /i/(.*)\.jpg {
> > ? ? ? ? ? ?set $folder $1-($1%100);
> > ? ? ? ? ? ?try_files /i/$folder/$1.jpg
> =400;
> > ? ? ? ?}
> >
> > but of course it does not work due to syntax
> error.
> >
> 
> The standard "set" directive does not support
> arithmetic operations at
> all. But it's an ideal use case for our ngx_lua
> module:
> 
>     location ~ /i/(.*)\.jpg {
>        set_by_lua $folder 'return ngx.var[1] -
> (ngx.var[1] % 100)';
>        try_files /i/$folder/$1.jpg =400;
>     }
> 
> See http://wiki.nginx.org/HttpLuaModule for
> details.
> 
> Regards,
> -agentzh
> 
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx

Thank you so much, i've just recompiled nginx with this module and it
just works!

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,219172,219260#msg-219260


From rami.essaid at gmail.com  Tue Nov 29 09:37:24 2011
From: rami.essaid at gmail.com (Rami Essaid)
Date: Tue, 29 Nov 2011 04:37:24 -0500
Subject: nginx behind load balancer
In-Reply-To: <CAFipNWzCzh4Zczj=VeDcRPLbL3MRs-rKJdEZcSbJ6C9s1X+GyA@mail.gmail.com>
References: <CAPMF7wjD8BYcac6NVjmOMSEVfwnKsT-_rrqh2sRn8YexPKZ4sA@mail.gmail.com>
	<20111121123822.GW95664@mdounin.ru>
	<CAPMF7wggdXCT=EQH57gGgM6SWP-o47pBmJWnsbCgSBCSk0OfQQ@mail.gmail.com>
	<20111121124602.GX95664@mdounin.ru>
	<CAPMF7wgk00ZX4jWo2EAgL2Fr6etD7pr6uHJGVSvxdC9Sm_k00A@mail.gmail.com>
	<20111125171440.GX95664@mdounin.ru>
	<CAPMF7wiggTJZUYz-tdBvw_frbfvNYYggznVXY-+bfwGU6sepqA@mail.gmail.com>
	<CALdpnrzFdfxvP0dvGKHg4cLSzvUZ6HHBr53FRgzzPVKLvRKGmw@mail.gmail.com>
	<CAPMF7wiwWKqu43FnUJY4R36ubSWO7Wh3TEkfkR2MOq_3JSUJ=A@mail.gmail.com>
	<CAPMF7wgD20dhJftsc_hQr63-30m+nVLpWQ9NxiuY7UBmcadTzQ@mail.gmail.com>
	<CAFipNWzCzh4Zczj=VeDcRPLbL3MRs-rKJdEZcSbJ6C9s1X+GyA@mail.gmail.com>
Message-ID: <CAPMF7wio1pTdJ3gz9T_YN8csZZ0tm4htEGT60=a_=MDbKOwwAg@mail.gmail.com>

Thanks for the note and the clever workaround. We were able to tweak this
to work with but then it still left a lot of the other functions we were
using such as deny/allow, limit_con, ect not working.  Instead we went back
to amazon and it turns out they were able to correct the behavior of their
load balancer.

I wanted to report back on the performance of putting nginx behind an elb.
 We compared elb to haproxy and on amazon's cloud we got better performance
through the elb than we did an haproxy instance. There is minimal impact on
end user performance for adding this extra step.  We did this for
redundancy to allow us to automatically fail over to another zone if the
current zone or instance go down.

Happy to answer any questions about the setup.

Rami


On Sat, Nov 26, 2011 at 3:44 PM, ?? <911csj at gmail.com> wrote:

> $remote_addr(OSI 3layer) and $http_x_forwarded_for(OSI 7layer) of these
> two variables is different. behind LB server may be you can use config like
> this.
> http {
>     map $proxy_add_x_forwarded_for $xff_pass {
>         default             "xff403";
>         ~^61\.152\.90\.8    "xffpass";
>         ~^61\.172\.241\.    "xffpass";
>     limit_req_zone          $http_x_forwarded_for zone=one:10m rate=200r/s;
>     ...
>     server {
>         if ($xff_pass !~* xffpass) {
>             return             403;
>         }
>         limit_req zone=one;
>         ...
>
>     }
> }
>
> 2011/11/26 Rami Essaid <rami.essaid at gmail.com>
>
>> As a quick update, it looks like this has happened before. The load
>> balancer bounces off of several internal IP's sometimes and nginx picks the
>> last one only. Does anyone know of a workaround to remove the last two
>> trusted IPs from the x-forwarded-for header?
>>
>> http://forum.nginx.org/read.php?11,26102,214069
>>
>> Rami
>>
>> On Fri, Nov 25, 2011 at 4:54 PM, Rami Essaid <rami.essaid at gmail.com>wrote:
>>
>>> In looking at the $proxy_add_x_forwarded_for variable I believe that the
>>> Loadbalancer is in fact passing the variables but nginx is taking the wrong
>>> value?
>>>
>>> Here is what i get from the variables.
>>> $proxy_add_x_forwarded_for: 217.27.244.18, 10.160.43.200, 10.160.43.200
>>>  $remote_addr: 10.160.43.200
>>>
>>> Does this mean that nginx is taking the last value in the
>>> x-forwarded-for?
>>>
>>> To answer your other question, yes, the Amazon plenty of sockets
>>> available to connect to nginx.
>>>
>>> The reason we added this was to be able to utilize Amazon's auto scaling
>>> feature for redundancy and scalability.  We have not been able to test
>>> performance yet since we are still trying to make it work but as soon as we
>>> do I will be sure to report back.
>>>
>>>
>>> On Fri, Nov 25, 2011 at 4:19 PM, Stefan Caunter <stef at caunter.ca> wrote:
>>>
>>>> On Fri, Nov 25, 2011 at 12:37 PM, Rami Essaid <rami.essaid at gmail.com>
>>>> wrote:
>>>> > Thanks for the help Maxim!  We disabled our limit_req and that seemed
>>>> to
>>>> > have fixed the problem.  Looking at the logs it seems that only 1/3
>>>> of the
>>>> > requests are correctly getting the new IP assigned via the realIP
>>>> module,
>>>> > the remainder are still logging the load balancer IP.  This probably
>>>> is more
>>>> > of an issue with the amazon load balancer but do you have any idea on
>>>> what
>>>> > may be going on?
>>>> > Also, where would you recommend as a place to start tracking and
>>>> fixing the
>>>> > other issue?
>>>> >
>>>>
>>>> You have not established that load balancer is setting X-Forwarded-For
>>>> or some other header to pass Real IP.
>>>>
>>>> Does the amazon pseudo device have enough sockets available to connect
>>>> to nginx?
>>>>
>>>> What do you expect from adding this layer? Were you hitting
>>>> performance limits? Has it improved or degraded performance?
>>>>
>>>> >
>>>> >
>>>> > On Fri, Nov 25, 2011 at 12:14 PM, Maxim Dounin <mdounin at mdounin.ru>
>>>> wrote:
>>>> >>
>>>> >> Hello!
>>>> >>
>>>> >> On Fri, Nov 25, 2011 at 09:54:14AM -0500, Rami Essaid wrote:
>>>> >>
>>>> >> > Hi Maxim,
>>>> >> >
>>>> >> > We implemented to module and still had some trouble.  A lot of the
>>>> >> > connections would return " 503 Service Temporarily Unavailable".
>>>>  Our
>>>> >> > configuration works fine without the load balancer but then gives
>>>> these
>>>> >> > 503
>>>> >> > errors behind the load balancer.
>>>> >>
>>>> >> nginx itself will only return 503 if it hits either limit_conn or
>>>> >> limit_req.
>>>> >>
>>>> >> If you see this returned by nginx, and it only happens with load
>>>> >> balancer, this may indicate you've not configured realip module
>>>> >> properly (or your load balancer doesn't provide appropriate
>>>> >> headers) and you are hitting per-ip limits configured due to all
>>>> >> requests appear to be from load balancer.
>>>> >>
>>>> >> Check if client's ip logged is really client's one, not an ip of
>>>> >> your load balancer.
>>>> >>
>>>> >> > Looking into the error logs I notice a lot of these errors both
>>>> with and
>>>> >> > without the load balancer "connect() failed (111: Connection
>>>> refused)
>>>> >> > while
>>>> >> > connecting to upstream".  Could this be the reason that we are
>>>> having
>>>> >> > issues?
>>>> >>
>>>> >> Unlikely, but It's a good idea to track and fix this in any case.
>>>> >>
>>>> >> Maxim Dounin
>>>> >>
>>>> >> >
>>>> >> > Thanks!
>>>> >> > Rami
>>>> >> >
>>>> >> >
>>>> >> > On Mon, Nov 21, 2011 at 7:46 AM, Maxim Dounin <mdounin at mdounin.ru>
>>>> >> > wrote:
>>>> >> >
>>>> >> > > Hello!
>>>> >> > >
>>>> >> > > On Mon, Nov 21, 2011 at 07:40:49AM -0500, Rami Essaid wrote:
>>>> >> > >
>>>> >> > > > Thanks Maxim,
>>>> >> > > >
>>>> >> > > > This looks like exactly what we need.  In your experience does
>>>> this
>>>> >> > > resolve
>>>> >> > > > most issues behind a load balancer?
>>>> >> > >
>>>> >> > > Yes.
>>>> >> > >
>>>> >> > > Maxim Dounin
>>>> >> > >
>>>> >> > > > On Mon, Nov 21, 2011 at 7:38 AM, Maxim Dounin <
>>>> mdounin at mdounin.ru>
>>>> >> > > wrote:
>>>> >> > > >
>>>> >> > > > > Hello!
>>>> >> > > > >
>>>> >> > > > > On Mon, Nov 21, 2011 at 07:25:39AM -0500, Rami Essaid wrote:
>>>> >> > > > >
>>>> >> > > > > > Hi Guys,
>>>> >> > > > > >
>>>> >> > > > > > This weekend for scalability we tried putting our nginx
>>>> servers
>>>> >> > > behind
>>>> >> > > > > > amazon's elastic load balancers and came across a road
>>>> block: it
>>>> >> > > does not
>>>> >> > > > > > transparently pass the user IP and header information to
>>>> nginx.
>>>> >> > > > > > This
>>>> >> > > > > caused
>>>> >> > > > > > issues with several pieces of nginx we use including the IP
>>>> >> > > > > > allow /
>>>> >> > > deny
>>>> >> > > > > > rules, the limit_req module, and the limit_con module.  Has
>>>> >> > > > > > anyone
>>>> >> > > > > > successfully put nginx behind a load balancer?  Any ideas
>>>> on how
>>>> >> > > > > > to
>>>> >> > > make
>>>> >> > > > > > this work?
>>>> >> > > > >
>>>> >> > > > > http://wiki.nginx.org/HttpRealIpModule
>>>> >> > > > >
>>>> >> > > > > 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
>>>> >>
>>>> >> _______________________________________________
>>>> >> 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/20111129/87e32ad7/attachment-0001.html>

From valery+nginxen at grid.net.ru  Tue Nov 29 10:29:52 2011
From: valery+nginxen at grid.net.ru (Valery Kholodkov)
Date: Tue, 29 Nov 2011 10:29:52 +0000 (GMT)
Subject: nginx-lua and nginx upload module
In-Reply-To: <CAB4Tn6PEtGMTH=vRrJmEOW95Pnz928CX7L4Zw=MP3DwF=KbqWA@mail.gmail.com>
Message-ID: <12986495.5059.1322562592926.JavaMail.root@zone.mtgsy.net>


----- agentzh <agentzh at gmail.com> wrote:
> On Mon, Nov 28, 2011 at 6:22 PM, Valery Kholodkov
> <valery+nginxen at grid.net.ru> wrote:
> >
> > I think we've been able to do that even before.
> 
> Sadly for nginx 0.8.41 ~ 0.8.53, "IO interruptions" are explicitly
> prohibited in rewrite phase handlers and for exactly the same reason
> your ngx_eval module does not work at all for these versions of Nginx
> :)
> 
> > But that's not point. The problem is not in what particular phase upload module will run, but that it must run whenever some module tries to read the request body. This kind of customisation is not supported by nginx yet.
> >
> 
> True. We definitely need an input filter mechanism for Nginx. I've
> cc'd Andrew Alexeev. Maybe he can help make this happen in the Nginx
> core ;)
> 
> >
> > Again, the point is not in what particular phase it will run... And by the way, the costs of internal redirects are negligible.
> >
> 
> Not really if the user has a lot of regex style locations defined in
> his nginx.conf. For now, the nginx core does not (yet) utilize the
> PCRE JIT feature for its pattern matching, so it does consume quite a
> few CPU cycles for complicated location patterns ;) And...if we *can*
> eliminate the cost of internal redirects altogether, why shouldn't we
> make that happen? ;)

Did you mean "quite a lot"?

> 
> Also, I believe it can simplify the user config file greatly and does
> not cause variable scope problems (like $uri and $args' scope). I know
> there is a upload_pass_args directive to forward the original URI
> query args, but if the $args scope issue does not exist in the first
> place, we do not need this config directive at all.
> 
> BTW, I notice that if ngx_upload's special variables like
> $upload_file_md5 are used in a wrong context, it will simply crash the
> nginx worker (segmentation faults) due to null module context pointer
> access (at least for the latest 2.2.0 release).

I'll add a check for that.

> Maybe it's more
> reasonable to check the ctx pointer and print out a helpful error
> message in functions like ngx_http_upload_md5_variable?

-- 
Regards,
Valery Kholodkov


From agentzh at gmail.com  Tue Nov 29 10:49:57 2011
From: agentzh at gmail.com (agentzh)
Date: Tue, 29 Nov 2011 18:49:57 +0800
Subject: nginx-lua and nginx upload module
In-Reply-To: <12986495.5059.1322562592926.JavaMail.root@zone.mtgsy.net>
References: <CAB4Tn6PEtGMTH=vRrJmEOW95Pnz928CX7L4Zw=MP3DwF=KbqWA@mail.gmail.com>
	<12986495.5059.1322562592926.JavaMail.root@zone.mtgsy.net>
Message-ID: <CAB4Tn6NbJQ5bjmyHqhnvioOCSSbtqT9b1BqZe1LR+KJ2g8hKPA@mail.gmail.com>

On Tue, Nov 29, 2011 at 6:29 PM, Valery Kholodkov >>
>> Not really if the user has a lot of regex style locations defined in
>> his nginx.conf. For now, the nginx core does not (yet) utilize the
>> PCRE JIT feature for its pattern matching, so it does consume quite a
>> few CPU cycles for complicated location patterns ;) And...if we *can*
>> eliminate the cost of internal redirects altogether, why shouldn't we
>> make that happen? ;)
>
> Did you mean "quite a lot"?
>

I didn't say it costs quite a lot in the real world settings. I just
said we should avoid wasting CPU cycles wherever possible (just like
what Marcus Clyne originally told me two years ago) :)

Regards,
-agentzh


From usirsiwal at verivue.com  Tue Nov 29 13:06:07 2011
From: usirsiwal at verivue.com (Sirsiwal, Umesh)
Date: Tue, 29 Nov 2011 08:06:07 -0500
Subject: Dealing with buffered data with upstream generated response
In-Reply-To: <4ED3BC56.4000200@verivue.com>
References: <4ED3BC56.4000200@verivue.com>
Message-ID: <B293D35E9E20694A8D9DFFE1BC655C55771598CA09@vvexch.verivue.com>

Any help on this will be greatly appreciated. 

Thank you!

-Umesh

________________________________________

I am developing a body filter which transforms outgoing stream. As part
of the function, the incoming stream is copied in new buffer and sent.
This in some cases results in a condition where my filter has busy
buffers but the upper layers don't have any busy buffer. In our case the
data is generated by upstream module. Since upstream module does not pay
attention to connection->buffered, the output_filter is never called
again to flush my busy buffers and the transfer just hangs.

Adding r->connection->buffered to the or condition solves the hang.

            if (u->out_bufs || u->busy_bufs) {
                 rc = ngx_http_output_filter(r, u->out_bufs);

                 if (rc == NGX_ERROR) {
                     ngx_http_upstream_finalize_request(r, u, 0);
                     return;
                 }

                 ngx_chain_update_chains(&u->free_bufs, &u->busy_bufs,
&u->out_bufs, u->output.tag);
             }

Is this nginx bug or am I missing something?

-Umesh


From andrew at nginx.com  Tue Nov 29 14:58:45 2011
From: andrew at nginx.com (Andrew Alexeev)
Date: Tue, 29 Nov 2011 18:58:45 +0400
Subject: nginx-lua and nginx upload module
In-Reply-To: <CAB4Tn6PEtGMTH=vRrJmEOW95Pnz928CX7L4Zw=MP3DwF=KbqWA@mail.gmail.com>
References: <CAB4Tn6MBC5Xn1oRKJezEX4hhwNtsL5QiVEAVukMbBvsDtOwD6w@mail.gmail.com>
	<5133784.4937.1322475745829.JavaMail.root@zone.mtgsy.net>
	<CAB4Tn6PEtGMTH=vRrJmEOW95Pnz928CX7L4Zw=MP3DwF=KbqWA@mail.gmail.com>
Message-ID: <A5C6BD66-F914-4B01-BDAC-AEEBD55E4F84@nginx.com>

Hi,

>> But that's not point. The problem is not in what particular phase upload module will run, but that it must run whenever some module tries to read the request body. This kind of customisation is not supported by nginx yet.
>> 
> 
> True. We definitely need an input filter mechanism for Nginx. I've
> cc'd Andrew Alexeev. Maybe he can help make this happen in the Nginx
> core ;)

I'm not a programmer (at least not anymore), so I personally can't implement this :) Input filters are planned in the next major release of nginx, where the core architecture would allow that.

> his nginx.conf. For now, the nginx core does not (yet) utilize the
> PCRE JIT feature for its pattern matching, so it does consume quite a

Btw, we have an almost ready implementation for PCRE JIT and it'll soon be released in the dev train.


From valery+nginxen at grid.net.ru  Tue Nov 29 15:35:23 2011
From: valery+nginxen at grid.net.ru (Valery Kholodkov)
Date: Tue, 29 Nov 2011 15:35:23 +0000 (GMT)
Subject: nginx upload module
In-Reply-To: <4ED4CD9E.6080803@urtext.ru>
Message-ID: <20189510.5144.1322580923798.JavaMail.root@zone.mtgsy.net>


----- Vladimir Zorin <vladimir at urtext.ru> wrote:
>  > Not really if the user has a lot of regex style locations defined in
>  > > his nginx.conf. For now, the nginx core does not (yet) utilize the
>  > > PCRE JIT feature for its pattern matching, so it does consume quite a
>  > > few CPU cycles for complicated location patterns ;) And...if we *can*
>  > > eliminate the cost of internal redirects altogether, why shouldn't we
>  > > make that happen? ;)
> 
> 
> VK> Did you mean "quite a lot"?
> 
> He did mean 'quite a lot'.  And if you had bothered to look up the 
> dictionary first, you'd've found that: "few - used in expressions such 
> as 'quite a few' or 'a good few' to mean quite a large number".
> 
> But russians are too fucking arrogant to learn english, sure thing.

Looks like it was worth completing the Advanced level of English at BLTC. At least I can respond to insults.

So here it comes: Gentleman, your speach hurts! 

-- 
Regards,
Valery Kholodkov


From valery+nginxen at grid.net.ru  Tue Nov 29 15:51:24 2011
From: valery+nginxen at grid.net.ru (Valery Kholodkov)
Date: Tue, 29 Nov 2011 15:51:24 +0000 (GMT)
Subject: nginx-lua and nginx upload module
In-Reply-To: <A5C6BD66-F914-4B01-BDAC-AEEBD55E4F84@nginx.com>
Message-ID: <19192068.5148.1322581884783.JavaMail.root@zone.mtgsy.net>


----- Andrew Alexeev <andrew at nginx.com> wrote:
> > True. We definitely need an input filter mechanism for Nginx. I've
> > cc'd Andrew Alexeev. Maybe he can help make this happen in the Nginx
> > core ;)
> 
> I'm not a programmer (at least not anymore), so I personally can't implement this :) Input filters are planned in the next major release of nginx, where the core architecture would allow that.

I'm waiting for that for 5 years already. Couple of years more is quite okay for me.

-- 
Regards,
Valery Kholodkov


From mdounin at mdounin.ru  Tue Nov 29 18:30:32 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Tue, 29 Nov 2011 22:30:32 +0400
Subject: Dealing with buffered data with upstream generated response
In-Reply-To: <4ED3BC56.4000200@verivue.com>
References: <4ED3BC56.4000200@verivue.com>
Message-ID: <20111129183031.GM67687@mdounin.ru>

Hello!

On Mon, Nov 28, 2011 at 11:52:38AM -0500, Umesh Sirsiwal wrote:

> I am developing a body filter which transforms outgoing stream. As
> part of the function, the incoming stream is copied in new buffer
> and sent. This in some cases results in a condition where my filter
> has busy buffers but the upper layers don't have any busy buffer. In
> our case the data is generated by upstream module. Since upstream
> module does not pay attention to connection->buffered, the
> output_filter is never called again to flush my busy buffers and the
> transfer just hangs.
> 
> Adding r->connection->buffered to the or condition solves the hang.
> 
>            if (u->out_bufs || u->busy_bufs) {
>                 rc = ngx_http_output_filter(r, u->out_bufs);
> 
>                 if (rc == NGX_ERROR) {
>                     ngx_http_upstream_finalize_request(r, u, 0);
>                     return;
>                 }
> 
>                 ngx_chain_update_chains(&u->free_bufs, &u->busy_bufs,
> &u->out_bufs, u->output.tag);
>             }
> 
> Is this nginx bug or am I missing something?

I tend to think this is a bug, and it should probably manifest 
itself with e.g. gzip filter as well.

(Well, as far as I see transfer shouldn't hang completely, but it 
will be only resumed as soon as upstream sends some more data or 
closes connection.)

It would be superb if you'll be able to provide test case which 
catches this for our test suite (http://mdounin.ru/hg/nginx-tests).

I'll look into this more closely as time permits.

Maxim Dounin


From usirsiwal at verivue.com  Tue Nov 29 19:19:10 2011
From: usirsiwal at verivue.com (Umesh Sirsiwal)
Date: Tue, 29 Nov 2011 14:19:10 -0500
Subject: Dealing with buffered data with upstream generated response
In-Reply-To: <20111129183031.GM67687@mdounin.ru>
References: <4ED3BC56.4000200@verivue.com> <20111129183031.GM67687@mdounin.ru>
Message-ID: <4ED5302E.9030709@verivue.com>

Thanks Maxim,

>> I am developing a body filter which transforms outgoing stream. As
>> part of the function, the incoming stream is copied in new buffer
>> and sent. This in some cases results in a condition where my filter
>> has busy buffers but the upper layers don't have any busy buffer. In
>> our case the data is generated by upstream module. Since upstream
>> module does not pay attention to connection->buffered, the
>> output_filter is never called again to flush my busy buffers and the
>> transfer just hangs.
>>
>> Adding r->connection->buffered to the or condition solves the hang.
>>
>>             if (u->out_bufs || u->busy_bufs) {
>>                  rc = ngx_http_output_filter(r, u->out_bufs);
>>
>>                  if (rc == NGX_ERROR) {
>>                      ngx_http_upstream_finalize_request(r, u, 0);
>>                      return;
>>                  }
>>
>>                  ngx_chain_update_chains(&u->free_bufs,&u->busy_bufs,
>> &u->out_bufs, u->output.tag);
>>              }
>>
>> Is this nginx bug or am I missing something?
> I tend to think this is a bug, and it should probably manifest
> itself with e.g. gzip filter as well.
>
> (Well, as far as I see transfer shouldn't hang completely, but it
> will be only resumed as soon as upstream sends some more data or
> closes connection.)
In our application we use Upstream KeepAlive module to keep the upstream 
connection open. So if this happens towards the end of the last 
subrequest response, the connection hangs.
>
> It would be superb if you'll be able to provide test case which
> catches this for our test suite (http://mdounin.ru/hg/nginx-tests).
I will try to see if I can reproduce it with gzip filter. I should be 
able to reproduce it with 
limit-rate+gzip-filter+proxy+upstream-keepalive module combination. But, 
I am not sure if the nginx-test test-suite includes upstream-keepalive.

>
> I'll look into this more closely as time permits.
>
> Maxim Dounin
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx


From kgorlo at gmail.com  Tue Nov 29 22:19:47 2011
From: kgorlo at gmail.com (Kamil Gorlo)
Date: Tue, 29 Nov 2011 23:19:47 +0100
Subject: limit_rate but for upload
Message-ID: <CAHzX=KBk_8EJGHKtOgtXuM5Y0Da27ZP4WYX1BTdj3J6S+70_8Q@mail.gmail.com>

Hello guys,

is there any way to limit upload rate (from user to server) like with
limit_rate directive for download?

I've found that this is possible with upload_module but I do not want
to use upload_module just for this feature (and also it will not work
in my application since I am using own method of uploading data). Here
is discussion: http://www.ruby-forum.com/topic/175565 which suggest
that it might be currently possible in nginx (but I can't find how).

If it's not currently possible, will it be hard to implement simple
module which will accomplish that (somehow - simplified version of
upload_module :))?

Cheers,

--
Kamil Gorlo


From nginx-forum at nginx.us  Wed Nov 30 02:05:37 2011
From: nginx-forum at nginx.us (adam)
Date: Tue, 29 Nov 2011 21:05:37 -0500
Subject: Premature failover to backup upstream
Message-ID: <48c5d85e94ec055d7e8eace0aadae0f3.NginxMailingListEnglish@forum.nginx.org>

Hello!

I am using Nginx 1.0.5 to reverse proxy to a pool of three upstreams. 
One of them is configured as a backup upstream.  From time to time, very
rarely, a request will be proxied to the backup upstream when it should
not be.  The two normal upstreams are completely healthy, and at
extremely low load.

Why might this be happening?

I see in the documentation
(http://wiki.nginx.org/HttpUpstreamModule#server) that the upstream
module will shunt to backup if the other upstreams are "down or busy." 
What is the definition of "busy" in this context?

Looking in the debug log, I see many many instances of "http upstream
recv(): -1 (11: Resource temporarily unavailable)".  Is this related?

One thing I haven't tried is to adjust the max_fails parameter, but
perhaps I should.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,219326,219326#msg-219326


From agentzh at gmail.com  Wed Nov 30 02:07:47 2011
From: agentzh at gmail.com (agentzh)
Date: Wed, 30 Nov 2011 10:07:47 +0800
Subject: nginx upload module
In-Reply-To: <20189510.5144.1322580923798.JavaMail.root@zone.mtgsy.net>
References: <4ED4CD9E.6080803@urtext.ru>
	<20189510.5144.1322580923798.JavaMail.root@zone.mtgsy.net>
Message-ID: <CAB4Tn6PmWnxb2jjLMNLUeT55nTO-MZVuWVVz-utURykxbbE65g@mail.gmail.com>

On Tue, Nov 29, 2011 at 11:35 PM, Valery Kholodkov
<valery+nginxen at grid.net.ru> wrote:
>
> ----- Vladimir Zorin <vladimir at urtext.ru> wrote:
>> ?> Not really if the user has a lot of regex style locations defined in
>> ?> > his nginx.conf. For now, the nginx core does not (yet) utilize the
>> ?> > PCRE JIT feature for its pattern matching, so it does consume quite a
>> ?> > few CPU cycles for complicated location patterns ;) And...if we *can*
>> ?> > eliminate the cost of internal redirects altogether, why shouldn't we
>> ?> > make that happen? ;)
>>
>>
>> VK> Did you mean "quite a lot"?
>>
>> He did mean 'quite a lot'. ?And if you had bothered to look up the
>> dictionary first, you'd've found that: "few - used in expressions such
>> as 'quite a few' or 'a good few' to mean quite a large number".
>>
>> But russians are too fucking arrogant to learn english, sure thing.
>
> Looks like it was worth completing the Advanced level of English at BLTC. At least I can respond to insults.
>
> So here it comes: Gentleman, your speach hurts!
>

I meant quite a lot of CPU cycles (or computer instructions), but not
"quite a lot" of time in the scale of real application response time.
There might be some misunderstanding here. Such debate is meaningless
for technical discussions and I apologize if I took you wrong.

Regards,
-agentzh


From piotr.sikora at frickle.com  Wed Nov 30 02:46:35 2011
From: piotr.sikora at frickle.com (Piotr Sikora)
Date: Wed, 30 Nov 2011 03:46:35 +0100
Subject: Premature failover to backup upstream
In-Reply-To: <48c5d85e94ec055d7e8eace0aadae0f3.NginxMailingListEnglish@forum.nginx.org>
References: <48c5d85e94ec055d7e8eace0aadae0f3.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <3C766E3FE05A4F67A2732D099AB1CD48@Desktop>

Hi,

> Looking in the debug log, I see many many instances of "http upstream
> recv(): -1 (11: Resource temporarily unavailable)".  Is this related?

Are you using UNIX sockets? If so, try using TCP sockets, it might help.

Best regards,
Piotr Sikora < piotr.sikora at frickle.com >


From hamano at osstech.co.jp  Wed Nov 30 05:19:47 2011
From: hamano at osstech.co.jp (Tsukasa Hamano)
Date: Wed, 30 Nov 2011 14:19:47 +0900
Subject: Nginx DSO(dynamic loadable module)
Message-ID: <877h2ikwzw.wl%hamano@osstech.co.jp>

Hello,

I wrote nginx dso module and tested GNU/Linux i386/x86_64.
https://github.com/hamano/nginx-mod-so

Also this module working independent from main unit, but It need
slightly modification of the nginx build script.

https://github.com/hamano/nginx-mod-so/blob/master/patches/nginx-1.0.9-mod-so.patch

Can you review and apply patches?

Thank you.

--
Open Source Solution Technology Corporation
Tsukasa Hamano <hamano at osstech.co.jp>
fingerprint = 2285 2111 6D34 3816 3C2E  A5B9 16BE D101 6069 BE55
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 197 bytes
Desc: OpenPGP Digital Signature
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20111130/21c0d478/attachment-0001.bin>

From nginx-forum at nginx.us  Wed Nov 30 06:39:02 2011
From: nginx-forum at nginx.us (adam)
Date: Wed, 30 Nov 2011 01:39:02 -0500
Subject: Premature failover to backup upstream
In-Reply-To: <48c5d85e94ec055d7e8eace0aadae0f3.NginxMailingListEnglish@forum.nginx.org>
References: <48c5d85e94ec055d7e8eace0aadae0f3.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <0e6a18b128b1c65be8b363a06e4ac884.NginxMailingListEnglish@forum.nginx.org>

> Are you using UNIX sockets? If so, try using TCP sockets, it might
help.

Is that a configuration option at compile time?

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,219326,219339#msg-219339


From nginx-forum at nginx.us  Wed Nov 30 10:18:57 2011
From: nginx-forum at nginx.us (c0re)
Date: Wed, 30 Nov 2011 05:18:57 -0500
Subject: proxy_cache/story and busylocks workaround
Message-ID: <0cef701999ab7cc09c6268a2d3d95758.NginxMailingListEnglish@forum.nginx.org>

proxy_cache workaround:

http{
        proxy_cache_path /cache keys_zone=localcache:10m;
        limit_zone perUri $uri 10m;
}
server {
        location /remote/ {
                try_files $uri @fake @proxy-remote;
        }
        location @fake {
                root /notexists;
        }
        location @proxy-remote {
                limit_conn perUri 1;
                proxy_cache localcache;
                proxy_cache_valid any 999d;
                proxy_cache_key "remote$request_uri"; # fix to suit your
needs
                proxy_set_header Range ""; # don't pass range requests
to upstream in this section, otherwise response won't be cached
                proxy_pass upstream;
                error_page 503 = @locked-remote;
        }
        location @locked-remote {
                # take your own. e.g.:
                proxy_buffering off;
                proxy_pass upstream;
        }
        # copy paste locations if you need (feel free to share common
@fake)
}

http://pastie.org/private/meuzkrrvanhgojuoz0mvxa

In case of proxy_store just remove @fake and set right root in /remote/

This is just a PoC, so you have to change it up to your needs.
It really works and nginx really rox =]
Any questions? - Welcome

icq:3016467 skype:imperial.roaming

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,219354,219354#msg-219354


From mdounin at mdounin.ru  Wed Nov 30 10:22:21 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Wed, 30 Nov 2011 14:22:21 +0400
Subject: Nginx DSO(dynamic loadable module)
In-Reply-To: <877h2ikwzw.wl%hamano@osstech.co.jp>
References: <877h2ikwzw.wl%hamano@osstech.co.jp>
Message-ID: <20111130102221.GN67687@mdounin.ru>

Hello!

On Wed, Nov 30, 2011 at 02:19:47PM +0900, Tsukasa Hamano wrote:

> Hello,
> 
> I wrote nginx dso module and tested GNU/Linux i386/x86_64.
> https://github.com/hamano/nginx-mod-so
> 
> Also this module working independent from main unit, but It need
> slightly modification of the nginx build script.
> 
> https://github.com/hamano/nginx-mod-so/blob/master/patches/nginx-1.0.9-mod-so.patch
> 
> Can you review and apply patches?

Unfortunately, this won't work even if implemented properly.  Most 
serious issue is that there should be ABI compatibility maintained 
and checked, this is not currently the case even within single 
unmodified nginx release: there are configure arguments which 
change layout of various internal structures.

Other problems include e.g. proper handling of reloads.

Maxim Dounin


From mdounin at mdounin.ru  Wed Nov 30 10:51:23 2011
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Wed, 30 Nov 2011 14:51:23 +0400
Subject: nginx-1.1.10
Message-ID: <20111130105123.GR67687@mdounin.ru>

Changes with nginx 1.1.10                                        30 Nov 2011

    *) Bugfix: a segmentation fault occured in a worker process if AIO was
       used on Linux; the bug had appeared in 1.1.9.


Maxim Dounin


From nginx-forum at nginx.us  Wed Nov 30 10:55:26 2011
From: nginx-forum at nginx.us (c0re)
Date: Wed, 30 Nov 2011 05:55:26 -0500
Subject: proxy_cache/story and busylocks workaround
In-Reply-To: <0cef701999ab7cc09c6268a2d3d95758.NginxMailingListEnglish@forum.nginx.org>
References: <0cef701999ab7cc09c6268a2d3d95758.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <f250390bc89af825c4789b3232bd44b5.NginxMailingListEnglish@forum.nginx.org>

Sorry guys.
There's problem with workaround for proxy_cache.
So it works with proxy_store only.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,219354,219362#msg-219362


From nginx-forum at nginx.us  Wed Nov 30 17:06:25 2011
From: nginx-forum at nginx.us (token)
Date: Wed, 30 Nov 2011 12:06:25 -0500
Subject: capture rewrite for every after question mark
Message-ID: <cfe732cf609f1c503d508ce600f1b106.NginxMailingListEnglish@forum.nginx.org>

Hello,

   I have found the following while searching it is the only one i can
get working with my config setup
   to send a 301 permanently if "/index.php" is directly accessed.


    if ($request_uri ~* ^(/home(/index)?|/index(.php)?)/?$)
    {
    rewrite ^(.*)$ / permanent;
    }


   GET http://www.example.com/index.php 301 Moved Permanently
   GET http://www.example.com/  200 OK

   ---------------------------------------
   above works as expected.
   
   But, now if i have the following results as 200 OK as expected.

   http://www.example.com/index.php?randomvalue  200 OK


   So how do you rewrite that if there is a question mark and or value
that follows "/index.php?" then returns 301.

   GET http://www.example.com/index.php?randomvalue  301 Moved
Permanently

   GET http://www.example.com/  200 OK

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,219385,219385#msg-219385


From appa at perusio.net  Wed Nov 30 17:12:35 2011
From: appa at perusio.net (=?UTF-8?B?QW50w7NuaW8=?= P. P. Almeida)
Date: Wed, 30 Nov 2011 17:12:35 +0000
Subject: capture rewrite for every after question mark
In-Reply-To: <cfe732cf609f1c503d508ce600f1b106.NginxMailingListEnglish@forum.nginx.org>
References: <cfe732cf609f1c503d508ce600f1b106.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <87fwh5lekc.wl%appa@perusio.net>

On 30 Nov 2011 17h06 WET, nginx-forum at nginx.us wrote:

> Hello,
>
> I have found the following while searching it is the only one i can
> get working with my config setup
> to send a 301 permanently if "/index.php" is directly accessed.
>
>
> if ($request_uri ~* ^(/home(/index)?|/index(.php)?)/?$)
> {
> rewrite ^(.*)$ / permanent;
> }
>
>
> GET http://www.example.com/index.php 301 Moved Permanently
> GET http://www.example.com/  200 OK
>
> ---------------------------------------
> above works as expected.
>
> But, now if i have the following results as 200 OK as expected.
>
> http://www.example.com/index.php?randomvalue  200 OK
>
>
> So how do you rewrite that if there is a question mark and or value
> that follows "/index.php?" then returns 301.
>
> GET http://www.example.com/index.php?randomvalue  301 Moved
> Permanently
>
> GET http://www.example.com/  200 OK

Try:

location = /index.php {
    return 301 /;
}

--- appa


From nginx-forum at nginx.us  Wed Nov 30 17:19:00 2011
From: nginx-forum at nginx.us (token)
Date: Wed, 30 Nov 2011 12:19:00 -0500
Subject: capture rewrite for every after question mark
In-Reply-To: <cfe732cf609f1c503d508ce600f1b106.NginxMailingListEnglish@forum.nginx.org>
References: <cfe732cf609f1c503d508ce600f1b106.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <91d71ba2390ef8ac29f7dfe480d52fb8.NginxMailingListEnglish@forum.nginx.org>

Thanks for suggestion appa, i apply what said to try and results in a
"Redirection Loop".

> Hello,
>
> I have found the following while searching it is the only one i can
> get working with my config setup
> to send a 301 permanently if "/index.php" is directly accessed.
>
>
> if ($request_uri ~* ^(/home(/index)?|/index(.php)?)/?$)
> {
> rewrite ^(.*)$ / permanent;
> }
>
>
> GET http://www.example.com/index.php 301 Moved Permanently
> GET http://www.example.com/ 200 OK
>
> ---------------------------------------
> above works as expected.
>
> But, now if i have the following results as 200 OK as expected.
>
> http://www.example.com/index.php?randomvalue 200 OK
>
>
> So how do you rewrite that if there is a question mark and or value
> that follows "/index.php?" then returns 301.
>
> GET http://www.example.com/index.php?randomvalue 301 Moved
> Permanently
>
> GET http://www.example.com/ 200 OK

> Try:

> location = /index.php {
> return 301 /;
> }

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,219385,219388#msg-219388


From appa at perusio.net  Wed Nov 30 17:46:34 2011
From: appa at perusio.net (=?UTF-8?B?QW50w7NuaW8=?= P. P. Almeida)
Date: Wed, 30 Nov 2011 17:46:34 +0000
Subject: capture rewrite for every after question mark
In-Reply-To: <91d71ba2390ef8ac29f7dfe480d52fb8.NginxMailingListEnglish@forum.nginx.org>
References: <cfe732cf609f1c503d508ce600f1b106.NginxMailingListEnglish@forum.nginx.org>
	<91d71ba2390ef8ac29f7dfe480d52fb8.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <87ehwplczp.wl%appa@perusio.net>

On 30 Nov 2011 17h19 WET, nginx-forum at nginx.us wrote:

> Thanks for suggestion appa, i apply what said to try and results in
> a "Redirection Loop".
>
>> Hello,
>>
>> I have found the following while searching it is the only one i can
>> get working with my config setup
>> to send a 301 permanently if "/index.php" is directly accessed.

Remove/comment out the if below and try:

http {
  
   map $redirect_direct_access $request_uri {
       default 0;
       ~/index(\.php)? 1;
       ~/home(/index)? 1;
  }
}


server {
    server_name www.example.com;

    root /path/to/root;

    location / {
        if ($redirect_direct_access) {
           return 301 /;
        }     
    }
}

> if ($request_uri ~* ^(/home(/index)?|/index(.php)?)/?$)
> {
> rewrite ^(.*)$ / permanent;
> }

--- appa


From nginx-forum at nginx.us  Wed Nov 30 17:51:45 2011
From: nginx-forum at nginx.us (token)
Date: Wed, 30 Nov 2011 12:51:45 -0500
Subject: capture rewrite for every after question mark
In-Reply-To: <cfe732cf609f1c503d508ce600f1b106.NginxMailingListEnglish@forum.nginx.org>
References: <cfe732cf609f1c503d508ce600f1b106.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <1422610818bc88e5e417460923b51139.NginxMailingListEnglish@forum.nginx.org>

> Thanks for suggestion appa, i apply what said to try and results in a
"Redirection Loop".

> Hello,
>
> I have found the following while searching it is the only one i can
> get working with my config setup
> to send a 301 permanently if "/index.php" is directly accessed.
>
>
> if ($request_uri ~* ^(/home(/index)?|/index(.php)?)/?$)
> {
> rewrite ^(.*)$ / permanent;
> }
>
>
> GET http://www.example.com/index.php 301 Moved Permanently
> GET http://www.example.com/ 200 OK
>
> ---------------------------------------
> above works as expected.
>
> But, now if i have the following results as 200 OK as expected.
>
> http://www.example.com/index.php?randomvalue 200 OK
>
>
> So how do you rewrite that if there is a question mark and or value
> that follows "/index.php?" then returns 301.
>
> GET http://www.example.com/index.php?randomvalue 301 Moved
> Permanently
>
> GET http://www.example.com/ 200 OK

> Try:

> location = /index.php {
> return 301 /;
> }

Update,

I am using 

	if ($request_uri ~* ^/index.php(.*)$){
	   rewrite ^(.*)$ / permanent;
	}
        
       The above resuls as which dose the 301 moved permanently with the
exception of retaining the the question and the the value. 

I guess from a search engines point of view its see's the 301 Moved
Permanently so in that case it ok, but it would be nice to not retain
the "?randomvalue".

       http://www.example.com/index.php?randomvalue  301 Moved
Permanently
       http://www.example.com/?randomvalue  200 OK

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,219385,219391#msg-219391


From nginx-forum at nginx.us  Wed Nov 30 18:12:42 2011
From: nginx-forum at nginx.us (token)
Date: Wed, 30 Nov 2011 13:12:42 -0500
Subject: capture rewrite for every after question mark
In-Reply-To: <1422610818bc88e5e417460923b51139.NginxMailingListEnglish@forum.nginx.org>
References: <cfe732cf609f1c503d508ce600f1b106.NginxMailingListEnglish@forum.nginx.org>
	<1422610818bc88e5e417460923b51139.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <d26c9bc93b4890dc601f63603d993085.NginxMailingListEnglish@forum.nginx.org>

token Wrote:
-------------------------------------------------------
> > Thanks for suggestion appa, i apply what said to
> try and results in a "Redirection Loop".
> 
> > Hello,
> >
> > I have found the following while searching it is
> the only one i can
> > get working with my config setup
> > to send a 301 permanently if "/index.php" is
> directly accessed.
> >
> >
> > if ($request_uri ~*
> ^(/home(/index)?|/index(.php)?)/?$)
> > {
> > rewrite ^(.*)$ / permanent;
> > }
> >
> >
> > GET http://www.example.com/index.php 301 Moved
> Permanently
> > GET http://www.example.com/ 200 OK
> >
> > ---------------------------------------
> > above works as expected.
> >
> > But, now if i have the following results as 200
> OK as expected.
> >
> > http://www.example.com/index.php?randomvalue 200
> OK
> >
> >
> > So how do you rewrite that if there is a
> question mark and or value
> > that follows "/index.php?" then returns 301.
> >
> > GET http://www.example.com/index.php?randomvalue
> 301 Moved
> > Permanently
> >
> > GET http://www.example.com/ 200 OK
> 
> > Try:
> 
> > location = /index.php {
> > return 301 /;
> > }
> 
> Update,
> 
> I am using 
> 
> 	if ($request_uri ~* ^/index.php(.*)$){
> 	   rewrite ^(.*)$ / permanent;
> 	}
>         
>        The above resuls as which dose the 301
> moved permanently with the exception of retaining
> the the question and the the value. 
> 
> I guess from a search engines point of view its
> see's the 301 Moved Permanently so in that case it
> ok, but it would be nice to not retain the
> "?randomvalue".
> 
>       
> http://www.example.com/index.php?randomvalue  301
> Moved Permanently
>        http://www.example.com/?randomvalue  200 OK


Final Update,  Resolved

	if ($request_uri ~* ^/index.php(.*)$){
		rewrite ^(.*)$ /? permanent;
	}

Seems to work now

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,219385,219392#msg-219392


From nginx-forum at nginx.us  Wed Nov 30 19:04:02 2011
From: nginx-forum at nginx.us (adam)
Date: Wed, 30 Nov 2011 14:04:02 -0500
Subject: Premature failover to backup upstream
In-Reply-To: <48c5d85e94ec055d7e8eace0aadae0f3.NginxMailingListEnglish@forum.nginx.org>
References: <48c5d85e94ec055d7e8eace0aadae0f3.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <949e0c4b2d5bee66ea454e4439200ca3.NginxMailingListEnglish@forum.nginx.org>

Oh, I see what you mean.  We're proxying to Tomcat upstreams on separate
hosts, not a local fcgi or anything like that.

So, still looking for some clues.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,219326,219394#msg-219394


From nginx-forum at nginx.us  Wed Nov 30 20:58:13 2011
From: nginx-forum at nginx.us (skyshade)
Date: Wed, 30 Nov 2011 15:58:13 -0500
Subject: [MOD] proxy_cache regexp batch purge modules 0.2
In-Reply-To: <CAENo=s15bjJj8C-u430kd-Aby9vtw2_U2VpLRcDdrxMfLbFQeg@mail.gmail.com>
References: <CAENo=s15bjJj8C-u430kd-Aby9vtw2_U2VpLRcDdrxMfLbFQeg@mail.gmail.com>
Message-ID: <bdb0845763d3588b9ced2fb21e273a34.NginxMailingListEnglish@forum.nginx.org>

Great jog, please keep updating and supporting this module, because this
a great feature for many nginx users.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218483,219397#msg-219397