From lyninx at gmail.com  Sun Apr  1 06:05:48 2012
From: lyninx at gmail.com (guo rong zhang)
Date: Sun, 1 Apr 2012 14:05:48 +0800
Subject: post
Message-ID: <CAF2q3zbGexaNuUUGRYBqyTDJuc2p-b5QD7mGgPobgMoSG=gRyQ@mail.gmail.com>


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20120401/42aae438/attachment.html>

From amoiz.shine at gmail.com  Sun Apr  1 13:33:23 2012
From: amoiz.shine at gmail.com (Sharl.Jimh.Tsin)
Date: Sun, 1 Apr 2012 21:33:23 +0800
Subject: [ANNOUNCE] Tengine-1.2.4
In-Reply-To: <CAOKiq2CbQUtcdwMa+y+RgDEQTSVaOJEooRdOEBYBSxYtsqxevg@mail.gmail.com>
References: <CAOKiq2CbQUtcdwMa+y+RgDEQTSVaOJEooRdOEBYBSxYtsqxevg@mail.gmail.com>
Message-ID: <CANXGsjzCm_AK=JtGYowQ17QiMfzDckRTxShH9QOn7JxCWN9rzg@mail.gmail.com>

are you sure that NOT a joke of April Fools' Day, like taolinux
(http://code.taobao.org/p/taobao_linux_server/src/trunk/taolinux/) ?

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.



2012/3/31 Joshua Zhu <zhuzhaoyuan at gmail.com>:
> Hi folks!
>
> We're pleased to announce that Tengine-1.2.4 has been released. You can
> either download the tar ball
> from?http://tengine.taobao.org/download/tengine-1.2.4.tar.gz ,?or checkout
> the code from https://github.com/taobao/tengine
>
> As this release, Tengine is based on Nginx-1.0.14. The highlight is a brand
> new 'user_agent' module which would be a good replacement for the 'browser'
> module in standard Nginx.?It's more efficient and flexible?than the
> 'browser' module. Other changes include:
> * Feature: added the 'log_escape' directive.
> * Bugfix: fixed a bug in the limit_req module.
> * Bugfix: fixed a bug in subrequest.
>
> Regards,
>
> --
> Joshua Zhu
> Senior Software Engineer
> Server Platforms Team at Taobao
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx


From smallfish.xy at gmail.com  Sun Apr  1 13:58:46 2012
From: smallfish.xy at gmail.com (smallfish)
Date: Sun, 1 Apr 2012 21:58:46 +0800
Subject: [ANNOUNCE] Tengine-1.2.4
In-Reply-To: <CANXGsjzCm_AK=JtGYowQ17QiMfzDckRTxShH9QOn7JxCWN9rzg@mail.gmail.com>
References: <CAOKiq2CbQUtcdwMa+y+RgDEQTSVaOJEooRdOEBYBSxYtsqxevg@mail.gmail.com>
	<CANXGsjzCm_AK=JtGYowQ17QiMfzDckRTxShH9QOn7JxCWN9rzg@mail.gmail.com>
Message-ID: <CACPpB2xrZfQqtwD=h2HSjMjp4BJmwVr9_z5=mBC_75EQYZ+VjQ@mail.gmail.com>

hah, perhaps just a joke. but nothing is impossable. :)
--
blog: http://chenxiaoyu.org



? 2012?4?1? ??9:33?Sharl.Jimh.Tsin <amoiz.shine at gmail.com>???

> are you sure that NOT a joke of April Fools' Day, like taolinux
> (http://code.taobao.org/p/taobao_linux_server/src/trunk/taolinux/) ?
>
> 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.
>
>
>
> 2012/3/31 Joshua Zhu <zhuzhaoyuan at gmail.com>:
> > Hi folks!
> >
> > We're pleased to announce that Tengine-1.2.4 has been released. You can
> > either download the tar ball
> > from http://tengine.taobao.org/download/tengine-1.2.4.tar.gz , or
> checkout
> > the code from https://github.com/taobao/tengine
> >
> > As this release, Tengine is based on Nginx-1.0.14. The highlight is a
> brand
> > new 'user_agent' module which would be a good replacement for the
> 'browser'
> > module in standard Nginx. It's more efficient and flexible than the
> > 'browser' module. Other changes include:
> > * Feature: added the 'log_escape' directive.
> > * Bugfix: fixed a bug in the limit_req module.
> > * Bugfix: fixed a bug in subrequest.
> >
> > Regards,
> >
> > --
> > Joshua Zhu
> > Senior Software Engineer
> > Server Platforms Team at Taobao
> >
> > _______________________________________________
> > 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/20120401/bd8de022/attachment.html>

From varia at e-healthexpert.org  Sun Apr  1 19:32:14 2012
From: varia at e-healthexpert.org (Mark Alan)
Date: Sun, 1 Apr 2012 20:32:14 +0100
Subject: proxmox: how to use nginx instead of apache
Message-ID: <20120401203214.1899b7c7@e-healthexpert.org>

Hello list,

I wonder if anybody in this list was successful at (completely)
replacing apache for nginx as the web server for the proxmox VE
management system <http://www.proxmox.com/products/proxmox-ve>.

As for simply proxying it, I know that the following has beeen used to
simply proxy clients to apache:

server {
	listen 80 default_server;
	server_name ~^(www\.)?(?<domain>.+)$;
	return 302 https://$domain$request_uri;
}
server {
	listen 443 default_server ssl;
	server_name _;
	ssl_certificate /etc/pve/local/pve-ssl.pem;
	ssl_certificate_key  /etc/pve/local/pve-ssl.key;
	location / {
		 proxy_pass https://127.0.0.1:8006 ;
	}
}


Thank you,
M.


From kevanstannard at gmail.com  Mon Apr  2 00:19:12 2012
From: kevanstannard at gmail.com (Kevan Stannard)
Date: Mon, 2 Apr 2012 10:19:12 +1000
Subject: Understanding host value in nginx error log files
Message-ID: <CAMgRdiXgjzjGH8ZrR4=NVMZMEBBn3h=x1Y6O94+rpVhrZFLmQQ@mail.gmail.com>

I typically see entries in my error.log files of hack attempts where the
host entry is the IP address of my server, but I also see some entries with
host values that are domain names I don't recognise.

Example 1:

012/04/01 06:12:49 [error] 644#0: *882
"/var/www/nginx-default/ab1e27867d53d8f4942095a891183ce0cacd3dbf/0d4fc9bfe7c5f26b02522d088dd98da95a9ed8d7/074977cbb342d6ffa7743ae477a5c0054fef5512/index.html"
is not found (2: No such file or directory), client: 150.70.75.37, server:
localhost, request: "GET
/ab1e27867d53d8f4942095a891183ce0cacd3dbf/0d4fc9bfe7c5f26b02522d088dd98da95a9ed8d7/074977cbb342d6ffa7743ae477a5c0054fef5512/
HTTP/1.0", host: "deepspacer.com"

Example 2:

2012/02/03 01:38:41 [error] 592#0: *14019 open()
"/var/www/nginx-default/home.php" failed (2: No such file or directory),
client: 216.104.15.130, server: localhost, request: "GET
/home.php?SES=517a4bfc0137889f05d67314df2715a1&from_diary=1&cpl=1&from=102_4
HTTP/1.0", host: "www.au.mytelecomsurvey.com"

Example 3:

2012/02/03 11:57:56 [error] 592#0: *18075 open()
"/var/www/nginx-default/sites/default/files/js/js_b3ffc00633d66887fcb4ecdfc2d1c13a.jsmin.js"
failed (2: No such file or directory), client: 150.70.64.197, server:
localhost, request: "GET
/sites/default/files/js/js_b3ffc00633d66887fcb4ecdfc2d1c13a.jsmin.js
HTTP/1.0", host: "www.formalites-juridiques.net"

I was hoping someone could explain what it means if I'm seeing these domain
names as host values and if it's something I need to be concerned about.

Thanks
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20120402/50aa680e/attachment.html>

From tdgh2323 at hotmail.com  Mon Apr  2 01:31:42 2012
From: tdgh2323 at hotmail.com (Micheal Wolfskill)
Date: Mon, 2 Apr 2012 01:31:42 +0000
Subject: How to limit_req depending if the requests has a REFERER or not.
Message-ID: <BLU144-W24E01412AED330F9E64F77D04C0@phx.gbl>


Hello.  

I have something like this... I need to be able to apply three different limit_req depending:

a.) If the referer to click.php is domain.com ... apply zone1
b.) If there is some other referer apply zone2 on click.php
c.) If there is no referer apply zone3 on click.php



  location /click.php { 
                                   limit_req zone=one nodelay; 
                                   proxy_pass http://22.136.1.121:80/; 
                                   }

Mike
 		 	   		  
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20120402/6cf314f1/attachment.html>

From tdgh2323 at hotmail.com  Mon Apr  2 04:09:14 2012
From: tdgh2323 at hotmail.com (Micheal Wolfskill)
Date: Mon, 2 Apr 2012 04:09:14 +0000
Subject: Anyway to assign # of workers per server {} ?
Message-ID: <BLU144-W21A74E565CD6B9354A9A11D04C0@phx.gbl>


Hello,

Id like to isolate the resources each one of my server {} consumes... in case any of my sites gets DoSed .. id like to try to contain  possible effects related to used worker processes within its respective site. Is there anyway to achieve this?


Regards,
Mike
 		 	   		  
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20120402/53c045fa/attachment.html>

From nginx-forum at nginx.us  Mon Apr  2 04:56:56 2012
From: nginx-forum at nginx.us (niraj)
Date: Mon, 2 Apr 2012 00:56:56 -0400 (EDT)
Subject: Not easy replacement of Apache with Ngnix due to error log
Message-ID: <69aa2109190fe2d487f2667e9bfb1765.NginxMailingListEnglish@forum.nginx.org>

Hi All

         I having the html files and cgi  files for that I was using the
apache, Its running fine. 

I heard about the Nginx and I was trying to setup the same in locally
first. but during that I was getting the error 


the error of Nginx not reflect the exact issue with the application. 

the error of Nginx is not system admin friendly at all.   from the below
error we are not able to conclude anything.

how to solve the such kind of issue by checking the error log ??????


2012/03/30 09:58:05 [error] 29619#0: *1 upstream closed prematurely
FastCGI stdout while reading response header from upstream, client:
192.168.1.44, server: localhost, request: "GET
/cgi-bin/baybuy_vendx/index.cgi?vendx_action=myoffice_commonuser&userid=53616c7465645f5f1fed4bcaa10a7da569ecb5a53e5a718a&page=1b662ecc4610f940ec530e3bc8762169&BusinessGrID=1
HTTP/1.1", upstream: "fastcgi://unix:/var/run/fcgiwrap.socket:", host:
"192.168.1.225:90", referrer:
"http://192.168.1.225:90/cgi-bin/baybuy_vendx/index.cgi?vendx_action=validate_both2"

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,224747,224747#msg-224747


From ru at nginx.com  Mon Apr  2 06:16:10 2012
From: ru at nginx.com (Ruslan Ermilov)
Date: Mon, 2 Apr 2012 10:16:10 +0400
Subject: Anyway to assign # of workers per server {} ?
In-Reply-To: <BLU144-W21A74E565CD6B9354A9A11D04C0@phx.gbl>
References: <BLU144-W21A74E565CD6B9354A9A11D04C0@phx.gbl>
Message-ID: <20120402061610.GA56549@lo0.su>

On Mon, Apr 02, 2012 at 04:09:14AM +0000, Micheal Wolfskill wrote:
> Id like to isolate the resources each one of my server {} consumes... in
> case any of my sites gets DoSed .. id like to try to contain  possible
> effects related to used worker processes within its respective site. Is
> there anyway to achieve this?

If your servers use different address/port pairs, an obvious
solution would be to run several instances of nginx, one per
address/port.


From ru at nginx.com  Mon Apr  2 06:28:11 2012
From: ru at nginx.com (Ruslan Ermilov)
Date: Mon, 2 Apr 2012 10:28:11 +0400
Subject: Understanding host value in nginx error log files
In-Reply-To: <CAMgRdiXgjzjGH8ZrR4=NVMZMEBBn3h=x1Y6O94+rpVhrZFLmQQ@mail.gmail.com>
References: <CAMgRdiXgjzjGH8ZrR4=NVMZMEBBn3h=x1Y6O94+rpVhrZFLmQQ@mail.gmail.com>
Message-ID: <20120402062811.GB56549@lo0.su>

On Mon, Apr 02, 2012 at 10:19:12AM +1000, Kevan Stannard wrote:
> I typically see entries in my error.log files of hack attempts where the
> host entry is the IP address of my server, but I also see some entries with
> host values that are domain names I don't recognise.
> 
> Example 1:
> 
> 012/04/01 06:12:49 [error] 644#0: *882
> "/var/www/nginx-default/ab1e27867d53d8f4942095a891183ce0cacd3dbf/0d4fc9bfe7c5f26b02522d088dd98da95a9ed8d7/074977cbb342d6ffa7743ae477a5c0054fef5512/index.html"
> is not found (2: No such file or directory), client: 150.70.75.37, server:
> localhost, request: "GET
> /ab1e27867d53d8f4942095a891183ce0cacd3dbf/0d4fc9bfe7c5f26b02522d088dd98da95a9ed8d7/074977cbb342d6ffa7743ae477a5c0054fef5512/
> HTTP/1.0", host: "deepspacer.com"
> 
> Example 2:
> 
> 2012/02/03 01:38:41 [error] 592#0: *14019 open()
> "/var/www/nginx-default/home.php" failed (2: No such file or directory),
> client: 216.104.15.130, server: localhost, request: "GET
> /home.php?SES=517a4bfc0137889f05d67314df2715a1&from_diary=1&cpl=1&from=102_4
> HTTP/1.0", host: "www.au.mytelecomsurvey.com"
> 
> Example 3:
> 
> 2012/02/03 11:57:56 [error] 592#0: *18075 open()
> "/var/www/nginx-default/sites/default/files/js/js_b3ffc00633d66887fcb4ecdfc2d1c13a.jsmin.js"
> failed (2: No such file or directory), client: 150.70.64.197, server:
> localhost, request: "GET
> /sites/default/files/js/js_b3ffc00633d66887fcb4ecdfc2d1c13a.jsmin.js
> HTTP/1.0", host: "www.formalites-juridiques.net"
> 
> I was hoping someone could explain what it means if I'm seeing these domain
> names as host values and if it's something I need to be concerned about.

It is just a value of the Host request header field.  This could happen
due to client's DNS misconfiguration.  This could equally be a sign of
malicious entity probing your site.


From mansab.uppal at gmail.com  Mon Apr  2 06:33:17 2012
From: mansab.uppal at gmail.com (Mansab Uppal)
Date: Mon, 2 Apr 2012 12:03:17 +0530
Subject: Encountered '403 Forbidden Error'
Message-ID: <CA+Vmo2Aen3XMPeE0cz8+B3HDoL+fxgVbsW2KmTm6CL_w45uq0Q@mail.gmail.com>

After configuring the nginx on windows, received  '403 Frobidden Error'
while opening the web page for the web site's 'index.php' page placed at
the document root of the web server: html/v4/index.php'

Any suggestions ?
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20120402/dda92247/attachment.html>

From nginx-forum at nginx.us  Mon Apr  2 07:50:38 2012
From: nginx-forum at nginx.us (mscheel)
Date: Mon, 2 Apr 2012 03:50:38 -0400 (EDT)
Subject: $https with SSL proxy
Message-ID: <7050ef8172d12a965c73c428d3950776.NginxMailingListEnglish@forum.nginx.org>

Dear all,

starting with version 1.1.11 the $https variable is activated by
default. This is good as many applications rely on them because apache
is using this header.
However, now i cannot set or map this variable (nginx: [emerg] the
duplicate "https" variable). We have a load balancer in front of the web
servers which does all SSL traffic. The nginx servers can only differ
between http and https via a header sent by the load balancer, the nginx
port is the same. That hasn't been a problem in < 1.1.11. Many plugins,
codings in applications use this $https header so it would be nice if i
could assign it manually.

Any ideas?

Thanks.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,224756,224756#msg-224756


From appa at perusio.net  Mon Apr  2 08:00:09 2012
From: appa at perusio.net (Antonio P.P. Almeida)
Date: Mon, 2 Apr 2012 10:00:09 +0200
Subject: $https with SSL proxy
In-Reply-To: <7050ef8172d12a965c73c428d3950776.NginxMailingListEnglish@forum.nginx.org>
References: <7050ef8172d12a965c73c428d3950776.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <8211b4188417939839a07dd6e1b9b56f.squirrel@damiao.org>

> Dear all,
>
> starting with version 1.1.11 the $https variable is activated by
> default. This is good as many applications rely on them because apache
> is using this header.
> However, now i cannot set or map this variable (nginx: [emerg] the
> duplicate "https" variable). We have a load balancer in front of the web
> servers which does all SSL traffic. The nginx servers can only differ
> between http and https via a header sent by the load balancer, the nginx
> port is the same. That hasn't been a problem in < 1.1.11. Many plugins,
> codings in applications use this $https header so it would be nice if i
> could assign it manually.

Have you tried explicitly setting the fastcgi_param, like in:

fastcgi_param HTTPS $my_https;

where $my_https is a variable that you set as you want it?

--appa


From nginx-forum at nginx.us  Mon Apr  2 08:05:16 2012
From: nginx-forum at nginx.us (mscheel)
Date: Mon, 2 Apr 2012 04:05:16 -0400 (EDT)
Subject: $https with SSL proxy
In-Reply-To: <8211b4188417939839a07dd6e1b9b56f.squirrel@damiao.org>
References: <8211b4188417939839a07dd6e1b9b56f.squirrel@damiao.org>
Message-ID: <face6af75bb04bcc28024d1d88689e2a.NginxMailingListEnglish@forum.nginx.org>

Antonio P.P. Almeida Wrote:
-------------------------------------------------------
> > Dear all,
> >
> > starting with version 1.1.11 the $https variable
> is activated by
> > default. This is good as many applications rely
> on them because apache
> > is using this header.
> > However, now i cannot set or map this variable
> (nginx: [emerg] the
> > duplicate "https" variable). We have a load
> balancer in front of the web
> > servers which does all SSL traffic. The nginx
> servers can only differ
> > between http and https via a header sent by the
> load balancer, the nginx
> > port is the same. That hasn't been a problem in
> < 1.1.11. Many plugins,
> > codings in applications use this $https header
> so it would be nice if i
> > could assign it manually.
> 
> Have you tried explicitly setting the
> fastcgi_param, like in:
> 
> fastcgi_param HTTPS $my_https;
> 
> where $my_https is a variable that you set as you
> want it?
> 
> --appa
> 
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx

The problem is that all applications rely on this "https" header,
changing the code of all applications makes no sense to me.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,224756,224758#msg-224758


From appa at perusio.net  Mon Apr  2 08:10:52 2012
From: appa at perusio.net (Antonio P.P. Almeida)
Date: Mon, 2 Apr 2012 10:10:52 +0200
Subject: $https with SSL proxy
In-Reply-To: <face6af75bb04bcc28024d1d88689e2a.NginxMailingListEnglish@forum.nginx.org>
References: <8211b4188417939839a07dd6e1b9b56f.squirrel@damiao.org>
	<face6af75bb04bcc28024d1d88689e2a.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <bf8317aa1cf3b37949663c7492eba8fa.squirrel@damiao.org>

> The problem is that all applications rely on this "https" header,
> changing the code of all applications makes no sense to me.

AFAIK it's not a header but rather a parameter that is passed through CGI
and in PHP, for example, you recover via $_SERVER['HTTPS']. This has
nothing to do with headers. Of course you can set a header based on it.

X-Forwarded-Proto $https;

or

X-Forwarded-Proto $my_https;

---appa


From appa at perusio.net  Mon Apr  2 08:35:35 2012
From: appa at perusio.net (Antonio P.P. Almeida)
Date: Mon, 2 Apr 2012 10:35:35 +0200
Subject: $https with SSL proxy
In-Reply-To: <bf8317aa1cf3b37949663c7492eba8fa.squirrel@damiao.org>
References: <8211b4188417939839a07dd6e1b9b56f.squirrel@damiao.org>
	<face6af75bb04bcc28024d1d88689e2a.NginxMailingListEnglish@forum.nginx.org>
	<bf8317aa1cf3b37949663c7492eba8fa.squirrel@damiao.org>
Message-ID: <eaa2ba3abdde971d1f910aee05ec6a76.squirrel@damiao.org>

>> The problem is that all applications rely on this "https" header,
>> changing the code of all applications makes no sense to me.
>
> AFAIK it's not a header but rather a parameter that is passed through CGI
> and in PHP, for example, you recover via $_SERVER['HTTPS']. This has
> nothing to do with headers. Of course you can set a header based on it.
>
> X-Forwarded-Proto $https;
>
> or
>
> X-Forwarded-Proto $my_https;

Oops. My bad. I meant:

X-Forwarded-Proto $scheme;


--appa


From nginx-forum at nginx.us  Mon Apr  2 08:39:55 2012
From: nginx-forum at nginx.us (mscheel)
Date: Mon, 2 Apr 2012 04:39:55 -0400 (EDT)
Subject: $https with SSL proxy
In-Reply-To: <7050ef8172d12a965c73c428d3950776.NginxMailingListEnglish@forum.nginx.org>
References: <7050ef8172d12a965c73c428d3950776.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <af386101762d936d8396d5296877ed35.NginxMailingListEnglish@forum.nginx.org>

Your right, setting it for fastcgi as param works fine. I had a problem
in the past setting variables in the fastcgi_params, they've just been
ignored. This seems to be gone.
Then i have to rewrite the nginx internal logic of my
substitions/mappings based on this variable.

I still think allowing a manual override of this variable should be
allowed. If any other plugin relies on this header users with SSL
Accelerators will definately have a problem.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,224756,224761#msg-224761


From nginx-forum at nginx.us  Mon Apr  2 09:23:30 2012
From: nginx-forum at nginx.us (shrinath.m)
Date: Mon, 2 Apr 2012 05:23:30 -0400 (EDT)
Subject: if proxy_read_timeout - then do something?
Message-ID: <8d456b14891f54100a6c46111ad2eba9.NginxMailingListEnglish@forum.nginx.org>

Hi, 

I would like to put a proxy_read_timeout and if it times out, I want to
do a specific action in php and return some response to the user. Is
this possible? How? 

Thanks, 
Shrinath

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,224764,224764#msg-224764


From john at jmsd.co.uk  Mon Apr  2 09:50:35 2012
From: john at jmsd.co.uk (John Moore)
Date: Mon, 02 Apr 2012 10:50:35 +0100
Subject: nginx as remote proxy - limit number of concurrent connections?
Message-ID: <4F79766B.2070909@jmsd.co.uk>

I'm using nginx as a remote proxy server in front of a bunch of Tomcat 
instances, and occasionally I'm running into a situation where all 
available threads are used up in the Tomcat instances, which causes 
various problems. Rather than trying to deal with these problems in that 
layer, I was wondering about preventing the situation from arising by 
limiting the number of concurrent connections to the upstream servers in 
nginx. I'm not entirely sure how to do this, but I have no doubt it's 
doable. Suggestions, please!


From appa at perusio.net  Mon Apr  2 09:54:04 2012
From: appa at perusio.net (Antonio P.P. Almeida)
Date: Mon, 2 Apr 2012 11:54:04 +0200
Subject: nginx as remote proxy - limit number of concurrent connections?
In-Reply-To: <4F79766B.2070909@jmsd.co.uk>
References: <4F79766B.2070909@jmsd.co.uk>
Message-ID: <bcb5b968d0bdd7bc47c30a1160ffcd63.squirrel@damiao.org>

> I'm using nginx as a remote proxy server in front of a bunch of Tomcat
> instances, and occasionally I'm running into a situation where all
> available threads are used up in the Tomcat instances, which causes
> various problems. Rather than trying to deal with these problems in that
> layer, I was wondering about preventing the situation from arising by
> limiting the number of concurrent connections to the upstream servers in
> nginx. I'm not entirely sure how to do this, but I have no doubt it's
> doable. Suggestions, please!

Try this: http://nginx.localdomain.pl/wiki/UpstreamFair

with the weight_mode=peak option.

--appa


From mdounin at mdounin.ru  Mon Apr  2 10:57:08 2012
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Mon, 2 Apr 2012 14:57:08 +0400
Subject: if proxy_read_timeout - then do something?
In-Reply-To: <8d456b14891f54100a6c46111ad2eba9.NginxMailingListEnglish@forum.nginx.org>
References: <8d456b14891f54100a6c46111ad2eba9.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20120402105708.GB13466@mdounin.ru>

Hello!

On Mon, Apr 02, 2012 at 05:23:30AM -0400, shrinath.m wrote:

> Hi, 
> 
> I would like to put a proxy_read_timeout and if it times out, I want to
> do a specific action in php and return some response to the user. Is
> this possible? How? 

http://nginx.org/r/error_page

You have to handle 504 (Gateway Timeout) errors.

Maxim Dounin


From mdounin at mdounin.ru  Mon Apr  2 11:17:54 2012
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Mon, 2 Apr 2012 15:17:54 +0400
Subject: Socket leaks., pread and [crit] SSL_Write() in 1.0.14
In-Reply-To: <jl813b$nn0$1@dough.gmane.org>
References: <3b3f2c2eddfdd3e59d242463275e502e.NginxMailingListEnglish@forum.nginx.org>
	<00cd479fdbb8bdf2f9f844e88607f8bd.NginxMailingListEnglish@forum.nginx.org>
	<20120326164754.GL13466@mdounin.ru> <jl813b$nn0$1@dough.gmane.org>
Message-ID: <20120402111754.GD13466@mdounin.ru>

Hello!

On Sat, Mar 31, 2012 at 06:39:37PM -0400, Floren Munteanu wrote:

> Hi Maxim,
> 
> On 3/26/2012 12:47 PM, Maxim Dounin wrote:
> >As already suggested - you may build nginx with any particular
> >openssl version statically, by using --with-openssl= configure
> >argument.
> 
> I followed your advice and built a backlevel RPM for libcripto.so6
> and libssl.so6 so none of the deps are broken in CentOS 5. Then, I
> built the OpenSSL 1.0.1 RPM's and rebuilt Nginx against the latest
> libs:
> # yum list openssl* nginx
> Loaded plugins: fastestmirror
> Loading mirror speeds from cached hostfile
>  * base: mirrors.manchester.icecolo.com
>  * extras: mirrors.manchester.icecolo.com
>  * updates: mirrors.manchester.icecolo.com
> Installed Packages
> nginx.x86_64		1.0.14-1.el5	installed
> openssl.x86_64		1.0.1-1.el5	installed
> openssl-libs.x86_64	1.0.1-1.el5	installed
> openssl098e.x86_64	0.9.8e-1.el5	installed
> 
> # nginx -V
> nginx version: nginx/1.0.14
> built by gcc 4.1.2 20080704 (Red Hat 4.1.2-52)
> TLS SNI support enabled
> configure arguments: --user=nginx --group=nginx
> --prefix=/usr/share/nginx --sbin-path=/usr/sbin/nginx
> --conf-path=/etc/nginx/nginx.conf --pid-path=/var/run/nginx.pid
> --error-log-path=/var/log/nginx/error.log
> --http-log-path=/var/log/nginx/access.log
> --http-client-body-temp-path=/var/lib/nginx/client
> --http-fastcgi-temp-path=/var/lib/nginx/fastcgi
> --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/subsys/nginx --with-cc-opt='-O3 -g -m64
> -mtune=nocona -m128bit-long-double -mmmx -msse3 -mfpmath=sse'
> --with-file-aio --with-http_addition_module --with-http_dav_module
> --with-http_degradation_module --with-http_flv_module
> --with-http_geoip_module --with-http_gzip_static_module
> --with-http_image_filter_module --with-http_mp4_module
> --with-http_perl_module --with-http_random_index_module
> --with-http_realip_module --with-http_secure_link_module
> --with-http_ssl_module --with-http_stub_status_module
> --with-http_sub_module --with-http_xslt_module --with-mail
> --with-mail_ssl_module --with-poll_module --with-rtsig_module
> --with-select_module

Please also check if nginx actually uses new openssl library, ldd 
should be helpful here.

> 
> http {
> 	...
> 	ssl_prefer_server_ciphers	on;
> 	ssl_ciphers			RC4:HIGH:!aNULL:!MD5;
> 	ssl_protocols			SSLv3 TLSv1 TLSv1.1 TLSv1.2;
> 	ssl_session_cache		shared:SSL:5m;
> 	ssl_session_timeout		10m;
> 	...
> 
> 	server {
> 		listen			192.168.1.3:443 ssl default_server;
> 		server_name		www.domain.com;
> 		access_log		off;
> 		error_log		/var/log/nginx/localhost.error.log      error;
> 		root			/var/www/domain.com;
> 		index			index.php index.html;
> 		ssl_certificate		domain.com.crt;
> 		ssl_certificate_key	domain.com.key;
> 		...
> 	}
> }
> 
> Even if I eliminated the OpenSSL version issues, I still have random
> [crit] SSL_write() failures at the same frequency as before. They
> are also accompanied by open socket alerts, of this format:
> [alert] 2380#0: open socket #34 left in connection 12
> 
> I'm looking forward to your suggestions.

As already suggested, it whould be cool to check if you see the 
same problem in 1.1.x.

And to proceed further we need debug log, see here:

http://wiki.nginx.org/Debugging

Note you'll need to recompile nginx with "--with-debug" configure 
argument to obtain one.

Maxim Dounin


From nginx-forum at nginx.us  Mon Apr  2 13:28:54 2012
From: nginx-forum at nginx.us (shrinath.m)
Date: Mon, 2 Apr 2012 09:28:54 -0400 (EDT)
Subject: if proxy_read_timeout - then do something?
In-Reply-To: <8d456b14891f54100a6c46111ad2eba9.NginxMailingListEnglish@forum.nginx.org>
References: <8d456b14891f54100a6c46111ad2eba9.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <98a3cd9dbeb8fda659db0536b7fca9e7.NginxMailingListEnglish@forum.nginx.org>

I tried to catch 504, not working. Anything wrong in the following? 


error_page 504 =200 @checkstatus;

location @checkstatus
{
    rewrite ^(.*)$ /k/binarystatus/check?original_uri=$1 break;
    proxy_pass https://shrinath.blah.com;
    proxy_hide_header Last-Modified;
    proxy_read_timeout 10s;
    proxy_intercept_errors on;
}

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,224764,224776#msg-224776


From nginx-forum at nginx.us  Mon Apr  2 13:37:01 2012
From: nginx-forum at nginx.us (shrinath.m)
Date: Mon, 2 Apr 2012 09:37:01 -0400 (EDT)
Subject: if proxy_read_timeout - then do something?
In-Reply-To: <98a3cd9dbeb8fda659db0536b7fca9e7.NginxMailingListEnglish@forum.nginx.org>
References: <8d456b14891f54100a6c46111ad2eba9.NginxMailingListEnglish@forum.nginx.org>
	<98a3cd9dbeb8fda659db0536b7fca9e7.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <10e88af288d55ad5b2e0a390b613fd47.NginxMailingListEnglish@forum.nginx.org>

Oops... I take it back, it works... 
Ignorant me had one more line few lines above the said code: 

error_page 500 504 =500 @error50x;

Solved now... Thanks :)

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,224764,224777#msg-224777


From admin at yqed.com  Mon Apr  2 16:15:10 2012
From: admin at yqed.com (Floren Munteanu)
Date: Mon, 02 Apr 2012 12:15:10 -0400
Subject: Socket leaks., pread and [crit] SSL_Write() in 1.0.14
In-Reply-To: <20120402111754.GD13466@mdounin.ru>
References: <3b3f2c2eddfdd3e59d242463275e502e.NginxMailingListEnglish@forum.nginx.org>
	<00cd479fdbb8bdf2f9f844e88607f8bd.NginxMailingListEnglish@forum.nginx.org>
	<20120326164754.GL13466@mdounin.ru> <jl813b$nn0$1@dough.gmane.org>
	<20120402111754.GD13466@mdounin.ru>
Message-ID: <jlcjag$a19$1@dough.gmane.org>

Hi Maxim,

On 4/2/2012 7:17 AM, Maxim Dounin wrote:
> As already suggested, it whould be cool to check if you see the
> same problem in 1.1.x.
>
> And to proceed further we need debug log, see here:
>
> http://wiki.nginx.org/Debugging
>
> Note you'll need to recompile nginx with "--with-debug" configure
> argument to obtain one.

How safe is to put in production a debug built of Nginx?
I could easily compile a different built, but I'm concerned of running 
this on a live environment as this is where the errors occur.

Regards,

Floren Munteanu



From mdounin at mdounin.ru  Mon Apr  2 17:06:37 2012
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Mon, 2 Apr 2012 21:06:37 +0400
Subject: Socket leaks., pread and [crit] SSL_Write() in 1.0.14
In-Reply-To: <jlcjag$a19$1@dough.gmane.org>
References: <3b3f2c2eddfdd3e59d242463275e502e.NginxMailingListEnglish@forum.nginx.org>
	<00cd479fdbb8bdf2f9f844e88607f8bd.NginxMailingListEnglish@forum.nginx.org>
	<20120326164754.GL13466@mdounin.ru> <jl813b$nn0$1@dough.gmane.org>
	<20120402111754.GD13466@mdounin.ru> <jlcjag$a19$1@dough.gmane.org>
Message-ID: <20120402170637.GJ13466@mdounin.ru>

Hello!

On Mon, Apr 02, 2012 at 12:15:10PM -0400, Floren Munteanu wrote:

> Hi Maxim,
> 
> On 4/2/2012 7:17 AM, Maxim Dounin wrote:
> >As already suggested, it whould be cool to check if you see the
> >same problem in 1.1.x.
> >
> >And to proceed further we need debug log, see here:
> >
> >http://wiki.nginx.org/Debugging
> >
> >Note you'll need to recompile nginx with "--with-debug" configure
> >argument to obtain one.
> 
> How safe is to put in production a debug built of Nginx?
> I could easily compile a different built, but I'm concerned of
> running this on a live environment as this is where the errors
> occur.

It's safe and only adds minimal overhead on logging level checks 
as long as debug log isn't enabled.

Note though that writing debug logs consume extra resources, and 
it may be noticeable under high load.  The debug_connection 
directive allows to enable debug logging only for selected ip 
addresses to minimize impact.

Maxim Dounin


From citrin at citrin.ru  Mon Apr  2 17:22:39 2012
From: citrin at citrin.ru (Anton Yuzhaninov)
Date: Mon, 02 Apr 2012 21:22:39 +0400
Subject: Not easy replacement of Apache with Ngnix due to error log
In-Reply-To: <69aa2109190fe2d487f2667e9bfb1765.NginxMailingListEnglish@forum.nginx.org>
References: <69aa2109190fe2d487f2667e9bfb1765.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <4F79E05F.1040302@citrin.ru>

On 04/02/12 08:56, niraj wrote:
> the error of Nginx is not system admin friendly at all.   from the below
> error we are not able to conclude anything.

nginx error logs is very friendly and useful.

> how to solve the such kind of issue by checking the error log ??????
>
>
> 2012/03/30 09:58:05 [error] 29619#0: *1 upstream closed prematurely
> FastCGI stdout while reading response header from upstream, client:
> 192.168.1.44, server: localhost, request: "GET
> /cgi-bin/baybuy_vendx/index.cgi?vendx_action=myoffice_commonuser&userid=53616c7465645f5f1fed4bcaa10a7da569ecb5a53e5a718a&page=1b662ecc4610f940ec530e3bc8762169&BusinessGrID=1
> HTTP/1.1", upstream: "fastcgi://unix:/var/run/fcgiwrap.socket:", host:
> "192.168.1.225:90", referrer:
> "http://192.168.1.225:90/cgi-bin/baybuy_vendx/index.cgi?vendx_action=validate_both2"

words "upstream closed prematurely FastCGI stdout" show, that problem is 
probably in FastCGI server.

See FastCGI manager error logs, and OS logs (one of usual problem is segfault in 
FastCGI process).

-- 
  Anton Yuzhaninov


From nginx-forum at nginx.us  Mon Apr  2 21:21:39 2012
From: nginx-forum at nginx.us (Nginx_User777)
Date: Mon, 2 Apr 2012 17:21:39 -0400 (EDT)
Subject: memc_flags_to_last_modified and srcache , no 304 possible ?
Message-ID: <6e2bc7fac8512842fa34ca858100f61b.NginxMailingListEnglish@forum.nginx.org>

Hello All, 

Is it possible to use the memc_flags_to_last_modified directive with
srcache ? 
It doesn't work for me. 

When a browser request nginx with a "If-Modified-Since" header, I would
like that nginx lookup in memcached to verify if the "Last Modified" is
the same and honor the 304 request without hit the proxied server (in
another country). 

Impossible to get a 304, so no bandwith saved ... But my provider is
happy :) 

Thanks for your help ! 
I working on for several days, i'll really go mad :) 


My config: 

#user  nobody;
worker_processes  1;

events {
    worker_connections  1024;
}


http {
    include       mime.types;
    default_type  application/octet-stream;
    sendfile        on;
    keepalive_timeout  0;
    server_tokens           off;

    upstream memc1 {
        server XX.XX.XX.XX:11211;
        server unix:/tmp/memcached.sock backup;
        keepalive 1024;
    }
    upstream memc2 {
        server XX.XX.XX.XX:11211;
        keepalive 1024;
    }
    upstream_list MemCluster memc1 memc2;

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

    server {
        listen       80;
        server_name  localhost;
        
     location = /memc {
            internal;

        set $memc_key $query_string;
        set $memc_exptime 0;
        set_hashed_upstream $backend MemCluster $memc_key;
        memc_pass $backend;
        memc_flags_to_last_modified on;
        }

        location / {
            set $key $uri;
            srcache_fetch GET /memc $key;
            srcache_store PUT /memc $key;

            proxy_pass http://the-far-far-away-server;
        }
        error_page   500 502 503 504  /50x.html;
        location = /50x.html {
            root   html;
        }
    }
}


Regards

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,224793,224793#msg-224793


From siefke_listen at web.de  Mon Apr  2 22:13:23 2012
From: siefke_listen at web.de (Silvio Siefke)
Date: Tue, 3 Apr 2012 00:13:23 +0200
Subject: PHP with FPM on Gentoo
Message-ID: <20120403001323.c03997dd.siefke_listen@web.de>

Hello,

i try run Nginx with PHP-FPM under Gentoo. Under my FreeBSD Rootserver 
it run without any problems but under Gentoo want not run. 

nginx.conf > http://nopaste.info/72a40e9c5e.html
php-fpm.conf > http://nopaste.info/94a9de62d5.html
emerge --info php nginx > http://nopaste.info/225b12e106.html

The last change in config is:

server {
listen 80 default;
server_name localhost;
access_log /var/log/nginx/localhost.access_log main;
error_log /var/log/nginx/localhost.error_log debug;
root /var/www/localhost/htdocs;
location ~ (.*)\.php$ {
include fastcgi.conf;
fastcgi_pass 127.0.0.1:9000; 
fastcgi_index index.php;
fastcgi_param SCRIPT_FILENAME /var/www/localhost/htdocs$fastcgi_script_name;
}        
}

The last Logs which i has:

<acces_log>
192.168.0.20 - - [02/Apr/2012:17:59:05 +0200] "GET / HTTP/1.1" 200 633 "-" "Mozilla/5.0 (Windows NT 5.1) AppleWebKit/535.19 (KHTML, like Gecko) Chrome/18.0.1025.142 Safari/535.19" "-"
192.168.0.20 - - [02/Apr/2012:17:59:05 +0200] "GET /favicon.ico HTTP/1.1" 200 1650 "-" "Mozilla/5.0 (Windows NT 5.1) AppleWebKit/535.19 (KHTML, like Gecko) Chrome/18.0.1025.142 Safari/535.19" "-"
192.168.0.20 - - [02/Apr/2012:17:59:21 +0200] "GET /info.php HTTP/1.1" 200 289 "-" "Mozilla/5.0 (Windows NT 5.1) AppleWebKit/535.19 (KHTML, like Gecko) Chrome/18.0.1025.142 Safari/535.19" "0.47"
192.168.0.20 - - [02/Apr/2012:17:59:24 +0200] "-" 400 0 "-" "-" "-"
</acces_log>

<error_log>
2012/04/02 17:59:24 [info] 21288#0: *4 client closed prematurely connection while reading client request line, client: 192.168.0.20, server: localhost
</error_log>

I really not understand where is my mistake. Has someone a idea?


Regards
Silvio


From kevanstannard at gmail.com  Mon Apr  2 23:15:25 2012
From: kevanstannard at gmail.com (Kevan Stannard)
Date: Tue, 3 Apr 2012 09:15:25 +1000
Subject: Understanding host value in nginx error log files
In-Reply-To: <20120402062811.GB56549@lo0.su>
References: <CAMgRdiXgjzjGH8ZrR4=NVMZMEBBn3h=x1Y6O94+rpVhrZFLmQQ@mail.gmail.com>
	<20120402062811.GB56549@lo0.su>
Message-ID: <CAMgRdiUh=Eo+42n=PW3B_P1DjKr3_gQrFtvs6o+hg1Re4h8xEQ@mail.gmail.com>

Thanks Ruslan

On 2 April 2012 16:28, Ruslan Ermilov <ru at nginx.com> wrote:

> On Mon, Apr 02, 2012 at 10:19:12AM +1000, Kevan Stannard wrote:
> > I typically see entries in my error.log files of hack attempts where the
> > host entry is the IP address of my server, but I also see some entries
> with
> > host values that are domain names I don't recognise.
> >
> > Example 1:
> >
> > 012/04/01 06:12:49 [error] 644#0: *882
> >
> "/var/www/nginx-default/ab1e27867d53d8f4942095a891183ce0cacd3dbf/0d4fc9bfe7c5f26b02522d088dd98da95a9ed8d7/074977cbb342d6ffa7743ae477a5c0054fef5512/index.html"
> > is not found (2: No such file or directory), client: 150.70.75.37,
> server:
> > localhost, request: "GET
> >
> /ab1e27867d53d8f4942095a891183ce0cacd3dbf/0d4fc9bfe7c5f26b02522d088dd98da95a9ed8d7/074977cbb342d6ffa7743ae477a5c0054fef5512/
> > HTTP/1.0", host: "deepspacer.com"
> >
> > Example 2:
> >
> > 2012/02/03 01:38:41 [error] 592#0: *14019 open()
> > "/var/www/nginx-default/home.php" failed (2: No such file or directory),
> > client: 216.104.15.130, server: localhost, request: "GET
> >
> /home.php?SES=517a4bfc0137889f05d67314df2715a1&from_diary=1&cpl=1&from=102_4
> > HTTP/1.0", host: "www.au.mytelecomsurvey.com"
> >
> > Example 3:
> >
> > 2012/02/03 11:57:56 [error] 592#0: *18075 open()
> >
> "/var/www/nginx-default/sites/default/files/js/js_b3ffc00633d66887fcb4ecdfc2d1c13a.jsmin.js"
> > failed (2: No such file or directory), client: 150.70.64.197, server:
> > localhost, request: "GET
> > /sites/default/files/js/js_b3ffc00633d66887fcb4ecdfc2d1c13a.jsmin.js
> > HTTP/1.0", host: "www.formalites-juridiques.net"
> >
> > I was hoping someone could explain what it means if I'm seeing these
> domain
> > names as host values and if it's something I need to be concerned about.
>
> It is just a value of the Host request header field.  This could happen
> due to client's DNS misconfiguration.  This could equally be a sign of
> malicious entity probing your site.
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx
>



-- 
Kevan Stannard
Mob: 0411 757 433
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20120403/f5972b54/attachment.html>

From admin at yqed.com  Tue Apr  3 01:08:47 2012
From: admin at yqed.com (Floren Munteanu)
Date: Mon, 02 Apr 2012 21:08:47 -0400
Subject: Socket leaks., pread and [crit] SSL_Write() in 1.0.14
In-Reply-To: <20120402170637.GJ13466@mdounin.ru>
References: <3b3f2c2eddfdd3e59d242463275e502e.NginxMailingListEnglish@forum.nginx.org>
	<00cd479fdbb8bdf2f9f844e88607f8bd.NginxMailingListEnglish@forum.nginx.org>
	<20120326164754.GL13466@mdounin.ru> <jl813b$nn0$1@dough.gmane.org>
	<20120402111754.GD13466@mdounin.ru> <jlcjag$a19$1@dough.gmane.org>
	<20120402170637.GJ13466@mdounin.ru>
Message-ID: <jldij1$k02$1@dough.gmane.org>

Hi Maxim,

On 4/2/2012 1:06 PM, Maxim Dounin wrote:
> It's safe and only adds minimal overhead on logging level checks
> as long as debug log isn't enabled.
>
> Note though that writing debug logs consume extra resources, and
> it may be noticeable under high load.  The debug_connection
> directive allows to enable debug logging only for selected ip
> addresses to minimize impact.

I've built a nginx-debug RPM but I'm tempted to try 1.1.18 now. :)
I will use instead 1.0.14, just in case is a nasty bug hidden somewhere 
in Nginx. In this way, we fix both versions. These are the compile 
cflags I used:
gcc -c -pipe  -O -W -Wall -Wpointer-arith -Wno-unused-parameter 
-Wunused-function -Wunused-variable -Wunused-value -Werror -g -O3 -g 
-m64 -mtune=nocona -m128bit-long-double -mmmx -msse3 -mfpmath=sse

I'm going to leave the error_log set on error and give it a 256k buffer 
so it should not affect the performance. Since we only deal [crit], I 
don't think we need to debug anything lower than error.

I'll add the following settings into http:
debug_points       abort;
worker_rlimit_core 512M;
working_directory  /var/lib/nginx/;

Let me know please if I should do anything extra.

Regards,

Floren


From tdgh2323 at hotmail.com  Tue Apr  3 04:56:20 2012
From: tdgh2323 at hotmail.com (Micheal Wolfskill)
Date: Tue, 3 Apr 2012 04:56:20 +0000
Subject: Possible trigger actio of clearing cache within location /clearcache
	{ }
Message-ID: <BLU144-W56667C01074F0569911D2D04D0@phx.gbl>


Is it possible to somehow (and without installing PHP) invoking the action of clearing the nginx cache (rm -fr /var/www/cache/*) by simply visiting a location {} of the server {} ?  Obviously I would allow only specific IPs to that location.


Regards,
Mike
 		 	   		  
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20120403/751e8f0d/attachment.html>

From nginx-forum at nginx.us  Tue Apr  3 05:47:53 2012
From: nginx-forum at nginx.us (niraj)
Date: Tue, 3 Apr 2012 01:47:53 -0400 (EDT)
Subject: Not easy replacement of Apache with Ngnix due to error log
In-Reply-To: <4F79E05F.1040302@citrin.ru>
References: <4F79E05F.1040302@citrin.ru>
Message-ID: <a13b9ae9a9d46496a448f4d3d4454f51.NginxMailingListEnglish@forum.nginx.org>

Hi  Anton 

               As u suggested,
               words "upstream closed prematurely FastCGI stdout" show,
that problem is
probably in FastCGI server.


                  This is what I am telling u that Nginx error log
misguide us.

              Actually Problem was one perl module was not installed in
that so I was getting the above error. 


     Earlier I used apache so if any module related error then it was
reflect in apache's error log ..............


that actually not happening with Nginx , this create a problem for
system admin.

If any idea to get details error logs for nginx  then please sugeest.

how to set  fcgiwraper error log ??????

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,224747,224814#msg-224814


From quintinpar at gmail.com  Tue Apr  3 06:43:35 2012
From: quintinpar at gmail.com (Quintin Par)
Date: Tue, 3 Apr 2012 12:13:35 +0530
Subject: Serve from cache but fire request to upstream server to increment
	page view counter
Message-ID: <CACW9FdNgfxGejBvzWoqbM7+ikNpZxum82MmmS0aFd6WsK3AOUw@mail.gmail.com>

Hi all,

I have a strange case, not sure if this is addressed by nginx yet

My site is cached as follows.

  location = / {

    proxy_pass  http://localhost:82;

    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 Accept-Encoding "";

    proxy_ignore_headers Set-Cookie;

    proxy_ignore_headers Cache-Control;

    proxy_ignore_headers Expires;

    proxy_ignore_headers X-Accel-Expires;

    add_header X-Cache-Status $upstream_cache_status;

    proxy_cache             cache;

    proxy_cache_key         $scheme$host$request_uri$cookie_site_sessionid;

    proxy_cache_valid       200 302 30s;

    proxy_cache_use_stale   updating;

  }

One issue I encounter with this is that I can?t increment the page view
counter that I maintain in redis.

Is there is way nginx can fire a request to the backend web ? to some URI
so that I can take the request and increment the associated counter?

I know I can achieve this with SSI, but I wanted to check if there is a
better pattern.

-Quintin
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20120403/2891d4f5/attachment.html>

From nginx-forum at nginx.us  Tue Apr  3 08:02:14 2012
From: nginx-forum at nginx.us (locojohn)
Date: Tue, 3 Apr 2012 04:02:14 -0400 (EDT)
Subject: PHP with FPM on Gentoo
In-Reply-To: <20120403001323.c03997dd.siefke_listen@web.de>
References: <20120403001323.c03997dd.siefke_listen@web.de>
Message-ID: <728e86fdb5c831482000d8acc06540fb.NginxMailingListEnglish@forum.nginx.org>

Hi,

Do you actually get any content, apart from errors in nginx error_log?  
The message "client closed prematurely connection while reading client
request line, client: ..." shouldn't be a problem itself.

"Some modern browsers tend to open 2 tcp connections to server from
start, and if you are loading a single static file - one of them remains
unused.  The browser eventually closes it without sending any request,
and this causes the above message in logs." - Maxim Dounin.

I can see from your access_log that requests are processed normally and
responses are OK (code 200) for all your files including PHP files
(request: /info.php).  

You can also use sockets for fastcgi connections, but that's a matter of
preference and probably won't solve your issue, which I do not fully
understand yet.

shell:

mkdir -p /var/run/fastcgi/php-fpm.sock

php-fpm.conf:

;listen = 127.0.0.1:9000
;listen.allowed_clients = 127.0.0.1
listen = /var/run/fastcgi/php-fpm.sock
listen.owner = nginx
listen.group = nginx

nginx.conf:

location \.php$ {
  include fastcgi_params;
  fastcgi_pass unix:/var/run/fastcgi/php-fpm.sock;
}

Andrejs

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,224804,224821#msg-224821


From gpakosz at yahoo.fr  Tue Apr  3 08:49:40 2012
From: gpakosz at yahoo.fr (=?ISO-8859-1?Q?Gr=E9gory_Pakosz?=)
Date: Tue, 3 Apr 2012 10:49:40 +0200
Subject: nginx + php-fpm "modern configuration way"
Message-ID: <CAC_01E2acR3-wVdXi4gX3PXrbi93JQvcYfQKzafdo-bvaU=7cQ@mail.gmail.com>

Hello,

Everyone knows there are many nginx tutorials out there and I would like to
know what's the "modern way" when writing a php location block, I'm doing
(nginx 1.1.:

location ~ \.php$ {
  try_files $uri =404;

  include fastcgi_params;
  fastcgi_pass unix:/var/run/php5-fpm/www-data.sock;
}

And it works pretty well. I quick tested phpinfo, drupal, joomla, wordpress
and they all run correctly with SEF urls etc.
However, I often see the following directives being used, how
important/useful are they?

fastcgi_split_path_info ^(.+\.php)(/.+)$;
fastcgi_param SCRIPT_FILENAME $document_root$fastcgi_script_name;
fastcgi_param SCRIPT_NAME $fastcgi_script_name;

Regards,
Gregory
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20120403/8476556f/attachment.html>

From edho at myconan.net  Tue Apr  3 08:51:30 2012
From: edho at myconan.net (Edho Arief)
Date: Tue, 3 Apr 2012 15:51:30 +0700
Subject: nginx + php-fpm "modern configuration way"
In-Reply-To: <CAC_01E2acR3-wVdXi4gX3PXrbi93JQvcYfQKzafdo-bvaU=7cQ@mail.gmail.com>
References: <CAC_01E2acR3-wVdXi4gX3PXrbi93JQvcYfQKzafdo-bvaU=7cQ@mail.gmail.com>
Message-ID: <CAAtReCmdXpCN_bAW26+_UWe-_16c9=z1g7EEugL_A65L7vwD1g@mail.gmail.com>

2012/4/3 Gr?gory Pakosz <gpakosz at yahoo.fr>:
> However, I often see the following directives being used, how
> important/useful are they?
>
> fastcgi_split_path_info ^(.+\.php)(/.+)$;
> fastcgi_param SCRIPT_FILENAME $document_root$fastcgi_script_name;
> fastcgi_param SCRIPT_NAME $fastcgi_script_name;
>

Those are to handle urls like www.example.com/index.php/some/para/meters.


From gpakosz at yahoo.fr  Tue Apr  3 09:02:07 2012
From: gpakosz at yahoo.fr (=?ISO-8859-1?Q?Gr=E9gory_Pakosz?=)
Date: Tue, 3 Apr 2012 11:02:07 +0200
Subject: nginx + php-fpm "modern configuration way"
In-Reply-To: <CAAtReCmdXpCN_bAW26+_UWe-_16c9=z1g7EEugL_A65L7vwD1g@mail.gmail.com>
References: <CAC_01E2acR3-wVdXi4gX3PXrbi93JQvcYfQKzafdo-bvaU=7cQ@mail.gmail.com>
	<CAAtReCmdXpCN_bAW26+_UWe-_16c9=z1g7EEugL_A65L7vwD1g@mail.gmail.com>
Message-ID: <CAC_01E0fnOhehvA=7vAOzueTZOuzzwb=_BuGJbXs5c_0zFnW+g@mail.gmail.com>

>
> >
> > fastcgi_split_path_info ^(.+\.php)(/.+)$;
> > fastcgi_param SCRIPT_FILENAME $document_root$fastcgi_script_name;
> > fastcgi_param SCRIPT_NAME $fastcgi_script_name;
> >
>
> Those are to handle urls like www.example.com/index.php/some/para/meters.
>
>
Well then Wordpress, Drupal and Joomla can do without them. From your
remark, I deduce some other PHP scripts can't

Regards,
Gregory
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20120403/cc3fbdf7/attachment.html>

From nginx-forum at nginx.us  Tue Apr  3 09:05:55 2012
From: nginx-forum at nginx.us (locojohn)
Date: Tue, 3 Apr 2012 05:05:55 -0400 (EDT)
Subject: nginx + php-fpm "modern configuration way"
In-Reply-To: <CAAtReCmdXpCN_bAW26+_UWe-_16c9=z1g7EEugL_A65L7vwD1g@mail.gmail.com>
References: <CAAtReCmdXpCN_bAW26+_UWe-_16c9=z1g7EEugL_A65L7vwD1g@mail.gmail.com>
Message-ID: <27b4608086c25028f86c8e29d09e102e.NginxMailingListEnglish@forum.nginx.org>

Edho Arief Wrote:
-------------------------------------------------------

> Those are to handle urls like
> www.example.com/index.php/some/para/meters.
> 

Or, in other words, where PHP scripts require PATH_INFO /
PATH_TRANSLATED environment variables to be set appropriately to the
request, such as shown above.  It isn't a popular practice these days,
though, because URLs can be something like:
/index.php/path/to/another/index.php/path...  

Andrejs

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,224829,224834#msg-224834


From appa at perusio.net  Tue Apr  3 09:07:16 2012
From: appa at perusio.net (Antonio P.P. Almeida)
Date: Tue, 3 Apr 2012 11:07:16 +0200
Subject: nginx + php-fpm "modern configuration way"
In-Reply-To: <CAC_01E2acR3-wVdXi4gX3PXrbi93JQvcYfQKzafdo-bvaU=7cQ@mail.gmail.com>
References: <CAC_01E2acR3-wVdXi4gX3PXrbi93JQvcYfQKzafdo-bvaU=7cQ@mail.gmail.com>
Message-ID: <74a27dfa34e47e04723227174f0a1acc.squirrel@damiao.org>

> Hello,
>
> Everyone knows there are many nginx tutorials out there and I would like
> to
> know what's the "modern way" when writing a php location block, I'm doing
> (nginx 1.1.:
>
> location ~ \.php$ {
>   try_files $uri =404;
>
>   include fastcgi_params;
>   fastcgi_pass unix:/var/run/php5-fpm/www-data.sock;
> }

This is a "generic" way. This way you don't enumerate all the PHP scripts
that are to be executed. The above is just the Nginx translation of
"default" Apache way.

There are better ways IMHO. But they require more effort to put in place.

Also there's no need to use fcgi_split_pathinfo even if your app uses
PATHINFO, like Chive for example. You can get the same employing named
captures with regex based locations.

--appa


From appa at perusio.net  Tue Apr  3 10:18:00 2012
From: appa at perusio.net (Antonio P.P. Almeida)
Date: Tue, 3 Apr 2012 12:18:00 +0200
Subject: Possible trigger actio of clearing cache within location
	/clearcache { }
In-Reply-To: <BLU144-W56667C01074F0569911D2D04D0@phx.gbl>
References: <BLU144-W56667C01074F0569911D2D04D0@phx.gbl>
Message-ID: <fa9af29fafffe6d05c5a93277c8e4bcb.squirrel@damiao.org>

> Is it possible to somehow (and without installing PHP) invoking the action
> of clearing the nginx cache (rm -fr /var/www/cache/*) by simply visiting a
> location {} of the server {} ?  Obviously I would allow only specific IPs
> to that location.

Yes with Lua. Try:

location = /clearcache {

    allow 127.0.0.1;
    allow ::1;
    deny all;
    content_by_lua 'local status = os.execute("rm -fr /var/www/cache/*")
                    if status == 0 then
                       ngx.say("Cache cleared")
                    else
                       ngx.say("Cache not cleared")
                    end';
}


--appa


From adrianhayter at gmail.com  Tue Apr  3 10:19:52 2012
From: adrianhayter at gmail.com (Adrian Hayter)
Date: Tue, 3 Apr 2012 11:19:52 +0100
Subject: nginx + php-fpm "modern configuration way"
In-Reply-To: <CAC_01E0fnOhehvA=7vAOzueTZOuzzwb=_BuGJbXs5c_0zFnW+g@mail.gmail.com>
References: <CAC_01E2acR3-wVdXi4gX3PXrbi93JQvcYfQKzafdo-bvaU=7cQ@mail.gmail.com>
	<CAAtReCmdXpCN_bAW26+_UWe-_16c9=z1g7EEugL_A65L7vwD1g@mail.gmail.com>
	<CAC_01E0fnOhehvA=7vAOzueTZOuzzwb=_BuGJbXs5c_0zFnW+g@mail.gmail.com>
Message-ID: <CAMa6G2PsWuTibHczoPFiZdHZPPNStLnf8G+46ZiNHPFK4jkK7g@mail.gmail.com>

A lot of forums use it for their "archive mode". Look at vBulletin / MyBB
for examples.

2012/4/3 Gr?gory Pakosz <gpakosz at yahoo.fr>

> >
>> > fastcgi_split_path_info ^(.+\.php)(/.+)$;
>> > fastcgi_param SCRIPT_FILENAME $document_root$fastcgi_script_name;
>> > fastcgi_param SCRIPT_NAME $fastcgi_script_name;
>> >
>>
>> Those are to handle urls like www.example.com/index.php/some/para/meters.
>>
>>
> Well then Wordpress, Drupal and Joomla can do without them. From your
> remark, I deduce some other PHP scripts can't
>
> Regards,
> Gregory
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx
>



-- 
Adrian Hayter
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20120403/50e2dd45/attachment.html>

From siefke_listen at web.de  Tue Apr  3 10:48:21 2012
From: siefke_listen at web.de (Silvio Siefke)
Date: Tue, 3 Apr 2012 12:48:21 +0200
Subject: PHP with FPM on Gentoo
In-Reply-To: <728e86fdb5c831482000d8acc06540fb.NginxMailingListEnglish@forum.nginx.org>
References: <20120403001323.c03997dd.siefke_listen@web.de>
	<728e86fdb5c831482000d8acc06540fb.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20120403124821.a25db2e3.siefke_listen@web.de>

Hello,

On Tue, 3 Apr 2012 04:02:14 -0400 (EDT)
"locojohn" <nginx-forum at nginx.us> wrote:

> mkdir -p /var/run/fastcgi/php-fpm.sock
> 
> php-fpm.conf:
> 
> ;listen = 127.0.0.1:9000
> ;listen.allowed_clients = 127.0.0.1
> listen = /var/run/fastcgi/php-fpm.sock
> listen.owner = nginx
> listen.group = nginx
> 
> nginx.conf:
> 
> location \.php$ {
>   include fastcgi_params;
>   fastcgi_pass unix:/var/run/fastcgi/php-fpm.sock;
> }

Thx for advice. But will not run. When i start nginx it come only 

gentoo-desk ~ # /etc/init.d/nginx start
 * Checking nginx' configuration ...
nginx: [emerg] invalid host in upstream "/var/run/fastcgi/php-fpm.sock" in /etc/nginx/nginx.conf:35
nginx: configuration file /etc/nginx/nginx.conf test failed
nginx: [emerg] invalid host in upstream "/var/run/fastcgi/php-fpm.sock" in /etc/nginx/nginx.conf:35
nginx: configuration file /etc/nginx/nginx.conf test failed
 * failed, please correct errors above                                    [ !! ]
 * ERROR: nginx failed to start


<error_log>
2012/04/03 12:44:52 [emerg] 2062#0: invalid host in upstream "/var/run/fastcgi/php-fpm.sock" in /etc/nginx/nginx.conf:35
2012/04/03 12:44:52 [emerg] 2063#0: invalid host in upstream "/var/run/fastcgi/php-fpm.sock" in /etc/nginx/nginx.conf:35
2012/04/03 12:45:02 [emerg] 2073#0: invalid host in upstream "/var/run/fastcgi/php-fpm.sock" in /etc/nginx/nginx.conf:35
2012/04/03 12:45:02 [emerg] 2074#0: invalid host in upstream "/var/run/fastcgi/php-fpm.sock" in /etc/nginx/nginx.conf:35
</error_log>


Mmh can it be that nginx and / or php not right build with emerge?



Regards 
Silvio 


From rainer at ultra-secure.de  Tue Apr  3 10:56:02 2012
From: rainer at ultra-secure.de (Rainer Duffner)
Date: Tue, 3 Apr 2012 12:56:02 +0200
Subject: PHP with FPM on Gentoo
In-Reply-To: <20120403124821.a25db2e3.siefke_listen@web.de>
References: <20120403001323.c03997dd.siefke_listen@web.de>
	<728e86fdb5c831482000d8acc06540fb.NginxMailingListEnglish@forum.nginx.org>
	<20120403124821.a25db2e3.siefke_listen@web.de>
Message-ID: <20120403125602.21849408@suse2.iptech.internal>

Am Tue, 3 Apr 2012 12:48:21 +0200
schrieb Silvio Siefke <siefke_listen at web.de>:

> Hello,
> 
> On Tue, 3 Apr 2012 04:02:14 -0400 (EDT)
> "locojohn" <nginx-forum at nginx.us> wrote:
> 
> > mkdir -p /var/run/fastcgi/php-fpm.sock



Shouldn't that only be:

mkdir -p /var/run/fastcgi/

though, on FreeBSD, I think this is created automatically.


Remove the php-fpm.sock directory.


Rainer


From nginx-forum at nginx.us  Tue Apr  3 10:59:24 2012
From: nginx-forum at nginx.us (locojohn)
Date: Tue, 3 Apr 2012 06:59:24 -0400 (EDT)
Subject: PHP with FPM on Gentoo
In-Reply-To: <20120403125602.21849408@suse2.iptech.internal>
References: <20120403125602.21849408@suse2.iptech.internal>
Message-ID: <18bfc0e6a8380be7be9f6eaf2e7120a6.NginxMailingListEnglish@forum.nginx.org>

Rainer Duffner Wrote:
-------------------------------------------------------
> Am Tue, 3 Apr 2012 12:48:21 +0200
> schrieb Silvio Siefke <siefke_listen at web.de>:
> 

> > > mkdir -p /var/run/fastcgi/php-fpm.sock
> 
> 
> 
> Shouldn't that only be:
> 
> mkdir -p /var/run/fastcgi/

Right, of course!   My stupid mistake.

Andrejs

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,224804,224850#msg-224850


From siefke_listen at web.de  Tue Apr  3 11:02:41 2012
From: siefke_listen at web.de (Silvio Siefke)
Date: Tue, 3 Apr 2012 13:02:41 +0200
Subject: PHP with FPM on Gentoo
In-Reply-To: <20120403125602.21849408@suse2.iptech.internal>
References: <20120403001323.c03997dd.siefke_listen@web.de>
	<728e86fdb5c831482000d8acc06540fb.NginxMailingListEnglish@forum.nginx.org>
	<20120403124821.a25db2e3.siefke_listen@web.de>
	<20120403125602.21849408@suse2.iptech.internal>
Message-ID: <20120403130241.e672847b.siefke_listen@web.de>

Hello,

On Tue, 3 Apr 2012 12:56:02 +0200
Rainer Duffner <rainer at ultra-secure.de> wrote:

> Shouldn't that only be:
> 
> mkdir -p /var/run/fastcgi/
> 
> though, on FreeBSD, I think this is created automatically.
> 
> 
> Remove the php-fpm.sock directory.
> 
> 
> Rainer
> 

Thank u, yes thats right i have only run mkdir -p /var/run/fastcgi, 
that i know from freebsd. But it wont not run. When i run complete
mkdir -p /var/run/fastcgi/php-fpm.sock it give mistakes have try both
ways. 


Regards
Silvio


From osa at FreeBSD.org.ru  Tue Apr  3 11:03:15 2012
From: osa at FreeBSD.org.ru (Sergey A. Osokin)
Date: Tue, 3 Apr 2012 15:03:15 +0400
Subject: [ANN] ngx_http_redis-0.3.6 released
Message-ID: <20120403110314.GC52498@FreeBSD.org.ru>

ngx_http_redis module version 0.3.6 released, available for immediate 
download at http://people.freebsd.org/~osa/ngx_http_redis-0.3.6.tar.gz

SHA256 (ngx_http_redis-0.3.6.tar.gz) = 9e5d8c0f1317a6910710c94d27f9aca4968c48ee1de271e5034525a1e9b6c70e
SIZE (ngx_http_redis-0.3.6.tar.gz) = 12099
 
<ChangeLog>

*) Feature: redis_gzip_flag.  Usefull if you are prefer to
   store data compressed in redis.  Works with ngx_http_gunzip_filter
   module.
   Thanks to Maxim Dounin.

*) Bugfix: ngx_http_redis_module might issue the error message
   "redis sent invalid trailer".
   Thanks to agentzh.

</ChangeLog>

-- 
Sergey A. Osokin
osa at FreeBSD.ORG.ru
osa at FreeBSD.ORG


From nginx-forum at nginx.us  Tue Apr  3 11:07:04 2012
From: nginx-forum at nginx.us (tikonen)
Date: Tue, 3 Apr 2012 07:07:04 -0400 (EDT)
Subject: Possible HTTP response read bug in Nginx 1.0.12
Message-ID: <dca0ee7c0edcbe874ca0b7939717a7b0.NginxMailingListEnglish@forum.nginx.org>

Hi,

I've nginx server setup that terminates SSL connections from clients and
forwards plain HTTP requests to backend servers. Some of these requests
can last pretty long (up to 1 minute) and usually get very short 15 byte
reply. (Ajax polling)

However, we see often these kind of errors that indicate that backend
servers would have disconnected socket while writing reply.

2012/03/03 06:46:38 [error] 13394#0: *2100 upstream prematurely closed
connection while reading response header from upstream, client:
XX.XX.XX.XX, server: secure.example.com, request: "GET
/url/example?par=12345 HTTP/1.1", upstream:
"http://127.0.0.1:8082/url/example?par=12345", host:
"secure.example.com", referrer: "https://secure.example.com/"

Strangely I can see correponding 200 OK responses from our server logs
at the same timestamp, so at least backend server is trying to respond
correctly. 
Then we captured the traffic between nginx and backend server with
wireshark and noticed that when this happens the response from server is
actually sent in two TCP packets.

The first packet has headers and the second packet has small 15 byte
HTTP body. I suspect that nginx fails to read the second packet in some
conditions and thinks backend server disconnected prematurely.

Any insights? Thanks!

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,224853,224853#msg-224853


From contact at jpluscplusm.com  Tue Apr  3 11:13:25 2012
From: contact at jpluscplusm.com (Jonathan Matthews)
Date: Tue, 3 Apr 2012 12:13:25 +0100
Subject: Not easy replacement of Apache with Ngnix due to error log
In-Reply-To: <a13b9ae9a9d46496a448f4d3d4454f51.NginxMailingListEnglish@forum.nginx.org>
References: <4F79E05F.1040302@citrin.ru>
	<a13b9ae9a9d46496a448f4d3d4454f51.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <CAKsTx7B6Av6MmKs+OvkHdXPCz0fZQt2quQNv70d-qUXS=GqHFQ@mail.gmail.com>

On 3 April 2012 06:47, niraj <nginx-forum at nginx.us> wrote:
> Hi ?Anton
>
> ? ? ? ? ? ? ? As u suggested,

It's "you", not "u". We're not teenagers swapping text messages here.

> ? ? ? ? ? ? ? words "upstream closed prematurely FastCGI stdout" show,
> that problem is
> probably in FastCGI server.
>
> ? ? ? ? ? ? ? ? ?This is what I am telling u that Nginx error log
> misguide us.

No, it told you exactly where to aim your troubleshooting lens. At the
FCGI process.

> ? ? ? ? ? ? ?Actually Problem was one perl module was not installed in
> that so I was getting the above error.
>
> ? ? Earlier I used apache so if any module related error then it was
> reflect in apache's error log ..............

That's because of an architectural decision for Apache to host the
perl process, and for nginx to outsource it. Each has their advantages
and disadvantages and both are valid approaches. Learn to use the
tools in your toolkit.

> that actually not happening with Nginx , this create a problem for
> system admin.

If your system administrator is unable to configure a debug log from
your perl process or its FCGI wrapper, after being given the earlier
information that the problem is *inside* the FCGI process, then your
problem is not with nginx: it is with your incompetent system
administrator.

> If any idea to get details error logs for nginx ?then please sugeest.
>
> how to set ?fcgiwraper error log ??????

I don't know if this thread is useful: http://www.ruby-forum.com/topic/180064
You may have to search for other people's solutions to getting the
verbose logs into nginx logs. It appears people have achieved this. I
don't have any pertinent information other than what I found searching
for the problem.

Jonathan
-- 
Jonathan Matthews
London, Oxford, UK
http://www.jpluscplusm.com/contact.html


From nginx-forum at nginx.us  Tue Apr  3 12:32:42 2012
From: nginx-forum at nginx.us (niraj)
Date: Tue, 3 Apr 2012 08:32:42 -0400 (EDT)
Subject: Not easy replacement of Apache with Ngnix due to error log
In-Reply-To: <CAKsTx7B6Av6MmKs+OvkHdXPCz0fZQt2quQNv70d-qUXS=GqHFQ@mail.gmail.com>
References: <CAKsTx7B6Av6MmKs+OvkHdXPCz0fZQt2quQNv70d-qUXS=GqHFQ@mail.gmail.com>
Message-ID: <0f7940628730140292547d241360dd5c.NginxMailingListEnglish@forum.nginx.org>

Hi

    First of all sorry for  ''u".


    I am the system admin. I was stuck when I was getting the above
error. I will try to find for the informative error log in Nginx.

   I just shared the thing that was happened with me. If error log was
informative like apache then It could be the solved earlier. 

      I totally agree with Each has their advantages and disadvantages
and both are valid approaches.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,224747,224858#msg-224858


From nginx-forum at nginx.us  Tue Apr  3 12:35:02 2012
From: nginx-forum at nginx.us (double)
Date: Tue, 3 Apr 2012 08:35:02 -0400 (EDT)
Subject: if( variable exists )
Message-ID: <2a3eaa6476032378a1a7e1be4c005723.NginxMailingListEnglish@forum.nginx.org>

Hello,

Is there a chance to test if a "$arg_variable" (empty value) exists?
The problem: "if" condition returns "false".

URL:
http://example.com/?variable

nginx.conf:
location / {
if( $arg_variable ) {
rewrite ^.*$ /other-location;
}
}

Thanks a lot
Marcus

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,224860,224860#msg-224860


From contact at jpluscplusm.com  Tue Apr  3 13:13:57 2012
From: contact at jpluscplusm.com (Jonathan Matthews)
Date: Tue, 3 Apr 2012 14:13:57 +0100
Subject: Config variable inheritance: docs and rules
Message-ID: <CAKsTx7DFmZvN3V44kBrSubzab_B=J7Yu8GxkfChKX71DFuc-HA@mail.gmail.com>

Hi all -

http://www.ruby-forum.com/topic/188742 seems to contain some
information I've not found explicitly stated anywhere else: that
config variable inheritance is an all or nothing mechanism. Have I
missed some docs somewhere?

Following on from this, I'd like to understand if there's anything I
can reliably determine from a variable's name about the impact that
defining it will have on *other* variables inheritance.

Can I, for instance, assume that defining an xyx_ prefixed variable in
a sub-scope will affect *only* the inheritance of other xyz_ prefixed
variables? Or can there be inheritance side-effects between, say $abc
and $def that I can't possibly determine without documentation or
trial and error?

Equally, can I assume that *all* xyz_ prefixed variables will have
their inheritance affected by the sub-scope definition of *any* other
xyz_ prefixed variable? Or are there examples where defining $xyz_foo
does *not* invalidate the inheritance of $xyz_bar ?

Thanks,
Jonathan
-- 
Jonathan Matthews
London, Oxford, UK
http://www.jpluscplusm.com/contact.html


From adrianhayter at gmail.com  Tue Apr  3 13:21:21 2012
From: adrianhayter at gmail.com (Adrian Hayter)
Date: Tue, 3 Apr 2012 14:21:21 +0100
Subject: Not easy replacement of Apache with Ngnix due to error log
In-Reply-To: <0f7940628730140292547d241360dd5c.NginxMailingListEnglish@forum.nginx.org>
References: <CAKsTx7B6Av6MmKs+OvkHdXPCz0fZQt2quQNv70d-qUXS=GqHFQ@mail.gmail.com>
	<0f7940628730140292547d241360dd5c.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <CAMa6G2NOXxK-RLmkqZ9u9y7A8mLAWR_7WrAX-h9EZ410+Qtd0A@mail.gmail.com>

>
>   I just shared the thing that was happened with me. If error log was
> informative like apache then It could be the solved earlier.
>

It was informative. It told you which process was actually having the error
(FastCGI). Nginx should be expected to handle it's own error messages well,
but if the error is in another process, it only has what that process gives
back to it.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20120403/de3612db/attachment.html>

From lists at ruby-forum.com  Tue Apr  3 14:56:31 2012
From: lists at ruby-forum.com (Ton Hoekstra)
Date: Tue, 03 Apr 2012 16:56:31 +0200
Subject: Rewrite URL's Glype
Message-ID: <b21c6a27054e71a4aba65aaa1ba38c03@ruby-forum.com>

Hi all,

I've a Glype proxy and I want to rewrite the URL's. All URLs on the page
are automatically converted to:
http://proxy.com/browse.php?u=URL HERE

Example: If I go to /browse on The Pirate Bay on my proxy I want to
convert the URL from this:
http://proxy.com/tpb/browse.php?u=http%3A%2F%2Fthepiratebay.se%2Fbrowse&b=0

To this:
http://proxy.com/tpb/browse

Also if i go to:
http://proxy.com/tpb/browse.php?u=http%3A%2F%2Fthepiratebay.se%2Ftop&b=0

Then I want to convert to:
http://proxy.com/tpb/top

As you can see, the whole part:
browse.php?u=http%3A%2F%2Fthepiratebay.se%2F

Is gone (and the `&b=0` what is behind the URL). And it has the same
domain structure as The Pirate Bay.

I've tried something like this, in the .htaccess file:
location /tpb/ {
    rewrite ^/browse.php?u=(.*)$ /$1? last;
    break;
}

But it is not working. Somebody has an answer? An other function is also
welcome. (Such as  `fastcgi_split_path_info` or something else what is
compatible with nginx)

(If you want see a example go to tpb.piratenpartij.nl but I'm not
sure if they are using Glype)

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


From francis at daoine.org  Tue Apr  3 15:52:03 2012
From: francis at daoine.org (Francis Daly)
Date: Tue, 3 Apr 2012 16:52:03 +0100
Subject: Config variable inheritance: docs and rules
In-Reply-To: <CAKsTx7DFmZvN3V44kBrSubzab_B=J7Yu8GxkfChKX71DFuc-HA@mail.gmail.com>
References: <CAKsTx7DFmZvN3V44kBrSubzab_B=J7Yu8GxkfChKX71DFuc-HA@mail.gmail.com>
Message-ID: <20120403155203.GA26052@craic.sysops.org>

On Tue, Apr 03, 2012 at 02:13:57PM +0100, Jonathan Matthews wrote:

Hi there,

> http://www.ruby-forum.com/topic/188742 seems to contain some
> information I've not found explicitly stated anywhere else: that
> config variable inheritance is an all or nothing mechanism. Have I
> missed some docs somewhere?

That thread refers to configuration directives, not variables.

The difference might not matter, but for clarity it's worth making sure
that we all understand the same thing.

In the above thread, the (wrong) expectation was that "fastcgi_param"
at one level would add to "fastcgi_param" from an enclosing level. In
fact, nginx doesn't add between levels.

That is how nginx configuration works.

Each request is handled by exactly one location.

Only the configuration in that location, and inherited into that location,
matters. Inheritance is global > http > server > location.

If a directive is used in that location, it overrides any
might-have-been-inherited values for that directive.

This is as true for single-valued directives like "root" as it is for
multi-valued directives like "fastcgi_param".

Whether a particular directive can be inherited depends on the module
that processes it -- which means "depends on the documentation". But if
it *can* be inherited, it should follow the above rules. (Or it's a bug.)

The rest of your mail talks specifically about defining variables;
I'm talking about directives. Someone else may be able to explain the
variable rules; if you do mean "directives", then the above should make
it clear what the answer in each case is.

There are some subtleties about what exactly is "a request"; and what
locations are involved when "if" is used; and how nested locations work.

But for trying to find what configuration applies to a particular
incoming http request, the "inherit or replace, but never add" rule
makes it straightforward to understand.

	f
-- 
Francis Daly        francis at daoine.org


From mdounin at mdounin.ru  Tue Apr  3 16:15:30 2012
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Tue, 3 Apr 2012 20:15:30 +0400
Subject: Possible HTTP response read bug in Nginx 1.0.12
In-Reply-To: <dca0ee7c0edcbe874ca0b7939717a7b0.NginxMailingListEnglish@forum.nginx.org>
References: <dca0ee7c0edcbe874ca0b7939717a7b0.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20120403161530.GS13466@mdounin.ru>

Hello!

On Tue, Apr 03, 2012 at 07:07:04AM -0400, tikonen wrote:

> Hi,
> 
> I've nginx server setup that terminates SSL connections from clients and
> forwards plain HTTP requests to backend servers. Some of these requests
> can last pretty long (up to 1 minute) and usually get very short 15 byte
> reply. (Ajax polling)
> 
> However, we see often these kind of errors that indicate that backend
> servers would have disconnected socket while writing reply.
> 
> 2012/03/03 06:46:38 [error] 13394#0: *2100 upstream prematurely closed
> connection while reading response header from upstream, client:
> XX.XX.XX.XX, server: secure.example.com, request: "GET
> /url/example?par=12345 HTTP/1.1", upstream:
> "http://127.0.0.1:8082/url/example?par=12345", host:
> "secure.example.com", referrer: "https://secure.example.com/"
> 
> Strangely I can see correponding 200 OK responses from our server logs
> at the same timestamp, so at least backend server is trying to respond
> correctly. 
> Then we captured the traffic between nginx and backend server with
> wireshark and noticed that when this happens the response from server is
> actually sent in two TCP packets.
> 
> The first packet has headers and the second packet has small 15 byte
> HTTP body. I suspect that nginx fails to read the second packet in some
> conditions and thinks backend server disconnected prematurely.
> 
> Any insights? Thanks!

Please try 1.1.x to see if this is already resolved.  If not, 
please provide traffic capture and corresponding debug log, see
http://wiki.nginx.org/Debugging

Maxim Dounin


From mdounin at mdounin.ru  Tue Apr  3 16:32:13 2012
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Tue, 3 Apr 2012 20:32:13 +0400
Subject: Socket leaks., pread and [crit] SSL_Write() in 1.0.14
In-Reply-To: <jldij1$k02$1@dough.gmane.org>
References: <3b3f2c2eddfdd3e59d242463275e502e.NginxMailingListEnglish@forum.nginx.org>
	<00cd479fdbb8bdf2f9f844e88607f8bd.NginxMailingListEnglish@forum.nginx.org>
	<20120326164754.GL13466@mdounin.ru> <jl813b$nn0$1@dough.gmane.org>
	<20120402111754.GD13466@mdounin.ru> <jlcjag$a19$1@dough.gmane.org>
	<20120402170637.GJ13466@mdounin.ru> <jldij1$k02$1@dough.gmane.org>
Message-ID: <20120403163212.GU13466@mdounin.ru>

Hello!

On Mon, Apr 02, 2012 at 09:08:47PM -0400, Floren Munteanu wrote:

> Hi Maxim,
> 
> On 4/2/2012 1:06 PM, Maxim Dounin wrote:
> >It's safe and only adds minimal overhead on logging level checks
> >as long as debug log isn't enabled.
> >
> >Note though that writing debug logs consume extra resources, and
> >it may be noticeable under high load.  The debug_connection
> >directive allows to enable debug logging only for selected ip
> >addresses to minimize impact.
> 
> I've built a nginx-debug RPM but I'm tempted to try 1.1.18 now. :)
> I will use instead 1.0.14, just in case is a nasty bug hidden
> somewhere in Nginx. In this way, we fix both versions. These are the

I'm afraid we aren't going to fix 1.0.x anyway.  The 1.1.x branch 
will become stable in serveral weeks, and 1.0.x is expected to get 
only security fixes.

> compile cflags I used:
> gcc -c -pipe  -O -W -Wall -Wpointer-arith -Wno-unused-parameter
> -Wunused-function -Wunused-variable -Wunused-value -Werror -g -O3 -g
> -m64 -mtune=nocona -m128bit-long-double -mmmx -msse3 -mfpmath=sse
> 
> I'm going to leave the error_log set on error and give it a 256k
> buffer so it should not affect the performance. Since we only deal

The error log doesn't support buffering.

> [crit], I don't think we need to debug anything lower than error.

I need complete request debug log to understand what's going on 
during request processing (and what's going wrong).

> I'll add the following settings into http:
> debug_points       abort;
> worker_rlimit_core 512M;
> working_directory  /var/lib/nginx/;
> 
> Let me know please if I should do anything extra.

For now, please capture debug log.  Anything else (including gdb 
dump of the request in question) isn't that important, but if you 
are going to obtain one it might be good idea to recompile nginx 
without -O3 as it might produce undesired side-effects in gdb 
output.  The -O used by default is good enough.

Maxim Dounin


From cliff at develix.com  Tue Apr  3 16:54:47 2012
From: cliff at develix.com (Cliff Wells)
Date: Tue, 03 Apr 2012 09:54:47 -0700
Subject: proxmox: how to use nginx instead of apache
In-Reply-To: <20120401203214.1899b7c7@e-healthexpert.org>
References: <20120401203214.1899b7c7@e-healthexpert.org>
Message-ID: <1333472087.2428.15.camel@portable-evil>

On Sun, 2012-04-01 at 20:32 +0100, Mark Alan wrote:
> Hello list,
> 
> I wonder if anybody in this list was successful at (completely)
> replacing apache for nginx as the web server for the proxmox VE
> management system <http://www.proxmox.com/products/proxmox-ve>.

AFAIK, Proxmox is implemented in mod_perl and uses mod_perl specific
features. It's tied to Apache.

Cliff



From ne at vbart.ru  Tue Apr  3 16:56:30 2012
From: ne at vbart.ru (Valentin V. Bartenev)
Date: Tue, 3 Apr 2012 20:56:30 +0400
Subject: if( variable exists )
In-Reply-To: <2a3eaa6476032378a1a7e1be4c005723.NginxMailingListEnglish@forum.nginx.org>
References: <2a3eaa6476032378a1a7e1be4c005723.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <201204032056.30563.ne@vbart.ru>

On Tuesday 03 April 2012 16:35:02 double wrote:
> Hello,
> 
> Is there a chance to test if a "$arg_variable" (empty value) exists?
> The problem: "if" condition returns "false".
> 
> URL:
> http://example.com/?variable
> 

  if ($args ~ variable) {
     ...
  }

 wbr, Valentin V. Bartenev


From contact at jpluscplusm.com  Tue Apr  3 17:07:13 2012
From: contact at jpluscplusm.com (Jonathan Matthews)
Date: Tue, 3 Apr 2012 18:07:13 +0100
Subject: Config variable inheritance: docs and rules
In-Reply-To: <20120403155203.GA26052@craic.sysops.org>
References: <CAKsTx7DFmZvN3V44kBrSubzab_B=J7Yu8GxkfChKX71DFuc-HA@mail.gmail.com>
	<20120403155203.GA26052@craic.sysops.org>
Message-ID: <CAKsTx7Bgph1J+o-QgKSG8wi+QHWVs+q4FvWAQjOc3SZof_wOSQ@mail.gmail.com>

On 3 April 2012 16:52, Francis Daly <francis at daoine.org> wrote:
> On Tue, Apr 03, 2012 at 02:13:57PM +0100, Jonathan Matthews wrote:
>
> Hi there,
>
>> http://www.ruby-forum.com/topic/188742 seems to contain some
>> information I've not found explicitly stated anywhere else: that
>> config variable inheritance is an all or nothing mechanism. Have I
>> missed some docs somewhere?
>
> That thread refers to configuration directives, not variables.
>
> The difference might not matter, but for clarity it's worth making sure
> that we all understand the same thing.

You're absolutely correct; I'm referring to config directives, but was
thinking too hard about the question to notice I'd mixed my
terminologies. I'll correct it, and resend the email in a new thread
for list archive niceness. It'd be great if you could cherry pick the
stuff you feel is relevant from your reply and resend it. If only for
those who come after ... :-)

Jonathan
-- 
Jonathan Matthews
London, Oxford, UK
http://www.jpluscplusm.com/contact.html


From contact at jpluscplusm.com  Tue Apr  3 17:20:13 2012
From: contact at jpluscplusm.com (Jonathan Matthews)
Date: Tue, 3 Apr 2012 18:20:13 +0100
Subject: Config directive inheritance: docs and rules (was: Config variable
	inheritance)
Message-ID: <CAKsTx7A88PTFp14208kb5iu5RK=KefEC7TY9GzZJOmBH1gtMAg@mail.gmail.com>

Hi all -

http://www.ruby-forum.com/topic/188742 seems to contain some
information I've not found explicitly stated anywhere else: that
config directive inheritance is an all or nothing mechanism. Have I
missed some docs somewhere?

Following on from this, I'd like to understand if there's anything I
can reliably determine from a directive's name about the impact that
defining it will have on *other* directives' inheritability.

Can I, for instance, assume that defining an xyx_ prefixed directive in
a sub-scope will affect *only* the inheritance of other xyz_ prefixed
directives? Or can there be inheritance side-effects between, say,
directive "abc" and "def" that I can't possibly determine without
documentation or trial and error?

Equally, can I assume that *all* xyz_ prefixed directives will have
their inheritance affected by the sub-scope definition of any *other*
xyz_ prefixed directive? Or are there examples where defining
"xyz_foo" does *not* impact the inheritance of "xyz_bar"?

I'm thinking specifically of proxy_* directives, whereas the thread I
link to above is talking about multiple fastcgi_param directives.
Here, I'm interested in how, say, proxy_set_header affects the
inheritance of higher scope's proxy_pass directive - i.e. where the
prefix matches but they're *not* the same directive. Does nginx (core)
derive or imply anything relating to the name-spacing of directives
before (in my example) the proxy module sees the config, or is it
entirely down to how each module implements the configuration
aggregation/inheritance?

Many thanks,
Jonathan
-- 
Jonathan Matthews
London, Oxford, UK
http://www.jpluscplusm.com/contact.html


From varia at e-healthexpert.org  Tue Apr  3 18:52:15 2012
From: varia at e-healthexpert.org (Mark Alan)
Date: Tue, 3 Apr 2012 19:52:15 +0100
Subject: proxmox: how to use nginx instead of apache
In-Reply-To: <1333472087.2428.15.camel@portable-evil>
References: <20120401203214.1899b7c7@e-healthexpert.org>
	<1333472087.2428.15.camel@portable-evil>
Message-ID: <20120403195215.65c4882b@e-healthexpert.org>

On Tue, 03 Apr 2012 09:54:47 -0700, Cliff Wells <cliff at develix.com>
wrote:
> > I wonder if anybody in this list was successful at (completely)
> > replacing apache for nginx as the web server for the proxmox VE
> > management system <http://www.proxmox.com/products/proxmox-ve>.
> 
> AFAIK, Proxmox is implemented in mod_perl and uses mod_perl specific
> features. It's tied to Apache.
> 
> Cliff

Thank you Cliff,

I was hoping that someone had already start some work (based on from
http://wiki.nginx.org/NginxEmbeddedPerlModule ?).

It would be much more light, secure and would, perhaps, avoid loading a
full perl instance for each apache child process.

Well, I won't give up on it yet.


M.


From gpakosz at yahoo.fr  Tue Apr  3 18:56:12 2012
From: gpakosz at yahoo.fr (=?ISO-8859-1?Q?Gr=E9gory_Pakosz?=)
Date: Tue, 3 Apr 2012 20:56:12 +0200
Subject: nginx + php-fpm "modern configuration way"
In-Reply-To: <CAMa6G2PsWuTibHczoPFiZdHZPPNStLnf8G+46ZiNHPFK4jkK7g@mail.gmail.com>
References: <CAC_01E2acR3-wVdXi4gX3PXrbi93JQvcYfQKzafdo-bvaU=7cQ@mail.gmail.com>
	<CAAtReCmdXpCN_bAW26+_UWe-_16c9=z1g7EEugL_A65L7vwD1g@mail.gmail.com>
	<CAC_01E0fnOhehvA=7vAOzueTZOuzzwb=_BuGJbXs5c_0zFnW+g@mail.gmail.com>
	<CAMa6G2PsWuTibHczoPFiZdHZPPNStLnf8G+46ZiNHPFK4jkK7g@mail.gmail.com>
Message-ID: <CAC_01E37M3iiA0nkD4u5FtqHFEgGfb5aRXHNZ1qOFdQCODs+bQ@mail.gmail.com>

Thank you all for your reply

So, adding
fastcgi_split_path_info ^(.+\.php)(/.+)$;
fastcgi_param SCRIPT_FILENAME $document_root$fastcgi_script_name;
fastcgi_param SCRIPT_NAME $fastcgi_script_name;

doesn't really hurt.

The box I'm running does shared hosting for a couple of users and I don't
really know in advance what they want to install on their account.
I'm going to keep those 3 lines in hope it helps some php scripts out there
to run correctly.

Gregory
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20120403/a7498f95/attachment.html>

From baishen.lists at gmail.com  Tue Apr  3 20:13:08 2012
From: baishen.lists at gmail.com (Bai Shen)
Date: Tue, 3 Apr 2012 16:13:08 -0400
Subject: Custom error pages
In-Reply-To: <20120328211619.GB3899@craic.sysops.org>
References: <CAJVBNiFtRpVr5vuWNzW2eue4pW-_KnrEWRzSHVvgZvF7oVhSrw@mail.gmail.com>
	<20120313234023.GJ18773@craic.sysops.org>
	<CAJVBNiE8U6MYy3_rkVgvmm0LGpjFptvLEFa_jYT1VU94QJezdw@mail.gmail.com>
	<20120315192409.GM18773@craic.sysops.org>
	<CAJVBNiH9abE7XKnf-7MNZfwnfOmTsR2wpLYJdCqXpC5zExcacw@mail.gmail.com>
	<20120316151856.GP18773@craic.sysops.org>
	<CAJVBNiGvP+YsVNSYdLL8+p=XpA+3UbMMraHtk9m0XnQ4HwgmZw@mail.gmail.com>
	<20120328211619.GB3899@craic.sysops.org>
Message-ID: <CAJVBNiG8HHQDzpGU8noE_M-jxtnnHk6Q_kfE1w=j2txK5qr-nw@mail.gmail.com>

My conf looks like this.

server {
   listen 192.168.1.1:80;
   server_name example.com;

  error_page 503 /maint/503.html;

  location /maint/503.html {
    internal;
  }

  location /maint/ {
    root /etc/nginx/html;
  }

server {
  listen 192.168.1.1:80;
  server_name example.com;

  location / {
    proxy_pass http://192.168.1.2;
  }
}


To show the 503 page instead of my site, I uncomment the top server block.
When I want to bring the site back, I comment the top block out.  The
/maint/ location was what I changed to get the css and images to work.
However, I'm not sure that it's the best solution.  Also, I'm assuming that
there's a better way to show/hide the 503 page than commenting out the
whole server block.

On Wed, Mar 28, 2012 at 5:16 PM, Francis Daly <francis at daoine.org> wrote:

> On Wed, Mar 28, 2012 at 03:42:45PM -0400, Bai Shen wrote:
>
> Hi there,
>
> > I got my error page to show up.  However, the css and image files don't
> > load.  I tried changing the location, but that doesn't seem to work.
> >
> > Any suggestions?
>
> Try changing the location again. Maybe it will work this time.
>
>
> Or alternatively:
>
> Please fill in the details:
>
> > > what did you do?
>
> That will be something like "curl -i http://myserver/myfile.css"
>
> > > what did you see?
>
> That will be whatever you got back -- the content of /var/tmp/myfile.css,
> or the content of another file, or some error message. There might be
> more details available in log files.
>
> > > what did you expect to see?
>
> That will be something like "http 200 and the content of
> /usr/local/nginx/html/myfile.css".
>
> > > and the more specific you are, the easier it will be for someone else
> to
> > > reproduce the problem you are reporting.
>
> Provide the relevant part of nginx.conf from a machine where you saw
> the problem. If you don't know what is the relevant part, provide the
> whole thing.
>
> It doesn't have to be your "live" config, but it should be a config that
> you ran on a test machine. The smaller the config, the better.
>
> If you make it easy for someone to run the same server config as you,
> and run the same client commands as you, and see the same error output
> as you; then you make it easy for that person to try to help resolve the
> problem you saw.
>
> If you don't, you don't.
>
>        f
> --
> Francis Daly        francis at daoine.org
>
> _______________________________________________
> 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/20120403/274ab8e9/attachment.html>

From yvldwt at gmail.com  Tue Apr  3 20:30:39 2012
From: yvldwt at gmail.com (=?UTF-8?B?RMW+ZW4=?=)
Date: Tue, 03 Apr 2012 22:30:39 +0200
Subject: try_files ignored in combination with error_page and named locations
Message-ID: <4F7B5DEF.8030107@gmail.com>

Hello,

For some reason the following nginx configuration snippet doesn't work
as expected, try_files seems to be ignored. When / is requested (or any
other URI which points to a file which doesn't exist), nginx's default
503 page is returned instead of the existing index.html. Requesting
/index.html works correctly.

	location @maintenance {
		root		/var/nginx/$hostid-maintenance;
		index		index.html;
		try_files	$uri $uri/ / =500;
	}

	error_page 503 @maintenance;

	if ( -d $document_root/$hostid-maintenance ) {
		return 503;
	}

Any thoughts? Running nginx version 1.0.12 here.

Kind regards

-- 
D?en


From caldcv at gmail.com  Tue Apr  3 20:59:09 2012
From: caldcv at gmail.com (Chris)
Date: Tue, 3 Apr 2012 16:59:09 -0400
Subject: proxmox: how to use nginx instead of apache
In-Reply-To: <20120403195215.65c4882b@e-healthexpert.org>
References: <20120401203214.1899b7c7@e-healthexpert.org>
	<1333472087.2428.15.camel@portable-evil>
	<20120403195215.65c4882b@e-healthexpert.org>
Message-ID: <CAPF5agdCLDTNNxtan=ndAJrFC_-rVS-gNE9CGPRih=U90yWR2A@mail.gmail.com>

You mean for the Proxmox VE Web GUI?


From nginx-forum at nginx.us  Tue Apr  3 21:12:33 2012
From: nginx-forum at nginx.us (locojohn)
Date: Tue, 3 Apr 2012 17:12:33 -0400 (EDT)
Subject: nginx + php-fpm "modern configuration way"
In-Reply-To: <CAC_01E2acR3-wVdXi4gX3PXrbi93JQvcYfQKzafdo-bvaU=7cQ@mail.gmail.com>
References: <CAC_01E2acR3-wVdXi4gX3PXrbi93JQvcYfQKzafdo-bvaU=7cQ@mail.gmail.com>
Message-ID: <cf53159d2ce36b6cac42ce385411bf13.NginxMailingListEnglish@forum.nginx.org>

Well, in that case you may as well support SCRIPT_URL / SCRIPT_URI, as
there are some [old] PHP scripts that rely on these variables set
normally by Apache.

nginx.conf:

http {
  ...
  map $uri $script_url {
          ~^(?<script_filename>.+\.(php|html))(?<path_info>.+)$
$path_info;
          ~^(?<script_filename>.+\.(php|html))$ $script_filename;
   }
  ...
}

fastcgi_php.conf:

include fastcgi_params;

          fastcgi_split_path_info ^((?U).+\.php)(/?.+)$;
          fastcgi_param SCRIPT_FILENAME
$document_root$fastcgi_script_name;
          fastcgi_param PATH_INFO $fastcgi_path_info;
          fastcgi_param PATH_TRANSLATED
$document_root$fastcgi_path_info;

          fastcgi_param  SCRIPT_URL $script_url;
          fastcgi_param  SCRIPT_URI $scheme://$http_host$script_url;

          try_files $fastcgi_script_name = 404;

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,224829,224900#msg-224900


From siefke_listen at web.de  Tue Apr  3 21:39:16 2012
From: siefke_listen at web.de (Silvio Siefke)
Date: Tue, 3 Apr 2012 23:39:16 +0200
Subject: PHP with FPM on Gentoo
In-Reply-To: <20120403001323.c03997dd.siefke_listen@web.de>
References: <20120403001323.c03997dd.siefke_listen@web.de>
Message-ID: <20120403233916.d0943f2e.siefke_listen@web.de>

Hello,


Could someone possibly provide me the USE flags available 
if someone has run Nginx, PHP and Fpm.

It were nice. 

Regards
Silvio


From tdgh2323 at hotmail.com  Tue Apr  3 22:09:29 2012
From: tdgh2323 at hotmail.com (Micheal Wolfskill)
Date: Tue, 3 Apr 2012 22:09:29 +0000
Subject: How to limit_req depending if the requests has a REFERER or not.
In-Reply-To: <BLU144-W24E01412AED330F9E64F77D04C0@phx.gbl>
References: <BLU144-W24E01412AED330F9E64F77D04C0@phx.gbl>
Message-ID: <BLU144-W5D2FBD5857518E8E8EF92D04D0@phx.gbl>








Hello.  

I have something like this... I need to be able to apply three different limit_req depending:

a.) If the referer to click.php is domain.com ... apply zone1
b.) If there is some other referer apply zone2 on click.php
c.) If there is no referer apply zone3 on click.php



  location /click.php { 
                                   limit_req zone=one nodelay; 
                                   proxy_pass http://22.136.1.121:80/; 
                                   }

Mike
 		 	   		  
 		 	   		  
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20120403/9e10fcca/attachment.html>

From jdorfman at netdna.com  Tue Apr  3 22:10:07 2012
From: jdorfman at netdna.com (Justin Dorfman)
Date: Tue, 3 Apr 2012 15:10:07 -0700
Subject: Location URI Wildcard
Message-ID: <CACDr0vGcTPMT2VW1j9dhAewRfjnnKzMCS3dDBw9gFY0f_f3MLA@mail.gmail.com>

Hello,

I am trying to find a way to redirect a bunch of jpg files to an html page.

This way is not working:

location = /attach/201204*.jpg {
            root /usr/local/nginx/html;
            index suspend.html;
}

Here is a real world file name: 201204031012088721.jpg

Thanks!

Regards,

Justin Dorfman <http://www.twitter.com/jdorfman>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20120403/33129fa3/attachment.html>

From mdounin at mdounin.ru  Tue Apr  3 22:20:40 2012
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Wed, 4 Apr 2012 02:20:40 +0400
Subject: Location URI Wildcard
In-Reply-To: <CACDr0vGcTPMT2VW1j9dhAewRfjnnKzMCS3dDBw9gFY0f_f3MLA@mail.gmail.com>
References: <CACDr0vGcTPMT2VW1j9dhAewRfjnnKzMCS3dDBw9gFY0f_f3MLA@mail.gmail.com>
Message-ID: <20120403222039.GV13466@mdounin.ru>

Hello!

On Tue, Apr 03, 2012 at 03:10:07PM -0700, Justin Dorfman wrote:

> Hello,
> 
> I am trying to find a way to redirect a bunch of jpg files to an html page.
> 
> This way is not working:
> 
> location = /attach/201204*.jpg {
>             root /usr/local/nginx/html;
>             index suspend.html;
> }
> 
> Here is a real world file name: 201204031012088721.jpg

location ~ ^/attach/201204.*\.jpg$ {
    alias /usr/local/nginx/html/suspend.html;
}

Maxim Dounin


From jdorfman at netdna.com  Tue Apr  3 22:26:57 2012
From: jdorfman at netdna.com (Justin Dorfman)
Date: Tue, 3 Apr 2012 15:26:57 -0700
Subject: Location URI Wildcard
In-Reply-To: <20120403222039.GV13466@mdounin.ru>
References: <CACDr0vGcTPMT2VW1j9dhAewRfjnnKzMCS3dDBw9gFY0f_f3MLA@mail.gmail.com>
	<20120403222039.GV13466@mdounin.ru>
Message-ID: <CACDr0vHDu3StCQg_HKYA5wAur1s+uuSEEh_3Baxs_-LV6NwH9Q@mail.gmail.com>

Perfect thanks Maxim!

Regards,

Justin Dorfman <http://www.twitter.com/jdorfman>

NetDNA <http://www.netdna.com>?
The Science of Acceleration?

Email / gtalk: jdorfman at netdna.com
P: 1.323.301.1400 x 272
Skype: netdna-justind
Twitter: @jdorfman <http://www.twitter.com/jdorfman>

www.NetDNA.com <http://www.netdna.com/> | www.MaxCDN.com<http://www.maxcdn.com/>

@NetDNA <http://twitter.com/NetDNA> | @MaxCDN <http://twitter.com/MaxCDN>



On Tue, Apr 3, 2012 at 3:20 PM, Maxim Dounin <mdounin at mdounin.ru> wrote:

> Hello!
>
> On Tue, Apr 03, 2012 at 03:10:07PM -0700, Justin Dorfman wrote:
>
> > Hello,
> >
> > I am trying to find a way to redirect a bunch of jpg files to an html
> page.
> >
> > This way is not working:
> >
> > location = /attach/201204*.jpg {
> >             root /usr/local/nginx/html;
> >             index suspend.html;
> > }
> >
> > Here is a real world file name: 201204031012088721.jpg
>
> location ~ ^/attach/201204.*\.jpg$ {
>    alias /usr/local/nginx/html/suspend.html;
> }
>
> Maxim Dounin
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20120403/2e1d752a/attachment.html>

From nginx-forum at nginx.us  Wed Apr  4 01:06:12 2012
From: nginx-forum at nginx.us (tikonen)
Date: Tue, 3 Apr 2012 21:06:12 -0400 (EDT)
Subject: Possible HTTP response read bug in Nginx 1.0.12
In-Reply-To: <20120403161530.GS13466@mdounin.ru>
References: <20120403161530.GS13466@mdounin.ru>
Message-ID: <0897eea6bf171e32cb305273ccc3d6a2.NginxMailingListEnglish@forum.nginx.org>

Hi,

Thanks for advice, I'll try next with 1.1.X and see if that helps.

In case 1.1.X does not solve the problem. Server load is pretty big and
only minor % of requests have the error so the debug log and capture
dump would be Huge to catch even one case. (Still these errors are
frequent enough to cause ip_hash to flag node failed and cause havoc in
the cluster).

To narrow down needed data, can you recommend proper debugging settings,
should I start server or global level?
Also, should I make debug log againts 1.1.X or 1.0.12 ?

Thanks, Teemu

Maxim Dounin Wrote:
-------------------------------------------------------
> Please try 1.1.x to see if this is already
> resolved.  If not, 
> please provide traffic capture and corresponding
> debug log, see
> http://wiki.nginx.org/Debugging

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,224853,224907#msg-224907


From agentzh at gmail.com  Wed Apr  4 03:23:12 2012
From: agentzh at gmail.com (agentzh)
Date: Wed, 4 Apr 2012 11:23:12 +0800
Subject: memc_flags_to_last_modified and srcache , no 304 possible ?
In-Reply-To: <6e2bc7fac8512842fa34ca858100f61b.NginxMailingListEnglish@forum.nginx.org>
References: <6e2bc7fac8512842fa34ca858100f61b.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <CAB4Tn6NaxFJEJeSbYN5YTh1hCnaM=8dyX+ZeX8kpYiANuT8j+g@mail.gmail.com>

On Tue, Apr 3, 2012 at 5:21 AM, Nginx_User777 <nginx-forum at nginx.us> wrote:
> Is it possible to use the memc_flags_to_last_modified directive with
> srcache ?
> It doesn't work for me.
>
> When a browser request nginx with a "If-Modified-Since" header, I would
> like that nginx lookup in memcached to verify if the "Last Modified" is
> the same and honor the 304 request without hit the proxied server (in
> another country).
>
> Impossible to get a 304, so no bandwith saved ... But my provider is
> happy :)
>
> Thanks for your help !
> I working on for several days, i'll really go mad :)
>

This is a TODO item for ngx_srcache. I'll look into this in the near
future. Thanks for reminding me of this :)

Best regards,
-agentzh


From quintinpar at gmail.com  Wed Apr  4 04:08:25 2012
From: quintinpar at gmail.com (Quintin Par)
Date: Wed, 4 Apr 2012 09:38:25 +0530
Subject: Serve from cache but fire request to upstream server to increment
	page view counter
In-Reply-To: <CACW9FdNgfxGejBvzWoqbM7+ikNpZxum82MmmS0aFd6WsK3AOUw@mail.gmail.com>
References: <CACW9FdNgfxGejBvzWoqbM7+ikNpZxum82MmmS0aFd6WsK3AOUw@mail.gmail.com>
Message-ID: <CACW9FdOxDwk_MPx1SD5-J-7z9TUAtm-Gpv3Vo23T=n4b5ZgBmQ@mail.gmail.com>

Hi all,

Can someone please help me on this? The pattern is similar to

proxy_cache_use_stale  updating

but just that the backend request is a fire-and-forget.

If I am to do this in SSI I?ll end up spending time to service the code
path in SSI whereas there is no much need to that urgently.

The next cache bypass will update it and that?s good enough for me.

-Quintin

On Tue, Apr 3, 2012 at 12:13 PM, Quintin Par <quintinpar at gmail.com> wrote:

> Hi all,
>
> I have a strange case, not sure if this is addressed by nginx yet
>
> My site is cached as follows.
>
>   location = / {
>
>     proxy_pass  http://localhost:82;
>
>     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 Accept-Encoding "";
>
>     proxy_ignore_headers Set-Cookie;
>
>     proxy_ignore_headers Cache-Control;
>
>     proxy_ignore_headers Expires;
>
>     proxy_ignore_headers X-Accel-Expires;
>
>     add_header X-Cache-Status $upstream_cache_status;
>
>     proxy_cache             cache;
>
>     proxy_cache_key
> $scheme$host$request_uri$cookie_site_sessionid;
>
>     proxy_cache_valid       200 302 30s;
>
>     proxy_cache_use_stale   updating;
>
>   }
>
> One issue I encounter with this is that I can?t increment the page view
> counter that I maintain in redis.
>
> Is there is way nginx can fire a request to the backend web ? to some URI
> so that I can take the request and increment the associated counter?
>
> I know I can achieve this with SSI, but I wanted to check if there is a
> better pattern.
>
> -Quintin
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20120404/3673dd6c/attachment-0001.html>

From nginx-forum at nginx.us  Wed Apr  4 06:14:41 2012
From: nginx-forum at nginx.us (niraj)
Date: Wed, 4 Apr 2012 02:14:41 -0400 (EDT)
Subject: Not easy replacement of Apache with Ngnix due to error log
In-Reply-To: <CAMa6G2NOXxK-RLmkqZ9u9y7A8mLAWR_7WrAX-h9EZ410+Qtd0A@mail.gmail.com>
References: <CAMa6G2NOXxK-RLmkqZ9u9y7A8mLAWR_7WrAX-h9EZ410+Qtd0A@mail.gmail.com>
Message-ID: <fcc9e5d435f9315eb2acdee832ef4889.NginxMailingListEnglish@forum.nginx.org>

Hi

     My view is that only if Nginx error log more informative then it
will be very userful for system admin who wants to switch from apache to
Nginx. Its need to be more informative.

     "As I am system admin and  I was face that issue and at that time
there were no error log which guide me to solve the issue.  who are new
to switch with  Nginx this logs not that much informative."


     My personal view if Nginx having more informative error log then
anybody can easily adopt the Nginx.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,224747,224911#msg-224911


From varia at e-healthexpert.org  Wed Apr  4 08:01:03 2012
From: varia at e-healthexpert.org (Mark Alan)
Date: Wed, 4 Apr 2012 09:01:03 +0100
Subject: proxmox: how to use nginx instead of apache
In-Reply-To: <CAPF5agdCLDTNNxtan=ndAJrFC_-rVS-gNE9CGPRih=U90yWR2A@mail.gmail.com>
References: <20120401203214.1899b7c7@e-healthexpert.org>
	<1333472087.2428.15.camel@portable-evil>
	<20120403195215.65c4882b@e-healthexpert.org>
	<CAPF5agdCLDTNNxtan=ndAJrFC_-rVS-gNE9CGPRih=U90yWR2A@mail.gmail.com>
Message-ID: <20120404090103.40a30405@e-healthexpert.org>

On Tue, 3 Apr 2012 16:59:09 -0400, Chris <caldcv at gmail.com> wrote:

> You mean for the Proxmox VE Web GUI?

Yes. The idea is to use Nginx Instead of Apache to serve the Proxmox VE
Web GUI.

M.


From appa at perusio.net  Wed Apr  4 08:09:43 2012
From: appa at perusio.net (Antonio P.P. Almeida)
Date: Wed, 4 Apr 2012 10:09:43 +0200
Subject: Serve from cache but fire request to upstream server to increment
	page view counter
In-Reply-To: <CACW9FdOxDwk_MPx1SD5-J-7z9TUAtm-Gpv3Vo23T=n4b5ZgBmQ@mail.gmail.com>
References: <CACW9FdNgfxGejBvzWoqbM7+ikNpZxum82MmmS0aFd6WsK3AOUw@mail.gmail.com>
	<CACW9FdOxDwk_MPx1SD5-J-7z9TUAtm-Gpv3Vo23T=n4b5ZgBmQ@mail.gmail.com>
Message-ID: <77590aa60cbf956e5fa84ac68220edca.squirrel@damiao.org>

There's the post_action directive. But AFAIK is not that reliable.

I would use Lua with the "new" cosocket API and do an HTTP request.

Here's an example of a library built around it that talks with a Redis
backend: https://github.com/agentzh/lua-resty-redis

As stated post_action is another option:

http://wiki.nginx.org/HttpCoreModule#post_action


--appa


From nginx-forum at nginx.us  Wed Apr  4 08:12:49 2012
From: nginx-forum at nginx.us (tikonen)
Date: Wed, 4 Apr 2012 04:12:49 -0400 (EDT)
Subject: Possible HTTP response read bug in Nginx 1.0.12
In-Reply-To: <dca0ee7c0edcbe874ca0b7939717a7b0.NginxMailingListEnglish@forum.nginx.org>
References: <dca0ee7c0edcbe874ca0b7939717a7b0.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <5f12ff38be70c57ffc2353ffac5ea907.NginxMailingListEnglish@forum.nginx.org>

I can confirm that problem exisits also with Nginx 1.1.18.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,224853,224917#msg-224917


From ru at nginx.com  Wed Apr  4 08:16:10 2012
From: ru at nginx.com (Ruslan Ermilov)
Date: Wed, 4 Apr 2012 12:16:10 +0400
Subject: try_files ignored in combination with error_page and named
	locations
In-Reply-To: <4F7B5DEF.8030107@gmail.com>
References: <4F7B5DEF.8030107@gmail.com>
Message-ID: <20120404081610.GA44137@lo0.su>

On Tue, Apr 03, 2012 at 10:30:39PM +0200, D?en wrote:
> For some reason the following nginx configuration snippet doesn't work
> as expected, try_files seems to be ignored. When / is requested (or any
> other URI which points to a file which doesn't exist), nginx's default
> 503 page is returned instead of the existing index.html. Requesting
> /index.html works correctly.
> 
> 	location @maintenance {
> 		root		/var/nginx/$hostid-maintenance;
> 		index		index.html;
> 		try_files	$uri $uri/ / =500;
> 	}
> 
> 	error_page 503 @maintenance;
> 
> 	if ( -d $document_root/$hostid-maintenance ) {
> 		return 503;
> 	}
> 
> Any thoughts? Running nginx version 1.0.12 here.

First, this is a wacky config, and I don't quite understand what
you're trying to achieve with it.  :)

Here's what happens with the request of "/", assuming that "if"
condition holds true.

The status code is set to 503, and further processing happends in
the named location @maintenance.

It tries "$uri", find that directory, but because argument does
not end with a slash (i.e., we're testing a file, not directory),
this variant is ignored.

It then tries "$uri/", and succeeds.  $uri is set to
"/$hostid-maintenance/".  As documentation of "try_files" says,
the further processing is done in thi same (named) location.

What happens next is since a request now ends with a slash, the
static module handler ignores it, and a request is processed by
the index module.  The index module then verifies that the
"/var/nginx/$hostid-maintenance/index.html" file exists and does
an internal redirect to a new $uri.  (The fact that it does an
internal redirect is also documented.)

This in turn gets processed by the server's default location
which is configured to return 503.  Because
"recursive_error_pages" aren't enabled, the server ends up
returning status code 503 with the standard content.

(What happens when you enable is left as an excercise to a
reader.)

When OTOH you request "/index.html", try_files checks that the
_file_ exists, and the processing is performed by the static
module in the context of location @maintenance.  There are no
more internal redirects involved, so the server ends up sending
you the status code 503 with the contents of index.html.

I agree that this might not be obvious, but this is how it works.

The important bit here is that requests that end up with "/", and
get processed by the index module, do an internal redirect.  This
is quite logical if you think about it, because they *usually*
get processed by the static module, as if a request ended up with
"/" followed by the name of index file.


From arissirajawali at gmail.com  Wed Apr  4 08:29:19 2012
From: arissirajawali at gmail.com (aris)
Date: Wed, 4 Apr 2012 15:29:19 +0700
Subject: install nginx dan php debian
In-Reply-To: <1215987271.1017595069@informatik.hu>
References: <CAD8siQ4sir_Y1tUq0ATMuyL8D01fR3zN_aXAJG+YYpG7Lg9HnQ@mail.gmail.com>
	<4F756EE7.401@esiee.fr>
	<CACZPkyopqd7HaZ=KHxKqHSE+5pfswKU5mE6Tv5xs_bq9+sv9ag@mail.gmail.com>
	<CAD8siQ4fwmGDVUWwZz9y_1bTQmDQ0zdN3Q2asAzJFd2NFjd_eg@mail.gmail.com>
	<1215987271.1017595069@informatik.hu>
Message-ID: <CAD8siQ5YsiZrMZXt8L=ziWaV+f8Fr-WDAzmPDMMBhNCjSGPR0g@mail.gmail.com>

hi all,
after I finished install nginx with apt-get install nginx, what steps should
I do so nginxcan walk?

-----------------------------

On Fri, Mar 30, 2012 at 10:17 PM, Vuki <szun at informatik.hu> wrote:

> Hi!
>
> Check file permissions and the index directive, and index files!
>
> Vukii
>
> ------- Original message -------
>
>> From: aris <arissirajawali at gmail.com>
>> To: surfchen at gmail.com
>> Cc: nginx at nginx.org
>> Sent: '12-03-30,  15:28
>>
>>
>> dear all,
>> why is my nginx 403 Forbidden?
>> Where such errors
>> please help me, thank you
>>
>>
>> On Fri, Mar 30, 2012 at 4:01 PM, Chen Ze <surfchen at gmail.com> wrote:
>>
>>> Google and RTFM
>>>
>>> On Fri, Mar 30, 2012 at 4:29 PM, Frank Bonnet <f.bonnet at esiee.fr> wrote:
>>>
>>>> On 03/30/2012 06:19 AM, aris wrote:
>>>>
>>>> please help me, I just wanted to start
>>>> the install nginx and php on **debian linux,
>>>> that I question is: how to install nginx and php on **debian, please send
>>>> it
>>>> to me once in the tutorial how to install nginx and php.
>>>>   Thank you so much before
>>>>
>>>> ______________________________**_________________
>>>> nginx mailing list
>>>> nginx at nginx.org
>>>> http://mailman.nginx.org/**mailman/listinfo/nginx<http://mailman.nginx.org/mailman/listinfo/nginx>
>>>>
>>>>
>>>> http://www.howtoforge.com/**installing-php-5.3-nginx-and-**
>>>> php-fpm-on-ubuntu-debian<http://www.howtoforge.com/installing-php-5.3-nginx-and-php-fpm-on-ubuntu-debian>
>>>>
>>>>
>>>> ______________________________**_________________
>>>> nginx mailing list
>>>> nginx at nginx.org
>>>> http://mailman.nginx.org/**mailman/listinfo/nginx<http://mailman.nginx.org/mailman/listinfo/nginx>
>>>>
>>>
>>>
> ______________________________**_________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/**mailman/listinfo/nginx<http://mailman.nginx.org/mailman/listinfo/nginx>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20120404/8ef0d0ed/attachment.html>

From yvldwt at gmail.com  Wed Apr  4 08:48:01 2012
From: yvldwt at gmail.com (=?utf-8?B?RMW+ZW4=?=)
Date: Wed, 4 Apr 2012 10:48:01 +0200
Subject: try_files ignored in combination with error_page and named
	locations
In-Reply-To: <20120404081610.GA44137@lo0.su>
References: <4F7B5DEF.8030107@gmail.com>
 <20120404081610.GA44137@lo0.su>
Message-ID: <20120404084801.GA11638@azarath.acceleris.ch>

Thank you very much for your clarifying explanation. Seems like the only
thing missing was "recursive_error_pages on" for it to work like I
wanted it to have.

On 04/04/12 12:16pm, Ruslan Ermilov wrote:
> [...]
> 
> First, this is a wacky config, and I don't quite understand what
> you're trying to achieve with it.  :)

Just to make you understand what I was trying to achieve with this
config: my maintenance page contains img tags pointing to images located
in $document_root. I used to add a simple

	try_files /$hostid-maintenance/ @backend;

to my configuration, but as expected a request to /foo.png returns the
contents of /$hostid-maintenance/index.html instead. Thus I was looking
for a solution to use a whole directory as an error page (being able to
request other files which are located inside the maintenance
$document_root).

> Here's what happens with the request of "/", assuming that "if"
> condition holds true.
> 
> The status code is set to 503, and further processing happends in
> the named location @maintenance.
> 
> It tries "$uri", find that directory, but because argument does
> not end with a slash (i.e., we're testing a file, not directory),
> this variant is ignored.

That's OK in my case, since the maintenance "environment" might contain
other directories.

> It then tries "$uri/", and succeeds.  $uri is set to
> "/$hostid-maintenance/".  As documentation of "try_files" says,
> the further processing is done in thi same (named) location.

I'm not really sure whether $uri is set to "/$hostid-maintenance/" if
"$uri/" is tried (since root is set to /var/nginx/$hostid-maintenance/
already). That works as expected and a request to / will actually just
try / relative to $document_root.

> What happens next is since a request now ends with a slash, the
> static module handler ignores it, and a request is processed by
> the index module.  The index module then verifies that the
> "/var/nginx/$hostid-maintenance/index.html" file exists and does
> an internal redirect to a new $uri.  (The fact that it does an
> internal redirect is also documented.)
> 
> This in turn gets processed by the server's default location
> which is configured to return 503.  Because
> "recursive_error_pages" aren't enabled, the server ends up
> returning status code 503 with the standard content.

Thanks, that's what I forgot about.

> (What happens when you enable is left as an excercise to a
> reader.)
> 
> When OTOH you request "/index.html", try_files checks that the
> _file_ exists, and the processing is performed by the static
> module in the context of location @maintenance.  There are no
> more internal redirects involved, so the server ends up sending
> you the status code 503 with the contents of index.html.
> 
> I agree that this might not be obvious, but this is how it works.
> 
> The important bit here is that requests that end up with "/", and
> get processed by the index module, do an internal redirect.  This
> is quite logical if you think about it, because they *usually*
> get processed by the static module, as if a request ended up with
> "/" followed by the name of index file.

Thanks again and kind regards

-- 
D?en


From edho at myconan.net  Wed Apr  4 09:03:30 2012
From: edho at myconan.net (Edho Arief)
Date: Wed, 4 Apr 2012 16:03:30 +0700
Subject: try_files ignored in combination with error_page and named
	locations
In-Reply-To: <20120404084801.GA11638@azarath.acceleris.ch>
References: <4F7B5DEF.8030107@gmail.com> <20120404081610.GA44137@lo0.su>
	<20120404084801.GA11638@azarath.acceleris.ch>
Message-ID: <CAAtReC=9A-435bYnN=MzXC+kNGHDGo=+-8ghAsSoeAP2CQfvYA@mail.gmail.com>

2012/4/4 D?en <yvldwt at gmail.com>:
> Thank you very much for your clarifying explanation. Seems like the only
> thing missing was "recursive_error_pages on" for it to work like I
> wanted it to have.
>
> On 04/04/12 12:16pm, Ruslan Ermilov wrote:
>> [...]
>>
>> First, this is a wacky config, and I don't quite understand what
>> you're trying to achieve with it. ?:)
>
> Just to make you understand what I was trying to achieve with this
> config: my maintenance page contains img tags pointing to images located
> in $document_root. I used to add a simple
>
> ? ? ? ?try_files /$hostid-maintenance/ @backend;
>
> to my configuration, but as expected a request to /foo.png returns the
> contents of /$hostid-maintenance/index.html instead. Thus I was looking
> for a solution to use a whole directory as an error page (being able to
> request other files which are located inside the maintenance
> $document_root).
>

location /maintenance-static/ {

}

location / {
  try_files /$hostid-maintenance/ @backend;
}


From nginx-forum at nginx.us  Wed Apr  4 09:33:47 2012
From: nginx-forum at nginx.us (Nginx_User777)
Date: Wed, 4 Apr 2012 05:33:47 -0400 (EDT)
Subject: memc_flags_to_last_modified and srcache , no 304 possible ?
In-Reply-To: <CAB4Tn6NaxFJEJeSbYN5YTh1hCnaM=8dyX+ZeX8kpYiANuT8j+g@mail.gmail.com>
References: <CAB4Tn6NaxFJEJeSbYN5YTh1hCnaM=8dyX+ZeX8kpYiANuT8j+g@mail.gmail.com>
Message-ID: <94dbd74a50f2b2281a62e87e00228e32.NginxMailingListEnglish@forum.nginx.org>

Hi Agentzh, 

Thanks for your answer. 

I'll continue to work with a srcache_fetch_skip to honor 304. 

And a really big thanks for these excellent modules, it's great !!!! 

Best Regards.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,224793,224923#msg-224923


From siefke_listen at web.de  Wed Apr  4 09:44:33 2012
From: siefke_listen at web.de (Silvio Siefke)
Date: Wed, 4 Apr 2012 11:44:33 +0200
Subject: PHP with FPM on Gentoo
In-Reply-To: <4F7BEA7A.8060304@gentoo.org>
References: <20120403001323.c03997dd.siefke_listen@web.de>
	<20120403233916.d0943f2e.siefke_listen@web.de>
	<4F7BEA7A.8060304@gentoo.org>
Message-ID: <20120404114433.49d17c07.siefke_listen@web.de>

Hello,

On Wed, 04 Apr 2012 08:30:18 +0200
Ole Markus With <olemarkus at gentoo.org> wrote:

> Did you see the Gentoo install instructions [1]?

Yes i have read.
 
> You only need the fpm USE flag enabled for php.
> Nginx require NGINX_MODULES_HTTP="fastcgi"

In PHP is activated and Nginx i have reinstalled yesterdays with fastcgi Flag.
I use Google normaly before. 

> With those set, fpm + nginx should work pretty much out of the box after
> adding the fastcgi_pass line to nginx.conf.

No work not pretty. On FreeBSD yes, on Gentoo nothing. 

> TCP sockets are the easiest to configure. If you want to use unix
> sockets, you have to make sure the file permissions are correct etc.

Has set again the IPs, which permissions must habe Unix sockets? Owner
is normal, nginx but chmod?


Regards
Silvio


From ol.morel at linkbynet.com  Wed Apr  4 09:58:49 2012
From: ol.morel at linkbynet.com (morel)
Date: Wed, 4 Apr 2012 11:58:49 +0200
Subject: Question about Vhost
Message-ID: <4F7C1B59.2060209@linkbynet.com>

hy

i have installed nginx on my web server, i can run the configtest on 
nginx.conf but i would want also check the configuration of my vhost .

When i try diverse handling  like this

  /usr/local/nginx/sbin/nginx -c -t 
/etc/nginx/conf/vhosts-available/Site1.conf

all the time i get the same error

  * nginx: [emerg] unknown directive "server" in
    /usr/local/nginx/vhosts-available/Site1.conf
  * nginx: configuration file
    /usr/local/nginx/vhosts-available/rSite1.conf test failed


My question is do you know theyre is  possibility to test vhost conf ?


Cordially

Oliver
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20120404/bdfefdf2/attachment.html>

From contact at jpluscplusm.com  Wed Apr  4 10:41:50 2012
From: contact at jpluscplusm.com (Jonathan Matthews)
Date: Wed, 4 Apr 2012 11:41:50 +0100
Subject: Question about Vhost
In-Reply-To: <4F7C1B59.2060209@linkbynet.com>
References: <4F7C1B59.2060209@linkbynet.com>
Message-ID: <CAKsTx7AuUrxeF5aGStGeY9hsOeuFQXpP3f4O5gy2T8-L7ZNQpA@mail.gmail.com>

On 4 April 2012 10:58, morel <ol.morel at linkbynet.com> wrote:
> i have installed nginx on my web server, i can run the configtest on
> nginx.conf but i would want also check the configuration of my vhost .

By testing your main config, you're automatically testing all included
files as well.

Jonathan
-- 
Jonathan Matthews
London, Oxford, UK
http://www.jpluscplusm.com/contact.html


From mdounin at mdounin.ru  Wed Apr  4 11:09:58 2012
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Wed, 4 Apr 2012 15:09:58 +0400
Subject: Possible HTTP response read bug in Nginx 1.0.12
In-Reply-To: <0897eea6bf171e32cb305273ccc3d6a2.NginxMailingListEnglish@forum.nginx.org>
References: <20120403161530.GS13466@mdounin.ru>
	<0897eea6bf171e32cb305273ccc3d6a2.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20120404110958.GW13466@mdounin.ru>

Hello!

On Tue, Apr 03, 2012 at 09:06:12PM -0400, tikonen wrote:

> Hi,
> 
> Thanks for advice, I'll try next with 1.1.X and see if that helps.
> 
> In case 1.1.X does not solve the problem. Server load is pretty big and
> only minor % of requests have the error so the debug log and capture
> dump would be Huge to catch even one case. (Still these errors are
> frequent enough to cause ip_hash to flag node failed and cause havoc in
> the cluster).
> 
> To narrow down needed data, can you recommend proper debugging settings,
> should I start server or global level?

It's better to have debugging activated at global level to be able 
to see events reported by the kernel.

> Also, should I make debug log againts 1.1.X or 1.0.12 ?

The 1.1.x would be better.

Maxim Dounin


From nginx-forum at nginx.us  Wed Apr  4 11:46:05 2012
From: nginx-forum at nginx.us (locojohn)
Date: Wed, 4 Apr 2012 07:46:05 -0400 (EDT)
Subject: PHP with FPM on Gentoo
In-Reply-To: <20120403233916.d0943f2e.siefke_listen@web.de>
References: <20120403233916.d0943f2e.siefke_listen@web.de>
Message-ID: <cb1fb20f89390f52039cb8fb33362dec.NginxMailingListEnglish@forum.nginx.org>

Silvio Siefke Wrote:
-------------------------------------------------------
> Hello,
> 
> 
> Could someone possibly provide me the USE flags
> available 
> if someone has run Nginx, PHP and Fpm.

/etc/make.conf:

NGINX_MODULES_HTTP="access autoindex browser charset \
  empty_gif fastcgi flv geoip gzip limit_zone \
  map proxy rewrite stub_status upstream_ip_hash \
  upload_progress"

/etc/portage/package.use:

www-servers/nginx aio

dev-lang/php -* fpm cli bcmath ssl ctype curl fileinfo filter ftp gd
hash iconv json mysqlnd mysql mysqli nls pcntl pcre posix readline
reflection session simplexml
 sharedmem spl sysvipc tokenizer truetype unicode xml zlib xmlreader
xmlwriter

---

www-servers/nginx-1.1.18::x-portage  USE="aio http http-cache pcre ssl
-debug -ipv6 -libatomic -pcre-jit -vim-syntax"
NGINX_MODULES_HTTP="access autoindex browser charset empty_gif fastcgi
flv geoip gzip limit_zone map proxy rewrite stub_status upload_progress
upstream_ip_hash -addition -auth_basic -cache_purge -dav -degradation
-geo -gzip_static -headers_more -image_filter -limit_req -memcached -mp4
-passenger -perl -push -random_index -realip -referer -scgi -secure_link
-slowfs_cache -split_clients -ssi -sub -upload -userid -uwsgi -xslt"
NGINX_MODULES_MAIL="-imap -pop3 -smtp" 0 kB

dev-lang/php-5.3.10-r2  USE="bcmath cli ctype curl fileinfo filter fpm
ftp gd hash iconv json mysql mysqli mysqlnd nls pcntl posix readline
session sharedmem simplexml ssl sysvipc tokenizer truetype unicode xml
xmlreader xmlwriter zlib -apache2 -berkdb -bzip2 -calendar -cdb -cgi
-cjk -crypt -curlwrappers -debug -doc -embed -enchant -exif -firebird
-flatfile (-frontbase) -gdbm -gmp -imap -inifile -intl -iodbc -ipv6
-kerberos -kolab -ldap -ldap-sasl -libedit -mhash -mssql
-oci8-instant-client -odbc -pdo -phar -pic -postgres -qdbm -recode -snmp
-soap -sockets -spell -sqlite -sqlite3 -suhosin (-sybase-ct) -threads
-tidy -wddx -xmlrpc -xpm -xsl -zip" 0 kB

> Has set again the IPs, which permissions must habe Unix sockets?
Owner
> is normal, nginx but chmod?

The socket file configured in php-fpm.conf should be readable and
writeable by nginx.  Naturally, if socket owner/group in php-fpm are set
to nginx:nginx, nginx should work with php fpm over socket connection
just fine.

One question: do you use selinux?  Any specific Gentoo installation? 
Please send me your /etc/make.conf file.
 

Andrejs

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,224804,224931#msg-224931


From quintinpar at gmail.com  Wed Apr  4 12:17:22 2012
From: quintinpar at gmail.com (Quintin Par)
Date: Wed, 4 Apr 2012 17:47:22 +0530
Subject: Serve from cache but fire request to upstream server to increment
	page view counter
In-Reply-To: <77590aa60cbf956e5fa84ac68220edca.squirrel@damiao.org>
References: <CACW9FdNgfxGejBvzWoqbM7+ikNpZxum82MmmS0aFd6WsK3AOUw@mail.gmail.com>
	<CACW9FdOxDwk_MPx1SD5-J-7z9TUAtm-Gpv3Vo23T=n4b5ZgBmQ@mail.gmail.com>
	<77590aa60cbf956e5fa84ac68220edca.squirrel@damiao.org>
Message-ID: <CACW9FdPv2mk+-7Nnc5hHcgSoAzwyPj_3=QORYq=QKr4Mvx03PA@mail.gmail.com>

Thanks Appa.

Why would you say this is not reliable? Is there data or experience to
suggest? The semantic is awesome. Precisely what I?ve been looking for.

-Cherian

On Wed, Apr 4, 2012 at 1:39 PM, Antonio P.P. Almeida <appa at perusio.net>wrote:

> There's the post_action directive. But AFAIK is not that reliable.
>
> I would use Lua with the "new" cosocket API and do an HTTP request.
>
> Here's an example of a library built around it that talks with a Redis
> backend: https://github.com/agentzh/lua-resty-redis
>
> As stated post_action is another option:
>
> http://wiki.nginx.org/HttpCoreModule#post_action
>
>
> --appa
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20120404/c4cb40c5/attachment.html>

From appa at perusio.net  Wed Apr  4 12:21:49 2012
From: appa at perusio.net (Antonio P.P. Almeida)
Date: Wed, 4 Apr 2012 14:21:49 +0200
Subject: Serve from cache but fire request to upstream server to increment
	page view counter
In-Reply-To: <CACW9FdPv2mk+-7Nnc5hHcgSoAzwyPj_3=QORYq=QKr4Mvx03PA@mail.gmail.com>
References: <CACW9FdNgfxGejBvzWoqbM7+ikNpZxum82MmmS0aFd6WsK3AOUw@mail.gmail.com>
	<CACW9FdOxDwk_MPx1SD5-J-7z9TUAtm-Gpv3Vo23T=n4b5ZgBmQ@mail.gmail.com>
	<77590aa60cbf956e5fa84ac68220edca.squirrel@damiao.org>
	<CACW9FdPv2mk+-7Nnc5hHcgSoAzwyPj_3=QORYq=QKr4Mvx03PA@mail.gmail.com>
Message-ID: <0b0d1d41b7b059cd1add3ed5f8755d1b.squirrel@damiao.org>

> Thanks Appa.
>
> Why would you say this is not reliable? Is there data or experience to
> suggest? The semantic is awesome. Precisely what I've been looking for.
>
> -Cherian

Well from remarks that core devs of Nginx have posted here before,
referring to it as "hack" (SIC). In fact it's not even documented in the
official docs.

I'm not sure if that's an overlook or has deeper meanings related to the
quality of the said directive concept/code.

--appa


From contact at jpluscplusm.com  Wed Apr  4 12:27:32 2012
From: contact at jpluscplusm.com (Jonathan Matthews)
Date: Wed, 4 Apr 2012 13:27:32 +0100
Subject: Serve from cache but fire request to upstream server to increment
	page view counter
In-Reply-To: <CACW9FdPv2mk+-7Nnc5hHcgSoAzwyPj_3=QORYq=QKr4Mvx03PA@mail.gmail.com>
References: <CACW9FdNgfxGejBvzWoqbM7+ikNpZxum82MmmS0aFd6WsK3AOUw@mail.gmail.com>
	<CACW9FdOxDwk_MPx1SD5-J-7z9TUAtm-Gpv3Vo23T=n4b5ZgBmQ@mail.gmail.com>
	<77590aa60cbf956e5fa84ac68220edca.squirrel@damiao.org>
	<CACW9FdPv2mk+-7Nnc5hHcgSoAzwyPj_3=QORYq=QKr4Mvx03PA@mail.gmail.com>
Message-ID: <CAKsTx7C=1Li5nSNHqUsncdRChvG00Ei9sw_3DW823DKEzfV1WA@mail.gmail.com>

On 4 April 2012 13:17, Quintin Par <quintinpar at gmail.com> wrote:
> Thanks Appa.
>
> Why would you say this is not reliable? Is there data or experience to
> suggest? The semantic is awesome. Precisely what I?ve been looking for.

I'd also be interested in discovering any unreliability of this
mechanism, as I'm planning to use it for some real-time logging
shortly.

The weaknesses that I've discovered thus far, which I think should all
be fixed -- or at least modified as indicated below -- are

* The client connection is held open until the post_action call completes
* The response code returned to the client is the post_action's response code
* The access.log entry logged reflects the post_action's URI, response
code and (possibly; I forget) content length

All of these render it difficult to use post_action as its name
implies - to be competed *after* the client request is completely
dealt with.

I hope that at some point in the future, we'll be able to use
post_action in a way which is invisible to the client, and to the
logs. Perhaps with a post_action directive flag, indicating that any
calls to this specific post_action URI should stay out of the logs and
the client's view.

J

> On Wed, Apr 4, 2012 at 1:39 PM, Antonio P.P. Almeida <appa at perusio.net>
> wrote:
>>
>> There's the post_action directive. But AFAIK is not that reliable.
>>
>> I would use Lua with the "new" cosocket API and do an HTTP request.
>>
>> Here's an example of a library built around it that talks with a Redis
>> backend: https://github.com/agentzh/lua-resty-redis
>>
>> As stated post_action is another option:
>>
>> http://wiki.nginx.org/HttpCoreModule#post_action
>>
>>
>> --appa
>>
>> _______________________________________________
>> 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



-- 
Jonathan Matthews
London, Oxford, UK
http://www.jpluscplusm.com/contact.html


From francis at daoine.org  Wed Apr  4 12:40:40 2012
From: francis at daoine.org (Francis Daly)
Date: Wed, 4 Apr 2012 13:40:40 +0100
Subject: Config directive inheritance: docs and rules (was: Config
	variable inheritance)
In-Reply-To: <CAKsTx7A88PTFp14208kb5iu5RK=KefEC7TY9GzZJOmBH1gtMAg@mail.gmail.com>
References: <CAKsTx7A88PTFp14208kb5iu5RK=KefEC7TY9GzZJOmBH1gtMAg@mail.gmail.com>
Message-ID: <20120404124040.GB26052@craic.sysops.org>

On Tue, Apr 03, 2012 at 06:20:13PM +0100, Jonathan Matthews wrote:

Hi there,

> http://www.ruby-forum.com/topic/188742 seems to contain some
> information I've not found explicitly stated anywhere else: that
> config directive inheritance is an all or nothing mechanism. Have I
> missed some docs somewhere?

That refers to inheritance of a *single* directive.

> Following on from this, I'd like to understand if there's anything I
> can reliably determine from a directive's name about the impact that
> defining it will have on *other* directives' inheritability.

Yes: it will have no impact on other directives.

> Can I, for instance, assume that defining an xyx_ prefixed directive in
> a sub-scope will affect *only* the inheritance of other xyz_ prefixed
> directives?

No. It will affect no other xyz_ prefixed directives.

It may make a difference to the *applicability* of other directives --
for example, if fastcgi_pass is not set, then it doesn't really matter
whether fastcgi_param is inherited or not, because it's not going to be
used. But fastcgi_pass does not affect whether fastcgi_param is inherited.

> Or can there be inheritance side-effects between, say,
> directive "abc" and "def" that I can't possibly determine without
> documentation or trial and error?

No, it will have no impact on other directives.

> Equally, can I assume that *all* xyz_ prefixed directives will have
> their inheritance affected by the sub-scope definition of any *other*
> xyz_ prefixed directive?

No: it will have no impact on other directives.

> Or are there examples where defining
> "xyz_foo" does *not* impact the inheritance of "xyz_bar"?

Yes: every example.

> I'm thinking specifically of proxy_* directives, whereas the thread I
> link to above is talking about multiple fastcgi_param directives.

Using a directive sets the value of that directive. It does not set the
value of different directives.

"Not using" a directive means the value of that directive is either
unset/default, or is the same value as the upper level.

> Here, I'm interested in how, say, proxy_set_header affects the
> inheritance of higher scope's proxy_pass directive - i.e. where the
> prefix matches but they're *not* the same directive.

No impact at all.

> Does nginx (core)
> derive or imply anything relating to the name-spacing of directives
> before (in my example) the proxy module sees the config, or is it
> entirely down to how each module implements the configuration
> aggregation/inheritance?

It's entirely down to how each module implements it. But if a module
implements inheritance of a directive as something other than "replace"
or "no inheritance", then that should probably be considered a bug.

	f
-- 
Francis Daly        francis at daoine.org


From mdounin at mdounin.ru  Wed Apr  4 13:58:46 2012
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Wed, 4 Apr 2012 17:58:46 +0400
Subject: Config directive inheritance: docs and rules (was: Config
	variable inheritance)
In-Reply-To: <20120404124040.GB26052@craic.sysops.org>
References: <CAKsTx7A88PTFp14208kb5iu5RK=KefEC7TY9GzZJOmBH1gtMAg@mail.gmail.com>
	<20120404124040.GB26052@craic.sysops.org>
Message-ID: <20120404135846.GX13466@mdounin.ru>

Hello!

On Wed, Apr 04, 2012 at 01:40:40PM +0100, Francis Daly wrote:

> On Tue, Apr 03, 2012 at 06:20:13PM +0100, Jonathan Matthews wrote:
> 
> Hi there,
> 
> > http://www.ruby-forum.com/topic/188742 seems to contain some
> > information I've not found explicitly stated anywhere else: that
> > config directive inheritance is an all or nothing mechanism. Have I
> > missed some docs somewhere?
> 
> That refers to inheritance of a *single* directive.
> 
> > Following on from this, I'd like to understand if there's anything I
> > can reliably determine from a directive's name about the impact that
> > defining it will have on *other* directives' inheritability.
> 
> Yes: it will have no impact on other directives.
> 
> > Can I, for instance, assume that defining an xyx_ prefixed directive in
> > a sub-scope will affect *only* the inheritance of other xyz_ prefixed
> > directives?
> 
> No. It will affect no other xyz_ prefixed directives.

Not really, there are cases where it will.  Examples include (not 
sure if there are others):

1) allow and deny directives of access module create an array of 
allow/deny rules, and this array is inherited as a whole;

2) recently introduced xslt_param and xslt_string_param do the 
same thing with xslt parameters.

All such cases are more or less obvious though.

[...]

Maxim Dounin


From francis at daoine.org  Wed Apr  4 16:54:44 2012
From: francis at daoine.org (Francis Daly)
Date: Wed, 4 Apr 2012 17:54:44 +0100
Subject: Config directive inheritance: docs and rules (was: Config
	variable inheritance)
In-Reply-To: <20120404135846.GX13466@mdounin.ru>
References: <CAKsTx7A88PTFp14208kb5iu5RK=KefEC7TY9GzZJOmBH1gtMAg@mail.gmail.com>
	<20120404124040.GB26052@craic.sysops.org>
	<20120404135846.GX13466@mdounin.ru>
Message-ID: <20120404165444.GA12819@craic.sysops.org>

On Wed, Apr 04, 2012 at 05:58:46PM +0400, Maxim Dounin wrote:
> On Wed, Apr 04, 2012 at 01:40:40PM +0100, Francis Daly wrote:
> > On Tue, Apr 03, 2012 at 06:20:13PM +0100, Jonathan Matthews wrote:

Hi there,

> > > Can I, for instance, assume that defining an xyx_ prefixed directive in
> > > a sub-scope will affect *only* the inheritance of other xyz_ prefixed
> > > directives?
> > 
> > No. It will affect no other xyz_ prefixed directives.
> 
> Not really, there are cases where it will.  Examples include (not 
> sure if there are others):
> 
> 1) allow and deny directives of access module create an array of 
> allow/deny rules, and this array is inherited as a whole;
> 
> 2) recently introduced xslt_param and xslt_string_param do the 
> same thing with xslt parameters.

Ah, thank you for the correction.

For the original poster: I was wrong, you *do* need to check or test each
time. So read "never" and "always" as "almost never" and "almost always".

Sorry about that.


I've done a quick analysis of the 1.1.18 codebase, to see if I can find
the current complete list (excluding 3rd party modules, of course).

The configuration directives are stored in a struct:

struct ngx_command_s {
    ngx_str_t             name;
    ngx_uint_t            type;
    char               *(*set)(ngx_conf_t *cf, ngx_command_t *cmd, void *conf);
    ngx_uint_t            conf;
    ngx_uint_t            offset;
    void                 *post;
};

"name" is the directive name; "*set" is the function to set the value
of the directive, and "offset" is the position relative to "conf" where
the value is stored.

I see 67 arrays of that struct (grep for "static ngx_command_t").

I see 449 "ngx_string" within those arrays.

110 have unique *set values.

Of the ones with non-unique *set values, the few that have the same
"offset" value are:

ngx_event_connections - worker_connections, connections; one deprecated
ngx_mail_block - mail, imap; one deprecated
ngx_conf_set_sec_slot - open_file_cache_valid, open_file_cache_retest;
  one deprecated

ngx_http_core_root - root, alias: single-valued, only one at a level
ngx_http_access_rule - allow, deny; both are valid
ngx_http_xslt_param - xslt_param, xslt_string_param; both are valid.


And separate from that, the directives which have a duplicate "offset"
value of "offsetof(something)" (as in, not-0 or a predefined constant),
are:

open_file_cache_valid, open_file_cache_retest - one deprecated
satisfy, satisfy_any - one deprecated
optimize_server_names, server_name_in_redirect - one deprecated

If that's an adequate analysis, it suggests that

  allow/deny and xslt_param/xslt_string_param

are the only distributed directives which don't follow the
replace-or-inherit-this-directive-only method.

> All such cases are more or less obvious though.

I wonder, is it worth having a note in the documentation for these
exceptions to the common case?

I see that there is such a note at
 
http://nginx.org/en/docs/http/ngx_http_xslt_module.html#xslt_param,

but not (yet) at

http://nginx.org/en/docs/http/ngx_http_access_module.html#allow

Thanks,

	f
-- 
Francis Daly        francis at daoine.org


From nginx-forum at nginx.us  Wed Apr  4 17:21:27 2012
From: nginx-forum at nginx.us (fiyah2011)
Date: Wed, 4 Apr 2012 13:21:27 -0400 (EDT)
Subject: POST DATA in cache_key
Message-ID: <cf4261773ab436bff2fcadc3f5d8eb08.NginxMailingListEnglish@forum.nginx.org>

is it possible to use a value found within POST DATA as part of the
cache key?

for example, the value of variable called 'username' to be part of the
cache key.

so i basically want the cache to key to simply hash against the
connecting IP and the 'username' POST variable.

proxy_cache_key "$remote_addr$POST[username]"

thanks in advance

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,224945,224945#msg-224945


From yvldwt at gmail.com  Wed Apr  4 17:27:35 2012
From: yvldwt at gmail.com (=?UTF-8?B?RMW+ZW4=?=)
Date: Wed, 04 Apr 2012 19:27:35 +0200
Subject: try_files ignored in combination with error_page and named
	locations
In-Reply-To: <CAAtReC=9A-435bYnN=MzXC+kNGHDGo=+-8ghAsSoeAP2CQfvYA@mail.gmail.com>
References: <4F7B5DEF.8030107@gmail.com> <20120404081610.GA44137@lo0.su>
	<20120404084801.GA11638@azarath.acceleris.ch>
	<CAAtReC=9A-435bYnN=MzXC+kNGHDGo=+-8ghAsSoeAP2CQfvYA@mail.gmail.com>
Message-ID: <4F7C8487.4030506@gmail.com>

On 04/04/2012 11:03, Edho Arief wrote:
[...]
> location /maintenance-static/ {
>
> }
>
> location / {
>    try_files /$hostid-maintenance/ @backend;
> }

Right, this would work just fine. However, requests to /maintenance-static/
would still be possible even if /$hostid-maintenance/ doesn't exist (minor
problem, but maybe the backend uses /maintenance-static/ for some reason).

Unfortunately, I still haven't found a solution which pleases me. Just to
clarify, this is how I want the request to be processed (el-cheapo pseudo
code):

	if the directory /maintenance/ exists, then
		set root to /maintenance/
		if the file $uri exists, then
			serve $uri with errcode 503
		otherwise
			serve index.html with errcode 503
	otherwise
		proxy_pass http://backend

My new attempt:

	recursive_error_pages on;
	error_page 599 @maintenance;

	if ( -d $document_root/$hostid-maintenance ) {
		return 599;
	}

	location @maintenance {
		root		/var/nginx/$hostid-maintenance;
		index		index.html;
		try_files	$uri $uri/ /index.html /;
	}

I used error code 599 because I noticed that for some reason the backend 
might
raise an 503 error although $document_root/$hostid-maintenance doesn't 
exist.
The result in such a situation would be unexpected (an infinite redirect 
loop?).
By using an unused error code like 599, nginx's default 503 error page 
will be
displayed instead, if the backend raises 503.

This configuration seems to work, but the 599 error code in the HTTP header
isn't that nice. Unfortunately, something like

	error_page 599 =503 @maintenance;

doesn't result in what I expected (it will again just return nginx's default
503 error page, although error 503 is correctly set in the HTTP header).

I still think that I misunderstood something, am I right?

Kind regards

-- 
D?en


From nginx-forum at nginx.us  Wed Apr  4 17:41:45 2012
From: nginx-forum at nginx.us (fiyah2011)
Date: Wed, 4 Apr 2012 13:41:45 -0400 (EDT)
Subject: POST DATA in cache_key
In-Reply-To: <cf4261773ab436bff2fcadc3f5d8eb08.NginxMailingListEnglish@forum.nginx.org>
References: <cf4261773ab436bff2fcadc3f5d8eb08.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <c58044de8f9f253b1349fbd744222d97.NginxMailingListEnglish@forum.nginx.org>

this looks like the solution i am looking for.

https://github.com/calio/form-input-nginx-module/blob/master/README

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,224945,224947#msg-224947


From mdounin at mdounin.ru  Wed Apr  4 18:18:42 2012
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Wed, 4 Apr 2012 22:18:42 +0400
Subject: Config directive inheritance: docs and rules (was: Config
	variable inheritance)
In-Reply-To: <20120404165444.GA12819@craic.sysops.org>
References: <CAKsTx7A88PTFp14208kb5iu5RK=KefEC7TY9GzZJOmBH1gtMAg@mail.gmail.com>
	<20120404124040.GB26052@craic.sysops.org>
	<20120404135846.GX13466@mdounin.ru>
	<20120404165444.GA12819@craic.sysops.org>
Message-ID: <20120404181842.GY13466@mdounin.ru>

Hello!

On Wed, Apr 04, 2012 at 05:54:44PM +0100, Francis Daly wrote:

[...]

> If that's an adequate analysis, it suggests that
> 
>   allow/deny and xslt_param/xslt_string_param
> 
> are the only distributed directives which don't follow the
> replace-or-inherit-this-directive-only method.

Mechanical analisys of the codebase might miss some cases with 
custom directive handlers used.  But I suppose the result is 
correct as it matches one from my memory.  :)

> > All such cases are more or less obvious though.
> 
> I wonder, is it worth having a note in the documentation for these
> exceptions to the common case?
> 
> I see that there is such a note at
>  
> http://nginx.org/en/docs/http/ngx_http_xslt_module.html#xslt_param,
> 
> but not (yet) at
> 
> http://nginx.org/en/docs/http/ngx_http_access_module.html#allow

Yes, this surely deserves a note.

Maxim Dounin


From siefke_listen at web.de  Wed Apr  4 20:19:45 2012
From: siefke_listen at web.de (Silvio Siefke)
Date: Wed, 4 Apr 2012 22:19:45 +0200
Subject: PHP with FPM on Gentoo
In-Reply-To: <20120403001323.c03997dd.siefke_listen@web.de>
References: <20120403001323.c03997dd.siefke_listen@web.de>
Message-ID: <20120404221945.265d10b0.siefke_listen@web.de>

Hello,


now i have build php and nginx from source direct, emerge --unmerge, 
have downloaded the sourcecode and compile it and same it run not. 

Thats really crazy, what is this? Is this maybe with eselect profile list
option? Because i use 

gentoo-desk imap # eselect profile list
Available profile symlink targets:
  [1]   default/linux/x86/10.0
  [2]   default/linux/x86/10.0/selinux
  [3]   default/linux/x86/10.0/desktop *
  [4]   default/linux/x86/10.0/desktop/gnome
  [5]   default/linux/x86/10.0/desktop/kde
  [6]   default/linux/x86/10.0/developer
  [7]   default/linux/x86/10.0/server
  [8]   hardened/linux/x86
  [9]   hardened/linux/x86/selinux

Is this maybe the mistake or where can be the error?


Regards
Silvio


From nginx-forum at nginx.us  Wed Apr  4 20:40:04 2012
From: nginx-forum at nginx.us (shoshomiga)
Date: Wed, 4 Apr 2012 16:40:04 -0400 (EDT)
Subject: limit_rate dynamically using $arg - security
Message-ID: <6421426a61f0bd2e0f8071c6705b87cd.NginxMailingListEnglish@forum.nginx.org>

I've been looking for a way to limit videos to their bitrate to save
bandwidth and I've come up with this code

            if ($arg_LIMITSPEED) {
              set $limit_rate $arg_LIMITSPEED;
            }

It works but I would like to know if this code would be secure to use on
a production server.

I am not worried about users setting their LIMITSPEED high on their own
because I am limiting speeds at the network level as well.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,224950,224950#msg-224950


From contact at jpluscplusm.com  Wed Apr  4 21:32:49 2012
From: contact at jpluscplusm.com (Jonathan Matthews)
Date: Wed, 4 Apr 2012 22:32:49 +0100
Subject: limit_rate dynamically using $arg - security
In-Reply-To: <6421426a61f0bd2e0f8071c6705b87cd.NginxMailingListEnglish@forum.nginx.org>
References: <6421426a61f0bd2e0f8071c6705b87cd.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <CAKsTx7ChGmajGvXK6UK6fnoR6kARkwFr97R7T52W8pxZKvSaoA@mail.gmail.com>

On 4 April 2012 21:40, shoshomiga <nginx-forum at nginx.us> wrote:
> I've been looking for a way to limit videos to their bitrate to save
> bandwidth and I've come up with this code
>
> ? ? ? ? ? ?if ($arg_LIMITSPEED) {
> ? ? ? ? ? ? ?set $limit_rate $arg_LIMITSPEED;
> ? ? ? ? ? ?}
>
> It works but I would like to know if this code would be secure to use on
> a production server.
>
> I am not worried about users setting their LIMITSPEED high on their own
> because I am limiting speeds at the network level as well.

To be honest, I'm not sure what definition of "insecure" you could be
thinking of that *isn't* "the user can override it trivially" :-)

If you're doing the rate limiting at the network level properly, then
why duplicate the effort? It's just one more place you have to change
when you upgrade the speed limits.

Personally, I'm prototyping a streaming service at the moment using
http://wiki.nginx.org/X-accel#X-Accel-Limit-Rate and a double
proxy_pass (via X-Accel-Redirect to an internal storage proxy_pass).
It all looks like it works nicely, and allows the dumb storage backend
to throw data at the nginx router as fast as nginx accepts it, and for
the first (intelligent) proxy_pass backend to *decide* the bitrate via
X-Accel-Limit-Rate. I'll blog it soonish :-)

Jonathan
-- 
Jonathan Matthews
London, Oxford, UK
http://www.jpluscplusm.com/contact.html


From albert at nn.iij4u.or.jp  Thu Apr  5 00:25:44 2012
From: albert at nn.iij4u.or.jp (shin fukuda)
Date: Thu, 5 Apr 2012 09:25:44 +0900
Subject: no response nginx
Message-ID: <20120405092544.f6c3c2cdd6d87bb9a320da8a@nn.iij4u.or.jp>

Hi.

I have a problem with nginx-1.1.16 uses proxy_cache_lock on. 
When high load or many request to nginx, Suddenly nginx seems hang, 
no response at all. 
In netstat "SYN_RECV","CLOSE_WAIT" too many. This problem needs 
restarting nginx, and "CLOSE_WAIT" clear.

There is no problem in front-end nginx servers, but backends one.

Any clue?


-- 
shin fukuda <albert at nn.iij4u.or.jp>


From quintinpar at gmail.com  Thu Apr  5 03:50:50 2012
From: quintinpar at gmail.com (Quintin Par)
Date: Thu, 5 Apr 2012 09:20:50 +0530
Subject: Serve from cache but fire request to upstream server to increment
	page view counter
In-Reply-To: <CAKsTx7C=1Li5nSNHqUsncdRChvG00Ei9sw_3DW823DKEzfV1WA@mail.gmail.com>
References: <CACW9FdNgfxGejBvzWoqbM7+ikNpZxum82MmmS0aFd6WsK3AOUw@mail.gmail.com>
	<CACW9FdOxDwk_MPx1SD5-J-7z9TUAtm-Gpv3Vo23T=n4b5ZgBmQ@mail.gmail.com>
	<77590aa60cbf956e5fa84ac68220edca.squirrel@damiao.org>
	<CACW9FdPv2mk+-7Nnc5hHcgSoAzwyPj_3=QORYq=QKr4Mvx03PA@mail.gmail.com>
	<CAKsTx7C=1Li5nSNHqUsncdRChvG00Ei9sw_3DW823DKEzfV1WA@mail.gmail.com>
Message-ID: <CACW9FdMXwY=toRmPkkv+=ucWNeJd7XHWHQQPGs65wrhdwMDiGg@mail.gmail.com>

This seems to very risky.

One more question:

The whole reason why I?d use post action is to take the time off from doing
a SSI. i.e. serve a page from cache in light speed and then let post action
take its own sweet time to complete.

But from Jonathan?s response I infer that a post actions response code is
passed to the client. Which means the client is made to wait till the post
action is complete. Right? If yes, that defeats the purpose.

On Wed, Apr 4, 2012 at 5:57 PM, Jonathan Matthews
<contact at jpluscplusm.com>wrote:

> On 4 April 2012 13:17, Quintin Par <quintinpar at gmail.com> wrote:
> > Thanks Appa.
> >
> > Why would you say this is not reliable? Is there data or experience to
> > suggest? The semantic is awesome. Precisely what I?ve been looking for.
>
> I'd also be interested in discovering any unreliability of this
> mechanism, as I'm planning to use it for some real-time logging
> shortly.
>
> The weaknesses that I've discovered thus far, which I think should all
> be fixed -- or at least modified as indicated below -- are
>
> * The client connection is held open until the post_action call completes
> * The response code returned to the client is the post_action's response
> code
> * The access.log entry logged reflects the post_action's URI, response
> code and (possibly; I forget) content length
>
> All of these render it difficult to use post_action as its name
> implies - to be competed *after* the client request is completely
> dealt with.
>
> I hope that at some point in the future, we'll be able to use
> post_action in a way which is invisible to the client, and to the
> logs. Perhaps with a post_action directive flag, indicating that any
> calls to this specific post_action URI should stay out of the logs and
> the client's view.
>
> J
>
> > On Wed, Apr 4, 2012 at 1:39 PM, Antonio P.P. Almeida <appa at perusio.net>
> > wrote:
> >>
> >> There's the post_action directive. But AFAIK is not that reliable.
> >>
> >> I would use Lua with the "new" cosocket API and do an HTTP request.
> >>
> >> Here's an example of a library built around it that talks with a Redis
> >> backend: https://github.com/agentzh/lua-resty-redis
> >>
> >> As stated post_action is another option:
> >>
> >> http://wiki.nginx.org/HttpCoreModule#post_action
> >>
> >>
> >> --appa
> >>
> >> _______________________________________________
> >> 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
>
>
>
> --
> Jonathan Matthews
> London, Oxford, UK
> http://www.jpluscplusm.com/contact.html
>
> _______________________________________________
> 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/20120405/9c1fe841/attachment.html>

From arissirajawali at gmail.com  Thu Apr  5 04:18:19 2012
From: arissirajawali at gmail.com (aris)
Date: Thu, 5 Apr 2012 11:18:19 +0700
Subject: nginx conf
Message-ID: <CAD8siQ5+Guommd9RaUPorN2o8Ho9Q9ny_d2_LxerUivoEfmgvw@mail.gmail.com>

Hi all,
how to configure nginx.conf, sites-available and sites-enabled
I get 403 Forbidden nginx/0.7.67
please check out where my configuration errors

*I nginx.conf following:*

user www-data;
worker_processes  1;

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

events {
    worker_connections  1024;
    # multi_accept on;
}

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

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

    sendfile        on;
    #tcp_nopush     on;

    #keepalive_timeout  0;
    keepalive_timeout  65;
    tcp_nodelay        on;

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

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

# mail {
#     # See sample authentication script at:
#     # http://wiki.nginx.org/NginxImapAuthenticateWithApachePhpScript
#
#     # auth_http localhost/auth.php;
#     # pop3_capabilities "TOP" "USER";
#     # imap_capabilities "IMAP4rev1" "UIDPLUS";
#
#     server {
#         listen     localhost:110;
#         protocol   pop3;
#         proxy      on;
#     }
#
#     server {
#         listen     localhost:143;
#         protocol   imap;
#         proxy      on;
#     }
# }


*The following sites-available I have :*
# You may add here your
# server {
#       ...
# }
# statements for each of your virtual hosts

server {

        listen   80; ## listen for ipv4
        server_name  localhost;
        access_log  /var/log/nginx/localhost.access.log;

        location / {
                root   /var/www;
                index  index.php;
        }

        location /doc {
                root   /usr/share;
                autoindex on;
                allow 127.0.0.1;
                deny all;
        }

        location /images {
                root   /usr/share;
                autoindex on;
        }



        #error_page  404  /404.html;

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

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

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

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


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

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


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

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

#ssl_session_timeout  5m;

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

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

*The following sites-enabled I have :*
# You may add here your
# server {
#       ...
# }
# statements for each of your virtual hosts

server {

        listen   80; ## listen for ipv4
        server_name  localhost;
        access_log  /var/log/nginx/localhost.access.log;

        location / {
                root   /var/www;
                index  index.php;
        }

        location /doc {
                root   /usr/share;
                autoindex on;
                allow 127.0.0.1;
                deny all;
        }

        location /images {
                root   /usr/share;
                autoindex on;
        }



        #error_page  404  /404.html;

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

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

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

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


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

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


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

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

#ssl_session_timeout  5m;

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

#location / {
#root   html;
#index  index.html index.htm;
#}
#}
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20120405/ff18ff86/attachment-0001.html>

From loco at d0pefish.de  Thu Apr  5 07:56:07 2012
From: loco at d0pefish.de (Peter Hinse)
Date: Thu, 05 Apr 2012 09:56:07 +0200
Subject: Testing the HttpLimitReqModule without blocking
Message-ID: <4F7D5017.5080808@d0pefish.de>

Hi all,

I want to implement the HttpLimitReqModule within our setup to get rid 
of people scraping the content of our page.

However, since some of our bigger clients are using the same proxy IP to 
connect to our page, I need to adjust the limit before really blocking 
users. Any chance to log the attempts without delivering a 503 page?

If some of the proxy users would fit into the blocking scheme, is there 
a chance to define a whitelist?

Regards,

	Peter.


From nginx-forum at nginx.us  Thu Apr  5 08:56:32 2012
From: nginx-forum at nginx.us (nenad)
Date: Thu, 5 Apr 2012 04:56:32 -0400 (EDT)
Subject: HttpSecureLinkModule timestamp inside uri
Message-ID: <ae368c74dcaf7f562eee8891e39d02f3.NginxMailingListEnglish@forum.nginx.org>

I would like to use HttpSecureLinkModule to restrict access to some
member photos. Now I would like to have the timestamp inside the path
but I'm not sure how to implement that.

The timestamp I could get with \/t(\d+)\/ from uri, must I do a rewrite
and if yes how? Or exist a "single" operation way?

Photo-URL:
www.example.aaa/260/7/4/3/4255/4543535/t1331894408/1234_x.jpg?a=1234567888

location ~ _x\.(jpg|jpeg|gif|png)$ {

  secure_link $arg_a,$???;
  secure_link_md5 mysecretword$uri$???;

  root /media/photos
  ...
}

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,224960,224960#msg-224960


From nginx-forum at nginx.us  Thu Apr  5 11:26:06 2012
From: nginx-forum at nginx.us (shoshomiga)
Date: Thu, 5 Apr 2012 07:26:06 -0400 (EDT)
Subject: limit_rate dynamically using $arg - security
In-Reply-To: <CAKsTx7ChGmajGvXK6UK6fnoR6kARkwFr97R7T52W8pxZKvSaoA@mail.gmail.com>
References: <CAKsTx7ChGmajGvXK6UK6fnoR6kARkwFr97R7T52W8pxZKvSaoA@mail.gmail.com>
Message-ID: <89fab4a3b8787ce2b1daf91159f681b5.NginxMailingListEnglish@forum.nginx.org>

Jonathan Matthews Wrote:
-------------------------------------------------------
> On 4 April 2012 21:40, shoshomiga
> <nginx-forum at nginx.us> wrote:
> > I've been looking for a way to limit videos to
> their bitrate to save
> > bandwidth and I've come up with this code
> >
> > ? ? ? ? ? ?if ($arg_LIMITSPEED) {
> > ? ? ? ? ? ? ?set $limit_rate
> $arg_LIMITSPEED;
> > ? ? ? ? ? ?}
> >
> > It works but I would like to know if this code
> would be secure to use on
> > a production server.
> >
> > I am not worried about users setting their
> LIMITSPEED high on their own
> > because I am limiting speeds at the network
> level as well.
> 
> To be honest, I'm not sure what definition of
> "insecure" you could be
> thinking of that *isn't* "the user can override it
> trivially" :-)
> 
> If you're doing the rate limiting at the network
> level properly, then
> why duplicate the effort? It's just one more place
> you have to change
> when you upgrade the speed limits.
> 
> Personally, I'm prototyping a streaming service at
> the moment using
> http://wiki.nginx.org/X-accel#X-Accel-Limit-Rate
> and a double
> proxy_pass (via X-Accel-Redirect to an internal
> storage proxy_pass).
> It all looks like it works nicely, and allows the
> dumb storage backend
> to throw data at the nginx router as fast as nginx
> accepts it, and for
> the first (intelligent) proxy_pass backend to
> *decide* the bitrate via
> X-Accel-Limit-Rate. I'll blog it soonish :-)
> 
> Jonathan
> -- 
> Jonathan Matthews
> London, Oxford, UK
> http://www.jpluscplusm.com/contact.html
> 
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx

By security I meant vulnerability to buffer overflows and other exploits
since limit_rate is probably not meant to recieve that kind of
unsanitized input.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,224950,224967#msg-224967


From ne at vbart.ru  Thu Apr  5 12:04:08 2012
From: ne at vbart.ru (Valentin V. Bartenev)
Date: Thu, 5 Apr 2012 16:04:08 +0400
Subject: HttpSecureLinkModule timestamp inside uri
In-Reply-To: <ae368c74dcaf7f562eee8891e39d02f3.NginxMailingListEnglish@forum.nginx.org>
References: <ae368c74dcaf7f562eee8891e39d02f3.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <201204051604.08712.ne@vbart.ru>

On Thursday 05 April 2012 12:56:32 nenad wrote:
> I would like to use HttpSecureLinkModule to restrict access to some
> member photos. Now I would like to have the timestamp inside the path
> but I'm not sure how to implement that.
> 
> The timestamp I could get with \/t(\d+)\/ from uri, must I do a rewrite
> and if yes how? Or exist a "single" operation way?
> 
> Photo-URL:
> www.example.aaa/260/7/4/3/4255/4543535/t1331894408/1234_x.jpg?a=1234567888
> 
> location ~ _x\.(jpg|jpeg|gif|png)$ {
> 
>   secure_link $arg_a,$???;
>   secure_link_md5 mysecretword$uri$???;
> 
>   root /media/photos
>   ...
> }
> 

    location ~ t(?P<t>\d+)/\d+_x\.(?:jpe?g|gif|png)$ {

        secure_link $arg_a,$t;
        secure_link_md5 mysecretword$uri$t;

        root /media/photos
        ...
    }

 wbr, Valentin V. Bartenev


From loco at d0pefish.de  Thu Apr  5 14:30:35 2012
From: loco at d0pefish.de (Peter Hinse)
Date: Thu, 05 Apr 2012 16:30:35 +0200
Subject: Combine if clauses for denial
Message-ID: <4F7DAC8B.3040304@d0pefish.de>

Hi all,

I know how to block requests with an empty user-agent or with an empty 
referrer - is it possible to combine these two criteria and deny access 
when a request has no referrer AND no user-agent?

Regards,

	Peter


From mdounin at mdounin.ru  Thu Apr  5 14:44:04 2012
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Thu, 5 Apr 2012 18:44:04 +0400
Subject: limit_rate dynamically using $arg - security
In-Reply-To: <89fab4a3b8787ce2b1daf91159f681b5.NginxMailingListEnglish@forum.nginx.org>
References: <CAKsTx7ChGmajGvXK6UK6fnoR6kARkwFr97R7T52W8pxZKvSaoA@mail.gmail.com>
	<89fab4a3b8787ce2b1daf91159f681b5.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20120405144404.GC13466@mdounin.ru>

Hello!

On Thu, Apr 05, 2012 at 07:26:06AM -0400, shoshomiga wrote:

> Jonathan Matthews Wrote:
> -------------------------------------------------------
> > On 4 April 2012 21:40, shoshomiga
> > <nginx-forum at nginx.us> wrote:
> > > I've been looking for a way to limit videos to
> > their bitrate to save
> > > bandwidth and I've come up with this code
> > >
> > > ? ? ? ? ? ?if ($arg_LIMITSPEED) {
> > > ? ? ? ? ? ? ?set $limit_rate
> > $arg_LIMITSPEED;
> > > ? ? ? ? ? ?}
> > >
> > > It works but I would like to know if this code
> > would be secure to use on
> > > a production server.

[...]

> By security I meant vulnerability to buffer overflows and other exploits
> since limit_rate is probably not meant to recieve that kind of
> unsanitized input.

It should be safe.  Note though that it will log error if there 
are invalid values passed, which may in turn be used as a DoS 
vector.

To be on safe side, I would recommend sanitizing the input, e.g. 
with map{}.  Something like this should work:

    map $arg_speed $speed {
        default 64k;
        64k     64k;
        128k    128k;
        256k    256k;
    }

    ...

    set $limit_rate $speed;

Maxim Dounin


From mdounin at mdounin.ru  Thu Apr  5 14:56:11 2012
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Thu, 5 Apr 2012 18:56:11 +0400
Subject: no response nginx
In-Reply-To: <20120405092544.f6c3c2cdd6d87bb9a320da8a@nn.iij4u.or.jp>
References: <20120405092544.f6c3c2cdd6d87bb9a320da8a@nn.iij4u.or.jp>
Message-ID: <20120405145610.GD13466@mdounin.ru>

Hello!

On Thu, Apr 05, 2012 at 09:25:44AM +0900, shin fukuda wrote:

> Hi.
> 
> I have a problem with nginx-1.1.16 uses proxy_cache_lock on. 
> When high load or many request to nginx, Suddenly nginx seems hang, 
> no response at all. 
> In netstat "SYN_RECV","CLOSE_WAIT" too many. This problem needs 
> restarting nginx, and "CLOSE_WAIT" clear.
> 
> There is no problem in front-end nginx servers, but backends one.
> 
> Any clue?

I would suggest you are hitting some system limits.  Check nginx 
and system logs, as well as various system stats.  Check the state 
of nginx worker processes, most importantly wait channel (wchan, 
"ps -alx" should show one at least on FreeBSD and Linux).

Maxim Dounin


From nginx-forum at nginx.us  Thu Apr  5 15:06:32 2012
From: nginx-forum at nginx.us (shoshomiga)
Date: Thu, 5 Apr 2012 11:06:32 -0400 (EDT)
Subject: limit_rate dynamically using $arg - security
In-Reply-To: <20120405144404.GC13466@mdounin.ru>
References: <20120405144404.GC13466@mdounin.ru>
Message-ID: <e5019e51daf113748ecd9c157d3ee778.NginxMailingListEnglish@forum.nginx.org>

Maxim Dounin Wrote:
-------------------------------------------------------
> Hello!
> 
> On Thu, Apr 05, 2012 at 07:26:06AM -0400,
> shoshomiga wrote:
> 
> > Jonathan Matthews Wrote:
> >
> --------------------------------------------------
> -----
> > > On 4 April 2012 21:40, shoshomiga
> > > <nginx-forum at nginx.us> wrote:
> > > > I've been looking for a way to limit videos
> to
> > > their bitrate to save
> > > > bandwidth and I've come up with this code
> > > >
> > > > ? ? ? ? ? ?if ($arg_LIMITSPEED) {
> > > > ? ? ? ? ? ? ?set $limit_rate
> > > $arg_LIMITSPEED;
> > > > ? ? ? ? ? ?}
> > > >
> > > > It works but I would like to know if this
> code
> > > would be secure to use on
> > > > a production server.
> 
> [...]
> 
> > By security I meant vulnerability to buffer
> overflows and other exploits
> > since limit_rate is probably not meant to
> recieve that kind of
> > unsanitized input.
> 
> It should be safe.  Note though that it will log
> error if there 
> are invalid values passed, which may in turn be
> used as a DoS 
> vector.
> 
> To be on safe side, I would recommend sanitizing
> the input, e.g. 
> with map{}.  Something like this should work:
> 
>     map $arg_speed $speed {
>         default 64k;
>         64k     64k;
>         128k    128k;
>         256k    256k;
>     }
> 
>     ...
> 
>     set $limit_rate $speed;

Is there a way to typecast to int instead?

map{} won't give me enough flexibility and it would probably be slower
than a typecast.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,224950,224974#msg-224974


From ft at falkotimme.com  Thu Apr  5 15:11:41 2012
From: ft at falkotimme.com (Falko Timme)
Date: Thu, 5 Apr 2012 17:11:41 +0200
Subject: Combine if clauses for denial
References: <4F7DAC8B.3040304@d0pefish.de>
Message-ID: <6CE56A81A763433C8B10A87FEFC97F7E@notebook>

This should give you the idea:

http://rosslawley.co.uk/2010/01/nginx-how-to-multiple-if-statements.html

----- Original Message ----- 
From: "Peter Hinse" <loco at d0pefish.de>
To: <nginx at nginx.org>
Sent: Thursday, April 05, 2012 4:30 PM
Subject: Combine if clauses for denial


> Hi all,
> 
> I know how to block requests with an empty user-agent or with an empty 
> referrer - is it possible to combine these two criteria and deny access 
> when a request has no referrer AND no user-agent?
> 
> Regards,
> 
> Peter
> 
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx


From admin at yqed.com  Thu Apr  5 15:17:20 2012
From: admin at yqed.com (Floren Munteanu)
Date: Thu, 05 Apr 2012 11:17:20 -0400
Subject: Socket leaks., pread and [crit] SSL_Write() in 1.0.14
In-Reply-To: <20120403163212.GU13466@mdounin.ru>
References: <3b3f2c2eddfdd3e59d242463275e502e.NginxMailingListEnglish@forum.nginx.org>
	<00cd479fdbb8bdf2f9f844e88607f8bd.NginxMailingListEnglish@forum.nginx.org>
	<20120326164754.GL13466@mdounin.ru> <jl813b$nn0$1@dough.gmane.org>
	<20120402111754.GD13466@mdounin.ru> <jlcjag$a19$1@dough.gmane.org>
	<20120402170637.GJ13466@mdounin.ru> <jldij1$k02$1@dough.gmane.org>
	<20120403163212.GU13466@mdounin.ru>
Message-ID: <jlkd1v$thv$1@dough.gmane.org>

Hi Maxim,

On 4/3/2012 12:32 PM, Maxim Dounin wrote:
> I'm afraid we aren't going to fix 1.0.x anyway.  The 1.1.x branch
> will become stable in serveral weeks, and 1.0.x is expected to get
> only security fixes.

Thanks, I will install a debug version of 1.1.18 into live server.

> For now, please capture debug log.  Anything else (including gdb
> dump of the request in question) isn't that important, but if you
> are going to obtain one it might be good idea to recompile nginx
> without -O3 as it might produce undesired side-effects in gdb
> output.  The -O used by default is good enough.

I'm afraid I will have to keep the -O3 for sake of compile consistency 
with other RPM's. I will do a debug capture on latest devel version, the 
[crit] errors still occur with 1.1.18 version and OpenSSL 1.0.1.

Regards,

Floren


From loco at d0pefish.de  Thu Apr  5 15:21:30 2012
From: loco at d0pefish.de (Peter Hinse)
Date: Thu, 05 Apr 2012 17:21:30 +0200
Subject: Combine if clauses for denial
In-Reply-To: <6CE56A81A763433C8B10A87FEFC97F7E@notebook>
References: <4F7DAC8B.3040304@d0pefish.de>
	<6CE56A81A763433C8B10A87FEFC97F7E@notebook>
Message-ID: <4F7DB87A.6010604@d0pefish.de>

Am 05.04.2012 17:11, schrieb Falko Timme:

> http://rosslawley.co.uk/2010/01/nginx-how-to-multiple-if-statements.html

Thanks, that will do it I guess.


From amoiz.shine at gmail.com  Thu Apr  5 15:26:37 2012
From: amoiz.shine at gmail.com (Sharl.Jimh.Tsin)
Date: Thu, 5 Apr 2012 23:26:37 +0800
Subject: [ANNOUNCE] Tengine-1.2.4
In-Reply-To: <CACPpB2xrZfQqtwD=h2HSjMjp4BJmwVr9_z5=mBC_75EQYZ+VjQ@mail.gmail.com>
References: <CAOKiq2CbQUtcdwMa+y+RgDEQTSVaOJEooRdOEBYBSxYtsqxevg@mail.gmail.com>
	<CANXGsjzCm_AK=JtGYowQ17QiMfzDckRTxShH9QOn7JxCWN9rzg@mail.gmail.com>
	<CACPpB2xrZfQqtwD=h2HSjMjp4BJmwVr9_z5=mBC_75EQYZ+VjQ@mail.gmail.com>
Message-ID: <CANXGsjzLS3RXRV3__Y8-L_fUfxxhoVZh64EL2D6icy6CPZFkCw@mail.gmail.com>

from your profile in LinkedIn website,i found that you are a student
of HuaiYin Normal University.right?

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.



? 2012?4?1? ??9:58?smallfish <smallfish.xy at gmail.com> ???
> hah, perhaps just a joke. but?nothing is impossable. :)
> --
> blog: http://chenxiaoyu.org
>
>
>
> ? 2012?4?1? ??9:33?Sharl.Jimh.Tsin <amoiz.shine at gmail.com>???
>>
>> are you sure that NOT a joke of April Fools' Day, like taolinux
>> (http://code.taobao.org/p/taobao_linux_server/src/trunk/taolinux/) ?
>>
>> 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.
>>
>>
>>
>> 2012/3/31 Joshua Zhu <zhuzhaoyuan at gmail.com>:
>> > Hi folks!
>> >
>> > We're pleased to announce that Tengine-1.2.4 has been released. You can
>> > either download the tar ball
>> > from?http://tengine.taobao.org/download/tengine-1.2.4.tar.gz ,?or
>> > checkout
>> > the code from https://github.com/taobao/tengine
>> >
>> > As this release, Tengine is based on Nginx-1.0.14. The highlight is a
>> > brand
>> > new 'user_agent' module which would be a good replacement for the
>> > 'browser'
>> > module in standard Nginx.?It's more efficient and flexible?than the
>> > 'browser' module. Other changes include:
>> > * Feature: added the 'log_escape' directive.
>> > * Bugfix: fixed a bug in the limit_req module.
>> > * Bugfix: fixed a bug in subrequest.
>> >
>> > Regards,
>> >
>> > --
>> > Joshua Zhu
>> > Senior Software Engineer
>> > Server Platforms Team at Taobao
>> >
>> > _______________________________________________
>> > 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 jdorfman at netdna.com  Thu Apr  5 18:55:43 2012
From: jdorfman at netdna.com (Justin Dorfman)
Date: Thu, 5 Apr 2012 11:55:43 -0700
Subject: nginx conf
In-Reply-To: <CAD8siQ5+Guommd9RaUPorN2o8Ho9Q9ny_d2_LxerUivoEfmgvw@mail.gmail.com>
References: <CAD8siQ5+Guommd9RaUPorN2o8Ho9Q9ny_d2_LxerUivoEfmgvw@mail.gmail.com>
Message-ID: <CACDr0vHKpAx6+U+ov6ukWad5VYK7_XZMqQntJe9BP75iUoKHqA@mail.gmail.com>

Hello,

What are the permissions on /etc/nginx/sites-enabled/?

You might need to give nginx rwx rights.

Regards,

Justin Dorfman <http://www.twitter.com/jdorfman>

NetDNA <http://www.netdna.com>?
The Science of Acceleration?

Email / gtalk: jdorfman at netdna.com
P: 1.323.301.1400 x 272
Skype: netdna-justind
Twitter: @jdorfman <http://www.twitter.com/jdorfman>

www.NetDNA.com <http://www.netdna.com/> | www.MaxCDN.com<http://www.maxcdn.com/>

@NetDNA <http://twitter.com/NetDNA> | @MaxCDN <http://twitter.com/MaxCDN>



On Wed, Apr 4, 2012 at 9:18 PM, aris <arissirajawali at gmail.com> wrote:

> Hi all,
> how to configure nginx.conf, sites-available and sites-enabled
> I get 403 Forbidden nginx/0.7.67
> please check out where my configuration errors
>
> *I nginx.conf following:*
>
> user www-data;
> worker_processes  1;
>
> error_log  /var/log/nginx/error.log;
> pid        /var/run/nginx.pid;
>
> events {
>     worker_connections  1024;
>     # multi_accept on;
> }
>
> http {
>     include       /etc/nginx/mime.types;
>
>     access_log  /var/log/nginx/access.log;
>
>     sendfile        on;
>     #tcp_nopush     on;
>
>     #keepalive_timeout  0;
>     keepalive_timeout  65;
>     tcp_nodelay        on;
>
>     gzip  on;
>     gzip_disable "MSIE [1-6]\.(?!.*SV1)";
>
>     include /etc/nginx/conf.d/*.conf;
>     include /etc/nginx/sites-enabled/*;
> }
>
> # mail {
> #     # See sample authentication script at:
> #     # http://wiki.nginx.org/NginxImapAuthenticateWithApachePhpScript
> #
> #     # auth_http localhost/auth.php;
> #     # pop3_capabilities "TOP" "USER";
> #     # imap_capabilities "IMAP4rev1" "UIDPLUS";
> #
> #     server {
> #         listen     localhost:110;
> #         protocol   pop3;
> #         proxy      on;
> #     }
> #
> #     server {
> #         listen     localhost:143;
> #         protocol   imap;
> #         proxy      on;
> #     }
> # }
>
>
> *The following sites-available I have :*
> # You may add here your
> # server {
> #       ...
> # }
> # statements for each of your virtual hosts
>
> server {
>
>         listen   80; ## listen for ipv4
>         server_name  localhost;
>         access_log  /var/log/nginx/localhost.access.log;
>
>         location / {
>                 root   /var/www;
>                 index  index.php;
>         }
>
>         location /doc {
>                 root   /usr/share;
>                 autoindex on;
>                 allow 127.0.0.1;
>                 deny all;
>         }
>
>         location /images {
>                 root   /usr/share;
>                 autoindex on;
>         }
>
>
>
>         #error_page  404  /404.html;
>
>         # redirect server error pages to the static page /50x.html
>         #
>         #error_page   500 502 503 504  /50x.html;
>         #location = /50x.html {
>         #       root   /var/www/nginx-default;
>         #}
>
>         # proxy the PHP scripts to Apache listening on 127.0.0.1:80
>         #
>         #location ~ \.php$ {
>                 #proxy_pass   http://127.0.0.1;
>         #}
>
>         # pass the PHP scripts to FastCGI server listening on
> 127.0.0.1:9000
>         #
>         #location ~ \.php$ {
>                 #fastcgi_pass   127.0.0.1:9000;
>                 #fastcgi_index  index.php;
>                 #fastcgi_param  SCRIPT_FILENAME
>  /scripts$fastcgi_script_name;
>                 #includefastcgi_params;
>        #}
>
>         # deny access to .htaccess files, if Apache's document root
>         # concurs with nginx's one
>         #
>         #location ~ /\.ht {
>                 #deny  all;
>         #}
> }
>
>
> # another virtual host using mix of IP-, name-, and port-based
> configuration
> #
> #server {
> #listen   8000;
> #listen   somename:8080;
> #server_name  somename  alias  another.alias;
>
> #location / {
> #root   html;
> #index  index.html index.htm;
> #}
> #}
>
>
> # HTTPS server
> #
> #server {
> #listen   443;
> #server_name  localhost;
>
> #ssl  on;
> #ssl_certificate  cert.pem;
> #ssl_certificate_key  cert.key;
>
> #ssl_session_timeout  5m;
>
> #ssl_protocols  SSLv3 TLSv1;
> #ssl_ciphers  ALL:!ADH:!EXPORT56:RC4+RSA:+HIGH:+MEDIUM:+LOW:+SSLv3:+EXP;
> #ssl_prefer_server_ciphers   on;
>
> #location / {
> #root   html;
> #index  index.html index.htm;
> #}
> #}
>
> *The following sites-enabled I have :*
> # You may add here your
> # server {
> #       ...
> # }
> # statements for each of your virtual hosts
>
> server {
>
>         listen   80; ## listen for ipv4
>         server_name  localhost;
>         access_log  /var/log/nginx/localhost.access.log;
>
>         location / {
>                 root   /var/www;
>                 index  index.php;
>         }
>
>         location /doc {
>                 root   /usr/share;
>                 autoindex on;
>                 allow 127.0.0.1;
>                 deny all;
>         }
>
>         location /images {
>                 root   /usr/share;
>                 autoindex on;
>         }
>
>
>
>         #error_page  404  /404.html;
>
>         # redirect server error pages to the static page /50x.html
>         #
>         #error_page   500 502 503 504  /50x.html;
>         #location = /50x.html {
>         #       root   /var/www/nginx-default;
>         #}
>
>         # proxy the PHP scripts to Apache listening on 127.0.0.1:80
>         #
>         #location ~ \.php$ {
>                 #proxy_pass   http://127.0.0.1;
>         #}
>
>         # pass the PHP scripts to FastCGI server listening on
> 127.0.0.1:9000
>         #
>         #location ~ \.php$ {
>                 #fastcgi_pass   127.0.0.1:9000;
>                 #fastcgi_index  index.php;
>                 #fastcgi_param  SCRIPT_FILENAME
>  /scripts$fastcgi_script_name;
>                 #includefastcgi_params;
>                #fastcgi_param  SCRIPT_FILENAME
>  /scripts$fastcgi_script_name;
>                 #includefastcgi_params;
>         #}
>
>         # deny access to .htaccess files, if Apache's document root
>         # concurs with nginx's one
>         #
>         #location ~ /\.ht {
>                 #deny  all;
>         #}
> }
>
>
> # another virtual host using mix of IP-, name-, and port-based
> configuration
> #
> #server {
> #listen   8000;
> #listen   somename:8080;
> #server_name  somename  alias  another.alias;
>
> #location / {
> #root   html;
> #index  index.html index.htm;
> #}
> #}
>
>
> # HTTPS server
> #
> #server {
> #listen   443;
> #server_name  localhost;
>
> #ssl  on;
> #ssl_certificate  cert.pem;
> #ssl_certificate_key  cert.key;
>
> #ssl_session_timeout  5m;
>
> #ssl_protocols  SSLv3 TLSv1;
> #ssl_ciphers  ALL:!ADH:!EXPORT56:RC4+RSA:+HIGH:+MEDIUM:+LOW:+SSLv3:+EXP;
> #ssl_prefer_server_ciphers   on;
>
> #location / {
> #root   html;
> #index  index.html index.htm;
> #}
> #}
>
>
> _______________________________________________
> 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/20120405/fa17f1a1/attachment-0001.html>

From ashishs.dev at gmail.com  Thu Apr  5 23:33:31 2012
From: ashishs.dev at gmail.com (Ashish S)
Date: Thu, 5 Apr 2012 16:33:31 -0700
Subject: multiget support in memcache/memc nginx modules
Message-ID: <CA+BpfUra9A3jy3rskKNu+D0scGJVoGRiuhvg=j1uiymsM+8r1g@mail.gmail.com>

Hi,

It looks like neither the nginx memc module, nor the memcache standard
module supports multi-get out of the box.  Is there an implementation
openly available, that supports the same?
How about support for larger keys? (If implementation is based on text
protocol, key-length seems limited to 250 chars vs 65K in binary
protocol)

Thanks,
Ashish


From albert at nn.iij4u.or.jp  Fri Apr  6 01:47:03 2012
From: albert at nn.iij4u.or.jp (shin fukuda)
Date: Fri, 6 Apr 2012 10:47:03 +0900
Subject: no response nginx
In-Reply-To: <20120405145610.GD13466@mdounin.ru>
References: <20120405092544.f6c3c2cdd6d87bb9a320da8a@nn.iij4u.or.jp>
	<20120405145610.GD13466@mdounin.ru>
Message-ID: <20120406104703.8d8a5d9d04229d835c791b0b@nn.iij4u.or.jp>

Hello, Maxim.

On Thu, 5 Apr 2012 18:56:11 +0400
Maxim Dounin <mdounin at mdounin.ru> wrote:

> Hello!
> 
> On Thu, Apr 05, 2012 at 09:25:44AM +0900, shin fukuda wrote:
> 
> > Hi.
> > 
> > I have a problem with nginx-1.1.16 uses proxy_cache_lock on. 
> > When high load or many request to nginx, Suddenly nginx seems hang, 
> > no response at all. 
> > In netstat "SYN_RECV","CLOSE_WAIT" too many. This problem needs 
> > restarting nginx, and "CLOSE_WAIT" clear.
> > 
> > There is no problem in front-end nginx servers, but backends one.
> > 
> > Any clue?
> 
> I would suggest you are hitting some system limits.  Check nginx 
> and system logs, as well as various system stats.  Check the state 
> of nginx worker processes, most importantly wait channel (wchan, 
> "ps -alx" should show one at least on FreeBSD and Linux).

Set kernel parameter like 

 ...
 net.ipv4.tcp_syncookies = 1
 net.ipv4.tcp_tw_reuse = 1
 net.ipv4.tcp_max_syn_backlog = 8192
 net.ipv4.tcp_syncookies = 1
 net.ipv4.tcp_tw_recycle = 1
 net.ipv4.tcp_fin_timeout = 15
 net.ipv4.tcp_keepalive_time = 300

But, no good result so far.

Anyway, I check workers details, and find this,

 15832 write(4, "2012/04/04 19:56:10 [info] 15832#0: sigtimedwait() 
 failed (4: Interrupted system"..., 87) = 87
 15832 rt_sigtimedwait([ALRM IO RT_12 RT_13], 0x691220) = -1 EAGAIN 
 (Resource temporarily unavailable)
 15832 rt_sigtimedwait([ALRM IO RT_12 RT_13], 0x691220) = -1 EAGAIN 
 (Resource temporarily unavailable)
  ....
  ....

Is this normal status ?


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


-- 
shin fukuda <albert at nn.iij4u.or.jp>


From lists at ruby-forum.com  Fri Apr  6 01:48:30 2012
From: lists at ruby-forum.com (MOHAMMED A.)
Date: Fri, 06 Apr 2012 03:48:30 +0200
Subject: I need the best nginx configuration please
Message-ID: <4f38466f869baed03ed613553a19acf0@ruby-forum.com>

Hello,

Please, I need the best nginx configuration for my servers
My servers for images hosting
In any server, From 2 to 5 million images
The problem is that images do not appear quickly, a long time, Slow
Please help me in this, I need the best nginx configuration

Thanks.

The configuration now

user  nobody;
# no need for more workers in the proxy mode
worker_processes  2;
error_log  /var/log/nginx/error.log info;
worker_rlimit_nofile 20480;
events {
 worker_connections 5120; # increase for busier servers
 use epoll; # you should use epoll here for Linux kernels 2.6.x
}
http {
 upload_progress_header id;
 upload_progress uploads 5m;
 server_name_in_redirect off;
 server_names_hash_max_size 10240;
 server_names_hash_bucket_size 1024;
 include    mime.types;
 default_type  application/octet-stream;
 server_tokens off;
 sendfile on;
 tcp_nopush on;
 tcp_nodelay on;
 keepalive_timeout  5;
 gzip on;
 gzip_vary on;
 gzip_disable "MSIE [1-6]\.";
 gzip_proxied any;
 gzip_http_version 1.1;
 gzip_min_length  1000;
 gzip_comp_level  6;
 gzip_buffers  16 8k;
# You can remove image/png image/x-icon image/gif image/jpeg if you have
slow CPU
 gzip_types    text/plain text/xml text/css application/x-javascript
application/xml image/png image/x-icon image/gif image/jpeg
application/xml+rss text/javascript application/atom+xml;
 ignore_invalid_headers on;
 client_header_timeout  360m;
 client_body_timeout 360m;
 send_timeout     360m;
 reset_timedout_connection on;
 connection_pool_size  256;
 client_header_buffer_size 256k;
 large_client_header_buffers 4 256k;
 client_max_body_size 2048M;
 client_body_buffer_size 1024k;
 request_pool_size  32k;
 output_buffers   4 32k;
 postpone_output  1460;
 proxy_temp_path  /home/matarNginx/nginx_proxy/;
 client_body_in_file_only on;
 log_format bytes_log "$msec $bytes_sent .";
 include "/etc/nginx/vhosts/*";
}

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


From nginx-forum at nginx.us  Fri Apr  6 09:15:23 2012
From: nginx-forum at nginx.us (haonhien)
Date: Fri, 6 Apr 2012 05:15:23 -0400 (EDT)
Subject: Problem add http_400 for module upstream
Message-ID: <d0802d688ca502e0a1436e7bb235c396.NginxMailingListEnglish@forum.nginx.org>

Hello every body!

I want to add http_400 in directive proxy_next_upstream of module
upstream:

1. Add code below in file src/http/ngx_http_upstream.h:
#define NGX_HTTP_UPSTREAM_FT_HTTP_502        0x00000020
#define NGX_HTTP_UPSTREAM_FT_HTTP_503        0x00000040
#define NGX_HTTP_UPSTREAM_FT_HTTP_504        0x00000080
#define NGX_HTTP_UPSTREAM_FT_HTTP_404        0x00000100

#define NGX_HTTP_UPSTREAM_FT_HTTP_400        0x00000120 //---> not
right

2.Add code below in file ngx_http_upstream.c:

   { 500, NGX_HTTP_UPSTREAM_FT_HTTP_500 },
   { 502, NGX_HTTP_UPSTREAM_FT_HTTP_502 },
   { 503, NGX_HTTP_UPSTREAM_FT_HTTP_503 },
   { 504, NGX_HTTP_UPSTREAM_FT_HTTP_504 },
   { 404, NGX_HTTP_UPSTREAM_FT_HTTP_404 },

   { 400, NGX_HTTP_UPSTREAM_FT_HTTP_400 }, //add code here

3.Add code below in file src/http/modules/ngx_http_proxy_module.c:

    { ngx_string("http_500"), NGX_HTTP_UPSTREAM_FT_HTTP_500 },
    { ngx_string("http_502"), NGX_HTTP_UPSTREAM_FT_HTTP_502 },
    { ngx_string("http_503"), NGX_HTTP_UPSTREAM_FT_HTTP_503 },
    { ngx_string("http_504"), NGX_HTTP_UPSTREAM_FT_HTTP_504 },
    { ngx_string("http_404"), NGX_HTTP_UPSTREAM_FT_HTTP_404 },

    { ngx_string("http_400"), NGX_HTTP_UPSTREAM_FT_HTTP_400 }, // add
code here

Plz help me 
Thank you so much.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225010,225010#msg-225010


From c.kworr at gmail.com  Fri Apr  6 08:06:54 2012
From: c.kworr at gmail.com (Volodymyr Kostyrko)
Date: Fri, 06 Apr 2012 11:06:54 +0300
Subject: I need the best nginx configuration please
In-Reply-To: <4f38466f869baed03ed613553a19acf0@ruby-forum.com>
References: <4f38466f869baed03ed613553a19acf0@ruby-forum.com>
Message-ID: <4F7EA41E.3090504@gmail.com>



MOHAMMED A. wrote:
> Hello,
>
> Please, I need the best nginx configuration for my servers
> My servers for images hosting
> In any server, From 2 to 5 million images
> The problem is that images do not appear quickly, a long time, Slow
> Please help me in this, I need the best nginx configuration
>
> Thanks.
>
> The configuration now
>
> user  nobody;
> # no need for more workers in the proxy mode
> worker_processes  2;
> error_log  /var/log/nginx/error.log info;
> worker_rlimit_nofile 20480;
> events {
>   worker_connections 5120; # increase for busier servers
>   use epoll; # you should use epoll here for Linux kernels 2.6.x
> }
> http {
>   upload_progress_header id;
>   upload_progress uploads 5m;
>   server_name_in_redirect off;
>   server_names_hash_max_size 10240;
>   server_names_hash_bucket_size 1024;
>   include    mime.types;
>   default_type  application/octet-stream;
>   server_tokens off;
>   sendfile on;
>   tcp_nopush on;
>   tcp_nodelay on;
>   keepalive_timeout  5;
>   gzip on;
>   gzip_vary on;
>   gzip_disable "MSIE [1-6]\.";
>   gzip_proxied any;
>   gzip_http_version 1.1;
>   gzip_min_length  1000;
>   gzip_comp_level  6;
>   gzip_buffers  16 8k;
> # You can remove image/png image/x-icon image/gif image/jpeg if you have
> slow CPU
>   gzip_types    text/plain text/xml text/css application/x-javascript
> application/xml image/png image/x-icon image/gif image/jpeg
> application/xml+rss text/javascript application/atom+xml;
>   ignore_invalid_headers on;
>   client_header_timeout  360m;
>   client_body_timeout 360m;
>   send_timeout     360m;
>   reset_timedout_connection on;
>   connection_pool_size  256;
>   client_header_buffer_size 256k;
>   large_client_header_buffers 4 256k;
>   client_max_body_size 2048M;
>   client_body_buffer_size 1024k;
>   request_pool_size  32k;
>   output_buffers   4 32k;
>   postpone_output  1460;
>   proxy_temp_path  /home/matarNginx/nginx_proxy/;
>   client_body_in_file_only on;
>   log_format bytes_log "$msec $bytes_sent .";

Everything looks good. Some options point to default values and can be 
omitted.

>   include "/etc/nginx/vhosts/*";
> }

This line includes all files under listed directory. Whole your config 
is only about nginx configuration and lack particular site 
configuration. Maybe posting those would give anyone better clue on 
whats going on with your system.

-- 
Sphinx of black quartz judge my vow.


From pobreiluso at gmail.com  Fri Apr  6 10:48:18 2012
From: pobreiluso at gmail.com (=?ISO-8859-1?Q?Antonio_Jerez_Guill=E9n?=)
Date: Fri, 6 Apr 2012 12:48:18 +0200
Subject: I need the best nginx configuration please
In-Reply-To: <4F7EA41E.3090504@gmail.com>
References: <4f38466f869baed03ed613553a19acf0@ruby-forum.com>
	<4F7EA41E.3090504@gmail.com>
Message-ID: <CAKNnf-XVC6H3XfZ6ExdGmSj17Tn9VZy-rXP9k_2OZpU+X1njkw@mail.gmail.com>

Hi there!

I guess, that your problem could be O.S. related and not Nginx related.
I meant, probably, you are using more files per directory than your O.S.
could support, in many cases, this limit is about 32.000 files per
directory, but it may depends of your filesystem.
take a look to the link below:
http://stackoverflow.com/questions/466521/how-many-files-in-a-directory-is-too-many

In order to solve this issue, you must add a two or three directory levels
structure. For example:

I guess your current directory structure is something like:
/img/abcdefghijklmn?opqrstuvwxyz.jpg

And you must try something like:
/img/*ab/cd/ef/*abcdefghijklmn?opqrstuvwxyz.jpg
Or also works for somebody, to add a modulus function to the directories,
for example:

/img/*01/02/*abcdefghijklmn?opqrstuvwxyz.jpg

Hope it helps.


El 6 de abril de 2012 10:06, Volodymyr Kostyrko <c.kworr at gmail.com>escribi?:

>
>
> MOHAMMED A. wrote:
>
>> Hello,
>>
>> Please, I need the best nginx configuration for my servers
>> My servers for images hosting
>> In any server, From 2 to 5 million images
>> The problem is that images do not appear quickly, a long time, Slow
>> Please help me in this, I need the best nginx configuration
>>
>> Thanks.
>>
>> The configuration now
>>
>> user  nobody;
>> # no need for more workers in the proxy mode
>> worker_processes  2;
>> error_log  /var/log/nginx/error.log info;
>> worker_rlimit_nofile 20480;
>> events {
>>  worker_connections 5120; # increase for busier servers
>>  use epoll; # you should use epoll here for Linux kernels 2.6.x
>> }
>> http {
>>  upload_progress_header id;
>>  upload_progress uploads 5m;
>>  server_name_in_redirect off;
>>  server_names_hash_max_size 10240;
>>  server_names_hash_bucket_size 1024;
>>  include    mime.types;
>>  default_type  application/octet-stream;
>>  server_tokens off;
>>  sendfile on;
>>  tcp_nopush on;
>>  tcp_nodelay on;
>>  keepalive_timeout  5;
>>  gzip on;
>>  gzip_vary on;
>>  gzip_disable "MSIE [1-6]\.";
>>  gzip_proxied any;
>>  gzip_http_version 1.1;
>>  gzip_min_length  1000;
>>  gzip_comp_level  6;
>>  gzip_buffers  16 8k;
>> # You can remove image/png image/x-icon image/gif image/jpeg if you have
>> slow CPU
>>  gzip_types    text/plain text/xml text/css application/x-javascript
>> application/xml image/png image/x-icon image/gif image/jpeg
>> application/xml+rss text/javascript application/atom+xml;
>>  ignore_invalid_headers on;
>>  client_header_timeout  360m;
>>  client_body_timeout 360m;
>>  send_timeout     360m;
>>  reset_timedout_connection on;
>>  connection_pool_size  256;
>>  client_header_buffer_size 256k;
>>  large_client_header_buffers 4 256k;
>>  client_max_body_size 2048M;
>>  client_body_buffer_size 1024k;
>>  request_pool_size  32k;
>>  output_buffers   4 32k;
>>  postpone_output  1460;
>>  proxy_temp_path  /home/matarNginx/nginx_proxy/;
>>  client_body_in_file_only on;
>>  log_format bytes_log "$msec $bytes_sent .";
>>
>
> Everything looks good. Some options point to default values and can be
> omitted.
>
>   include "/etc/nginx/vhosts/*";
>> }
>>
>
> This line includes all files under listed directory. Whole your config is
> only about nginx configuration and lack particular site configuration.
> Maybe posting those would give anyone better clue on whats going on with
> your system.
>
> --
> Sphinx of black quartz judge my vow.
>
>
> ______________________________**_________________
> 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/20120406/635b3433/attachment.html>

From nginx-forum at nginx.us  Fri Apr  6 12:32:16 2012
From: nginx-forum at nginx.us (shrinath.m)
Date: Fri, 6 Apr 2012 08:32:16 -0400 (EDT)
Subject: proxy_buffering on - doesn't return files with large data?
Message-ID: <8e10eb48ce3782d66f10e4ae8386362e.NginxMailingListEnglish@forum.nginx.org>

Hi, 

We have setup like so : 

1. Nginx is configured as server which recieves user request
2. The user request is proxy_passed to another server which is also
Nginx. 

Now when user requests for a file(401481 bytes) over AJAX, it is not
being served. This problem doesn't happen for every file, just for
random files. The same file doesn't load every time. 

I tried disabling proxy_buffering off, it worked. 
BTW, it is not the file size problem. It is working for larger file than
the one that is not loading. I would love to know why. 


Thanks, 
Shrinath

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225015,225015#msg-225015


From bruno.premont at restena.lu  Fri Apr  6 13:24:52 2012
From: bruno.premont at restena.lu (Bruno =?UTF-8?B?UHLDqW1vbnQ=?=)
Date: Fri, 6 Apr 2012 15:24:52 +0200
Subject: limit_req_zone and limit_req: rate too strict
Message-ID: <20120406152452.5729f6e0@pluto.restena.lu>

Hi,

I have started using limit_req_zone + limit_req but am seeing more delayed
log entries than I would expect.

I have the following:

  limit_req_zone $binary_remote_addr zone=one:10m rate=3r/s;
  ...
  server {
    ...
    location ~ .*\.php$ {
        limit_req zone=one burst=10;
        ...
    }


In the logs I see that when 2 requests happen near simultaneously from the
same IP address (e.g. page redirect to second PHP page or single
PHP-generated image referenced by page) the second and third request get
delayed if they arrive too early during the same second.


I would wish to limit the user to 3 requests within a 1s window though
not setting any constraint as to spacing between those requests, how can I
accomplish it or what changes to limit_req module would be needed to
not have a minimum 0.33s spacing between requests for above configuration?

Thanks,
Bruno


From nginx-forum at nginx.us  Fri Apr  6 13:25:53 2012
From: nginx-forum at nginx.us (nenad)
Date: Fri, 6 Apr 2012 09:25:53 -0400 (EDT)
Subject: HttpSecureLinkModule timestamp inside uri
In-Reply-To: <ae368c74dcaf7f562eee8891e39d02f3.NginxMailingListEnglish@forum.nginx.org>
References: <ae368c74dcaf7f562eee8891e39d02f3.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <5086ab5410a967ea6017e179cd4884b6.NginxMailingListEnglish@forum.nginx.org>

thank you, that worked!

can you also say me how i can point to correct root without /t{time},
now it try to open the file in a folder with timestamp in. 

www.example.aaa/260/7/4/3/4255/4543535/t1331894408/1234_x.jpg?a=1234567888

means it try to open it in:
/media/photos/260/7/4/3/4255/4543535/t1331894408/1234_x.jpg instead
/srv/media/260/7/4/3/4255/4543535/1234_x.jpg

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,224960,225018#msg-225018


From nginx-forum at nginx.us  Fri Apr  6 13:41:59 2012
From: nginx-forum at nginx.us (zuckbin)
Date: Fri, 6 Apr 2012 09:41:59 -0400 (EDT)
Subject: serve static content on static domain
Message-ID: <d8832eaa542bb1f75db0d4b6749833d2.NginxMailingListEnglish@forum.nginx.org>

Hi,

How to serve static content on static domain cookieless.

I use ngix to serve static content then apache to dynamic content.

Want to serve on static.mydomain.com all static content: jpeg, css, js,
...

Here my actual conf:

	server {
		listen 80;
		client_max_body_size 50M;
		server_name www.mydoamin.com;
		access_log /var/log/nginx/localhost.access_log main;
		error_log /var/log/nginx/localhost.error_log info;
	location / {
            proxy_pass http://IP:8080/;
		proxy_redirect default;
	proxy_set_header Host $host;
	proxy_set_header X-Real-IP $remote_addr;
	proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;

	}
	location ~*
^.+.(jpg|jpeg|gif|png|ico|css|zip|tgz|gz|rar|bz2|doc|xls|exe|pdf|ppt|txt|tar|mid|midi|wav|bmp|rtf|js)$
{
            root   /home/www/domain.com;
			expires 30d;
			add_header Pragma public;
			add_header Cache-Control "public";
        }
	}



Want all my static content will be delivered only with:
static.domain.com
(no duplicate content)

Thanks for your help
Bye

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225020,225020#msg-225020


From edho at myconan.net  Fri Apr  6 13:46:26 2012
From: edho at myconan.net (Edho Arief)
Date: Fri, 6 Apr 2012 20:46:26 +0700
Subject: serve static content on static domain
In-Reply-To: <d8832eaa542bb1f75db0d4b6749833d2.NginxMailingListEnglish@forum.nginx.org>
References: <d8832eaa542bb1f75db0d4b6749833d2.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <CAAtReCkvD6-mwj5--o9hzSeTmcRUgcgVMJoz2AMa-5ix80iqwA@mail.gmail.com>

2012/4/6 zuckbin <nginx-forum at nginx.us>:
> Hi,
>
> How to serve static content on static domain cookieless.
>
> I use ngix to serve static content then apache to dynamic content.
>
> Want to serve on static.mydomain.com all static content: jpeg, css, js,
> ...
>

Then you should tell your app to link static contents at
static.mydomain.com instead of mydomain.com and then add its server
block.


From ne at vbart.ru  Fri Apr  6 16:03:19 2012
From: ne at vbart.ru (Valentin V. Bartenev)
Date: Fri, 6 Apr 2012 20:03:19 +0400
Subject: HttpSecureLinkModule timestamp inside uri
In-Reply-To: <5086ab5410a967ea6017e179cd4884b6.NginxMailingListEnglish@forum.nginx.org>
References: <ae368c74dcaf7f562eee8891e39d02f3.NginxMailingListEnglish@forum.nginx.org>
	<5086ab5410a967ea6017e179cd4884b6.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <201204062003.19924.ne@vbart.ru>

On Friday 06 April 2012 17:25:53 nenad wrote:
> thank you, that worked!
> 
> can you also say me how i can point to correct root without /t{time},
> now it try to open the file in a folder with timestamp in.
> 
> www.example.aaa/260/7/4/3/4255/4543535/t1331894408/1234_x.jpg?a=1234567888
> 
> means it try to open it in:
> /media/photos/260/7/4/3/4255/4543535/t1331894408/1234_x.jpg instead
> /srv/media/260/7/4/3/4255/4543535/1234_x.jpg
> 

http://nginx.org/r/alias


    location ~ ^(.+)/t(?P<t>\d+)(/\d+_x\.(?:jpe?g|gif|png))$ {

        secure_link $arg_a,$t;
        secure_link_md5 mysecretword$uri$t;

        alias /srv/media$1$3;
        ...
    }


 wbr, Valentin V. Bartenev


From ne at vbart.ru  Fri Apr  6 16:06:36 2012
From: ne at vbart.ru (Valentin V. Bartenev)
Date: Fri, 6 Apr 2012 20:06:36 +0400
Subject: limit_req_zone and limit_req: rate too strict
In-Reply-To: <20120406152452.5729f6e0@pluto.restena.lu>
References: <20120406152452.5729f6e0@pluto.restena.lu>
Message-ID: <201204062006.36424.ne@vbart.ru>

On Friday 06 April 2012 17:24:52 Bruno Pr?mont wrote:
> Hi,
> 
> I have started using limit_req_zone + limit_req but am seeing more delayed
> log entries than I would expect.
> 
> I have the following:
> 
>   limit_req_zone $binary_remote_addr zone=one:10m rate=3r/s;
>   ...
>   server {
>     ...
>     location ~ .*\.php$ {
>         limit_req zone=one burst=10;
>         ...
>     }
> 
> 
> In the logs I see that when 2 requests happen near simultaneously from the
> same IP address (e.g. page redirect to second PHP page or single
> PHP-generated image referenced by page) the second and third request get
> delayed if they arrive too early during the same second.
> 
> 
> I would wish to limit the user to 3 requests within a 1s window though
> not setting any constraint as to spacing between those requests, how can I
> accomplish it or what changes to limit_req module would be needed to
> not have a minimum 0.33s spacing between requests for above configuration?
> 

http://nginx.org/r/limit_req

    limit_req zone=one burst=10 nodelay;


 wbr, Valentin V. Bartenev


From mdounin at mdounin.ru  Fri Apr  6 16:35:25 2012
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Fri, 6 Apr 2012 20:35:25 +0400
Subject: proxy_buffering on - doesn't return files with large data?
In-Reply-To: <8e10eb48ce3782d66f10e4ae8386362e.NginxMailingListEnglish@forum.nginx.org>
References: <8e10eb48ce3782d66f10e4ae8386362e.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20120406163525.GK13466@mdounin.ru>

Hello!

On Fri, Apr 06, 2012 at 08:32:16AM -0400, shrinath.m wrote:

> Hi, 
> 
> We have setup like so : 
> 
> 1. Nginx is configured as server which recieves user request
> 2. The user request is proxy_passed to another server which is also
> Nginx. 
> 
> Now when user requests for a file(401481 bytes) over AJAX, it is not
> being served. This problem doesn't happen for every file, just for
> random files. The same file doesn't load every time. 
> 
> I tried disabling proxy_buffering off, it worked. 
> BTW, it is not the file size problem. It is working for larger file than
> the one that is not loading. I would love to know why. 

First of all try looking into error logs.  If it doesn't help, 
please provide debug log, see here for details:

http://wiki.nginx.org/Debugging

Maxim Dounin


From bruno.premont at restena.lu  Fri Apr  6 19:38:59 2012
From: bruno.premont at restena.lu (Bruno =?UTF-8?B?UHLDqW1vbnQ=?=)
Date: Fri, 6 Apr 2012 21:38:59 +0200
Subject: limit_req_zone and limit_req: rate too strict
In-Reply-To: <201204062006.36424.ne@vbart.ru>
References: <20120406152452.5729f6e0@pluto.restena.lu>
	<201204062006.36424.ne@vbart.ru>
Message-ID: <20120406213859.61e809bd@neptune>

On Fri, 06 April 2012 "Valentin V. Bartenev" <ne at vbart.ru> wrote:
> On Friday 06 April 2012 17:24:52 Bruno Pr?mont wrote:
> > Hi,
> > 
> > I have started using limit_req_zone + limit_req but am seeing more delayed
> > log entries than I would expect.
> > 
> > I have the following:
> > 
> >   limit_req_zone $binary_remote_addr zone=one:10m rate=3r/s;
> >   ...
> >   server {
> >     ...
> >     location ~ .*\.php$ {
> >         limit_req zone=one burst=10;
> >         ...
> >     }
> > 
> > 
> > In the logs I see that when 2 requests happen near simultaneously from the
> > same IP address (e.g. page redirect to second PHP page or single
> > PHP-generated image referenced by page) the second and third request get
> > delayed if they arrive too early during the same second.
> > 
> > 
> > I would wish to limit the user to 3 requests within a 1s window though
> > not setting any constraint as to spacing between those requests, how can I
> > accomplish it or what changes to limit_req module would be needed to
> > not have a minimum 0.33s spacing between requests for above configuration?
> > 
> 
> http://nginx.org/r/limit_req
> 
>     limit_req zone=one burst=10 nodelay;

In that case legitimate (from my point of view) requests get rejected
which is even worse!

I wish those that fall under rate to not be delayed to get nice distribution
(in my case 1s sample-rate!), but to have those between rate and burst being
delayed to the following seconds.

As an example of what I expect (rate=3r/s, burst=5):
   Time   Request Id
   0.0s     1           initial request
   0.1s     2           second request, should not get delayed
   0.1s     3           third request, should not get delayed
   3.5s     4           new initial request
   3.6s     5           new second request, should not get delayed
   3.7s     6           new third request, should not get delayed
   3.8s     7           new fourth request, should get delayed until t=4.6s
   3.9s     8           new fifth request, should get delayed until t=4.7s
   4.0s     9           new sixth request, should get delayed until t=4.8s
   4.1s    10           new seventh request, should get delayed until t=5.6s
   4.2s    11           new eighth request, should get delayed until t=5.7s
   4.3s    12           new ninth request, should get rejected with 503 (exceeds burst)
   4.4s    13           new tenth request, should get rejected with 503 (exceeds burst)

but I do get like:
   0.0s     1           initial request
   0.1s     2           second request, delayed until 0.33s
   0.1s     3           third request, delayed until 0.66s

Bruno


From lists at ruby-forum.com  Fri Apr  6 20:27:17 2012
From: lists at ruby-forum.com (Ashish S Dev)
Date: Fri, 06 Apr 2012 22:27:17 +0200
Subject: Memcache module implementation with multi-get support?
Message-ID: <d48770f138305a228da2cbe3368d9d68@ruby-forum.com>

Hi,

Is there a memcache module implementation with multi-get support?

Thanks,
Ashish

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


From ne at vbart.ru  Fri Apr  6 20:56:05 2012
From: ne at vbart.ru (Valentin V. Bartenev)
Date: Sat, 7 Apr 2012 00:56:05 +0400
Subject: limit_req_zone and limit_req: rate too strict
In-Reply-To: <20120406213859.61e809bd@neptune>
References: <20120406152452.5729f6e0@pluto.restena.lu>
	<201204062006.36424.ne@vbart.ru> <20120406213859.61e809bd@neptune>
Message-ID: <201204070056.05706.ne@vbart.ru>

On Friday 06 April 2012 23:38:59 Bruno Pr?mont wrote:
> >
> > http://nginx.org/r/limit_req
> > 
> >     limit_req zone=one burst=10 nodelay;
> 
> In that case legitimate (from my point of view) requests get rejected
> which is even worse!

No, with "nodelay" parameter they won't be rejected or delayed until the 
overflow of burst.

> I wish those that fall under rate to not be delayed to get nice
> distribution (in my case 1s sample-rate!), but to have those between rate
> and burst being delayed to the following seconds.
[...]

There's no discrete "sample-rate". The limit_req module implements classical 
"leaky bucket" algorithm, which is simple, cheap and very common in the telecom 
world.

http://en.wikipedia.org/wiki/Leaky_bucket

 wbr, Valentin V. Bartenev


From brian at akins.org  Fri Apr  6 21:40:14 2012
From: brian at akins.org (Brian Akins)
Date: Fri, 6 Apr 2012 17:40:14 -0400
Subject: Memcache module implementation with multi-get support?
In-Reply-To: <d48770f138305a228da2cbe3368d9d68@ruby-forum.com>
References: <d48770f138305a228da2cbe3368d9d68@ruby-forum.com>
Message-ID: <26880F2B-B366-40C5-ABDE-A1B8B525CAE8@akins.org>


On Apr 6, 2012, at 4:27 PM, Ashish S Dev wrote:
> 
> Is there a memcache module implementation with multi-get support?

Probably be pretty simple to add it to https://github.com/agentzh/lua-resty-memcached



From matthieu.tourne at gmail.com  Fri Apr  6 22:00:33 2012
From: matthieu.tourne at gmail.com (Matthieu Tourne)
Date: Fri, 6 Apr 2012 15:00:33 -0700
Subject: Memcache module implementation with multi-get support?
In-Reply-To: <26880F2B-B366-40C5-ABDE-A1B8B525CAE8@akins.org>
References: <d48770f138305a228da2cbe3368d9d68@ruby-forum.com>
	<26880F2B-B366-40C5-ABDE-A1B8B525CAE8@akins.org>
Message-ID: <CAAGYmFd-3Da7+yfBCZt3Kg2ao+L=UJBQTEVe9iW9nQw7HbrS0g@mail.gmail.com>

On Fri, Apr 6, 2012 at 2:40 PM, Brian Akins <brian at akins.org> wrote:

>
> On Apr 6, 2012, at 4:27 PM, Ashish S Dev wrote:
> >
> > Is there a memcache module implementation with multi-get support?
>
> Probably be pretty simple to add it to
> https://github.com/agentzh/lua-resty-memcached
>
>
I think it already supports it if you pass an array of keys to the
memc:get() function

Matt
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20120406/08e1dc04/attachment.html>

From agentzh at gmail.com  Sat Apr  7 00:42:07 2012
From: agentzh at gmail.com (agentzh)
Date: Sat, 7 Apr 2012 08:42:07 +0800
Subject: Memcache module implementation with multi-get support?
In-Reply-To: <CAAGYmFd-3Da7+yfBCZt3Kg2ao+L=UJBQTEVe9iW9nQw7HbrS0g@mail.gmail.com>
References: <d48770f138305a228da2cbe3368d9d68@ruby-forum.com>
	<26880F2B-B366-40C5-ABDE-A1B8B525CAE8@akins.org>
	<CAAGYmFd-3Da7+yfBCZt3Kg2ao+L=UJBQTEVe9iW9nQw7HbrS0g@mail.gmail.com>
Message-ID: <CAB4Tn6NnuRz9h7utzmaD-YvEQaqT6WMPUgxbKOa-fp8h5npATQ@mail.gmail.com>

On Sat, Apr 7, 2012 at 6:00 AM, Matthieu Tourne
<matthieu.tourne at gmail.com> wrote:
> On Fri, Apr 6, 2012 at 2:40 PM, Brian Akins <brian at akins.org> wrote:
>> On Apr 6, 2012, at 4:27 PM, Ashish S Dev wrote:
>> >
>> > Is there a memcache module implementation with multi-get support?
>>
>> Probably be pretty simple to add it to
>> https://github.com/agentzh/lua-resty-memcached
>>
>
> I think it already supports it if you pass an array of keys to the
> memc:get() function
>

Yes, it does.

Below is a test case for this feature:

https://github.com/agentzh/lua-resty-memcached/blob/master/t/sanity.t#L1413

The lua-resty-memcached library is not an nginx module itself, but
rather a (nonblocking) Lua library that is intended to be used with
the ngx_lua module :)

Best regards,
-agentzh


From nginx-forum at nginx.us  Mon Apr  9 08:26:01 2012
From: nginx-forum at nginx.us (niraj)
Date: Mon, 9 Apr 2012 04:26:01 -0400 (EDT)
Subject: rewrite proxy flag set problem
In-Reply-To: <eb683e094059327e61cfe41a8b0f5742.NginxMailingListEnglish@forum.nginx.org>
References: <20120327085508.GA12789@nginx.com>
	<93f51c1c0f37415b621ae6a794042639.NginxMailingListEnglish@forum.nginx.org>
	<eb683e094059327e61cfe41a8b0f5742.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <7e72081cefac23b42757af8f94d6912d.NginxMailingListEnglish@forum.nginx.org>

Hi

       As mentioned above the different condition for Nginx, If anybody
know in details then please  guide me.       
     
It will help me to understand the Nginx rewrite rule in details.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,224274,225057#msg-225057


From lists at ruby-forum.com  Mon Apr  9 11:42:41 2012
From: lists at ruby-forum.com (Tariq wali)
Date: Mon, 09 Apr 2012 13:42:41 +0200
Subject: nginx upstream prematurely closed connection
In-Reply-To: <COL106-W45FCC10FEE7344300CA673B9480@phx.gbl>
References: <COL106-W45FCC10FEE7344300CA673B9480@phx.gbl>
Message-ID: <d1e46dfbde5bb554073a9e5a36b8aea4@ruby-forum.com>

Maxim,

I see the same error on few of our lamp servers and nginx serving static 
content .
I have looked for clues in apache error logs and syslog messages but 
can't establish a connection ..

2012/03/21 00:00:01 [error] 569#0: *10867863 upstream prematurely closed 
connection while reading response header from upstream, client: 
121.243.22.130, server: _, request: "GET / HTTP/1.1", upstream: 
"http://127.0.0.1:8080/", host: "servername.net"
2012/03/22 00:00:01 [error] 568#0: *10869967 upstream prematurely closed 
connection while reading response header from upstream, client: 
121.243.22.130, server: _, request: "GET / HTTP/1.1", upstream: 
"http://127.0.0.1:8080/", host: "servername.net"
2012/03/23 00:00:01 [error] 569#0: *10872221 upstream prematurely closed 
connection while reading response header from upstream, client: 
121.243.22.130, server: _, request: "GET / HTTP/1.1", upstream: 
"http://127.0.0.1:8080/", host: "servername.net"

I would appreciate a direction I can take this issue in ..'


T

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


From nbubingo at gmail.com  Mon Apr  9 13:28:26 2012
From: nbubingo at gmail.com (=?GB2312?B?0qbOsLHz?=)
Date: Mon, 9 Apr 2012 21:28:26 +0800
Subject: [ANNOUNCE] Tengine-1.2.4
In-Reply-To: <CANXGsjzLS3RXRV3__Y8-L_fUfxxhoVZh64EL2D6icy6CPZFkCw@mail.gmail.com>
References: <CAOKiq2CbQUtcdwMa+y+RgDEQTSVaOJEooRdOEBYBSxYtsqxevg@mail.gmail.com>
	<CANXGsjzCm_AK=JtGYowQ17QiMfzDckRTxShH9QOn7JxCWN9rzg@mail.gmail.com>
	<CACPpB2xrZfQqtwD=h2HSjMjp4BJmwVr9_z5=mBC_75EQYZ+VjQ@mail.gmail.com>
	<CANXGsjzLS3RXRV3__Y8-L_fUfxxhoVZh64EL2D6icy6CPZFkCw@mail.gmail.com>
Message-ID: <CAGxuLK2dVw8vOLFrFt5Rq3qe6+1AnsoQbkdrgLON1qpiN=Z6Qw@mail.gmail.com>

Hello,

Can you just focus on the technical problem? If you have any detail
question to ask, we are pleased to serve you.

Thanks.

2012/4/5 Sharl.Jimh.Tsin <amoiz.shine at gmail.com>:
> from your profile in LinkedIn website,i found that you are a student
> of HuaiYin Normal University.right?
>
> 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.
>
>
>
> ? 2012?4?1? ??9:58?smallfish <smallfish.xy at gmail.com> ???
>> hah, perhaps just a joke. but nothing is impossable. :)
>> --
>> blog: http://chenxiaoyu.org
>>
>>
>>
>> ? 2012?4?1? ??9:33?Sharl.Jimh.Tsin <amoiz.shine at gmail.com>???
>>>
>>> are you sure that NOT a joke of April Fools' Day, like taolinux
>>> (http://code.taobao.org/p/taobao_linux_server/src/trunk/taolinux/) ?
>>>
>>> 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.
>>>
>>>
>>>
>>> 2012/3/31 Joshua Zhu <zhuzhaoyuan at gmail.com>:
>>> > Hi folks!
>>> >
>>> > We're pleased to announce that Tengine-1.2.4 has been released. You can
>>> > either download the tar ball
>>> > from http://tengine.taobao.org/download/tengine-1.2.4.tar.gz , or
>>> > checkout
>>> > the code from https://github.com/taobao/tengine
>>> >
>>> > As this release, Tengine is based on Nginx-1.0.14. The highlight is a
>>> > brand
>>> > new 'user_agent' module which would be a good replacement for the
>>> > 'browser'
>>> > module in standard Nginx. It's more efficient and flexible than the
>>> > 'browser' module. Other changes include:
>>> > * Feature: added the 'log_escape' directive.
>>> > * Bugfix: fixed a bug in the limit_req module.
>>> > * Bugfix: fixed a bug in subrequest.
>>> >
>>> > Regards,
>>> >
>>> > --
>>> > Joshua Zhu
>>> > Senior Software Engineer
>>> > Server Platforms Team at Taobao
>>> >
>>> > _______________________________________________
>>> > 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 nbubingo at gmail.com  Mon Apr  9 13:34:56 2012
From: nbubingo at gmail.com (=?GB2312?B?0qbOsLHz?=)
Date: Mon, 9 Apr 2012 21:34:56 +0800
Subject: nginx upstream prematurely closed connection
In-Reply-To: <d1e46dfbde5bb554073a9e5a36b8aea4@ruby-forum.com>
References: <COL106-W45FCC10FEE7344300CA673B9480@phx.gbl>
	<d1e46dfbde5bb554073a9e5a36b8aea4@ruby-forum.com>
Message-ID: <CAGxuLK2L8BC8JRjTkiw-ib7NdjMPt70RYv3tw9CzveHt4Fn3Uw@mail.gmail.com>

This is also proxy log, not the static content request log.

You should be sure you have enough resources in your backend server,
like memory, connection...

2012/4/9 Tariq wali <lists at ruby-forum.com>:
> Maxim,
>
> I see the same error on few of our lamp servers and nginx serving static
> content .
> I have looked for clues in apache error logs and syslog messages but
> can't establish a connection ..
>
> 2012/03/21 00:00:01 [error] 569#0: *10867863 upstream prematurely closed
> connection while reading response header from upstream, client:
> 121.243.22.130, server: _, request: "GET / HTTP/1.1", upstream:
> "http://127.0.0.1:8080/", host: "servername.net"
> 2012/03/22 00:00:01 [error] 568#0: *10869967 upstream prematurely closed
> connection while reading response header from upstream, client:
> 121.243.22.130, server: _, request: "GET / HTTP/1.1", upstream:
> "http://127.0.0.1:8080/", host: "servername.net"
> 2012/03/23 00:00:01 [error] 569#0: *10872221 upstream prematurely closed
> connection while reading response header from upstream, client:
> 121.243.22.130, server: _, request: "GET / HTTP/1.1", upstream:
> "http://127.0.0.1:8080/", host: "servername.net"
>
> I would appreciate a direction I can take this issue in ..'
>
>
> T
>
> --
> 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  Mon Apr  9 13:41:54 2012
From: nginx-forum at nginx.us (auxbuss)
Date: Mon, 9 Apr 2012 09:41:54 -0400 (EDT)
Subject: Rewrites for sites without SSL on same IP address as those using SSL
Message-ID: <980ba766245be3167fb1f1a2770fa2d6.NginxMailingListEnglish@forum.nginx.org>

I have a few sites on a single IP and a single nginx instance. Some
domains have SSL certificates (and listen on ports 80 and 443) and some
don't (and only listen on port 80).

For those domains that don't have an SSL certificate, including a server
block listening on port 443 to rewrite to port 80 does not work. I
understand this is because the SSL exchange happens prior to the server
name being inspected.

When browsing over https to one of the sites without an SSL certificate,
the current behaviour is that a security warning is presented to the
client that lists the certificate for the default_server and not the
appropriate domain.

Is there a way to rewrite https requests to http (for the domains with
an SSL certificate) while retaining the appropriate domain? 

If not, is the best way to resolve this to add an additional IP and use
one for those sites requiring ports 80 and 443 and another for those
only using port 80?

Thanks.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225064,225064#msg-225064


From igor at sysoev.ru  Mon Apr  9 13:55:15 2012
From: igor at sysoev.ru (Igor Sysoev)
Date: Mon, 9 Apr 2012 17:55:15 +0400
Subject: Rewrites for sites without SSL on same IP address as those using
	SSL
In-Reply-To: <980ba766245be3167fb1f1a2770fa2d6.NginxMailingListEnglish@forum.nginx.org>
References: <980ba766245be3167fb1f1a2770fa2d6.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20120409135515.GA49392@nginx.com>

On Mon, Apr 09, 2012 at 09:41:54AM -0400, auxbuss wrote:
> I have a few sites on a single IP and a single nginx instance. Some
> domains have SSL certificates (and listen on ports 80 and 443) and some
> don't (and only listen on port 80).
> 
> For those domains that don't have an SSL certificate, including a server
> block listening on port 443 to rewrite to port 80 does not work. I
> understand this is because the SSL exchange happens prior to the server
> name being inspected.
> 
> When browsing over https to one of the sites without an SSL certificate,
> the current behaviour is that a security warning is presented to the
> client that lists the certificate for the default_server and not the
> appropriate domain.
> 
> Is there a way to rewrite https requests to http (for the domains with
> an SSL certificate) while retaining the appropriate domain? 

Do you mean the domains WITHOUT an SSL certificate ?

> If not, is the best way to resolve this to add an additional IP and use
> one for those sites requiring ports 80 and 443 and another for those
> only using port 80?

Yes, in this case Firefox will show
-----
Unable to connect
Firefox can't establish a connection to the server at ...
-----


-- 
Igor Sysoev


From nginx-forum at nginx.us  Mon Apr  9 13:57:46 2012
From: nginx-forum at nginx.us (auxbuss)
Date: Mon, 9 Apr 2012 09:57:46 -0400 (EDT)
Subject: Rewrites for sites without SSL on same IP address as those using
	SSL
In-Reply-To: <980ba766245be3167fb1f1a2770fa2d6.NginxMailingListEnglish@forum.nginx.org>
References: <980ba766245be3167fb1f1a2770fa2d6.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <d7327cdccfc5217fa7f4527f15e74030.NginxMailingListEnglish@forum.nginx.org>

Yes, I mean domains without an SSL certificate.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225064,225066#msg-225066


From jdorfman at netdna.com  Mon Apr  9 19:49:43 2012
From: jdorfman at netdna.com (Justin Dorfman)
Date: Mon, 9 Apr 2012 12:49:43 -0700
Subject: could not build the referers_hash
In-Reply-To: <1333144207.2204.8.camel@portable-evil>
References: <CACDr0vF2ZYQ4urpqhgV3dBUaudgZd6NXp0qRbwccEA66ip-vXQ@mail.gmail.com>
	<1333071717.2082.3.camel@portable-evil>
	<CACDr0vErxKZgTYomxjsG++ZMvWu1_R=jEifaO15SAniqAvcjaQ@mail.gmail.com>
	<1333120642.2082.7.camel@portable-evil>
	<CACDr0vF7N4mNx0q63ZEVUfY8wzzTwys2ZvV7PGXECKdbmz=+KA@mail.gmail.com>
	<1333144207.2204.8.camel@portable-evil>
Message-ID: <CACDr0vF2=-J2cHnVRytmshPz35AMvbncFC3EnR6m2wx=CYDV3A@mail.gmail.com>

Our senior engineer figured how to fix it.

What he did was edited this file: src/http/modules/ngx_http_referer_module.c

and changed this line:
prev->referer_hash_bucket_size, 64);

to:
prev->referer_hash_bucket_size, 128);

Then recompiled nginx.

Hope that helps someone in the future.

Regards,

Justin Dorfman <http://www.twitter.com/jdorfman>




On Fri, Mar 30, 2012 at 2:50 PM, Cliff Wells <cliff at develix.com> wrote:

> On Fri, 2012-03-30 at 09:08 -0700, Justin Dorfman wrote:
> > Hey Cliff,
> >
> >
> > Here are the current values:
> >
> >
> > server_names_hash_max_size 65536;
> > server_names_hash_bucket_size 256;
>
> These are not the hashes you are looking for =)  Your error message
> specified "you should increase referers_hash_bucket_size", indicating
> that it is currently 64.
>
> Regards,
> Cliff
>
>
> _______________________________________________
> 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/20120409/dbd2a491/attachment-0001.html>

From contact at jpluscplusm.com  Mon Apr  9 20:00:05 2012
From: contact at jpluscplusm.com (Jonathan Matthews)
Date: Mon, 9 Apr 2012 21:00:05 +0100
Subject: Chunked input module problem with http_dav
Message-ID: <CAKsTx7C5NS6aJOsFoALewVASY5t7Hw=MBeTwrOKz3nSkBBwUBg@mail.gmail.com>

Hi all -

I'm running the latest tagged version of the chunked input body module
as per https://github.com/agentzh/chunkin-nginx-module/tags (0.23rc2)
and nginx 1.0.14 with the http_dav module. I'm seeing some failures
which are, to my eye, stopping chunkin work at /all/ in this
combination.

I'm using http_dav to provide an some upload space. After an upload
completes (I started at 400MB, and still see the following problem
with 4MB uploads), I get the following in the debug/error logs:

  2012/04/09 20:38:02 [crit] 1485#0: *3 chmod() "/srv/tmp/0000000001"
failed (2: No such file or directory), client: 10.0.10.183, server:
upload.server, request: "PUT /external/testing123 HTTP/1.1", host:
"upload.server:81"
  2012/04/09 20:38:02 [crit] 1485#0: *3 unlink() "/srv/tmp/0000000001"
failed (2: No such file or directory), client: 10.0.10.183, server:
upload.server, request: "PUT /external/testing123 HTTP/1.1", host:
"upload.server:81"

Then the client receives a 500.
"/srv/tmp" (the client body temp location) *is* on the same partition
as the final resting place as the data.
If I *don't* use chunked input, then the upload works fine.

Here's the site config:
---------------------------------------------------------------------
server {
  listen *:81;
  server_name upload.server;
  chunkin on;
  error_page 411 = @my_411_error;
  location @my_411_error {
      chunkin_resume;
  }
  location / {
    root /srv/data/;
    client_body_temp_path /srv/tmp/
    client_max_body_size 500m;
    dav_methods  PUT DELETE MKCOL COPY MOVE;
    create_full_put_path   off;
    dav_access             group:rw  all:r;
  }
}

... and here's the nginx-V output:

nginx version: nginx/1.0.14
built by gcc 4.4.5 (Debian 4.4.5-8)
configure arguments: --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_dav_module --with-http_realip_module
--with-http_stub_status_module --with-http_sub_module
--add-module=/root/src/nginx-modules/agentzh-chunkin-nginx-module-ddc0dd5
--prefix=/usr

Any ideas what's causing this and how I can fix it? Or is chunkin just
not supposed to be run in this way with http_dav?

Cheers,
Jonathan
-- 
Jonathan Matthews
London, Oxford, UK
http://www.jpluscplusm.com/contact.html


From nginx-forum at nginx.us  Tue Apr 10 01:28:23 2012
From: nginx-forum at nginx.us (spacerobot)
Date: Mon, 9 Apr 2012 21:28:23 -0400 (EDT)
Subject: Nginx returning 414 even when large_client_header_buffers is set
Message-ID: <8b19edbb6b2d137c94eae907213f378f.NginxMailingListEnglish@forum.nginx.org>

My nginx server needs a request whose URI is about 4k long, it currently
returns 414 with message Request-URI Too Long.

I looked online and added:

large_client_header_buffers 4 16k;

to the http block in nginx.conf, restarted nginx, still returns 414. 

I also tried to add client_header_buffer_size and set to 16k and well
but didn't help.

I have nginx 1.0.12 and built with the following options:

  "--with-http_ssl_module",
  "--with-http_gzip_static_module",
  "--add-module=../git/nginx-x-rid-header",
  "--with-ld-opt=-luuid",
  "--with-http_stub_status_module"

Anybody has any insights on the 414 problem?

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225093,225093#msg-225093


From ashishs.dev at gmail.com  Tue Apr 10 01:44:24 2012
From: ashishs.dev at gmail.com (Ashish S)
Date: Mon, 9 Apr 2012 18:44:24 -0700
Subject: Memcache module implementation with multi-get support?
In-Reply-To: <CAB4Tn6NnuRz9h7utzmaD-YvEQaqT6WMPUgxbKOa-fp8h5npATQ@mail.gmail.com>
References: <d48770f138305a228da2cbe3368d9d68@ruby-forum.com>
	<26880F2B-B366-40C5-ABDE-A1B8B525CAE8@akins.org>
	<CAAGYmFd-3Da7+yfBCZt3Kg2ao+L=UJBQTEVe9iW9nQw7HbrS0g@mail.gmail.com>
	<CAB4Tn6NnuRz9h7utzmaD-YvEQaqT6WMPUgxbKOa-fp8h5npATQ@mail.gmail.com>
Message-ID: <CA+BpfUomkqqXzuOQQ7jUfhfLfpGoN2cXDYHQEbYucCokpimm5A@mail.gmail.com>

Thanks, everyone. This works.

--Ashish

On Fri, Apr 6, 2012 at 5:42 PM, agentzh <agentzh at gmail.com> wrote:
> On Sat, Apr 7, 2012 at 6:00 AM, Matthieu Tourne
> <matthieu.tourne at gmail.com> wrote:
>> On Fri, Apr 6, 2012 at 2:40 PM, Brian Akins <brian at akins.org> wrote:
>>> On Apr 6, 2012, at 4:27 PM, Ashish S Dev wrote:
>>> >
>>> > Is there a memcache module implementation with multi-get support?
>>>
>>> Probably be pretty simple to add it to
>>> https://github.com/agentzh/lua-resty-memcached
>>>
>>
>> I think it already supports it if you pass an array of keys to the
>> memc:get() function
>>
>
> Yes, it does.
>
> Below is a test case for this feature:
>
> https://github.com/agentzh/lua-resty-memcached/blob/master/t/sanity.t#L1413
>
> The lua-resty-memcached library is not an nginx module itself, but
> rather a (nonblocking) Lua library that is intended to be used with
> the ngx_lua module :)
>
> Best regards,
> -agentzh
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx


From ashishs.dev at gmail.com  Tue Apr 10 02:02:10 2012
From: ashishs.dev at gmail.com (Ashish S)
Date: Mon, 9 Apr 2012 19:02:10 -0700
Subject: Program hangs when modifying u->buffer while using proxy module to
	talk to upstream backend
Message-ID: <CA+BpfUrJsVQFBio8Am-0wre-CG79Xm9iiexT4W1GkcpECetY8g@mail.gmail.com>

Hi,

I currently use upstream to talk to a back-end, which sends me some
response. In my module, i am trying to re-format the plain-text
upstream backend response, to XML, and I also add some new data to it,
based on an in-memory lookup within my module.  What would be the best
way to do this?

In my setup, I am able to parse response from upstream (u->buffer),
construct a modified string and assign it back to u->buffer, and am I
setting  u->headers_in.content_length_n and
r->headers_out.content_length_n correctly. The issue i face is, every
*second* request hangs, if i am trying the request through the
browser. (However, trying through command-line & curl, works
everytime). And I always see this error message in the logs: "client
prematurely closed connection, so upstream connection is closed too
while sending to client, client: 127.0.0.1".   What am i doing wrong?

Thanks,
Ashish


From nbubingo at gmail.com  Tue Apr 10 05:17:17 2012
From: nbubingo at gmail.com (=?GB2312?B?0qbOsLHz?=)
Date: Tue, 10 Apr 2012 13:17:17 +0800
Subject: Program hangs when modifying u->buffer while using proxy module
	to talk to upstream backend
In-Reply-To: <CA+BpfUrJsVQFBio8Am-0wre-CG79Xm9iiexT4W1GkcpECetY8g@mail.gmail.com>
References: <CA+BpfUrJsVQFBio8Am-0wre-CG79Xm9iiexT4W1GkcpECetY8g@mail.gmail.com>
Message-ID: <CAGxuLK0EB-5uxVsucq6F6Gq-mbomcjaT2Kw+Qq96yjmsLa3hoA@mail.gmail.com>

see the fastcgi module as an example

2012/4/10 Ashish S <ashishs.dev at gmail.com>:
> Hi,
>
> I currently use upstream to talk to a back-end, which sends me some
> response. In my module, i am trying to re-format the plain-text
> upstream backend response, to XML, and I also add some new data to it,
> based on an in-memory lookup within my module. ?What would be the best
> way to do this?
>
> In my setup, I am able to parse response from upstream (u->buffer),
> construct a modified string and assign it back to u->buffer, and am I
> setting ?u->headers_in.content_length_n and
> r->headers_out.content_length_n correctly. The issue i face is, every
> *second* request hangs, if i am trying the request through the
> browser. (However, trying through command-line & curl, works
> everytime). And I always see this error message in the logs: "client
> prematurely closed connection, so upstream connection is closed too
> while sending to client, client: 127.0.0.1". ? What am i doing wrong?
>
> Thanks,
> Ashish
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx


From nginx-forum at nginx.us  Tue Apr 10 07:47:12 2012
From: nginx-forum at nginx.us (zuckbin)
Date: Tue, 10 Apr 2012 03:47:12 -0400 (EDT)
Subject: serve static content on static domain
In-Reply-To: <d8832eaa542bb1f75db0d4b6749833d2.NginxMailingListEnglish@forum.nginx.org>
References: <d8832eaa542bb1f75db0d4b6749833d2.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <7b3d5fd92177e2f34fe20ec621a93193.NginxMailingListEnglish@forum.nginx.org>

What is the server block ?

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225020,225098#msg-225098


From ne at vbart.ru  Tue Apr 10 08:52:17 2012
From: ne at vbart.ru (Valentin V. Bartenev)
Date: Tue, 10 Apr 2012 12:52:17 +0400
Subject: serve static content on static domain
In-Reply-To: <7b3d5fd92177e2f34fe20ec621a93193.NginxMailingListEnglish@forum.nginx.org>
References: <d8832eaa542bb1f75db0d4b6749833d2.NginxMailingListEnglish@forum.nginx.org>
	<7b3d5fd92177e2f34fe20ec621a93193.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <201204101252.17233.ne@vbart.ru>

On Tuesday 10 April 2012 11:47:12 zuckbin wrote:
> What is the server block ?
> 

http://nginx.org/r/server

 wbr, Valentin V. Bartenev


From nginx-forum at nginx.us  Tue Apr 10 10:07:40 2012
From: nginx-forum at nginx.us (sibsoft)
Date: Tue, 10 Apr 2012 06:07:40 -0400 (EDT)
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: <9d8ab115b3e9176b7c4111e95478bbc7.NginxMailingListEnglish@forum.nginx.org>

Dealing with this trouble again (nginx 1.0.12, 1.0.14):

2012/04/10 12:06:59 [alert] 22812#0: *1571 pread() read only 0 of
1048576 from "/var/www/cgi-bin/uploads/03/00364/z46j20qdp6gt" while
sending mp4 to client, client: 95.131.10.226, server: gorillavid.com,
request: "GET
/mgobz2aj6gu4rqukwztlpbbrdqzwmtcoid2bk6t6xenwqd3fwvt6z3pyqy/video.mp4?start=1190.84
HTTP/1.1", host: "85.17.30.216:8182"
2012/04/10 12:07:00 [error] 22815#0: *1630 start time is out mp4 stts
samples in "/var/www/cgi-bin/uploads/03/00001/k2hhvgyfdmmu", client:
86.18.83.8, server: gorillavid.com, request: "GET
/mgobznkw5ku4tqukwyf3nfjxilouln5pl5abqotvftlq7rqggvegaydxva/video.flv?start=0
HTTP/1.1", host: "85.17.30.216:8182"

I believe that I have this patch already applied:
        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);
            goto found;
        }

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218451,225103#msg-225103


From nginx-forum at nginx.us  Tue Apr 10 10:10:34 2012
From: nginx-forum at nginx.us (sibsoft)
Date: Tue, 10 Apr 2012 06:10:34 -0400 (EDT)
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: <a4fc266375e06fc3d706cc49744df7c9.NginxMailingListEnglish@forum.nginx.org>

Dealing with this trouble again (nginx 1.0.12, 1.0.14):

2012/04/10 12:06:59 [alert] 22812#0: *1571 pread() read only 0 of
1048576 from "/var/www/cgi-bin/uploads/03/00364/z46j20qdp6gt" while
sending mp4 to client, client: ....., server: ....., request: "GET
/mgobz2aj6gu4rqukwztlpbbrdqzwmtcoid2bk6t6xenwqd3fwvt6z3pyqy/video.mp4?start=1190.84
HTTP/1.1", host: "....."
2012/04/10 12:07:00 [error] 22815#0: *1630 start time is out mp4 stts
samples in "/var/www/cgi-bin/uploads/03/00001/k2hhvgyfdmmu", client:
....., server: ....., request: "GET
/mgobznkw5ku4tqukwyf3nfjxilouln5pl5abqotvftlq7rqggvegaydxva/video.flv?start=0
HTTP/1.1", host: "....."

I believe that I have this patch already applied:
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);
goto found;
}

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,218451,225104#msg-225104


From rajnesh.siwal at gmail.com  Tue Apr 10 10:53:14 2012
From: rajnesh.siwal at gmail.com (Rajnesh Kumar Siwal)
Date: Tue, 10 Apr 2012 16:23:14 +0530
Subject: KeepAlive Not Working as expected for a Reverse Proxy Scenario
Message-ID: <CACTshoT9GAZ+LQD9YDbQxpW=nxrVGNEA+7XyRSuBPms5d2Aaxg@mail.gmail.com>

Hi,

The KeepAlive Timeout is set to 2 hours, but we can see that the
connections start terminating after a few minutes.
The KeepAlive module does not seems to be behaving as expected.
The connections are created on port :343.
The configuration file nginx.conf is :-
-------------------------------------------------------------------
user              nginx;
worker_processes  16;
error_log  /var/log/nginx/error.log info;
pid        /var/run/nginx.pid;
events {
        #worker_connections  16192;
        #worker_connections  161920;
        worker_connections  40960;
        use epoll;
        multi_accept on;
}
http {
    include       /etc/nginx/mime.types;
    default_type  application/octet-stream;
    log_format  main  '$remote_addr - $remote_user [$time_local] "$request"
'
                      '$status $body_bytes_sent "$http_referer" '
                      '"$http_user_agent" "$http_x_forwarded_for"';
    access_log  /var/log/nginx/access.log  main;
    sendfile        on;
    #tcp_nopush     on;
    keepalive_timeout  7200 7200;
tcp_keepalive on;
tcp_keepcnt 10;
tcp_keepidle 3600;
tcp_keepintvl 1m;
    #gzip  on;

    #
    # The default server
    #
        server {
                listen 4443;
                server_name _;
                location /nginx_status {
  # copied from
http://blog.kovyrin.net/2006/04/29/monitoring-nginx-with-rrdtool/
                stub_status on;
                access_log   off;
                }
        }
    server {
        listen       80;
        server_name  _;
    #   keepalive_timeout  7200 7200;
        #charset koi8-r;
        access_log  logs/host.access.log  main;
        location / {
                proxy_pass      http://172.31.253.50:180/;
                include         /etc/nginx/proxy.conf;
        }
        }
        server {
        listen 443 ssl;
        ssl on;
        ssl_verify_client on;
        server_name _;
        ssl_certificate      /home/rsiwal/cps.chargepoint.net.crt;
        #ssl_client_certificate  /etc/httpd/certs/
chargepoint.net/server/certs/cps.chargepoint.net.crt;
        ssl_client_certificate  /home/rsiwal/chargepoint.net.crt;
        ssl_certificate_key  /home/rsiwal/cps.chargepoint.net.pem;
        location / {
                 proxy_pass      http://172.31.253.50:1443/;
                include         /etc/nginx/proxy.conf;
        }
        }
        server {
        listen 343 ssl;
        ssl on;
        ssl_verify_client on;
        server_name _;
        #ssl_session_cache shared:SSL:40m;    # Works fine upto 15K
        ssl_session_cache shared:SSL:70m;
        #######ssl_session_timeout 5m;     #
        ssl_session_timeout 120m;      #
tcp_keepalive on;
tcp_keepcnt 10;
tcp_keepidle 7200;
tcp_keepintvl 1m;
        access_log      logs/ssl-access.log main;
        error_log       logs/ssl-error.log debug;
        keepalive_requests    10000;
        #keepalive_disable none
        sendfile on;
        tcp_nopush on;
        tcp_nodelay on;
        client_header_timeout 300;
        client_body_timeout 300;
        ssl_certificate      /home/rsiwal/cps.chargepoint.net.crt;
        ssl_certificate_key  /home/rsiwal/cps.chargepoint.net.pem;
        ssl_client_certificate  /home/rsiwal/chargepoint.net.crt;
        location / {
                 proxy_pass      http://172.31.253.50:1343/;
                include         /etc/nginx/proxy.conf;
        }
        }
    include /etc/nginx/conf.d/*.conf;
}
-----------------------------------------------------------------------------------------------------------------------------

Thanks in advance

-- 
Regards,
Rajnesh Kumar Siwal
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20120410/96908704/attachment-0001.html>

From amoiz.shine at gmail.com  Tue Apr 10 10:55:36 2012
From: amoiz.shine at gmail.com (Sharl.Jimh.Tsin)
Date: Tue, 10 Apr 2012 18:55:36 +0800
Subject: [ANNOUNCE] Tengine-1.2.4
In-Reply-To: <CAGxuLK2dVw8vOLFrFt5Rq3qe6+1AnsoQbkdrgLON1qpiN=Z6Qw@mail.gmail.com>
References: <CAOKiq2CbQUtcdwMa+y+RgDEQTSVaOJEooRdOEBYBSxYtsqxevg@mail.gmail.com>
	<CANXGsjzCm_AK=JtGYowQ17QiMfzDckRTxShH9QOn7JxCWN9rzg@mail.gmail.com>
	<CACPpB2xrZfQqtwD=h2HSjMjp4BJmwVr9_z5=mBC_75EQYZ+VjQ@mail.gmail.com>
	<CANXGsjzLS3RXRV3__Y8-L_fUfxxhoVZh64EL2D6icy6CPZFkCw@mail.gmail.com>
	<CAGxuLK2dVw8vOLFrFt5Rq3qe6+1AnsoQbkdrgLON1qpiN=Z6Qw@mail.gmail.com>
Message-ID: <CANXGsjy6zadyuRwiwPQBjXtQzwtnMXuR=UTVUsuHY2-5rEzReg@mail.gmail.com>

sorry for those non-technical posts,maybe i should send him directly
instead of mailing-list.

BTW,is there any improvements of tengine on CGI?

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.



? 2012?4?9? ??9:28???? <nbubingo at gmail.com> ???
> Hello,
>
> Can you just focus on the technical problem? If you have any detail
> question to ask, we are pleased to serve you.
>
> Thanks.
>
> 2012/4/5 Sharl.Jimh.Tsin <amoiz.shine at gmail.com>:
>> from your profile in LinkedIn website,i found that you are a student
>> of HuaiYin Normal University.right?
>>
>> 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.
>>
>>
>>
>> ? 2012?4?1? ??9:58?smallfish <smallfish.xy at gmail.com> ???
>>> hah, perhaps just a joke. but nothing is impossable. :)
>>> --
>>> blog: http://chenxiaoyu.org
>>>
>>>
>>>
>>> ? 2012?4?1? ??9:33?Sharl.Jimh.Tsin <amoiz.shine at gmail.com>???
>>>>
>>>> are you sure that NOT a joke of April Fools' Day, like taolinux
>>>> (http://code.taobao.org/p/taobao_linux_server/src/trunk/taolinux/) ?
>>>>
>>>> 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.
>>>>
>>>>
>>>>
>>>> 2012/3/31 Joshua Zhu <zhuzhaoyuan at gmail.com>:
>>>> > Hi folks!
>>>> >
>>>> > We're pleased to announce that Tengine-1.2.4 has been released. You can
>>>> > either download the tar ball
>>>> > from http://tengine.taobao.org/download/tengine-1.2.4.tar.gz , or
>>>> > checkout
>>>> > the code from https://github.com/taobao/tengine
>>>> >
>>>> > As this release, Tengine is based on Nginx-1.0.14. The highlight is a
>>>> > brand
>>>> > new 'user_agent' module which would be a good replacement for the
>>>> > 'browser'
>>>> > module in standard Nginx. It's more efficient and flexible than the
>>>> > 'browser' module. Other changes include:
>>>> > * Feature: added the 'log_escape' directive.
>>>> > * Bugfix: fixed a bug in the limit_req module.
>>>> > * Bugfix: fixed a bug in subrequest.
>>>> >
>>>> > Regards,
>>>> >
>>>> > --
>>>> > Joshua Zhu
>>>> > Senior Software Engineer
>>>> > Server Platforms Team at Taobao
>>>> >
>>>> > _______________________________________________
>>>> > 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  Tue Apr 10 11:16:04 2012
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Tue, 10 Apr 2012 15:16:04 +0400
Subject: could not build the referers_hash
In-Reply-To: <CACDr0vF2=-J2cHnVRytmshPz35AMvbncFC3EnR6m2wx=CYDV3A@mail.gmail.com>
References: <CACDr0vF2ZYQ4urpqhgV3dBUaudgZd6NXp0qRbwccEA66ip-vXQ@mail.gmail.com>
	<1333071717.2082.3.camel@portable-evil>
	<CACDr0vErxKZgTYomxjsG++ZMvWu1_R=jEifaO15SAniqAvcjaQ@mail.gmail.com>
	<1333120642.2082.7.camel@portable-evil>
	<CACDr0vF7N4mNx0q63ZEVUfY8wzzTwys2ZvV7PGXECKdbmz=+KA@mail.gmail.com>
	<1333144207.2204.8.camel@portable-evil>
	<CACDr0vF2=-J2cHnVRytmshPz35AMvbncFC3EnR6m2wx=CYDV3A@mail.gmail.com>
Message-ID: <20120410111604.GO13466@mdounin.ru>

Hello!

On Mon, Apr 09, 2012 at 12:49:43PM -0700, Justin Dorfman wrote:

> Our senior engineer figured how to fix it.
> 
> What he did was edited this file: src/http/modules/ngx_http_referer_module.c
> 
> and changed this line:
> prev->referer_hash_bucket_size, 64);
> 
> to:
> prev->referer_hash_bucket_size, 128);
> 
> Then recompiled nginx.
> 
> Hope that helps someone in the future.

There is configuration directive to adjust this, 
referer_hash_bucket_size.  You don't need to edit the code.

Maxim Dounin

> 
> Regards,
> 
> Justin Dorfman <http://www.twitter.com/jdorfman>
> 
> 
> 
> 
> On Fri, Mar 30, 2012 at 2:50 PM, Cliff Wells <cliff at develix.com> wrote:
> 
> > On Fri, 2012-03-30 at 09:08 -0700, Justin Dorfman wrote:
> > > Hey Cliff,
> > >
> > >
> > > Here are the current values:
> > >
> > >
> > > server_names_hash_max_size 65536;
> > > server_names_hash_bucket_size 256;
> >
> > These are not the hashes you are looking for =)  Your error message
> > specified "you should increase referers_hash_bucket_size", indicating
> > that it is currently 64.
> >
> > Regards,
> > Cliff
> >
> >
> > _______________________________________________
> > nginx mailing list
> > nginx at nginx.org
> > http://mailman.nginx.org/mailman/listinfo/nginx
> >

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


From mdounin at mdounin.ru  Tue Apr 10 12:13:38 2012
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Tue, 10 Apr 2012 16:13:38 +0400
Subject: Program hangs when modifying u->buffer while using proxy module
	to talk to upstream backend
In-Reply-To: <CA+BpfUrJsVQFBio8Am-0wre-CG79Xm9iiexT4W1GkcpECetY8g@mail.gmail.com>
References: <CA+BpfUrJsVQFBio8Am-0wre-CG79Xm9iiexT4W1GkcpECetY8g@mail.gmail.com>
Message-ID: <20120410121337.GR13466@mdounin.ru>

Hello!

On Mon, Apr 09, 2012 at 07:02:10PM -0700, Ashish S wrote:

> Hi,
> 
> I currently use upstream to talk to a back-end, which sends me some
> response. In my module, i am trying to re-format the plain-text
> upstream backend response, to XML, and I also add some new data to it,
> based on an in-memory lookup within my module.  What would be the best
> way to do this?

Unless you are working on custom protocol module, you may want to 
use filter module to re-format data instead.

> In my setup, I am able to parse response from upstream (u->buffer),
> construct a modified string and assign it back to u->buffer, and am I
> setting  u->headers_in.content_length_n and
> r->headers_out.content_length_n correctly. The issue i face is, every
> *second* request hangs, if i am trying the request through the
> browser. (However, trying through command-line & curl, works
> everytime). And I always see this error message in the logs: "client
> prematurely closed connection, so upstream connection is closed too
> while sending to client, client: 127.0.0.1".   What am i doing wrong?

It's hard to say anything without seeing the code, but most likely 
you fail to set u->length properly.  See e.g. memcached module for 
a simple example of non-buffered protocol handler.

Maxim Dounin


From mdounin at mdounin.ru  Tue Apr 10 13:03:03 2012
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Tue, 10 Apr 2012 17:03:03 +0400
Subject: KeepAlive Not Working as expected for a Reverse Proxy Scenario
In-Reply-To: <CACTshoT9GAZ+LQD9YDbQxpW=nxrVGNEA+7XyRSuBPms5d2Aaxg@mail.gmail.com>
References: <CACTshoT9GAZ+LQD9YDbQxpW=nxrVGNEA+7XyRSuBPms5d2Aaxg@mail.gmail.com>
Message-ID: <20120410130303.GT13466@mdounin.ru>

Hello!

On Tue, Apr 10, 2012 at 04:23:14PM +0530, Rajnesh Kumar Siwal wrote:

> The KeepAlive Timeout is set to 2 hours, but we can see that the
> connections start terminating after a few minutes.
> The KeepAlive module does not seems to be behaving as expected.
> The connections are created on port :343.
> The configuration file nginx.conf is :-

[...]

>     keepalive_timeout  7200 7200;

How do you test?

Note: many (most?) browsers won't keep connections open for such a 
long time and will close them after a minute or so.  E.g. Chrome 
unconditionally closes connections after 5 minutes.

Note well: nginx itself might close keepalive connections before 
timeout passes if it's short on worker_connections.  Though I 
suppose you are hitting browsers behaviour instead.

Maxim Dounin


From ne at vbart.ru  Tue Apr 10 13:03:14 2012
From: ne at vbart.ru (Valentin V. Bartenev)
Date: Tue, 10 Apr 2012 17:03:14 +0400
Subject: KeepAlive Not Working as expected for a Reverse Proxy Scenario
In-Reply-To: <CACTshoT9GAZ+LQD9YDbQxpW=nxrVGNEA+7XyRSuBPms5d2Aaxg@mail.gmail.com>
References: <CACTshoT9GAZ+LQD9YDbQxpW=nxrVGNEA+7XyRSuBPms5d2Aaxg@mail.gmail.com>
Message-ID: <201204101703.14125.ne@vbart.ru>

On Tuesday 10 April 2012 14:53:14 Rajnesh Kumar Siwal wrote:
> Hi,
> 
> The KeepAlive Timeout is set to 2 hours, but we can see that the
> connections start terminating after a few minutes.
> The KeepAlive module does not seems to be behaving as expected.

So, what's in the error log? Are you sure that the client doesn't close the 
connection? Some browsers actually do after a minute or few.

[...]
> tcp_keepalive on;
> tcp_keepcnt 10;
> tcp_keepidle 7200;
> tcp_keepintvl 1m;

There are no such directives in the official nginx distribution. But you can set 
SO_KEEPALIVE by parameter of the listen directive:

  listen 443 ssl so_keepalive=2h:1m:10;

Please, take a look at the docs: http://nginx.org/r/listen

  wbr, Valentin V. Bartenev


From mp3geek at gmail.com  Tue Apr 10 13:53:08 2012
From: mp3geek at gmail.com (Ryan Brown)
Date: Wed, 11 Apr 2012 01:53:08 +1200
Subject: could not build the referers_hash
In-Reply-To: <20120410111604.GO13466@mdounin.ru>
References: <CACDr0vF2ZYQ4urpqhgV3dBUaudgZd6NXp0qRbwccEA66ip-vXQ@mail.gmail.com>
	<1333071717.2082.3.camel@portable-evil>
	<CACDr0vErxKZgTYomxjsG++ZMvWu1_R=jEifaO15SAniqAvcjaQ@mail.gmail.com>
	<1333120642.2082.7.camel@portable-evil>
	<CACDr0vF7N4mNx0q63ZEVUfY8wzzTwys2ZvV7PGXECKdbmz=+KA@mail.gmail.com>
	<1333144207.2204.8.camel@portable-evil>
	<CACDr0vF2=-J2cHnVRytmshPz35AMvbncFC3EnR6m2wx=CYDV3A@mail.gmail.com>
	<20120410111604.GO13466@mdounin.ru>
Message-ID: <CAG_yTXxUK=DK1oScBWPcgj1sLFoCGgQz0KLF-aAjvbZPxd6UFA@mail.gmail.com>

> There is configuration directive to adjust this,
> referer_hash_bucket_size. ?You don't need to edit the code.
>
> Maxim Dounin

I got caught by this bug too,

Seems a bit strange why the defaults (if not set) complains that its
not enough.. I can see why people edit the code directly.

If 64 isn't clearly enough when referer_hash_bucket_size isn't set,
why not just up the default to 128 to avoid this unnecessary complaint
by nginx.


From mdounin at mdounin.ru  Tue Apr 10 14:05:00 2012
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Tue, 10 Apr 2012 18:05:00 +0400
Subject: MP4 - start time is out mp4 stts samples
In-Reply-To: <9d8ab115b3e9176b7c4111e95478bbc7.NginxMailingListEnglish@forum.nginx.org>
References: <20111117050603.GA10385@nginx.com>
	<5e238ac1a229e14931b6ea72c107997c.NginxMailingListEnglish@forum.nginx.org>
	<9d8ab115b3e9176b7c4111e95478bbc7.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20120410140500.GV13466@mdounin.ru>

Hello!

On Tue, Apr 10, 2012 at 06:07:40AM -0400, sibsoft wrote:

> Dealing with this trouble again (nginx 1.0.12, 1.0.14):
> 
> 2012/04/10 12:06:59 [alert] 22812#0: *1571 pread() read only 0 of
> 1048576 from "/var/www/cgi-bin/uploads/03/00364/z46j20qdp6gt" while
> sending mp4 to client, client: 95.131.10.226, server: gorillavid.com,
> request: "GET
> /mgobz2aj6gu4rqukwztlpbbrdqzwmtcoid2bk6t6xenwqd3fwvt6z3pyqy/video.mp4?start=1190.84
> HTTP/1.1", host: "85.17.30.216:8182"

The message suggests video.mp4 is corrupted (or just truncated).

> 2012/04/10 12:07:00 [error] 22815#0: *1630 start time is out mp4 stts
> samples in "/var/www/cgi-bin/uploads/03/00001/k2hhvgyfdmmu", client:
> 86.18.83.8, server: gorillavid.com, request: "GET
> /mgobznkw5ku4tqukwyf3nfjxilouln5pl5abqotvftlq7rqggvegaydxva/video.flv?start=0
> HTTP/1.1", host: "85.17.30.216:8182"

The "video.flv" file isn't likely to be an mp4 file.  You have to 
use flv module for flv pseudo-streaming, not mp4.

http://nginx.org/r/flv

Maxim Dounin


From mdounin at mdounin.ru  Tue Apr 10 14:44:28 2012
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Tue, 10 Apr 2012 18:44:28 +0400
Subject: could not build the referers_hash
In-Reply-To: <CAG_yTXxUK=DK1oScBWPcgj1sLFoCGgQz0KLF-aAjvbZPxd6UFA@mail.gmail.com>
References: <CACDr0vF2ZYQ4urpqhgV3dBUaudgZd6NXp0qRbwccEA66ip-vXQ@mail.gmail.com>
	<1333071717.2082.3.camel@portable-evil>
	<CACDr0vErxKZgTYomxjsG++ZMvWu1_R=jEifaO15SAniqAvcjaQ@mail.gmail.com>
	<1333120642.2082.7.camel@portable-evil>
	<CACDr0vF7N4mNx0q63ZEVUfY8wzzTwys2ZvV7PGXECKdbmz=+KA@mail.gmail.com>
	<1333144207.2204.8.camel@portable-evil>
	<CACDr0vF2=-J2cHnVRytmshPz35AMvbncFC3EnR6m2wx=CYDV3A@mail.gmail.com>
	<20120410111604.GO13466@mdounin.ru>
	<CAG_yTXxUK=DK1oScBWPcgj1sLFoCGgQz0KLF-aAjvbZPxd6UFA@mail.gmail.com>
Message-ID: <20120410144428.GW13466@mdounin.ru>

Hello!

On Wed, Apr 11, 2012 at 01:53:08AM +1200, Ryan Brown wrote:

> > There is configuration directive to adjust this,
> > referer_hash_bucket_size. ?You don't need to edit the code.
> >
> > Maxim Dounin
> 
> I got caught by this bug too,
> 
> Seems a bit strange why the defaults (if not set) complains that its
> not enough.. I can see why people edit the code directly.
> 
> If 64 isn't clearly enough when referer_hash_bucket_size isn't set,
> why not just up the default to 128 to avoid this unnecessary complaint
> by nginx.

The default is enough in most cases.  There are some specific 
situations (long values in hash, multiple colliding values) which 
trigger the message, and the message itself explains what to do 
(and we have the page at http://nginx.org/en/docs/hash.html which 
explains details).

We might want to better (read: automatically) handle some 
situation like non-colliding long values, where the only sensible 
solution is to increase ..._hash_bucket_size, but this has nothing 
to do with defaults change.  With any default one will be able to 
write a configuration which triggers the "could not build hash" 
message.

Maxim Dounin


From ashishs.dev at gmail.com  Tue Apr 10 18:16:10 2012
From: ashishs.dev at gmail.com (Ashish S)
Date: Tue, 10 Apr 2012 11:16:10 -0700
Subject: Program hangs when modifying u->buffer while using proxy module
	to talk to upstream backend
In-Reply-To: <20120410121337.GR13466@mdounin.ru>
References: <CA+BpfUrJsVQFBio8Am-0wre-CG79Xm9iiexT4W1GkcpECetY8g@mail.gmail.com>
	<20120410121337.GR13466@mdounin.ru>
Message-ID: <CA+BpfUqNv3L83Tr-U=YnVQmtd9Ox2r1DQgqHpn-EYBnN1X0N-w@mail.gmail.com>

Hi Maxim,

Here is a simple test code i wrote, (for my process header method),
which shows the same behavior.

============
ngx_int_t
ngx_http_my_process_header(ngx_http_request_t *r)
{
    ngx_http_upstream_t *u = r->upstream;

    std::string tmpString ("Test string");
    ngx_memcpy(u->buffer.start, tmpString.c_str(), tmpString.length());
    u->buffer.pos = u->buffer.start;
    u->buffer.last = u->buffer.pos + tmpString.length();

    u->length = u->headers_in.content_length_n = tmpString.length();
    r->headers_out.content_length_n = tmpString.length();
    u->headers_in.status_n = NGX_HTTP_OK;
    u->state->status = NGX_HTTP_OK;

    return NGX_OK;
}
============

Thanks,
Ashish


On Tue, Apr 10, 2012 at 5:13 AM, Maxim Dounin <mdounin at mdounin.ru> wrote:
> Hello!
>
> On Mon, Apr 09, 2012 at 07:02:10PM -0700, Ashish S wrote:
>
>> Hi,
>>
>> I currently use upstream to talk to a back-end, which sends me some
>> response. In my module, i am trying to re-format the plain-text
>> upstream backend response, to XML, and I also add some new data to it,
>> based on an in-memory lookup within my module. ?What would be the best
>> way to do this?
>
> Unless you are working on custom protocol module, you may want to
> use filter module to re-format data instead.
>
>> In my setup, I am able to parse response from upstream (u->buffer),
>> construct a modified string and assign it back to u->buffer, and am I
>> setting ?u->headers_in.content_length_n and
>> r->headers_out.content_length_n correctly. The issue i face is, every
>> *second* request hangs, if i am trying the request through the
>> browser. (However, trying through command-line & curl, works
>> everytime). And I always see this error message in the logs: "client
>> prematurely closed connection, so upstream connection is closed too
>> while sending to client, client: 127.0.0.1". ? What am i doing wrong?
>
> It's hard to say anything without seeing the code, but most likely
> you fail to set u->length properly. ?See e.g. memcached module for
> a simple example of non-buffered protocol handler.
>
> Maxim Dounin
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx


From nginx-forum at nginx.us  Tue Apr 10 20:03:50 2012
From: nginx-forum at nginx.us (spacerobot)
Date: Tue, 10 Apr 2012 16:03:50 -0400 (EDT)
Subject: Nginx returning 414 even when large_client_header_buffers is set
In-Reply-To: <8b19edbb6b2d137c94eae907213f378f.NginxMailingListEnglish@forum.nginx.org>
References: <8b19edbb6b2d137c94eae907213f378f.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <dec42a2b3d7bf00ebf638b3dfdd97392.NginxMailingListEnglish@forum.nginx.org>

It appears that no matter how big I set the value of
large_client_header_buffers to be, nginx just doesn't care of the
setting and still returns 414 on a long request. 

I tried to make it 16k, 32k, 256k, and 512k, etc and POSTing a request
with 1.5k long URL returns 414. It works when I reduce the request URI
length to about 1k, regardless of the large_client_header_buffers value
as well.

I also tried to set proxy_buffers to a large value and didn't help.

Is there any other settings I need to look at to make the nginx box take
longer request URI without returning 414?

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225093,225135#msg-225135


From mdounin at mdounin.ru  Tue Apr 10 20:09:56 2012
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Wed, 11 Apr 2012 00:09:56 +0400
Subject: Nginx returning 414 even when large_client_header_buffers is set
In-Reply-To: <dec42a2b3d7bf00ebf638b3dfdd97392.NginxMailingListEnglish@forum.nginx.org>
References: <8b19edbb6b2d137c94eae907213f378f.NginxMailingListEnglish@forum.nginx.org>
	<dec42a2b3d7bf00ebf638b3dfdd97392.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20120410200956.GB13466@mdounin.ru>

Hello!

On Tue, Apr 10, 2012 at 04:03:50PM -0400, spacerobot wrote:

> It appears that no matter how big I set the value of
> large_client_header_buffers to be, nginx just doesn't care of the
> setting and still returns 414 on a long request. 
> 
> I tried to make it 16k, 32k, 256k, and 512k, etc and POSTing a request
> with 1.5k long URL returns 414. It works when I reduce the request URI
> length to about 1k, regardless of the large_client_header_buffers value
> as well.
> 
> I also tried to set proxy_buffers to a large value and didn't help.
> 
> Is there any other settings I need to look at to make the nginx box take
> longer request URI without returning 414?

Most likely you are trying to configure 
client_header_buffer_size/large_client_header_buffers in a pure 
virtual server{}.  This won't work as request headers parsing 
happens before Host header is known (and virtual server is 
selected), hence parseing happens in a context of the default 
server for a listen socket.

You have to configure client_header_buffer_size/large_client_header_buffers 
in a default server (or at http level).

Maxim Dounin


From nginx-forum at nginx.us  Tue Apr 10 20:26:25 2012
From: nginx-forum at nginx.us (spacerobot)
Date: Tue, 10 Apr 2012 16:26:25 -0400 (EDT)
Subject: Nginx returning 414 even when large_client_header_buffers is set
In-Reply-To: <20120410200956.GB13466@mdounin.ru>
References: <20120410200956.GB13466@mdounin.ru>
Message-ID: <dcf20d7a54ecd5c7b6b0aa75817f5602.NginxMailingListEnglish@forum.nginx.org>

> Most likely you are trying to configure 
> client_header_buffer_size/large_client_header_buff
> ers in a pure 
> virtual server{}.  This won't work as request
> headers parsing 
> happens before Host header is known (and virtual
> server is 
> selected), hence parseing happens in a context of
> the default 
> server for a listen socket.
> 
> You have to configure
> client_header_buffer_size/large_client_header_buff
> ers 
> in a default server (or at http level).
> 

I tried to set it in the http context of the main /etc/nginx/nginx.conf
file:

http {
  large_client_header_buffers  8  512k;

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

And my server contexts are defined in a conf file in sites-enabled. But
large_client_header_buffers still appears to be ignored.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225093,225138#msg-225138


From massimo.ferrari at gmail.com  Tue Apr 10 20:32:31 2012
From: massimo.ferrari at gmail.com (massimo ferrari)
Date: Tue, 10 Apr 2012 22:32:31 +0200
Subject: Redis2_pass with <host>:<port> from $args fail [maxf3r]
Message-ID: <CAA-=2De5_CqgTtbzF5=Hn+Geu4dSoYsG99Ap=KjDwocz7FALYg@mail.gmail.com>

I'm trying to address several redis instances, dispatching it on a
host:port basis (one single backend location)

here part of my nginx.conf location
...
location /redis_backend {
internal;
set_unescape_uri $verb $arg_verb;
set_unescape_uri $key $arg_key;
set_unescape_uri $r_host $arg_host;
set_unescape_uri $r_port $arg_port;
redis2_query $verb $key;
redis2_pass $r_host:$r_port;
}
...

I call /redis_backend location from a lua like this:

....
r_instance["redis_host"] = '127.0.0.1'
r_instance["redis_port"] = '33001'
....
local res = ngx.location.capture (
"/redis_backend",
{ args =
{ verb = r_verb,
key = redis_key,
host = r_instance["redis_host"],
port = r_instance["redis_port"],
}
}
)
...

in nginx error log (i use the openresty suite ngx_openresty/1.0.10.44), I
got this error

*2012/04/07 19:25:03 [error] 11715#0: *1 redis2: upstream "127.0.0.1:33001"
not found, client: 127.0.0.1, server: ....*

I'm sure redis instance is up&running. I'm able to connect to her via
redis-cli or, simply setting redis2_pass 127.0.0.1:33001;

ciao
massimo
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20120410/2cf6dfb0/attachment.html>

From ashishs.dev at gmail.com  Tue Apr 10 21:03:09 2012
From: ashishs.dev at gmail.com (Ashish S)
Date: Tue, 10 Apr 2012 14:03:09 -0700
Subject: Program hangs when modifying u->buffer while using proxy module
	to talk to upstream backend
In-Reply-To: <CA+BpfUqNv3L83Tr-U=YnVQmtd9Ox2r1DQgqHpn-EYBnN1X0N-w@mail.gmail.com>
References: <CA+BpfUrJsVQFBio8Am-0wre-CG79Xm9iiexT4W1GkcpECetY8g@mail.gmail.com>
	<20120410121337.GR13466@mdounin.ru>
	<CA+BpfUqNv3L83Tr-U=YnVQmtd9Ox2r1DQgqHpn-EYBnN1X0N-w@mail.gmail.com>
Message-ID: <CA+BpfUr_aNktaQ5dxJPp11hoqC23ERdmiD+Fo1+SoV5_W1_pVQ@mail.gmail.com>

Hi, I figured the issue.   Maxim's intuition was right. u->length was
not set correctly (the u->length set in the code i copy-pasted
earlier, was modified later by a filter method). After fixing the
issue, the code works correctly, no more hangs.

Thanks,
Ashish


On Tue, Apr 10, 2012 at 11:16 AM, Ashish S <ashishs.dev at gmail.com> wrote:
> Hi Maxim,
>
> Here is a simple test code i wrote, (for my process header method),
> which shows the same behavior.
>
> ============
> ngx_int_t
> ngx_http_my_process_header(ngx_http_request_t *r)
> {
> ? ?ngx_http_upstream_t *u = r->upstream;
>
> ? ?std::string tmpString ("Test string");
> ? ?ngx_memcpy(u->buffer.start, tmpString.c_str(), tmpString.length());
> ? ?u->buffer.pos = u->buffer.start;
> ? ?u->buffer.last = u->buffer.pos + tmpString.length();
>
> ? ?u->length = u->headers_in.content_length_n = tmpString.length();
> ? ?r->headers_out.content_length_n = tmpString.length();
> ? ?u->headers_in.status_n = NGX_HTTP_OK;
> ? ?u->state->status = NGX_HTTP_OK;
>
> ? ?return NGX_OK;
> }
> ============
>
> Thanks,
> Ashish
>
>
> On Tue, Apr 10, 2012 at 5:13 AM, Maxim Dounin <mdounin at mdounin.ru> wrote:
>> Hello!
>>
>> On Mon, Apr 09, 2012 at 07:02:10PM -0700, Ashish S wrote:
>>
>>> Hi,
>>>
>>> I currently use upstream to talk to a back-end, which sends me some
>>> response. In my module, i am trying to re-format the plain-text
>>> upstream backend response, to XML, and I also add some new data to it,
>>> based on an in-memory lookup within my module. ?What would be the best
>>> way to do this?
>>
>> Unless you are working on custom protocol module, you may want to
>> use filter module to re-format data instead.
>>
>>> In my setup, I am able to parse response from upstream (u->buffer),
>>> construct a modified string and assign it back to u->buffer, and am I
>>> setting ?u->headers_in.content_length_n and
>>> r->headers_out.content_length_n correctly. The issue i face is, every
>>> *second* request hangs, if i am trying the request through the
>>> browser. (However, trying through command-line & curl, works
>>> everytime). And I always see this error message in the logs: "client
>>> prematurely closed connection, so upstream connection is closed too
>>> while sending to client, client: 127.0.0.1". ? What am i doing wrong?
>>
>> It's hard to say anything without seeing the code, but most likely
>> you fail to set u->length properly. ?See e.g. memcached module for
>> a simple example of non-buffered protocol handler.
>>
>> Maxim Dounin
>>
>> _______________________________________________
>> nginx mailing list
>> nginx at nginx.org
>> http://mailman.nginx.org/mailman/listinfo/nginx


From nginx-forum at nginx.us  Tue Apr 10 23:45:21 2012
From: nginx-forum at nginx.us (spacerobot)
Date: Tue, 10 Apr 2012 19:45:21 -0400 (EDT)
Subject: Nginx returning 414 even when large_client_header_buffers is set
In-Reply-To: <dcf20d7a54ecd5c7b6b0aa75817f5602.NginxMailingListEnglish@forum.nginx.org>
References: <20120410200956.GB13466@mdounin.ru>
	<dcf20d7a54ecd5c7b6b0aa75817f5602.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <f4f12b08c913420b5636a5355bfa0b4e.NginxMailingListEnglish@forum.nginx.org>

Hello,

spacerobot Wrote:
-------------------------------------------------------
> > Most likely you are trying to configure 
> >
> client_header_buffer_size/large_client_header_buff
> 
> > ers in a pure 
> > virtual server{}.  This won't work as request
> > headers parsing 
> > happens before Host header is known (and
> virtual
> > server is 
> > selected), hence parseing happens in a context
> of
> > the default 
> > server for a listen socket.
> > 
> > You have to configure
> >
> client_header_buffer_size/large_client_header_buff
> 
> > ers 
> > in a default server (or at http level).
> > 
> 
> I tried to set it in the http context of the main
> /etc/nginx/nginx.conf file:
> 
> http {
>   large_client_header_buffers  8  512k;
> 
>   ...
>   include /etc/nginx/sites-enabled/*;
> }
> 
> And my server contexts are defined in a conf file
> in sites-enabled. But large_client_header_buffers
> still appears to be ignored.

I further tried the following things, which didn't help either:
- took out most of the configs in my nginx.conf and now I have the
simplest configuration possible, as below.
- upgraded to 1.0.14 stable from 1.0.12 development.

My new simplified nginx.conf:

worker_processes  2;
daemon off;

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

events {
  worker_connections  2048;
}

http {
  client_header_buffer_size 256k;
  large_client_header_buffers 8 1024k;

  upstream foo.com {
    server 10.0.0.99:16234;
    server 10.0.0.20:16234;
  }

  server {
    listen   443;

    ssl on;
    ssl_certificate /etc/foo.com.crt;
    ssl_certificate_key /etc//foo.com.key;

    server_name  foo;

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

    set $foo "foo.com";

    location / {
      proxy_set_header x-foo-rid $request_id;
      proxy_redirect http://$foo /;
      proxy_pass http://$foo;
    }
  }
}

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225093,225146#msg-225146


From ianevans at digitalhit.com  Tue Apr 10 23:52:56 2012
From: ianevans at digitalhit.com (Ian M. Evans)
Date: Tue, 10 Apr 2012 19:52:56 -0400
Subject: Non-root install of Wordpress in an already running nginx site
Message-ID: <55bffb6bd356c42141918488c2b8967c.squirrel@www.digitalhit.com>

Been running nginx on our server for years. php-fpm, multiple location
sections, etc., all working fine.

However, for one project we're launching my partner wants to run a
wordpress blog from a subdirecory, i.e., example.com/ournewblog

Can anyone point me to an example of a location block for the wordpress
blog that handles all the wordpress needs while not futzing with all the
other locations or causing php problems elsewhere?

Thanks!


From edho at myconan.net  Wed Apr 11 00:02:56 2012
From: edho at myconan.net (Edho Arief)
Date: Wed, 11 Apr 2012 07:02:56 +0700
Subject: Non-root install of Wordpress in an already running nginx site
In-Reply-To: <55bffb6bd356c42141918488c2b8967c.squirrel@www.digitalhit.com>
References: <55bffb6bd356c42141918488c2b8967c.squirrel@www.digitalhit.com>
Message-ID: <CAAtReCnptc=Vut6jZ5WqMU=Nf_Zm6bna01R86E=wGv=L5wnbfg@mail.gmail.com>

2012/4/11 Ian M. Evans <ianevans at digitalhit.com>:
> Been running nginx on our server for years. php-fpm, multiple location
> sections, etc., all working fine.
>
> However, for one project we're launching my partner wants to run a
> wordpress blog from a subdirecory, i.e., example.com/ournewblog
>
> Can anyone point me to an example of a location block for the wordpress
> blog that handles all the wordpress needs while not futzing with all the
> other locations or causing php problems elsewhere?
>

location /ournewblog/ {
  root ...;
  try_files $uri /ournewblog/index.php?q=$uri;
  location ~ \.php$ {
    fastcgi_pass ...
    ...
  }
}

Hopefully you don't have regex based location on server { } block.
Otherwise you can try location ~ ^/ournewblog/ instead.


From ianevans at digitalhit.com  Wed Apr 11 00:22:45 2012
From: ianevans at digitalhit.com (Ian M. Evans)
Date: Tue, 10 Apr 2012 20:22:45 -0400
Subject: Non-root install of Wordpress in an already running nginx site
In-Reply-To: <CAAtReCnptc=Vut6jZ5WqMU=Nf_Zm6bna01R86E=wGv=L5wnbfg@mail.gmail.com>
References: <55bffb6bd356c42141918488c2b8967c.squirrel@www.digitalhit.com>
	<CAAtReCnptc=Vut6jZ5WqMU=Nf_Zm6bna01R86E=wGv=L5wnbfg@mail.gmail.com>
Message-ID: <2ae0445f93b65b54c15ed98cd78c6450.squirrel@www.digitalhit.com>

On Tue, April 10, 2012 8:02 pm, Edho Arief wrote:
> location /ournewblog/ {
>   root ...;
>   try_files $uri /ournewblog/index.php?q=$uri;
>   location ~ \.php$ {
>     fastcgi_pass ...
>     ...
>   }
> }
>
> Hopefully you don't have regex based location on server { } block.
> Otherwise you can try location ~ ^/ournewblog/ instead.

I've always been a bit unsure of handling nginx locations so how can I
tell if I have a "regex based location"?

Most of my locations are like:

location / {
index index.shtml index.php;
}
location ~ \.(shtml|php|inc)$ {
fastcgi_pass 127.0.0.1:10004;
}
# start of extensionless stuff
location ~ ^/(bunch|of|site|sections)(/.*$|$) {
...
}
location ~ /(more|extensionless|sections)(/|$) {
...
}

Thanks.


From edho at myconan.net  Wed Apr 11 00:26:52 2012
From: edho at myconan.net (Edho Arief)
Date: Wed, 11 Apr 2012 07:26:52 +0700
Subject: Non-root install of Wordpress in an already running nginx site
In-Reply-To: <2ae0445f93b65b54c15ed98cd78c6450.squirrel@www.digitalhit.com>
References: <55bffb6bd356c42141918488c2b8967c.squirrel@www.digitalhit.com>
	<CAAtReCnptc=Vut6jZ5WqMU=Nf_Zm6bna01R86E=wGv=L5wnbfg@mail.gmail.com>
	<2ae0445f93b65b54c15ed98cd78c6450.squirrel@www.digitalhit.com>
Message-ID: <CAAtReCn0X124XQwsgSBax5b=QT+uqwVOAWNSZMmkthtiuqD1gw@mail.gmail.com>

2012/4/11 Ian M. Evans <ianevans at digitalhit.com>:
> On Tue, April 10, 2012 8:02 pm, Edho Arief wrote:
>> location /ournewblog/ {
>> ? root ...;
>> ? try_files $uri /ournewblog/index.php?q=$uri;
>> ? location ~ \.php$ {
>> ? ? fastcgi_pass ...
>> ? ? ...
>> ? }
>> }
>>
>> Hopefully you don't have regex based location on server { } block.
>> Otherwise you can try location ~ ^/ournewblog/ instead.
>
> I've always been a bit unsure of handling nginx locations so how can I
> tell if I have a "regex based location"?
>
> Most of my locations are like:
>
> location / {
> index index.shtml index.php;
> }
> location ~ \.(shtml|php|inc)$ {
> fastcgi_pass 127.0.0.1:10004;
> }

this is regex based location block. It gets higher priority when
accessing /ournewblog/index.php .

You need location ~ ^/ournewblog/ instead and put it as the first
location block (before other locations).


From nginx-forum at nginx.us  Wed Apr 11 01:36:22 2012
From: nginx-forum at nginx.us (CheezItMan)
Date: Tue, 10 Apr 2012 21:36:22 -0400 (EDT)
Subject: Nginx and Wordpress Super Cache
Message-ID: <be339508cbde58be9bdbad24d8709b5e.NginxMailingListEnglish@forum.nginx.org>

'm trying to setup a new server with nginx instead of LAMP using
php-fpm. I've got wordpress setup and running, but I'm at a loss to
figure out how to get Wordpress Super Cache working. I'd appreciate any
help. I'm also trying to figure out how to get rid of the if statements
in the wordpress Multisite configuration file. I do know about the
ifisevil page.

I'd taken my initial configuration from:
http://codex.wordpress.org/Nginx


My Sites-available file
---
# Redirect everything to the main site.
server {
listen 80 default;
server_name *.students.scisdragons.net;
rewrite ^ http://students.scisdragons.net$request_uri permanent;
}

server {
server_name students.scisdragons.net;
root /var/www/students.scisdragons.net;

include global/restrictions.conf;
include global/wordpress-ms-subdir.conf;
}


---


My global wordpress file
---

# WordPress multisite subdirectory rules.
# Designed to be included in any server {} block.

# This order might seem weird - this is attempted to match last if rules
below fail.
# http://wiki.nginx.org/HttpCoreModule
location / {
try_files $uri $uri/ /index.php?$args;
}

# Add trailing slash to */wp-admin requests.
rewrite /wp-admin$ $scheme://$host$uri/ permanent;

# Directives to send expires headers and turn off 404 error logging.
location ~* \.(js|css|png|jpg|jpeg|gif|ico)$ {
expires 48h;
log_not_found off;
}

# Pass uploaded files to wp-includes/ms-files.php.
rewrite /files/$ /index.php last;

# For multisite: Use a caching plugin/script that creates symlinks to
the correct subdirectory structure to get some performance gains.
set $cachetest
"$document_root/wp-content/cache/ms-filemap/${host}${uri}";
if ($uri ~ /$) {
set $cachetest "";
}
if (-f $cachetest) {
# Rewrites the URI and stops rewrite processing so it doesn't start over
and attempt to pass it to the next rule.
rewrite ^ /wp-content/cache/ms-filemap/${host}${uri} break;
}

if ($uri !~ wp-content/plugins) {
rewrite /files/(.+)$ /wp-includes/ms-files.php?file=$1 last;
}

# Uncomment one of the lines below for the appropriate caching plugin
(if used).
# include global/wordpress-ms-subdir-wp-super-cache.conf;
# include global/wordpress-ms-subdir-w3-total-cache.conf;

# Rewrite multisite '.../wp-.*' and '.../*.php'.
if (!-e $request_filename) {
rewrite ^/[_0-9a-zA-Z-]+(/wp-.*) $1 last;
rewrite ^/[_0-9a-zA-Z-]+(/.*\.php)$ $1 last;
}

# Pass all .php files onto a php-fpm/php-fcgi server.
location ~ \.php$ {
# Zero-day exploit defense.
# http://forum.nginx.org/read.php?2,88845,page=3
# Won't work properly (404 error) if the file is not stored on this
server, which is entirely possible with php-fpm/php-fcgi.
# Comment the 'try_files' line out if you set up php-fpm/php-fcgi on
another machine. And then cross your fingers that you won't get hacked.
try_files $uri =404;

fastcgi_split_path_info ^(.+\.php)(/.+)$;
include fastcgi_params;
fastcgi_index index.php;
fastcgi_param SCRIPT_FILENAME $document_root$fastcgi_script_name;
# fastcgi_intercept_errors on;
fastcgi_pass php;
}

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225151,225151#msg-225151


From nginx-forum at nginx.us  Wed Apr 11 01:38:07 2012
From: nginx-forum at nginx.us (activa)
Date: Tue, 10 Apr 2012 21:38:07 -0400 (EDT)
Subject: nginx + FollowSymLinks owner verification
Message-ID: <d28b11bfc06a3323b9fab10b68e80d1f.NginxMailingListEnglish@forum.nginx.org>

i have made a test security and i have found the fallowing : 

we have the fallowing synoposis : 


[root at server4 www]# ls -alh 
total 144K 
drwxr-x---  6 usertest nobody   4.0K Apr 10 20:09 . 
drwx--x--x 13 usertest usertest 4.0K Apr  7 02:16 .. 
-rw-r--r--  1 usertest usertest  184 Apr  6 21:29 .htaccess 
lrwxrwxrwx  1 usertest usertest   38 Apr  6 22:48 im1.txt ->
/home/anotheruser/public_html/config.php 
-rw-r--r--  1 usertest usertest    3 May  3  2011 index.html  

i can read the file of other user without any probleme !!! 

normally it should verify the ownership of files before handel them .

NOTE , i use nginx as proxy of apache . when i use just apache a get a
403 error (this is a normal result) , with nginx i can read the file ,
becuase nginx hadler the static files + images 


anyfix for this ?

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225152,225152#msg-225152


From edho at myconan.net  Wed Apr 11 01:42:58 2012
From: edho at myconan.net (Edho Arief)
Date: Wed, 11 Apr 2012 08:42:58 +0700
Subject: nginx + FollowSymLinks owner verification
In-Reply-To: <d28b11bfc06a3323b9fab10b68e80d1f.NginxMailingListEnglish@forum.nginx.org>
References: <d28b11bfc06a3323b9fab10b68e80d1f.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <CAAtReC=VUXz4O85HWiRw6GvaytNvaXnc2cLOT+xhuER6t04i5Q@mail.gmail.com>

2012/4/11 activa <nginx-forum at nginx.us>:
>
> anyfix for this ?
>

disable_symlinks if_not_owner;


From cliff at develix.com  Wed Apr 11 01:47:46 2012
From: cliff at develix.com (Cliff Wells)
Date: Tue, 10 Apr 2012 18:47:46 -0700
Subject: Non-root install of Wordpress in an already running nginx site
In-Reply-To: <2ae0445f93b65b54c15ed98cd78c6450.squirrel@www.digitalhit.com>
References: <55bffb6bd356c42141918488c2b8967c.squirrel@www.digitalhit.com>
	<CAAtReCnptc=Vut6jZ5WqMU=Nf_Zm6bna01R86E=wGv=L5wnbfg@mail.gmail.com>
	<2ae0445f93b65b54c15ed98cd78c6450.squirrel@www.digitalhit.com>
Message-ID: <4F84E2C2.6040603@develix.com>

On 04/10/2012 05:22 PM, Ian M. Evans wrote:
> On Tue, April 10, 2012 8:02 pm, Edho Arief wrote:
>> location /ournewblog/ {
>>    root ...;
>>    try_files $uri /ournewblog/index.php?q=$uri;
>>    location ~ \.php$ {
>>      fastcgi_pass ...
>>      ...
>>    }
>> }
>>
>> Hopefully you don't have regex based location on server { } block.
>> Otherwise you can try location ~ ^/ournewblog/ instead.
> I've always been a bit unsure of handling nginx locations so how can I
> tell if I have a "regex based location"?


http://nginx.org/en/docs/http/request_processing.html

Regards,
Cliff


From nginx-forum at nginx.us  Wed Apr 11 02:25:48 2012
From: nginx-forum at nginx.us (activa)
Date: Tue, 10 Apr 2012 22:25:48 -0400 (EDT)
Subject: nginx + FollowSymLinks owner verification
In-Reply-To: <d28b11bfc06a3323b9fab10b68e80d1f.NginxMailingListEnglish@forum.nginx.org>
References: <d28b11bfc06a3323b9fab10b68e80d1f.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <7a33c7c8e452dfc4a668b22d7a1c0b44.NginxMailingListEnglish@forum.nginx.org>

where shod i add this ? in nginx conf or in vhost conf ?

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225152,225155#msg-225155


From nginx-forum at nginx.us  Wed Apr 11 02:46:31 2012
From: nginx-forum at nginx.us (activa)
Date: Tue, 10 Apr 2012 22:46:31 -0400 (EDT)
Subject: nginx + FollowSymLinks owner verification
In-Reply-To: <d28b11bfc06a3323b9fab10b68e80d1f.NginxMailingListEnglish@forum.nginx.org>
References: <d28b11bfc06a3323b9fab10b68e80d1f.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <3a58114175502301cd1c9d9be4b64605.NginxMailingListEnglish@forum.nginx.org>

as i have found n is only available in developper version , not the
stable version .

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225152,225156#msg-225156


From nginx-forum at nginx.us  Wed Apr 11 04:18:48 2012
From: nginx-forum at nginx.us (activa)
Date: Wed, 11 Apr 2012 00:18:48 -0400 (EDT)
Subject: nginx + FollowSymLinks owner verification
In-Reply-To: <d28b11bfc06a3323b9fab10b68e80d1f.NginxMailingListEnglish@forum.nginx.org>
References: <d28b11bfc06a3323b9fab10b68e80d1f.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <04a7dc7ec690ac6de5bf7dc953d31448.NginxMailingListEnglish@forum.nginx.org>

installed nginx-1.1.18 and probleme was resolved .

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225152,225158#msg-225158


From nginx-forum at nginx.us  Wed Apr 11 04:50:59 2012
From: nginx-forum at nginx.us (activa)
Date: Wed, 11 Apr 2012 00:50:59 -0400 (EDT)
Subject: all visitor have same IP (my server IP)
Message-ID: <dfc3399bb73ac5a3e13c98115020c4e2.NginxMailingListEnglish@forum.nginx.org>

i'm using nginx as proxy fro apache .

my users ask why 90% of thier vistors has the same IP of the servers .

and some php application block the server IP adresse as flood ip . 

[root at server4 ]# netstat -lanp | grep :8081 | awk '{print $5}' | cut -d:
-f1 | sort | uniq -c | sort -n
      1 0.0.0.0
      3 127.0.0.1
   1588 xxx.xxx.xxx.xxx (this is server ip) 


[root at server4 ]# netstat -lanp | grep :80 | awk '{print $5}' | cut -d:
-f1 | sort | uniq -c | sort -n
      1 94.111.122.97
      2 150.70.172.203
      2 197.162.58.71
      2 41.141.1.104
      2 41.250.191.48
      2 41.250.200.157
      3 0.0.0.0
      3 196.217.214.235
      3 41.137.75.212
      3 41.248.144.207
      3 41.92.15.186
      4 41.250.91.158
      5 196.12.254.12
      6 37.105.108.27
      6 41.143.100.215
      7 41.141.169.127
      9 197.194.27.175
      9 67.167.101.181
   1679 xxx.xxx.xxx.xxx (this is server ip) 

the RPAF configurations well done (include the ip servers) . 

nginx 1.1.18  config 

user  nobody;
# no need for more workers in the proxy mode
worker_processes  2;
error_log  /var/log/nginx/error.log info;
worker_rlimit_nofile 20480;
events {
 worker_connections 5120; # increase for busier servers
 use epoll; # you should use epoll here for Linux kernels 2.6.x
}
http {
 server_name_in_redirect off;
 server_names_hash_max_size 10240;
 server_names_hash_bucket_size 1024;
 include    mime.types;
 default_type  application/octet-stream;
 server_tokens off;
 sendfile on;
 tcp_nopush on;
 tcp_nodelay on;
 keepalive_timeout  5;
 gzip on;
 gzip_vary on;
 gzip_disable "MSIE [1-6]\.";
 gzip_proxied any;
 gzip_http_version 1.1;
 gzip_min_length  1000;
 gzip_comp_level  6;
 gzip_buffers  16 8k;
# You can remove image/png image/x-icon image/gif image/jpeg if you have
slow CPU
 gzip_types    text/plain text/xml text/css application/x-javascript
application/xml image/png image/x-icon image/gif image/jpeg
application/xml+rss text/javascript application/atom+xml;
 ignore_invalid_headers on;
 client_header_timeout  3m;
 client_body_timeout 3m;
 send_timeout     3m;
 reset_timedout_connection on;
 connection_pool_size  256;
 client_header_buffer_size 256k;
 large_client_header_buffers 4 256k;
 client_max_body_size 200M; 
 client_body_buffer_size 128k;
 request_pool_size  32k;
 output_buffers   4 32k;
 postpone_output  1460;
 proxy_temp_path  /tmp/nginx_proxy/;
 client_body_in_file_only on;
 disable_symlinks if_not_owner;
 log_format bytes_log "$msec $bytes_sent .";
 include "/etc/nginx/vhosts/*";
}



is this have relation to mod realip ? 

any solution to make nginx forward the proccess with the visitor ip not
with the server ip ?

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225160,225160#msg-225160


From ru at nginx.com  Wed Apr 11 05:24:28 2012
From: ru at nginx.com (Ruslan Ermilov)
Date: Wed, 11 Apr 2012 09:24:28 +0400
Subject: all visitor have same IP (my server IP)
In-Reply-To: <dfc3399bb73ac5a3e13c98115020c4e2.NginxMailingListEnglish@forum.nginx.org>
References: <dfc3399bb73ac5a3e13c98115020c4e2.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20120411052428.GA89268@lo0.su>

On Wed, Apr 11, 2012 at 12:50:59AM -0400, activa wrote:
> i'm using nginx as proxy fro apache .
> 
> my users ask why 90% of thier vistors has the same IP of the servers .
> 
> and some php application block the server IP adresse as flood ip . 
> 
> [root at server4 ]# netstat -lanp | grep :8081 | awk '{print $5}' | cut -d:
> -f1 | sort | uniq -c | sort -n
>       1 0.0.0.0
>       3 127.0.0.1
>    1588 xxx.xxx.xxx.xxx (this is server ip) 
> 
> 
> [root at server4 ]# netstat -lanp | grep :80 | awk '{print $5}' | cut -d:
> -f1 | sort | uniq -c | sort -n
>       1 94.111.122.97
>       2 150.70.172.203
>       2 197.162.58.71
>       2 41.141.1.104
>       2 41.250.191.48
>       2 41.250.200.157
>       3 0.0.0.0
>       3 196.217.214.235
>       3 41.137.75.212
>       3 41.248.144.207
>       3 41.92.15.186
>       4 41.250.91.158
>       5 196.12.254.12
>       6 37.105.108.27
>       6 41.143.100.215
>       7 41.141.169.127
>       9 197.194.27.175
>       9 67.167.101.181
>    1679 xxx.xxx.xxx.xxx (this is server ip) 
> 
> the RPAF configurations well done (include the ip servers) . 
> 
> nginx 1.1.18  config 
> 
> user  nobody;
> # no need for more workers in the proxy mode
> worker_processes  2;
> error_log  /var/log/nginx/error.log info;
> worker_rlimit_nofile 20480;
> events {
>  worker_connections 5120; # increase for busier servers
>  use epoll; # you should use epoll here for Linux kernels 2.6.x
> }
> http {
>  server_name_in_redirect off;
>  server_names_hash_max_size 10240;
>  server_names_hash_bucket_size 1024;
>  include    mime.types;
>  default_type  application/octet-stream;
>  server_tokens off;
>  sendfile on;
>  tcp_nopush on;
>  tcp_nodelay on;
>  keepalive_timeout  5;
>  gzip on;
>  gzip_vary on;
>  gzip_disable "MSIE [1-6]\.";
>  gzip_proxied any;
>  gzip_http_version 1.1;
>  gzip_min_length  1000;
>  gzip_comp_level  6;
>  gzip_buffers  16 8k;
> # You can remove image/png image/x-icon image/gif image/jpeg if you have
> slow CPU
>  gzip_types    text/plain text/xml text/css application/x-javascript
> application/xml image/png image/x-icon image/gif image/jpeg
> application/xml+rss text/javascript application/atom+xml;
>  ignore_invalid_headers on;
>  client_header_timeout  3m;
>  client_body_timeout 3m;
>  send_timeout     3m;
>  reset_timedout_connection on;
>  connection_pool_size  256;
>  client_header_buffer_size 256k;
>  large_client_header_buffers 4 256k;
>  client_max_body_size 200M; 
>  client_body_buffer_size 128k;
>  request_pool_size  32k;
>  output_buffers   4 32k;
>  postpone_output  1460;
>  proxy_temp_path  /tmp/nginx_proxy/;
>  client_body_in_file_only on;
>  disable_symlinks if_not_owner;
>  log_format bytes_log "$msec $bytes_sent .";
>  include "/etc/nginx/vhosts/*";
> }
> 
> 
> 
> is this have relation to mod realip ? 
> 
> any solution to make nginx forward the proccess with the visitor ip not
> with the server ip ?

proxy_set_header X-Real-IP       $remote_addr;
proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;

http://nginx.org/en/docs/http/ngx_http_proxy_module.html#proxy_set_header
http://nginx.org/en/docs/http/ngx_http_proxy_module.html#variables


From rajnesh.siwal at gmail.com  Wed Apr 11 05:40:18 2012
From: rajnesh.siwal at gmail.com (Rajnesh Kumar Siwal)
Date: Wed, 11 Apr 2012 11:10:18 +0530
Subject: KeepAlive Not Working as expected for a Reverse Proxy Scenario
In-Reply-To: <CACTshoT9GAZ+LQD9YDbQxpW=nxrVGNEA+7XyRSuBPms5d2Aaxg@mail.gmail.com>
References: <CACTshoT9GAZ+LQD9YDbQxpW=nxrVGNEA+7XyRSuBPms5d2Aaxg@mail.gmail.com>
Message-ID: <CACTshoSFg4Syb6pYYqK0a7hdE6mSnpR8CkZASog2rSg9WoNSiA@mail.gmail.com>

Thanks Valentin, but it still is not accepting the parameters:-

[root at api1.nos2 conf]# /etc/init.d/nginxfromsrc start
Starting nginx: nginx: [emerg] invalid parameter "so_keepalive=2h:1m:10" in
/usr/local/nginx-1.0.14/conf/nginx.conf:
96

Secondly, I don't see the message that the client is terminating the
connections.
In case I use elinks to open the URL and close it, it displays me that the
KeepAlive connection is terminated by the client.
So, I am pretty sure that the connections are not terminated by the client.

Additionally, we are using CURL library to check it.
We have a simulator that create the connections and sends data using CURL
library.
The same scenario is working fine with Apache and the connections are
persistent.


On Tue, Apr 10, 2012 at 4:23 PM, Rajnesh Kumar Siwal <
rajnesh.siwal at gmail.com> wrote:

> Hi,
>
> The KeepAlive Timeout is set to 2 hours, but we can see that the
> connections start terminating after a few minutes.
> The KeepAlive module does not seems to be behaving as expected.
> The connections are created on port :343.
> The configuration file nginx.conf is :-
> -------------------------------------------------------------------
> user              nginx;
> worker_processes  16;
> error_log  /var/log/nginx/error.log info;
> pid        /var/run/nginx.pid;
> events {
>         #worker_connections  16192;
>         #worker_connections  161920;
>         worker_connections  40960;
>         use epoll;
>         multi_accept on;
> }
> http {
>     include       /etc/nginx/mime.types;
>     default_type  application/octet-stream;
>     log_format  main  '$remote_addr - $remote_user [$time_local]
> "$request" '
>                       '$status $body_bytes_sent "$http_referer" '
>                       '"$http_user_agent" "$http_x_forwarded_for"';
>     access_log  /var/log/nginx/access.log  main;
>     sendfile        on;
>     #tcp_nopush     on;
>     keepalive_timeout  7200 7200;
> tcp_keepalive on;
> tcp_keepcnt 10;
> tcp_keepidle 3600;
> tcp_keepintvl 1m;
>     #gzip  on;
>
>     #
>     # The default server
>     #
>         server {
>                 listen 4443;
>                 server_name _;
>                 location /nginx_status {
>   # copied from
> http://blog.kovyrin.net/2006/04/29/monitoring-nginx-with-rrdtool/
>                 stub_status on;
>                 access_log   off;
>                 }
>         }
>     server {
>         listen       80;
>         server_name  _;
>     #   keepalive_timeout  7200 7200;
>         #charset koi8-r;
>         access_log  logs/host.access.log  main;
>         location / {
>                 proxy_pass      http://172.31.253.50:180/;
>                 include         /etc/nginx/proxy.conf;
>         }
>         }
>         server {
>         listen 443 ssl;
>         ssl on;
>         ssl_verify_client on;
>         server_name _;
>         ssl_certificate      /home/rsiwal/cps.chargepoint.net.crt;
>         #ssl_client_certificate  /etc/httpd/certs/
> chargepoint.net/server/certs/cps.chargepoint.net.crt;
>         ssl_client_certificate  /home/rsiwal/chargepoint.net.crt;
>         ssl_certificate_key  /home/rsiwal/cps.chargepoint.net.pem;
>         location / {
>                  proxy_pass      http://172.31.253.50:1443/;
>                 include         /etc/nginx/proxy.conf;
>         }
>         }
>         server {
>         listen 343 ssl;
>         ssl on;
>         ssl_verify_client on;
>         server_name _;
>         #ssl_session_cache shared:SSL:40m;    # Works fine upto 15K
>         ssl_session_cache shared:SSL:70m;
>         #######ssl_session_timeout 5m;     #
>         ssl_session_timeout 120m;      #
> tcp_keepalive on;
> tcp_keepcnt 10;
> tcp_keepidle 7200;
> tcp_keepintvl 1m;
>         access_log      logs/ssl-access.log main;
>         error_log       logs/ssl-error.log debug;
>         keepalive_requests    10000;
>         #keepalive_disable none
>         sendfile on;
>         tcp_nopush on;
>         tcp_nodelay on;
>         client_header_timeout 300;
>         client_body_timeout 300;
>         ssl_certificate      /home/rsiwal/cps.chargepoint.net.crt;
>         ssl_certificate_key  /home/rsiwal/cps.chargepoint.net.pem;
>         ssl_client_certificate  /home/rsiwal/chargepoint.net.crt;
>         location / {
>                  proxy_pass      http://172.31.253.50:1343/;
>                 include         /etc/nginx/proxy.conf;
>         }
>         }
>     include /etc/nginx/conf.d/*.conf;
> }
>
> -----------------------------------------------------------------------------------------------------------------------------
>
> Thanks in advance
>
> --
> Regards,
> Rajnesh Kumar Siwal
>



-- 
Regards,
Rajnesh Kumar Siwal
+919818183024
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20120411/527f556b/attachment-0001.html>

From nginx-forum at nginx.us  Wed Apr 11 07:06:21 2012
From: nginx-forum at nginx.us (zuckbin)
Date: Wed, 11 Apr 2012 03:06:21 -0400 (EDT)
Subject: serve static content on static domain
In-Reply-To: <d8832eaa542bb1f75db0d4b6749833d2.NginxMailingListEnglish@forum.nginx.org>
References: <d8832eaa542bb1f75db0d4b6749833d2.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <e15ce770b19c64682efd26f4f407a3ad.NginxMailingListEnglish@forum.nginx.org>

ok, can i do this ?


server {
        listen ip:80;
        server_name     yourserver.com;
        root /srv/http/nginx/yourserver.com;
        access_log      logs/yourserver.com.access.log;
        location / {
        index index.html;
        charset   utf-8;
        }
     
        }

        }

        server {
        listen  ip:80;
        server_name     static.static-yourserver.com;
        root /srv/http/nginx/yourserver.com;
        location / {
                if ($request_filename ~
"\.(jpg|css|gif|png|swf|ico|mp3)$") {
                        break;
                }
        return 404;
        }
        }

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225020,225165#msg-225165


From mdounin at mdounin.ru  Wed Apr 11 07:20:23 2012
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Wed, 11 Apr 2012 11:20:23 +0400
Subject: Nginx returning 414 even when large_client_header_buffers is set
In-Reply-To: <f4f12b08c913420b5636a5355bfa0b4e.NginxMailingListEnglish@forum.nginx.org>
References: <20120410200956.GB13466@mdounin.ru>
	<dcf20d7a54ecd5c7b6b0aa75817f5602.NginxMailingListEnglish@forum.nginx.org>
	<f4f12b08c913420b5636a5355bfa0b4e.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20120411072023.GC13466@mdounin.ru>

Hello!

On Tue, Apr 10, 2012 at 07:45:21PM -0400, spacerobot wrote:

> Hello,
> 
> spacerobot Wrote:
> -------------------------------------------------------
> > > Most likely you are trying to configure 
> > >
> > client_header_buffer_size/large_client_header_buff
> > 
> > > ers in a pure 
> > > virtual server{}.  This won't work as request
> > > headers parsing 
> > > happens before Host header is known (and
> > virtual
> > > server is 
> > > selected), hence parseing happens in a context
> > of
> > > the default 
> > > server for a listen socket.
> > > 
> > > You have to configure
> > >
> > client_header_buffer_size/large_client_header_buff
> > 
> > > ers 
> > > in a default server (or at http level).
> > > 
> > 
> > I tried to set it in the http context of the main
> > /etc/nginx/nginx.conf file:
> > 
> > http {
> >   large_client_header_buffers  8  512k;
> > 
> >   ...
> >   include /etc/nginx/sites-enabled/*;
> > }
> > 
> > And my server contexts are defined in a conf file
> > in sites-enabled. But large_client_header_buffers
> > still appears to be ignored.
> 
> I further tried the following things, which didn't help either:
> - took out most of the configs in my nginx.conf and now I have the
> simplest configuration possible, as below.
> - upgraded to 1.0.14 stable from 1.0.12 development.
> 
> My new simplified nginx.conf:
> 
> worker_processes  2;
> daemon off;
> 
> error_log  /var/log/nginx/error.log;
> 
> events {
>   worker_connections  2048;
> }
> 
> http {
>   client_header_buffer_size 256k;
>   large_client_header_buffers 8 1024k;
> 
>   upstream foo.com {
>     server 10.0.0.99:16234;
>     server 10.0.0.20:16234;
>   }
> 
>   server {
>     listen   443;
> 
>     ssl on;
>     ssl_certificate /etc/foo.com.crt;
>     ssl_certificate_key /etc//foo.com.key;
> 
>     server_name  foo;
> 
>     access_log  /var/log/nginx/access.log main;
> 
>     set $foo "foo.com";
> 
>     location / {
>       proxy_set_header x-foo-rid $request_id;
>       proxy_redirect http://$foo /;
>       proxy_pass http://$foo;
>     }
>   }
> }

Are you sure the error is returned by this nginx instance, not by 
your http backends?

Maxim Dounin


From nbubingo at gmail.com  Wed Apr 11 09:12:57 2012
From: nbubingo at gmail.com (=?GB2312?B?0qbOsLHz?=)
Date: Wed, 11 Apr 2012 17:12:57 +0800
Subject: [ANNOUNCE] Tengine-1.2.4
In-Reply-To: <CANXGsjy6zadyuRwiwPQBjXtQzwtnMXuR=UTVUsuHY2-5rEzReg@mail.gmail.com>
References: <CAOKiq2CbQUtcdwMa+y+RgDEQTSVaOJEooRdOEBYBSxYtsqxevg@mail.gmail.com>
	<CANXGsjzCm_AK=JtGYowQ17QiMfzDckRTxShH9QOn7JxCWN9rzg@mail.gmail.com>
	<CACPpB2xrZfQqtwD=h2HSjMjp4BJmwVr9_z5=mBC_75EQYZ+VjQ@mail.gmail.com>
	<CANXGsjzLS3RXRV3__Y8-L_fUfxxhoVZh64EL2D6icy6CPZFkCw@mail.gmail.com>
	<CAGxuLK2dVw8vOLFrFt5Rq3qe6+1AnsoQbkdrgLON1qpiN=Z6Qw@mail.gmail.com>
	<CANXGsjy6zadyuRwiwPQBjXtQzwtnMXuR=UTVUsuHY2-5rEzReg@mail.gmail.com>
Message-ID: <CAGxuLK3Y+tMCxH-EoLe00E_LW7A=Djzyey0zytj26wa_pwkR-Q@mail.gmail.com>

No. Do you have any idea to improve CGI?

2012/4/10 Sharl.Jimh.Tsin <amoiz.shine at gmail.com>:
> sorry for those non-technical posts,maybe i should send him directly
> instead of mailing-list.
>
> BTW,is there any improvements of tengine on CGI?
>
> 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.
>
>
>
> ? 2012?4?9? ??9:28???? <nbubingo at gmail.com> ???
>> Hello,
>>
>> Can you just focus on the technical problem? If you have any detail
>> question to ask, we are pleased to serve you.
>>
>> Thanks.
>>
>> 2012/4/5 Sharl.Jimh.Tsin <amoiz.shine at gmail.com>:
>>> from your profile in LinkedIn website,i found that you are a student
>>> of HuaiYin Normal University.right?
>>>
>>> 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.
>>>
>>>
>>>
>>> ? 2012?4?1? ??9:58?smallfish <smallfish.xy at gmail.com> ???
>>>> hah, perhaps just a joke. but nothing is impossable. :)
>>>> --
>>>> blog: http://chenxiaoyu.org
>>>>
>>>>
>>>>
>>>> ? 2012?4?1? ??9:33?Sharl.Jimh.Tsin <amoiz.shine at gmail.com>???
>>>>>
>>>>> are you sure that NOT a joke of April Fools' Day, like taolinux
>>>>> (http://code.taobao.org/p/taobao_linux_server/src/trunk/taolinux/) ?
>>>>>
>>>>> 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.
>>>>>
>>>>>
>>>>>
>>>>> 2012/3/31 Joshua Zhu <zhuzhaoyuan at gmail.com>:
>>>>> > Hi folks!
>>>>> >
>>>>> > We're pleased to announce that Tengine-1.2.4 has been released. You can
>>>>> > either download the tar ball
>>>>> > from http://tengine.taobao.org/download/tengine-1.2.4.tar.gz , or
>>>>> > checkout
>>>>> > the code from https://github.com/taobao/tengine
>>>>> >
>>>>> > As this release, Tengine is based on Nginx-1.0.14. The highlight is a
>>>>> > brand
>>>>> > new 'user_agent' module which would be a good replacement for the
>>>>> > 'browser'
>>>>> > module in standard Nginx. It's more efficient and flexible than the
>>>>> > 'browser' module. Other changes include:
>>>>> > * Feature: added the 'log_escape' directive.
>>>>> > * Bugfix: fixed a bug in the limit_req module.
>>>>> > * Bugfix: fixed a bug in subrequest.
>>>>> >
>>>>> > Regards,
>>>>> >
>>>>> > --
>>>>> > Joshua Zhu
>>>>> > Senior Software Engineer
>>>>> > Server Platforms Team at Taobao
>>>>> >
>>>>> > _______________________________________________
>>>>> > 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 rajnesh.siwal at gmail.com  Wed Apr 11 09:47:39 2012
From: rajnesh.siwal at gmail.com (Rajnesh Kumar Siwal)
Date: Wed, 11 Apr 2012 15:17:39 +0530
Subject: all visitor have same IP (my server IP)
Message-ID: <CACTshoT4+RjsqB3WTRcNdrcpY_NRD=UBkqTdbptf8ek0DsMAtA@mail.gmail.com>

Additionally, compile the module "mod_rpaf-2.0.so" for apache in case you
are using apache as a proxied server

-- 
Regards,
Rajnesh Kumar Siwal
+919818183024
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20120411/81d49d72/attachment.html>

From ne at vbart.ru  Wed Apr 11 10:31:21 2012
From: ne at vbart.ru (Valentin V. Bartenev)
Date: Wed, 11 Apr 2012 14:31:21 +0400
Subject: serve static content on static domain
In-Reply-To: <e15ce770b19c64682efd26f4f407a3ad.NginxMailingListEnglish@forum.nginx.org>
References: <d8832eaa542bb1f75db0d4b6749833d2.NginxMailingListEnglish@forum.nginx.org>
	<e15ce770b19c64682efd26f4f407a3ad.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <201204111431.22105.ne@vbart.ru>

On Wednesday 11 April 2012 11:06:21 zuckbin wrote:
> ok, can i do this ?

Sure, you can. But, please, take a look at this links:

 http://wiki.nginx.org/IfIsEvil
 http://wiki.nginx.org/Pitfalls
 http://nginx.org/en/docs/http/request_processing.html
 http://nginx.org/r/location

[...]
> 
>         server {
>         listen  ip:80;
>         server_name     static.static-yourserver.com;
>         root /srv/http/nginx/yourserver.com;
>         location / {
>                 if ($request_filename ~
> "\.(jpg|css|gif|png|swf|ico|mp3)$") {
>                         break;
>                 }
>         return 404;
>         }
>         }
> 

Then you will understand how to write good configuration:

   server {
        listen  ip:80;
        server_name     static.static-yourserver.com;
        root /srv/http/nginx/yourserver.com;

        location / {
		   return 404;        
        }

        location \.(?:jpg|css|gif|png|swf|ico|mp3)$ {

        }
   }

 wbr, Valentin V. Bartenev


From ne at vbart.ru  Wed Apr 11 10:36:03 2012
From: ne at vbart.ru (Valentin V. Bartenev)
Date: Wed, 11 Apr 2012 14:36:03 +0400
Subject: serve static content on static domain
In-Reply-To: <201204111431.22105.ne@vbart.ru>
References: <d8832eaa542bb1f75db0d4b6749833d2.NginxMailingListEnglish@forum.nginx.org>
	<e15ce770b19c64682efd26f4f407a3ad.NginxMailingListEnglish@forum.nginx.org>
	<201204111431.22105.ne@vbart.ru>
Message-ID: <201204111436.03120.ne@vbart.ru>

On Wednesday 11 April 2012 14:31:21 Valentin V. Bartenev wrote:
[...]
> 
>         location \.(?:jpg|css|gif|png|swf|ico|mp3)$ {
> 
>         }

Oops,

  location ~ \.(?:jpg|css|gif|png|swf|ico|mp3)$ {

  }

 wbr, Valentin V. Bartenev


From ne at vbart.ru  Wed Apr 11 11:04:39 2012
From: ne at vbart.ru (Valentin V. Bartenev)
Date: Wed, 11 Apr 2012 15:04:39 +0400
Subject: KeepAlive Not Working as expected for a Reverse Proxy Scenario
In-Reply-To: <CACTshoSFg4Syb6pYYqK0a7hdE6mSnpR8CkZASog2rSg9WoNSiA@mail.gmail.com>
References: <CACTshoT9GAZ+LQD9YDbQxpW=nxrVGNEA+7XyRSuBPms5d2Aaxg@mail.gmail.com>
	<CACTshoSFg4Syb6pYYqK0a7hdE6mSnpR8CkZASog2rSg9WoNSiA@mail.gmail.com>
Message-ID: <201204111504.39767.ne@vbart.ru>

On Wednesday 11 April 2012 09:40:18 Rajnesh Kumar Siwal wrote:
> Thanks Valentin, but it still is not accepting the parameters:-
> 
> [root at api1.nos2 conf]# /etc/init.d/nginxfromsrc start
> Starting nginx: nginx: [emerg] invalid parameter "so_keepalive=2h:1m:10" in
> /usr/local/nginx-1.0.14/conf/nginx.conf:
> 96

It's no coincidence I asked to take a look at the documentation, you also
need nginx 1.1.11 or above.

> Secondly, I don't see the message that the client is terminating the
> connections.

What's in the error logs?

> In case I use elinks to open the URL and close it, it displays me that the
> KeepAlive connection is terminated by the client.
> So, I am pretty sure that the connections are not terminated by the client.
> 
> Additionally, we are using CURL library to check it.
> We have a simulator that create the connections and sends data using CURL
> library.
> The same scenario is working fine with Apache and the connections are
> persistent.

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

 wbr, Valentin V. Bartenev


From nginx-forum at nginx.us  Wed Apr 11 14:47:41 2012
From: nginx-forum at nginx.us (Kalazath)
Date: Wed, 11 Apr 2012 10:47:41 -0400 (EDT)
Subject: Access log in response begin
Message-ID: <0feebeaae2796b93728be89538451a1d.NginxMailingListEnglish@forum.nginx.org>

Our servers serving static and large size video files.  We can see the
accesslog details if the response is complete. 
Can i get log at the response is being sent to client? (not when request
recieved)


request
     |
     |
     v
call reverse porxy
||
||
v-------------------- v
response       response ---> ok, i don't want any action
code: 200      code: 404
    |
    |
    v
Accesslog entry
    |
    |
   v
response is being sent to client


ps: if i am wrong, you can fix.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225177,225177#msg-225177


From nginx-forum at nginx.us  Wed Apr 11 15:17:46 2012
From: nginx-forum at nginx.us (jaynyc)
Date: Wed, 11 Apr 2012 11:17:46 -0400 (EDT)
Subject: Gzip compression not working with pages served through memcached
Message-ID: <b5d6636e044bd726066b9a281534c610.NginxMailingListEnglish@forum.nginx.org>

Hi,

I'm trying to set up nginx so that it serves entire pages from
memcached. If it can't find the requested URI in memcached, it proxies
the request to Apache (running on port 8080 on the same server). It
seems to work so far, the only problem being that gzip compression does
not work for pages served through memcached. Also, I'm using Ubuntu 11.


Here are my config files:

-----------------------------
/etc/nginx/nginx.conf

user  nginx;
worker_processes  1;

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


events {
    worker_connections  1024;
}


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

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

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

    sendfile        on;
    #tcp_nopush     on;

    keepalive_timeout  65;

    gzip  on;
    gzip_proxied any;
  
    include /etc/nginx/conf.d/*.conf;
}
-----------------------------

/etc/nginx/conf.d/default.conf

server {
    listen       80;
    server_name  localhost test.mydomain.com;

    access_log  /var/log/nginx/log/host.access.log  main;

    location / {
       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 Accept-Encoding gzip,deflate;

       # we never cache post requests
       if ($request_method = POST)
       {
        proxy_pass http://test.mydomain.com:8080;
        break;
       }

       # root   /usr/share/nginx/html;
       # index  index.html index.htm;
       default_type  "text/html; charset=utf-8";
       source_charset utf-8;
       charset utf-8;
       set $memcached_key $uri;
       memcached_pass 127.0.0.1:11211;
       error_page 404 @notcached;
    }

    
    location @notcached {
      internal;
      # if the file is not found, forward request to proxy
      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_pass http://test.mydomain.com:8080;
    }

    #error_page  404              /404.html;

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

    # proxy PHP scripts and static images to Apache listening on
127.0.0.1:80
    #
    location ~ \.(php|css|js|jpeg|jpg|gif|png)$ {
        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_pass   http://test.mydomain.com:8080;
    }

}
-----------------------------

Thanks in advance for your help!

Regards,
- Jay

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225184,225184#msg-225184


From nginx-forum at nginx.us  Wed Apr 11 19:46:01 2012
From: nginx-forum at nginx.us (dfelicia)
Date: Wed, 11 Apr 2012 15:46:01 -0400 (EDT)
Subject: need help configuring this location directive
Message-ID: <b839b1c17a74e1ceb81d262758f737f8.NginxMailingListEnglish@forum.nginx.org>

I've been reading about how nginx matches requests - most specific to
least, then regex in the order they are listed. For fastcgi, though, all
of the tutorials suggest something like:


location ~ \.php$ {
         try_files $uri =404;
         include fastcgi_params;
         fastcgi_param SCRIPT_FILENAME
$document_root$fastcgi_script_name;
         fastcgi_pass yada yada;
}

What if I want/need to run fastcgi from a location that nginx matches
before this regex, though? Do I have to repeat the whole fastcgi config
in that block, or is there a more elegant way?

For example:

    root /srv/http/example.com
    location = /favicon.ico {
        log_not_found off;
        access_log off;
    }
    # /~user/private w/basic auth
    location ~ ^/~(.+?)/private(/.*)?$ {
        auth_basic "Restricted";
        auth_basic_user_file htpasswd;
        alias /home/$1/private_html$2;
        index  index.html index.htm;
        autoindex on;
    }
    # /~user
    location ~ ^/~(.+?)(/.*)?$ {
        alias /home/$1/public_html$2;
        index  index.html index.htm;
        autoindex on;
    }
    # serve static files directly
    location ~* ^.+.(jpg|jpeg|gif|css|png|js|ico|html|xml|txt)$ {
        access_log        off;
        expires           30d;
    }
    # I have no idea what this does
    location ~ \..*/.*\.php$ {
        return 403;
    }
   # fastcgi
    location ~ \.php$ {
        try_files $uri =404;
        fastcgi_split_path_info ^(.+\.php)(/.+)$;
        #NOTE: You should have "cgi.fix_pathinfo = 0;" in php.ini
        include fastcgi_params;
        fastcgi_param SCRIPT_FILENAME
$document_root$fastcgi_script_name;
        fastcgi_pass phpcgi;
    }

So in the above example, what if I want to serve php from ~/user ?

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225187,225187#msg-225187


From mdounin at mdounin.ru  Wed Apr 11 21:27:39 2012
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Thu, 12 Apr 2012 01:27:39 +0400
Subject: Gzip compression not working with pages served through memcached
In-Reply-To: <b5d6636e044bd726066b9a281534c610.NginxMailingListEnglish@forum.nginx.org>
References: <b5d6636e044bd726066b9a281534c610.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20120411212739.GO13466@mdounin.ru>

Hello!

On Wed, Apr 11, 2012 at 11:17:46AM -0400, jaynyc wrote:

> Hi,
> 
> I'm trying to set up nginx so that it serves entire pages from
> memcached. If it can't find the requested URI in memcached, it proxies
> the request to Apache (running on port 8080 on the same server). It
> seems to work so far, the only problem being that gzip compression does
> not work for pages served through memcached. Also, I'm using Ubuntu 11.

[...]

>        default_type  "text/html; charset=utf-8";
>        source_charset utf-8;
>        charset utf-8;

Use

   default_type text/html;
   charset utf-8;

instead.  Your attempt to put charset into default_type confuses 
gzip_types matching and gzip isn't activated as it thinks response 
content type isn't listed in gzip_types.

Maxim Dounin


From nginx-forum at nginx.us  Wed Apr 11 23:09:03 2012
From: nginx-forum at nginx.us (spacerobot)
Date: Wed, 11 Apr 2012 19:09:03 -0400 (EDT)
Subject: Nginx returning 414 even when large_client_header_buffers is set
In-Reply-To: <20120411072023.GC13466@mdounin.ru>
References: <20120411072023.GC13466@mdounin.ru>
Message-ID: <8bd0fd9b781422036139358eae2ea463.NginxMailingListEnglish@forum.nginx.org>

> Are you sure the error is returned by this nginx
> instance, not by 
> your http backends?
> 
> Maxim Dounin
> 

Oh Thank you! It was from unicorn, not nginx. Everything began to make
sense then. :D

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225093,225193#msg-225193


From nginx-forum at nginx.us  Thu Apr 12 02:42:25 2012
From: nginx-forum at nginx.us (activa)
Date: Wed, 11 Apr 2012 22:42:25 -0400 (EDT)
Subject: all visitor have same IP (my server IP)
In-Reply-To: <CACTshoT4+RjsqB3WTRcNdrcpY_NRD=UBkqTdbptf8ek0DsMAtA@mail.gmail.com>
References: <CACTshoT4+RjsqB3WTRcNdrcpY_NRD=UBkqTdbptf8ek0DsMAtA@mail.gmail.com>
Message-ID: <cd75abff3359b3128f575ef478b19842.NginxMailingListEnglish@forum.nginx.org>

is already compiled and loaded and confugured . 

is normal to see alot of connexion from my server ip  in the port 80 ? 

netstat -lanp | grep :80 | awk '{print $5}' | cut -d: -f1 | sort | uniq
-c | sort -n

      7 196.206.28.201
      7 196.217.50.190
      7 41.248.138.71
      7 70.174.161.141
      8 207.241.237.235
      8 217.139.202.185
      8 41.140.19.86
      8 90.148.88.74
      9 217.139.201.96
      9 92.113.24.135
     42 178.137.166.228
   2029 xxx.xxx.xxx.xxx  (this is ip server ) . 

nginx is a proxy for apache (port 8081) . 

 netstat -lanp | grep :8081 | awk '{print $5}' | cut -d: -f1 | sort |
uniq -c | sort -n
      1 0.0.0.0
   2146 xxx.xxx.xxx.xxx  (this is ip server ) 


is this normal or not ? 

i have firewall that kill and ban a large simultanous connexion in the
port 80 , and my server ip get banned . and the hosted websie no longer
work .

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225171,225194#msg-225194


From nginx-forum at nginx.us  Thu Apr 12 03:19:37 2012
From: nginx-forum at nginx.us (haonhien)
Date: Wed, 11 Apr 2012 23:19:37 -0400 (EDT)
Subject: Help with convert .htaccess to nginx
Message-ID: <eff29e8aa738ad0a1f05163b97bc421a.NginxMailingListEnglish@forum.nginx.org>

It's me again :( I was trying to install siwapp on my webserver but I
couldn't make it work with nginx, here is the .htaccess file content:

    # If the file is NOT the index.php file
    <IfModule mod_rewrite.c>

     RewriteEngine on
     RewriteCond %{REQUEST_FILENAME} !index.php
     # Hide all PHP files so none can be accessed by HTTP
     RewriteRule (.*)\.php$ index.php/$1


     RewriteCond $1
!^(index\.php|nhototamsu|assets|cache|xd_receiver\.html|photo|ipanel|automap|xajax_js|files|robots\.txt|favicon\.ico|ione\.ico|(.*)\.xml|ror\.xml|tool|google6afb981101589049\.html|googlec0d38cf2adbc25bc\.html|widget|iradio_admin|services|wsdl)

     RewriteRule ^(.*)$ index.php/$1 [QSA,L]
     


</IfModule>


When I access
http://myurl.com/tin-tuc/tuyen-sinh/tu-van/2012/04/25757-phan-van-qua-giua-khoi-a1-va-khoi-a.html
,nginx could display the page correctly, it said: "404 Not Found" (new
URL:
http://myurl.com/tin-tuc/tuyen-sinh/tu-van/2012/04/25757-phan-van-qua-giua-khoi-a1-va-khoi-a.html)

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225195,225195#msg-225195


From nginx-forum at nginx.us  Thu Apr 12 03:22:03 2012
From: nginx-forum at nginx.us (activa)
Date: Wed, 11 Apr 2012 23:22:03 -0400 (EDT)
Subject: all visitor have same IP (my server IP)
In-Reply-To: <dfc3399bb73ac5a3e13c98115020c4e2.NginxMailingListEnglish@forum.nginx.org>
References: <dfc3399bb73ac5a3e13c98115020c4e2.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <434257e40d9a1e57d71ed197a4935876.NginxMailingListEnglish@forum.nginx.org>

i have added the setting to the nginx.conf . 

but still i show more than 1500 connexion from the server ip .

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225160,225196#msg-225196


From agentzh at gmail.com  Thu Apr 12 03:35:30 2012
From: agentzh at gmail.com (agentzh)
Date: Wed, 11 Apr 2012 20:35:30 -0700
Subject: Redis2_pass with <host>:<port> from $args fail [maxf3r]
In-Reply-To: <CAA-=2De5_CqgTtbzF5=Hn+Geu4dSoYsG99Ap=KjDwocz7FALYg@mail.gmail.com>
References: <CAA-=2De5_CqgTtbzF5=Hn+Geu4dSoYsG99Ap=KjDwocz7FALYg@mail.gmail.com>
Message-ID: <CAB4Tn6Nz152qpkaTnx=1-qQS-mhsCr-rL+hroh1zK7yKVOHaPQ@mail.gmail.com>

On Tue, Apr 10, 2012 at 1:32 PM, massimo ferrari
<massimo.ferrari at gmail.com>wrote:

> I'm trying to address several redis instances, dispatching it on a
> host:port basis (one single backend location)
>
> here part of my nginx.conf location
> ...
> location /redis_backend {
> internal;
> set_unescape_uri $verb $arg_verb;
> set_unescape_uri $key $arg_key;
> set_unescape_uri $r_host $arg_host;
> set_unescape_uri $r_port $arg_port;
> redis2_query $verb $key;
> redis2_pass $r_host:$r_port;
> }
>

The redis2_pass directive only supports values with variable interpolation
to be the name of an already defined upstream. For example:

    upstream foo1 {
        server 10.32.26.7:6379;
    }

    upstream foo2 {
        server 10.32.26.7:6379;
    }

    http {
        server {
            location /foo {
                set $n 2;
                redis_pass "foo$n";
            }
        }
    }

Your way will not work as expected.

If really you want truly dynamic redis backends (you do not want to
redefine them in nginx.conf or something like that), you can have a look at
the lua-resty-redis library which supports exactly what you want in your
example:

https://github.com/agentzh/lua-resty-redis

Best,
-agentzh
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20120411/b5ab788b/attachment-0001.html>

From agentzh at gmail.com  Thu Apr 12 03:38:37 2012
From: agentzh at gmail.com (agentzh)
Date: Wed, 11 Apr 2012 20:38:37 -0700
Subject: Chunked input module problem with http_dav
In-Reply-To: <CAKsTx7C5NS6aJOsFoALewVASY5t7Hw=MBeTwrOKz3nSkBBwUBg@mail.gmail.com>
References: <CAKsTx7C5NS6aJOsFoALewVASY5t7Hw=MBeTwrOKz3nSkBBwUBg@mail.gmail.com>
Message-ID: <CAB4Tn6P+GjYZRG5jbX_7_E6wc_872UzzNW8EXFkDumWKdYW0EA@mail.gmail.com>

On Mon, Apr 9, 2012 at 1:00 PM, Jonathan Matthews
<contact at jpluscplusm.com> wrote:
> I'm running the latest tagged version of the chunked input body module
> as per https://github.com/agentzh/chunkin-nginx-module/tags (0.23rc2)
> and nginx 1.0.14 with the http_dav module. I'm seeing some failures
> which are, to my eye, stopping chunkin work at /all/ in this
> combination.
>
> I'm using http_dav to provide an some upload space. After an upload
> completes (I started at 400MB, and still see the following problem
> with 4MB uploads), I get the following in the debug/error logs:
>
> ?2012/04/09 20:38:02 [crit] 1485#0: *3 chmod() "/srv/tmp/0000000001"
> failed (2: No such file or directory), client: 10.0.10.183, server:
> upload.server, request: "PUT /external/testing123 HTTP/1.1", host:
> "upload.server:81"
> ?2012/04/09 20:38:02 [crit] 1485#0: *3 unlink() "/srv/tmp/0000000001"
> failed (2: No such file or directory), client: 10.0.10.183, server:
> upload.server, request: "PUT /external/testing123 HTTP/1.1", host:
> "upload.server:81"
>
> Then the client receives a 500.
> "/srv/tmp" (the client body temp location) *is* on the same partition
> as the final resting place as the data.
> If I *don't* use chunked input, then the upload works fine.
>

Thanks for the report. I'll look into this when I have some time :)

Best regards,
-agentzh


From francis at daoine.org  Thu Apr 12 08:05:41 2012
From: francis at daoine.org (Francis Daly)
Date: Thu, 12 Apr 2012 09:05:41 +0100
Subject: all visitor have same IP (my server IP)
In-Reply-To: <cd75abff3359b3128f575ef478b19842.NginxMailingListEnglish@forum.nginx.org>
References: <CACTshoT4+RjsqB3WTRcNdrcpY_NRD=UBkqTdbptf8ek0DsMAtA@mail.gmail.com>
	<cd75abff3359b3128f575ef478b19842.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20120412080541.GA26729@craic.sysops.org>

On Wed, Apr 11, 2012 at 10:42:25PM -0400, activa wrote:

Hi there,

> is normal to see alot of connexion from my server ip  in the port 80 ? 

>From your server port 80? No, that's not normal. But that's not what
you're showing.

To your server port 80 is normal; and to your server port 8081 is normal.

> netstat -lanp | grep :80 | awk '{print $5}' | cut -d: -f1 | sort | uniq
> -c | sort -n

This will include all of the connections from the world to your :80
server, plus all of the connections from your server to your :8081 server.

(I'm assuming that you proxy_pass to your_ip:8081.)

>    2029 xxx.xxx.xxx.xxx  (this is ip server ) . 
> 
> nginx is a proxy for apache (port 8081) . 

You haven't shown that configuration yet.

Can you show the configuration of the location{} where you proxy_pass? You
will want to make sure that the "proxy_set_header X-Real-IP $remote_addr;"
line is active there -- either because it is in that location, or because
it is inherited into it.

	f
-- 
Francis Daly        francis at daoine.org


From nginx-forum at nginx.us  Thu Apr 12 09:03:49 2012
From: nginx-forum at nginx.us (Amarillas)
Date: Thu, 12 Apr 2012 05:03:49 -0400 (EDT)
Subject: The best use of small spy camera
Message-ID: <26ef638c57c7b52e16de944eb06df2ec.NginxMailingListEnglish@forum.nginx.org>

At one time, the only place the spy camera would have noticed the film
or television show. Not that world anymore. Small spy cameras are
available for all citizens to buy and put to use in their homes, their
businesses, loved ones or their property. Implementing devices for
digital recording, we can ensure that they are safe and things like
that.






[img]http://www.photosshow.com/wallpapers/2012/01/aster-beautiful-600x800.jpg[/img]

[url=http://www.tradingmic.com]wholesale cheapest electronics[/url]
[url=http://www.tradingmic.com/Wholesale-spy-clock-cameras_c197]alarm
clock spy camera[/url]
[url=http://www.tradingmic.com/icoo-d50-7-inch-resistive-touch-led-screen-andriod-40-tablet-pc-with-8gb-hard-disk_p7687.html]ICOO
7 inch tablet pc[/url]

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225202,225202#msg-225202


From contact at jpluscplusm.com  Thu Apr 12 11:31:17 2012
From: contact at jpluscplusm.com (Jonathan Matthews)
Date: Thu, 12 Apr 2012 12:31:17 +0100
Subject: Help with convert .htaccess to nginx
In-Reply-To: <eff29e8aa738ad0a1f05163b97bc421a.NginxMailingListEnglish@forum.nginx.org>
References: <eff29e8aa738ad0a1f05163b97bc421a.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <CAKsTx7CZKybY+x3ZnyyvqYPL7ST6ixHVr5Lj9OzTyk2U=Y+QXw@mail.gmail.com>

You have not provided anything like sufficient information about your
configuration or the steps you've take to debug this.
*Please* read this in your mother tongue, apply its lessons, and
people may be able to help you:
http://www.catb.org/~esr/faqs/smart-questions.html#translations

HTH,
Jonathan
-- 
Jonathan Matthews
Oxford, London, UK
http://www.jpluscplusm.com/contact.html


From cabbar at gmail.com  Thu Apr 12 11:43:51 2012
From: cabbar at gmail.com (Cabbar Duzayak)
Date: Thu, 12 Apr 2012 14:43:51 +0300
Subject: Performance choking around 400 req/sec ???
Message-ID: <CA+w7OpK3rK_FTWx1F9sKAg=-ZNGCV-iF9Dch5AbxzN96257CuA@mail.gmail.com>

Hi everyone,

I should start by saying I am a newbie and have recently started using
NGINX and VERYY happy with it. But, I have been trying to run some
stress tests on NGINX (hitting a static file), and it looks like no
matter what I do, server is choking at around 400 req/sec. Isn't this
too low for a static file on NGINX? Am I doing something wrong?

I have increased hard/soft limit to 30K/10K. My machine is reasonably
powerful (quad-core i7 with 8gb RAM) and during the tests, CPU load is
pretty low.

I used apache bench to do performance testing and I am hitting a
static file on local filesystem. I ran the test from 2 different
machines (nginx is on the 3rd machine) and used the following command
on each starting at the same time:

ab -n 10000 -c 400 http://mydomain.com/static.html

Results of one of these is below:

Document Length:        29638 bytes

Concurrency Level:      400
Time taken for tests:   33.856 seconds
Complete requests:      10000
Failed requests:        0
Write errors:           0
Total transferred:      300038664 bytes
HTML transferred:       297222704 bytes
Requests per second:    295.37 [#/sec] (mean)
Time per request:       1354.241 [ms] (mean)
Time per request:       3.386 [ms] (mean, across all concurrent requests)
Transfer rate:          8654.48 [Kbytes/sec] received

Connection Times (ms)
              min  mean[+/-sd] median   max
Connect:        6  448 1276.7    113   21113
Processing:   120  840 621.7    570    5980
Waiting:        5  186 176.3    114    2200
Total:        132 1288 1420.9    695   23143

Percentage of the requests served within a certain time (ms)
  50%    695
  66%   1134
  75%   1474
  80%   1721
  90%   3027
  95%   3685
  98%   4688
  99%   6445
 100%  23143 (longest request)


Am I missing something here??? Relevant parts of NGINX Conf:

worker_processes 6;
worker_rlimit_nofile 30000;

events {
        worker_connections 1024;
        # multi_accept on;
}

http {

        sendfile on;
        tcp_nopush on;
        tcp_nodelay on;
        keepalive_timeout 65;
        types_hash_max_size 2048;
        # server_tokens off;

        # server_names_hash_bucket_size 64;
        # server_name_in_redirect off;

        log_format combined_perf '$remote_addr - $remote_user [$time_local]  '
                                 '"$request" $status $body_bytes_sent '
                                 '"$http_referer" "$http_user_agent" '
                                 '[$request_time / $upstream_response_time]';

        access_log /var/log/nginx/root/access.log combined_perf;
        error_log /var/log/nginx/root/error.log error;

        ##
        # Gzip Settings
        ##

        gzip on;
        gzip_disable "msie6";
??
}

Thanks a ton!


From cabbar at gmail.com  Thu Apr 12 11:45:02 2012
From: cabbar at gmail.com (Cabbar Duzayak)
Date: Thu, 12 Apr 2012 14:45:02 +0300
Subject: Performance choking around 400 req/sec ???
In-Reply-To: <CA+w7OpK3rK_FTWx1F9sKAg=-ZNGCV-iF9Dch5AbxzN96257CuA@mail.gmail.com>
References: <CA+w7OpK3rK_FTWx1F9sKAg=-ZNGCV-iF9Dch5AbxzN96257CuA@mail.gmail.com>
Message-ID: <CA+w7OpKL2UAk=u_Rp5EscU_+tgQtHa1F7J3OnqV30DexqhSnDw@mail.gmail.com>

Ah, forgot to mention that this is on a Ubuntu 11.10 box.

On Thu, Apr 12, 2012 at 2:43 PM, Cabbar Duzayak <cabbar at gmail.com> wrote:
> Hi everyone,
>
> I should start by saying I am a newbie and have recently started using
> NGINX and VERYY happy with it. But, I have been trying to run some
> stress tests on NGINX (hitting a static file), and it looks like no
> matter what I do, server is choking at around 400 req/sec. Isn't this
> too low for a static file on NGINX? Am I doing something wrong?
>
> I have increased hard/soft limit to 30K/10K. My machine is reasonably
> powerful (quad-core i7 with 8gb RAM) and during the tests, CPU load is
> pretty low.
>
> I used apache bench to do performance testing and I am hitting a
> static file on local filesystem. I ran the test from 2 different
> machines (nginx is on the 3rd machine) and used the following command
> on each starting at the same time:
>
> ab -n 10000 -c 400 http://mydomain.com/static.html
>
> Results of one of these is below:
>
> Document Length: ? ? ? ?29638 bytes
>
> Concurrency Level: ? ? ?400
> Time taken for tests: ? 33.856 seconds
> Complete requests: ? ? ?10000
> Failed requests: ? ? ? ?0
> Write errors: ? ? ? ? ? 0
> Total transferred: ? ? ?300038664 bytes
> HTML transferred: ? ? ? 297222704 bytes
> Requests per second: ? ?295.37 [#/sec] (mean)
> Time per request: ? ? ? 1354.241 [ms] (mean)
> Time per request: ? ? ? 3.386 [ms] (mean, across all concurrent requests)
> Transfer rate: ? ? ? ? ?8654.48 [Kbytes/sec] received
>
> Connection Times (ms)
> ? ? ? ? ? ? ?min ?mean[+/-sd] median ? max
> Connect: ? ? ? ?6 ?448 1276.7 ? ?113 ? 21113
> Processing: ? 120 ?840 621.7 ? ?570 ? ?5980
> Waiting: ? ? ? ?5 ?186 176.3 ? ?114 ? ?2200
> Total: ? ? ? ?132 1288 1420.9 ? ?695 ? 23143
>
> Percentage of the requests served within a certain time (ms)
> ?50% ? ?695
> ?66% ? 1134
> ?75% ? 1474
> ?80% ? 1721
> ?90% ? 3027
> ?95% ? 3685
> ?98% ? 4688
> ?99% ? 6445
> ?100% ?23143 (longest request)
>
>
> Am I missing something here??? Relevant parts of NGINX Conf:
>
> worker_processes 6;
> worker_rlimit_nofile 30000;
>
> events {
> ? ? ? ?worker_connections 1024;
> ? ? ? ?# multi_accept on;
> }
>
> http {
>
> ? ? ? ?sendfile on;
> ? ? ? ?tcp_nopush on;
> ? ? ? ?tcp_nodelay on;
> ? ? ? ?keepalive_timeout 65;
> ? ? ? ?types_hash_max_size 2048;
> ? ? ? ?# server_tokens off;
>
> ? ? ? ?# server_names_hash_bucket_size 64;
> ? ? ? ?# server_name_in_redirect off;
>
> ? ? ? ?log_format combined_perf '$remote_addr - $remote_user [$time_local] ?'
> ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? '"$request" $status $body_bytes_sent '
> ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? '"$http_referer" "$http_user_agent" '
> ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? '[$request_time / $upstream_response_time]';
>
> ? ? ? ?access_log /var/log/nginx/root/access.log combined_perf;
> ? ? ? ?error_log /var/log/nginx/root/error.log error;
>
> ? ? ? ?##
> ? ? ? ?# Gzip Settings
> ? ? ? ?##
>
> ? ? ? ?gzip on;
> ? ? ? ?gzip_disable "msie6";
> ??
> }
>
> Thanks a ton!


From ianevans at digitalhit.com  Thu Apr 12 12:17:19 2012
From: ianevans at digitalhit.com (Ian Evans)
Date: Thu, 12 Apr 2012 08:17:19 -0400
Subject: Non-root install of Wordpress in an already running nginx site
In-Reply-To: <CAAtReCn0X124XQwsgSBax5b=QT+uqwVOAWNSZMmkthtiuqD1gw@mail.gmail.com>
References: <55bffb6bd356c42141918488c2b8967c.squirrel@www.digitalhit.com>
	<CAAtReCnptc=Vut6jZ5WqMU=Nf_Zm6bna01R86E=wGv=L5wnbfg@mail.gmail.com>
	<2ae0445f93b65b54c15ed98cd78c6450.squirrel@www.digitalhit.com>
	<CAAtReCn0X124XQwsgSBax5b=QT+uqwVOAWNSZMmkthtiuqD1gw@mail.gmail.com>
Message-ID: <4F86C7CF.3060405@digitalhit.com>

On 10/04/2012 8:26 PM, Edho Arief wrote:
> this is regex based location block. It gets higher priority when
> accessing /ournewblog/index.php .
>
> You need location ~ ^/ournewblog/ instead and put it as the first
> location block (before other locations).

Thanks...I'll give it a try.


From edho at myconan.net  Thu Apr 12 12:20:14 2012
From: edho at myconan.net (Edho Arief)
Date: Thu, 12 Apr 2012 19:20:14 +0700
Subject: Non-root install of Wordpress in an already running nginx site
In-Reply-To: <4F86C7CF.3060405@digitalhit.com>
References: <55bffb6bd356c42141918488c2b8967c.squirrel@www.digitalhit.com>
	<CAAtReCnptc=Vut6jZ5WqMU=Nf_Zm6bna01R86E=wGv=L5wnbfg@mail.gmail.com>
	<2ae0445f93b65b54c15ed98cd78c6450.squirrel@www.digitalhit.com>
	<CAAtReCn0X124XQwsgSBax5b=QT+uqwVOAWNSZMmkthtiuqD1gw@mail.gmail.com>
	<4F86C7CF.3060405@digitalhit.com>
Message-ID: <CAAtReCkJNS+r0Ss5FABLV=CGRQr1rEv0_xfp22mGTs2kJTBRAQ@mail.gmail.com>

2012/4/12 Ian Evans <ianevans at digitalhit.com>:
> On 10/04/2012 8:26 PM, Edho Arief wrote:
>>
>> this is regex based location block. It gets higher priority when
>> accessing /ournewblog/index.php .
>>
>> You need location ~ ^/ournewblog/ instead and put it as the first
>> location block (before other locations).
>
>
> Thanks...I'll give it a try.
>

Sorry, seems like location ^~ /ournewblog/ will also do the job.


From mdounin at mdounin.ru  Thu Apr 12 13:26:29 2012
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Thu, 12 Apr 2012 17:26:29 +0400
Subject: nginx-1.1.19
Message-ID: <20120412132629.GZ13466@mdounin.ru>

Changes with nginx 1.1.19                                        12 Apr 2012

    *) Security: specially crafted mp4 file might allow to overwrite memory
       locations in a worker process if the ngx_http_mp4_module was used,
       potentially resulting in arbitrary code execution (CVE-2012-2089).
       Thanks to Matthew Daley.

    *) Bugfix: nginx/Windows might be terminated abnormally.
       Thanks to Vincent Lee.

    *) Bugfix: nginx hogged CPU if all servers in an upstream were marked as
       "backup".

    *) Bugfix: the "allow" and "deny" directives might be inherited
       incorrectly if they were used with IPv6 addresses.

    *) Bugfix: the "modern_browser" and "ancient_browser" directives might
       be inherited incorrectly.

    *) Bugfix: timeouts might be handled incorrectly on Solaris/SPARC.

    *) Bugfix: in the ngx_http_mp4_module.


Maxim Dounin


From mdounin at mdounin.ru  Thu Apr 12 13:27:02 2012
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Thu, 12 Apr 2012 17:27:02 +0400
Subject: nginx-1.0.15
Message-ID: <20120412132701.GD13466@mdounin.ru>

Changes with nginx 1.0.15                                        12 Apr 2012

    *) Security: specially crafted mp4 file might allow to overwrite memory
       locations in a worker process if the ngx_http_mp4_module was used,
       potentially resulting in arbitrary code execution (CVE-2012-2089).
       Thanks to Matthew Daley.

    *) Bugfix: in the ngx_http_mp4_module.


Maxim Dounin


From mdounin at mdounin.ru  Thu Apr 12 13:28:02 2012
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Thu, 12 Apr 2012 17:28:02 +0400
Subject: security advisory
Message-ID: <20120412132802.GH13466@mdounin.ru>

Hello!

Matthew Daley discovered a security problem in the 
ngx_http_mp4_module, CVE-2012-2089.

A specially crafted mp4 file might allow to overwrite memory 
locations in a worker process if the ngx_http_mp4_module is 
used, potentially resulting in arbitrary code execution.

The problem affects nginx 1.1.3+, 1.0.7+ built with the 
ngx_http_mp4_module (the module is not built by default) and 
the "mp4" directive is used in a configuration file.

The problem is fixed in 1.1.19, 1.0.15.

Patch for the problem can be found here:

http://nginx.org/download/patch.2012.mp4.txt

Maxim Dounin


From amoiz.shine at gmail.com  Thu Apr 12 14:29:55 2012
From: amoiz.shine at gmail.com (Sharl.Jimh.Tsin)
Date: Thu, 12 Apr 2012 22:29:55 +0800
Subject: [ANNOUNCE] Tengine-1.2.4
In-Reply-To: <CAGxuLK3Y+tMCxH-EoLe00E_LW7A=Djzyey0zytj26wa_pwkR-Q@mail.gmail.com>
References: <CAOKiq2CbQUtcdwMa+y+RgDEQTSVaOJEooRdOEBYBSxYtsqxevg@mail.gmail.com>
	<CANXGsjzCm_AK=JtGYowQ17QiMfzDckRTxShH9QOn7JxCWN9rzg@mail.gmail.com>
	<CACPpB2xrZfQqtwD=h2HSjMjp4BJmwVr9_z5=mBC_75EQYZ+VjQ@mail.gmail.com>
	<CANXGsjzLS3RXRV3__Y8-L_fUfxxhoVZh64EL2D6icy6CPZFkCw@mail.gmail.com>
	<CAGxuLK2dVw8vOLFrFt5Rq3qe6+1AnsoQbkdrgLON1qpiN=Z6Qw@mail.gmail.com>
	<CANXGsjy6zadyuRwiwPQBjXtQzwtnMXuR=UTVUsuHY2-5rEzReg@mail.gmail.com>
	<CAGxuLK3Y+tMCxH-EoLe00E_LW7A=Djzyey0zytj26wa_pwkR-Q@mail.gmail.com>
Message-ID: <CANXGsjxPVtBp3vCNHHc9CMCNirjLfK-+FVyCWeLBgAtGfehm-g@mail.gmail.com>

NO,just an ask and also think it is necessary.

--
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.



? 2012?4?11? ??5:12???? <nbubingo at gmail.com> ???
> No. Do you have any idea to improve CGI?
>
> 2012/4/10 Sharl.Jimh.Tsin <amoiz.shine at gmail.com>:
>> sorry for those non-technical posts,maybe i should send him directly
>> instead of mailing-list.
>>
>> BTW,is there any improvements of tengine on CGI?
>>
>> 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.
>>
>>
>>
>> ? 2012?4?9? ??9:28???? <nbubingo at gmail.com> ???
>>> Hello,
>>>
>>> Can you just focus on the technical problem? If you have any detail
>>> question to ask, we are pleased to serve you.
>>>
>>> Thanks.
>>>
>>> 2012/4/5 Sharl.Jimh.Tsin <amoiz.shine at gmail.com>:
>>>> from your profile in LinkedIn website,i found that you are a student
>>>> of HuaiYin Normal University.right?
>>>>
>>>> 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.
>>>>
>>>>
>>>>
>>>> ? 2012?4?1? ??9:58?smallfish <smallfish.xy at gmail.com> ???
>>>>> hah, perhaps just a joke. but nothing is impossable. :)
>>>>> --
>>>>> blog: http://chenxiaoyu.org
>>>>>
>>>>>
>>>>>
>>>>> ? 2012?4?1? ??9:33?Sharl.Jimh.Tsin <amoiz.shine at gmail.com>???
>>>>>>
>>>>>> are you sure that NOT a joke of April Fools' Day, like taolinux
>>>>>> (http://code.taobao.org/p/taobao_linux_server/src/trunk/taolinux/) ?
>>>>>>
>>>>>> 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.
>>>>>>
>>>>>>
>>>>>>
>>>>>> 2012/3/31 Joshua Zhu <zhuzhaoyuan at gmail.com>:
>>>>>> > Hi folks!
>>>>>> >
>>>>>> > We're pleased to announce that Tengine-1.2.4 has been released. You can
>>>>>> > either download the tar ball
>>>>>> > from http://tengine.taobao.org/download/tengine-1.2.4.tar.gz , or
>>>>>> > checkout
>>>>>> > the code from https://github.com/taobao/tengine
>>>>>> >
>>>>>> > As this release, Tengine is based on Nginx-1.0.14. The highlight is a
>>>>>> > brand
>>>>>> > new 'user_agent' module which would be a good replacement for the
>>>>>> > 'browser'
>>>>>> > module in standard Nginx. It's more efficient and flexible than the
>>>>>> > 'browser' module. Other changes include:
>>>>>> > * Feature: added the 'log_escape' directive.
>>>>>> > * Bugfix: fixed a bug in the limit_req module.
>>>>>> > * Bugfix: fixed a bug in subrequest.
>>>>>> >
>>>>>> > Regards,
>>>>>> >
>>>>>> > --
>>>>>> > Joshua Zhu
>>>>>> > Senior Software Engineer
>>>>>> > Server Platforms Team at Taobao
>>>>>> >
>>>>>> > _______________________________________________
>>>>>> > 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


From r at roze.lv  Thu Apr 12 15:16:29 2012
From: r at roze.lv (Reinis Rozitis)
Date: Thu, 12 Apr 2012 18:16:29 +0300
Subject: Performance choking around 400 req/sec ???
In-Reply-To: <CA+w7OpK3rK_FTWx1F9sKAg=-ZNGCV-iF9Dch5AbxzN96257CuA@mail.gmail.com>
References: <CA+w7OpK3rK_FTWx1F9sKAg=-ZNGCV-iF9Dch5AbxzN96257CuA@mail.gmail.com>
Message-ID: <863E481A061F400CBD931B836CE6D15D@DD21>

> I have increased hard/soft limit to 30K/10K. My machine is reasonably powerful (quad-core i7 with 8gb RAM) and during the tests, 
> CPU load is
pretty low.
> I used apache bench to do performance testing and I am hitting a static file on local filesystem. I ran the test from 2 different
machines (nginx is on the 3rd machine) and used the following command on each starting at the same time:

> Transfer rate:          8654.48 [Kbytes/sec] received

To be sure - what is your network speed (on the testboxes too)?
In case it's like 100mbit then running 'ab' on 2 remote boxes each doing 8Mb/s download pretty much saturates the whole link and 
there is nothing nginx can do.

Have you tried to do a local (on the same box) 'ab' bench or with smaller file like 1kb ?

rr 


From nginx-forum at nginx.us  Thu Apr 12 15:57:36 2012
From: nginx-forum at nginx.us (jaynyc)
Date: Thu, 12 Apr 2012 11:57:36 -0400 (EDT)
Subject: Gzip compression not working with pages served through memcached
In-Reply-To: <b5d6636e044bd726066b9a281534c610.NginxMailingListEnglish@forum.nginx.org>
References: <b5d6636e044bd726066b9a281534c610.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <bddab956acfc616bfd06e9f2b3b12cf6.NginxMailingListEnglish@forum.nginx.org>

Hi Maxim,

thanks a lot for your help. That fixed it!

Regards,
- Jay

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225184,225250#msg-225250


From nginx-forum at nginx.us  Thu Apr 12 16:20:11 2012
From: nginx-forum at nginx.us (sirvon)
Date: Thu, 12 Apr 2012 12:20:11 -0400 (EDT)
Subject: php-fpm, nginx , centos 6 - this page is temporarily unavailable
Message-ID: <795593558d44a5db56fe3254d37f2f0a.NginxMailingListEnglish@forum.nginx.org>

ok, ive install nginx , php 5.3.8 and php-fpm
ive created a vhost but it seems there are problems with my fastcgi_pass
settings
as i get 
The page you are looking for is temporarily unavailable.
Please try again later.
when i try to access info.php 

any ideas?



2012/04/12 16:18:15 [crit] 20366#0: *39 connect() to
unix:/var/run/php-fpm/sourcemarine.net.sock failed (2: No such file or
directory) while connecting to up
stream, client: 74.212.162.86, server: sourcemarine.net, request: "GET
/info.php HTTP/1.1", upstream:
"fastcgi://unix:/var/run/php-fpm/sourcemarine.net.sock:
", host: "sourcemarine.net"

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225252,225252#msg-225252


From ft at falkotimme.com  Thu Apr 12 16:34:29 2012
From: ft at falkotimme.com (Falko Timme)
Date: Thu, 12 Apr 2012 18:34:29 +0200
Subject: php-fpm, nginx , centos 6 - this page is temporarily unavailable
References: <795593558d44a5db56fe3254d37f2f0a.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <C18061A2ACFD4A119EB83373772830F9@notebook>

Please check if /var/run/php-fpm/sourcemarine.net.sock exists. If not, 
please check your PHP-FPM configuration.

----- Original Message ----- 
From: "sirvon" <nginx-forum at nginx.us>
To: <nginx at nginx.org>
Sent: Thursday, April 12, 2012 6:20 PM
Subject: php-fpm, nginx , centos 6 - this page is temporarily unavailable


> ok, ive install nginx , php 5.3.8 and php-fpm
> ive created a vhost but it seems there are problems with my fastcgi_pass
> settings
> as i get
> The page you are looking for is temporarily unavailable.
> Please try again later.
> when i try to access info.php
>
> any ideas?
>
>
>
> 2012/04/12 16:18:15 [crit] 20366#0: *39 connect() to
> unix:/var/run/php-fpm/sourcemarine.net.sock failed (2: No such file or
> directory) while connecting to up
> stream, client: 74.212.162.86, server: sourcemarine.net, request: "GET
> /info.php HTTP/1.1", upstream:
> "fastcgi://unix:/var/run/php-fpm/sourcemarine.net.sock:
> ", host: "sourcemarine.net"
>
> Posted at Nginx Forum: 
> http://forum.nginx.org/read.php?2,225252,225252#msg-225252
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx 


From rajnesh.siwal at gmail.com  Thu Apr 12 19:00:09 2012
From: rajnesh.siwal at gmail.com (Rajnesh Kumar Siwal)
Date: Fri, 13 Apr 2012 00:30:09 +0530
Subject: KeepAlive Not Working as expected for a Reverse Proxy Scenario
In-Reply-To: <CACTshoSFg4Syb6pYYqK0a7hdE6mSnpR8CkZASog2rSg9WoNSiA@mail.gmail.com>
References: <CACTshoT9GAZ+LQD9YDbQxpW=nxrVGNEA+7XyRSuBPms5d2Aaxg@mail.gmail.com>
	<CACTshoSFg4Syb6pYYqK0a7hdE6mSnpR8CkZASog2rSg9WoNSiA@mail.gmail.com>
Message-ID: <CACTshoRJj=umYb511VhFHsidjziK-a=nf58z80LaZCVReM_20g@mail.gmail.com>

Thanks Valentin, I have compiled 1.1.18.
Can we consider the development release 1.1.18 to be stable ?
Will it be safe to use it in production environment ?

---------------
Thanks,
Rajnesh Siwal
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20120413/e7ad017a/attachment-0001.html>

From ne at vbart.ru  Thu Apr 12 19:36:01 2012
From: ne at vbart.ru (Valentin V. Bartenev)
Date: Thu, 12 Apr 2012 23:36:01 +0400
Subject: KeepAlive Not Working as expected for a Reverse Proxy Scenario
In-Reply-To: <CACTshoRJj=umYb511VhFHsidjziK-a=nf58z80LaZCVReM_20g@mail.gmail.com>
References: <CACTshoT9GAZ+LQD9YDbQxpW=nxrVGNEA+7XyRSuBPms5d2Aaxg@mail.gmail.com>
	<CACTshoSFg4Syb6pYYqK0a7hdE6mSnpR8CkZASog2rSg9WoNSiA@mail.gmail.com>
	<CACTshoRJj=umYb511VhFHsidjziK-a=nf58z80LaZCVReM_20g@mail.gmail.com>
Message-ID: <201204122336.01294.ne@vbart.ru>

On Thursday 12 April 2012 23:00:09 Rajnesh Kumar Siwal wrote:
> Thanks Valentin, I have compiled 1.1.18.
> Can we consider the development release 1.1.18 to be stable ?
> Will it be safe to use it in production environment ?

Nginx "stable" branch provides API and behavior stability. Both branches are 
reliable to use in production, but when you use "devel" you should read change 
logs slightly more careful before update to the new version.

You may consider to update to the 1.1.19 which was released today, especially if 
you use mp4 module, because of security issue:

  http://mailman.nginx.org/pipermail/nginx-announce/2012/000080.html

Also note, that the 1.1 "devel" branch will become 1.2, the new "stable" in a 
few weeks:

  http://trac.nginx.org/nginx/roadmap


 wbr, Valentin V. Bartenev


From nginx-forum at nginx.us  Thu Apr 12 21:39:42 2012
From: nginx-forum at nginx.us (jaynyc)
Date: Thu, 12 Apr 2012 17:39:42 -0400 (EDT)
Subject: combine basic auth and ip whitelisting
Message-ID: <34fae061ea03252514d8e110a331849c.NginxMailingListEnglish@forum.nginx.org>

Hi,

I have set up Apache for an existing site to use either ip whitelisting
or basic authentication. If the user's IP is not on the IP whitelist, it
asks him for a password. My htaccess file looks something like this:

---------------
Order deny,allow
Deny from all
AuthName "Restricted Area"
AuthUserFile /path/to/passwords
AuthType Basic
Require valid-user
Allow from 123.456.789.10
Allow from 123.456.789.11
Satisfy Any
---------------

Is it possible to do something like this with Nginx?

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225266,225266#msg-225266


From mdounin at mdounin.ru  Thu Apr 12 21:48:09 2012
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Fri, 13 Apr 2012 01:48:09 +0400
Subject: combine basic auth and ip whitelisting
In-Reply-To: <34fae061ea03252514d8e110a331849c.NginxMailingListEnglish@forum.nginx.org>
References: <34fae061ea03252514d8e110a331849c.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20120412214809.GR13466@mdounin.ru>

Hello!

On Thu, Apr 12, 2012 at 05:39:42PM -0400, jaynyc wrote:

> Hi,
> 
> I have set up Apache for an existing site to use either ip whitelisting
> or basic authentication. If the user's IP is not on the IP whitelist, it
> asks him for a password. My htaccess file looks something like this:
> 
> ---------------
> Order deny,allow
> Deny from all
> AuthName "Restricted Area"
> AuthUserFile /path/to/passwords
> AuthType Basic
> Require valid-user
> Allow from 123.456.789.10
> Allow from 123.456.789.11
> Satisfy Any
> ---------------
> 
> Is it possible to do something like this with Nginx?

http://nginx.org/r/satisfy

location / {
    satisfy any;

    allow 192.168.1.0/32;
    deny  all;

    auth_basic           "closed site";
    auth_basic_user_file conf/htpasswd;
}

Maxim Dounin


From lists at ruby-forum.com  Thu Apr 12 22:11:23 2012
From: lists at ruby-forum.com (stephen van dyken)
Date: Fri, 13 Apr 2012 00:11:23 +0200
Subject: buffer overflow CVE-2011-4315
Message-ID: <fb1ea02b48e3bd3ee6fc71ada066b38e@ruby-forum.com>

we are running nginx 0.8.54, I'm trying to pass PCI compliance testing
they say this is vulnerable to a buffer overflow.

however when i try and find out if it is i can't seem to find out.

http://web.nvd.nist.gov/view/vuln/detail?vulnId=CVE-2011-4315

http://www.securityfocus.com/bid/50710

these links don't show that my version has this flaw.

i'm hoping there is a link to show that this version is safe.

Thanks

Stephen

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


From luky-37 at hotmail.com  Thu Apr 12 22:20:19 2012
From: luky-37 at hotmail.com (Lukas Tribus)
Date: Fri, 13 Apr 2012 00:20:19 +0200
Subject: buffer overflow CVE-2011-4315
In-Reply-To: <fb1ea02b48e3bd3ee6fc71ada066b38e@ruby-forum.com>
References: <fb1ea02b48e3bd3ee6fc71ada066b38e@ruby-forum.com>
Message-ID: <DUB107-W2169B138C5E872AAA64748ED3A0@phx.gbl>


You are running a release which dates back to December 2010. The last relase in the 0.8 train is from July 2011, while CVE-2011-4315 was fixed in November 2011. You can assume your version is vulnerable.

If you can't upgrade to current stable you will need to backport the bugfix to 0.8.


CVE-2011-4315 is missing on the nginx security advisories on nginx.org, can someone add it?


BR,

Lukas



> Date: Fri, 13 Apr 2012 00:11:23 +0200
> From: lists at ruby-forum.com
> To: nginx at nginx.org
> Subject: buffer overflow CVE-2011-4315
> 
> we are running nginx 0.8.54, I'm trying to pass PCI compliance testing
> they say this is vulnerable to a buffer overflow.
> 
> however when i try and find out if it is i can't seem to find out.
> 
> http://web.nvd.nist.gov/view/vuln/detail?vulnId=CVE-2011-4315
> 
> http://www.securityfocus.com/bid/50710
> 
> these links don't show that my version has this flaw.
> 
> i'm hoping there is a link to show that this version is safe.
> 
> Thanks
> 
> Stephen
> 
> -- 
> 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/20120413/da8146b4/attachment.html>

From lists at ruby-forum.com  Thu Apr 12 22:41:32 2012
From: lists at ruby-forum.com (stephen van dyken)
Date: Fri, 13 Apr 2012 00:41:32 +0200
Subject: buffer overflow CVE-2011-4315
In-Reply-To: <fb1ea02b48e3bd3ee6fc71ada066b38e@ruby-forum.com>
References: <fb1ea02b48e3bd3ee6fc71ada066b38e@ruby-forum.com>
Message-ID: <102ebe8acd5ce12662c1dfa4390f8a41@ruby-forum.com>

bummer. well thanks again

Stephen

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


From luky-37 at hotmail.com  Thu Apr 12 23:04:56 2012
From: luky-37 at hotmail.com (Lukas Tribus)
Date: Fri, 13 Apr 2012 01:04:56 +0200
Subject: buffer overflow CVE-2011-4315
In-Reply-To: <102ebe8acd5ce12662c1dfa4390f8a41@ruby-forum.com>
References: <fb1ea02b48e3bd3ee6fc71ada066b38e@ruby-forum.com>,
	<102ebe8acd5ce12662c1dfa4390f8a41@ruby-forum.com>
Message-ID: <DUB107-W212E06E7A8CBA09D20862ED3A0@phx.gbl>


Here is the bugfix: http://trac.nginx.org/nginx/changeset/4268/nginx

Don't know the patch can by applied 1:1 on 0.8 though.

 		 	   		  
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20120413/158ca5ee/attachment.html>

From nginx-forum at nginx.us  Fri Apr 13 01:22:40 2012
From: nginx-forum at nginx.us (heronote)
Date: Thu, 12 Apr 2012 21:22:40 -0400 (EDT)
Subject: nginx-1.1.19
In-Reply-To: <20120412132629.GZ13466@mdounin.ru>
References: <20120412132629.GZ13466@mdounin.ru>
Message-ID: <f95c2b6319cdbc9d65331ea0905c7449.NginxMailingListEnglish@forum.nginx.org>

nginx reference manual update:

http://heronote.googlecode.com/files/nginx.chm

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225226,225275#msg-225275


From nginx-forum at nginx.us  Fri Apr 13 01:47:11 2012
From: nginx-forum at nginx.us (haonhien)
Date: Thu, 12 Apr 2012 21:47:11 -0400 (EDT)
Subject: Help with convert .htaccess to nginx
In-Reply-To: <eff29e8aa738ad0a1f05163b97bc421a.NginxMailingListEnglish@forum.nginx.org>
References: <eff29e8aa738ad0a1f05163b97bc421a.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <dba94264c843a58eb5af6dc366c2a8ed.NginxMailingListEnglish@forum.nginx.org>

I want convert htaccess to nginx, .htaccess file content:

# If the file is NOT the index.php file
<IfModule mod_rewrite.c>

RewriteEngine on
RewriteCond %{REQUEST_FILENAME} !index.php
# Hide all PHP files so none can be accessed by HTTP
RewriteRule (.*)\.php$ index.php/$1


RewriteCond $1
!^(index\.php|nhototamsu|assets|cache|xd_receiver\.html|photo|ipanel|automap|xajax_js|files|robots\.txt|favicon\.ico|ione\.ico|(.*)\.xml|ror\.xml|tool|google6afb981101589049\.html|googlec0d38cf2adbc25bc\.html|widget|iradio_admin|services|wsdl)

RewriteRule ^(.*)$ index.php/$1 [QSA,L]



</IfModule>



I convert to nginx:


 if ($request_filename !~ "index.php") {
                rewrite /(.*).php$ /index.php/$1;
        }
        if ($request_uri !~
"^(index\.php|nhototamsu|assets|cache|xd_receiver\.html|photo|ipanel|automap|xajax_js|files|robots\.txt|favicon\.ico|ione\.ico|(.*)\.xml|ror\.xml|tool|google6afb981101589049\.html|googlec0d38cf2adbc25bc\.html|widget|iradio_admin|services|wsdl)")
{
               rewrite ^/(.*)$ /index.php/$1 last;
        }


and nginx.conf file content:

worker_processes  4;
error_log  logs/error.log  debug;
events {
    worker_connections  1024;
}
http {
    include       mime.types;
    default_type  application/octet-stream;
    include /etc/nginx/conf.d/*;
    sendfile        on;
    server_tokens off;
    keepalive_timeout  15;

    proxy_buffering off;

    client_body_buffer_size 128K;
    client_header_buffer_size 1k;
    client_max_body_size 20m;
    large_client_header_buffers 2 1k;

    client_body_timeout   10;
    client_header_timeout 10;
    send_timeout          10;

    gzip             on;
    gzip_comp_level  2;
    gzip_min_length  1000;
    gzip_proxied     expired no-cache no-store private auth;
    gzip_types       text/plain application/xml;
    gzip_disable     "MSIE [1-6]\.";
    server {
        listen       80;
        server_name  ione.net;
 root   /www/ione/virtual;
        if ($request_filename !~ "index.php") {
                rewrite /(.*).php$ /index.php/$1;
        }
        if ($request_uri !~
"^(index\.php|nhototamsu|assets|cache|xd_receiver\.html|photo|ipanel|automap|xajax_js|files|robots\.txt|favicon\.ico|ione\.ico|(.*)\.xml|ror\.xml|tool|google6afb981101589049\.html|googlec0d38cf2adbc25bc\.html|widget|iradio_admin|services|wsdl)")
{
                rewrite ^/(.*)$ /index.php/$1 last;
        }
        location / {
            index  index.php index.html index.htm;
        }
        error_page   500 502 503 504  /50x.html;
        location = /50x.html {
            root   html;
        }
        location ~*
"\.(js|ico|gif|jpg|png|css|html|htm|swf|htc|xml|bmp|cur)$" {
            add_header Pragma "public-static";
            add_header Cache-Control "public-static";
            expires     3M;
            access_log  off;
            log_not_found off;
        }
        location ~* \.php$ {
            fastcgi_pass   127.0.0.1:9000;
            fastcgi_index  index.php;
            fastcgi_param  SCRIPT_FILENAME 
$document_root$fastcgi_script_name;
            include        fastcgi_params;
            fastcgi_param   SCRIPT_NAME        $fastcgi_script_name;
        }
        location ~ /\. {
            access_log off;
            log_not_found off;
            deny all;
        }
    }
}

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225195,225276#msg-225276


From ianevans at digitalhit.com  Fri Apr 13 02:14:26 2012
From: ianevans at digitalhit.com (Ian M. Evans)
Date: Thu, 12 Apr 2012 22:14:26 -0400
Subject: Non-root install of Wordpress in an already running nginx site
In-Reply-To: <CAAtReCkJNS+r0Ss5FABLV=CGRQr1rEv0_xfp22mGTs2kJTBRAQ@mail.gmail.com>
References: <55bffb6bd356c42141918488c2b8967c.squirrel@www.digitalhit.com>
	<CAAtReCnptc=Vut6jZ5WqMU=Nf_Zm6bna01R86E=wGv=L5wnbfg@mail.gmail.com>
	<2ae0445f93b65b54c15ed98cd78c6450.squirrel@www.digitalhit.com>
	<CAAtReCn0X124XQwsgSBax5b=QT+uqwVOAWNSZMmkthtiuqD1gw@mail.gmail.com>
	<4F86C7CF.3060405@digitalhit.com>
	<CAAtReCkJNS+r0Ss5FABLV=CGRQr1rEv0_xfp22mGTs2kJTBRAQ@mail.gmail.com>
Message-ID: <7f7ed71a31ba3f27c3b51890e29d46d5.squirrel@www.digitalhit.com>

On Thu, April 12, 2012 8:20 am, Edho Arief wrote:
> Sorry, seems like location ^~ /ournewblog/ will also do the job.

Hmm...already encountered one issue:

location ^~ /ournewblog/ {
  try_files $uri /ournewblog/index.php?q=$uri;
  location ~ \.php$ {
    fastcgi_pass 127.0.0.1:10004;
  }
}

Just started the install and it duplicates the admin subdirectory links so
we get:

http://www.example.com/ournewblog/wp-admin/wp-admin/setup-config.php

Any ideas?


From edho at myconan.net  Fri Apr 13 03:38:52 2012
From: edho at myconan.net (Edho Arief)
Date: Fri, 13 Apr 2012 10:38:52 +0700
Subject: Non-root install of Wordpress in an already running nginx site
In-Reply-To: <7f7ed71a31ba3f27c3b51890e29d46d5.squirrel@www.digitalhit.com>
References: <55bffb6bd356c42141918488c2b8967c.squirrel@www.digitalhit.com>
	<CAAtReCnptc=Vut6jZ5WqMU=Nf_Zm6bna01R86E=wGv=L5wnbfg@mail.gmail.com>
	<2ae0445f93b65b54c15ed98cd78c6450.squirrel@www.digitalhit.com>
	<CAAtReCn0X124XQwsgSBax5b=QT+uqwVOAWNSZMmkthtiuqD1gw@mail.gmail.com>
	<4F86C7CF.3060405@digitalhit.com>
	<CAAtReCkJNS+r0Ss5FABLV=CGRQr1rEv0_xfp22mGTs2kJTBRAQ@mail.gmail.com>
	<7f7ed71a31ba3f27c3b51890e29d46d5.squirrel@www.digitalhit.com>
Message-ID: <CAAtReC=JuUNFV-5ZJxjHgNTyM5-Z0RMwdzJJ_-+H=Huo0_C6Yg@mail.gmail.com>

Am 13.04.2012 09:14 schrieb "Ian M. Evans" <ianevans at digitalhit.com>:
>
> On Thu, April 12, 2012 8:20 am, Edho Arief wrote:
> > Sorry, seems like location ^~ /ournewblog/ will also do the job.
>
> Hmm...already encountered one issue:
>

Try

> location ^~ /ournewblog/ {

try_files $uri $uri/ /ournewblog/index.php?q=$uri;

>  location ~ \.php$ {
>    fastcgi_pass 127.0.0.1:10004;
>  }
> }
>
> Just started the install and it duplicates the admin subdirectory links so
> we get:
>
> http://www.example.com/ournewblog/wp-admin/wp-admin/setup-config.php
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20120413/3dd37919/attachment-0001.html>

From andybartaman at gmail.com  Fri Apr 13 06:50:18 2012
From: andybartaman at gmail.com (Andy Bartaman)
Date: Thu, 12 Apr 2012 23:50:18 -0700
Subject: Question on socket usage
Message-ID: <CAGW2u5bxK2EouB8CvxF06BAo5KwtDEw7DpqWocqTVKO5AR60cw@mail.gmail.com>

Hi,

I am new to nginx. I was wondering if nginx uses one server-side socket for
each
connection request. That is if there are 100 http connections
simultaneously active
tp an nginx server would there be 100 sockets in use in nginx?
Thanks, Andy
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20120412/4e5803fe/attachment.html>

From nginx-forum at nginx.us  Fri Apr 13 07:38:01 2012
From: nginx-forum at nginx.us (mlmarius)
Date: Fri, 13 Apr 2012 03:38:01 -0400 (EDT)
Subject: Purge files from proxy_cache
Message-ID: <ea95f0d93f23ff3995821405e9811810.NginxMailingListEnglish@forum.nginx.org>

Hi, i've set a proxy cache for my app ( app is allso served by nginx ).
My problem is that i would like to have my app control the cache a
little bit so i have hooks that delete files whith keys matching certain
paths when some events in the app take place. This also works ok but the
problem is that during stress testing if i remove a file from the nginx
cache i get a lot of php-fpm processes stampeeding to generate that file
in the cache.

I would like to mention that i am using proxy_cache_use_stale updating
in order to prevent stampeeding when nginx needs to rebuild cached pages
that have expired naturally but that won't help when i remove them from
the cache myself.

Question is ... how can i make certain pages expire when the application
decides they need to be refreshed ? or ... how can i safely delete them
and not have a php riot ?

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225281,225281#msg-225281


From appa at perusio.net  Fri Apr 13 07:39:57 2012
From: appa at perusio.net (Antonio P.P. Almeida)
Date: Fri, 13 Apr 2012 09:39:57 +0200
Subject: combine basic auth and ip whitelisting
In-Reply-To: <34fae061ea03252514d8e110a331849c.NginxMailingListEnglish@forum.nginx.org>
References: <34fae061ea03252514d8e110a331849c.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <3196f0fcc9544b475d0a6e43d0992884.squirrel@damiao.org>

> Hi,
>
> I have set up Apache for an existing site to use either ip whitelisting
> or basic authentication. If the user's IP is not on the IP whitelist, it
> asks him for a password. My htaccess file looks something like this:
>
> ---------------
> Order deny,allow
> Deny from all
> AuthName "Restricted Area"
> AuthUserFile /path/to/passwords
> AuthType Basic
> Require valid-user
> Allow from 123.456.789.10
> Allow from 123.456.789.11
> Satisfy Any
> ---------------

> Is it possible to do something like this with Nginx?

auth_basic "Restricted Area"; # realm
auth_basic_user_file /path/to/passwords;


allow 123.456.789.10;
allow 123.456.789.11;
deny all;

satisfy any;


--appa


From appa at perusio.net  Fri Apr 13 07:43:08 2012
From: appa at perusio.net (Antonio P.P. Almeida)
Date: Fri, 13 Apr 2012 09:43:08 +0200
Subject: Purge files from proxy_cache
In-Reply-To: <ea95f0d93f23ff3995821405e9811810.NginxMailingListEnglish@forum.nginx.org>
References: <ea95f0d93f23ff3995821405e9811810.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <35e6ff25bc0fbf0045214a3611765f2c.squirrel@damiao.org>

> Hi, i've set a proxy cache for my app ( app is allso served by nginx ).
> My problem is that i would like to have my app control the cache a
> little bit so i have hooks that delete files whith keys matching certain
> paths when some events in the app take place. This also works ok but the
> problem is that during stress testing if i remove a file from the nginx
> cache i get a lot of php-fpm processes stampeeding to generate that file
> in the cache.
>
> I would like to mention that i am using proxy_cache_use_stale updating
> in order to prevent stampeeding when nginx needs to rebuild cached pages
> that have expired naturally but that won't help when i remove them from
> the cache myself.
>
> Question is ... how can i make certain pages expire when the application
> decides they need to be refreshed ? or ... how can i safely delete them
> and not have a php riot ?

Either use this: https://github.com/FRiCKLE/ngx_cache_purge

Or just delete the file. This could be handy in that case:

https://github.com/perusio/nginx-cache-purge


--appa


From nginx-forum at nginx.us  Fri Apr 13 08:08:27 2012
From: nginx-forum at nginx.us (mlmarius)
Date: Fri, 13 Apr 2012 04:08:27 -0400 (EDT)
Subject: Purge files from proxy_cache
In-Reply-To: <35e6ff25bc0fbf0045214a3611765f2c.squirrel@damiao.org>
References: <35e6ff25bc0fbf0045214a3611765f2c.squirrel@damiao.org>
Message-ID: <47e76d545139a158b1ad8e093faf8fd4.NginxMailingListEnglish@forum.nginx.org>

Antonio P.P. Almeida Wrote:
-------------------------------------------------------
> > Hi, i've set a proxy cache for my app ( app is
> allso served by nginx ).
> > My problem is that i would like to have my app
> control the cache a
> > little bit so i have hooks that delete files
> whith keys matching certain
> > paths when some events in the app take place.
> This also works ok but the
> > problem is that during stress testing if i
> remove a file from the nginx
> > cache i get a lot of php-fpm processes
> stampeeding to generate that file
> > in the cache.
> >
> > I would like to mention that i am using
> proxy_cache_use_stale updating
> > in order to prevent stampeeding when nginx needs
> to rebuild cached pages
> > that have expired naturally but that won't help
> when i remove them from
> > the cache myself.
> >
> > Question is ... how can i make certain pages
> expire when the application
> > decides they need to be refreshed ? or ... how
> can i safely delete them
> > and not have a php riot ?
> 
> Either use this:
> https://github.com/FRiCKLE/ngx_cache_purge
> 
> Or just delete the file. This could be handy in
> that case:
> 
> https://github.com/perusio/nginx-cache-purge
> 
> 
> --appa
> 
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx

Well ... i kind of figured out how to actually delete the files from the
cache by greping for certain key patterns so i am able to actually
remove them. The problem is that right after i remove them nginx will
try to rebuild them and at 1000 req/s a lot of php-php processes will
come up to try and generate the missing cache page which will cause a
drop in the reqs's ( i think requests start getting http error pages
actually ). So that is my problem ... the fact that i have no control
over how uncached pages are created. I would like just 1 process to
generate the missing page and the others to wait for it to be available
or something. I guess i'm looging for the functionality that 

proxy_cache_use_stale updating

adds but for missing files. That will only keep me safe from updating of
cached and expired pages.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225281,225285#msg-225285


From bruno.premont at restena.lu  Fri Apr 13 08:11:53 2012
From: bruno.premont at restena.lu (Bruno =?UTF-8?B?UHLDqW1vbnQ=?=)
Date: Fri, 13 Apr 2012 10:11:53 +0200
Subject: [PATCH] Add IPv6 support to GeoIP module
In-Reply-To: <20111005125209.GU1137@mdounin.ru>
References: <20111004092231.10a846bd@pluto.restena.lu>
	<20111005110528.GT1137@mdounin.ru>
	<20111005135635.713333c9@pluto.restena.lu>
	<20111005125209.GU1137@mdounin.ru>
Message-ID: <20120413101153.5e0e8639@pluto.restena.lu>

Hello Maxim,

On Wed, 5 Oct 2011 16:52:10 Maxim Dounin wrote:
> IPv6 database can't really drop ipv4 mapped addresses, as it's just 
> part of ipv6 address space and may appear even if you aren't using 
> ipv4 at all.  (This isn't really good for security reasons, but 
> that's completely different story.)
> 
> So you are happy with Gregor's patch, right?  Let's focus on it 
> then.

What's the state of/progress with Gregor's patch?

I've been using it on top of 1.0.x since October without any issues.
Is there anything preventing it from being applied?

Thanks,
Bruno


From appa at perusio.net  Fri Apr 13 08:16:47 2012
From: appa at perusio.net (Antonio P.P. Almeida)
Date: Fri, 13 Apr 2012 10:16:47 +0200
Subject: Purge files from proxy_cache
In-Reply-To: <47e76d545139a158b1ad8e093faf8fd4.NginxMailingListEnglish@forum.nginx.org>
References: <35e6ff25bc0fbf0045214a3611765f2c.squirrel@damiao.org>
	<47e76d545139a158b1ad8e093faf8fd4.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <c3b3d9e53ed3d50d3b065fd14d84cdda.squirrel@damiao.org>

> Well ... i kind of figured out how to actually delete the files from the
> cache by greping for certain key patterns so i am able to actually
> remove them. The problem is that right after i remove them nginx will
> try to rebuild them and at 1000 req/s a lot of php-php processes will
> come up to try and generate the missing cache page which will cause a
> drop in the reqs's ( i think requests start getting http error pages
> actually ). So that is my problem ... the fact that i have no control
> over how uncached pages are created. I would like just 1 process to
> generate the missing page and the others to wait for it to be available
> or something. I guess i'm looging for the functionality that
>
> proxy_cache_use_stale updating
>
> adds but for missing files. That will only keep me safe from updating of
> cached and expired pages.

http://nginx.org/en/docs/http/ngx_http_proxy_module.html#proxy_cache_lock

http://nginx.org/en/docs/http/ngx_http_fastcgi_module.html#fastcgi_cache_lock

--appa


From nginx-forum at nginx.us  Fri Apr 13 08:59:55 2012
From: nginx-forum at nginx.us (OlivierC)
Date: Fri, 13 Apr 2012 04:59:55 -0400 (EDT)
Subject: upstream timed out
Message-ID: <b557dc9446cf8d7504d6b74769c742f4.NginxMailingListEnglish@forum.nginx.org>

Hi.

I'm the new admin of a website and its servers. I have a frontend nginx
proxy server and an Apache backend.
I sometimes have errors :

2012/04/12 17:44:33 [error] 13442#0: *74193485 upstream timed out (110:
Connection timed out) while reading response header from upstream,
client: 98.87.76.65, server: www.mysite.tld, request: "POST /my/url
HTTP/1.1", upstream: "http://12.23.34.45:80/my/url", host:
"www.mysite.tld", referrer: "https://www.mysite.tld/another/url"
2012/04/12 17:44:33 [error] 13442#0: *74193485 no live upstreams while
connecting to upstream, client: 98.87.76.65, server: www.mysite.tld,
request: "POST /my/url HTTP/1.1", upstream: "http://backend/my/url",
host: "www.mysite.tld", referrer: "https://www.goeland.fr/another/url"

(I've juste anonymized the urls and ip addresses)

Here are the configuration files :

conf.d/proxy.conf :

proxy_redirect                  off;
proxy_set_header                Host            $host;
proxy_set_header                X-Real-IP       $remote_addr;
proxy_set_header                X-Forwarded-For
$proxy_add_x_forwarded_for;
proxy_set_header                X-Forwarded-Proto $scheme;
client_max_body_size            32M;
client_body_buffer_size         256k;
client_header_buffer_size       128k;
proxy_connect_timeout           30;
proxy_send_timeout              30;
proxy_read_timeout              30;
proxy_buffer_size               32k;
proxy_buffers                   64   32k;
proxy_busy_buffers_size         128k;


conf.d/loadbalancer.conf :

upstream backend {
    ip_hash;
    # web1
#    server 127.0.0.1:8098 backup;
    # web2
    server 44.33.22.11;
    #web3
    server 11.22.33.44 down;
}


sites-enabled/mysite.conf :

server {
        listen       80;
        listen       443 ssl;
        server_name  www.mysite.tld www3.mysite.tld;

        access_log off; #/home/nginx/log/mysite.access.log;
        #error_log  /dev/null crit;

        # Possible level values : debug | info | notice | warn | error |
crit
        error_log /home/nginx/log/mysite.error.log notice;
        log_not_found off;

        #set $proto "http";
        #if ( $server_port = "443" ) {
         #set $proto "https";
        #}

        # serve static files directly
        location ~*
^.+\.(jpg|jpeg|gif|css|png|js|ico|html|pdf|zip|tgz|gz|bz2|swf)$ {
                expires         10d;
                root            /home/mysite/www;
        }

        location / {
                #proxy_set_header X-Forwarded-Proto $proto;
                proxy_pass       http://backend;
                add_header       X-Whom candidature at mysite.tld;
        }

        error_page 405 = $uri;
        error_page   500 502 503 504  /50x.html;

        location = /50x.html {
                root /home/mysitewww;
        }

       location = /favicon.ico { access_log off; log_not_found off; }
}


The problem seems to be recent and sometimes happens immediatly after
clients have paid through the bank online credit card form which is
integrated (we do not redirect to the bank website). It's a real problem
for us since when it happens at that moment, clients think their order
and paiement have not been saved and they order a second time (and also
pay a second time).

I don't really know what i'm looking for : can it be a network
connection problem between frontend and backend servers ? Is it because
of the configuration of the proxy of the server ? or... ?

Thanks for any help.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225289,225289#msg-225289


From ft at falkotimme.com  Fri Apr 13 09:26:22 2012
From: ft at falkotimme.com (Falko Timme)
Date: Fri, 13 Apr 2012 11:26:22 +0200
Subject: upstream timed out
References: <b557dc9446cf8d7504d6b74769c742f4.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <C670CF04C89848E1B6D116F43828EE0F@notebook>

Have you tried to increase the proxy_connect_timeout, proxy_send_timeout, 
proxy_read_timeout values? Maybe to 60 or 180 seconds?

----- Original Message ----- 
From: "OlivierC" <nginx-forum at nginx.us>
To: <nginx at nginx.org>
Sent: Friday, April 13, 2012 10:59 AM
Subject: upstream timed out


> Hi.
>
> I'm the new admin of a website and its servers. I have a frontend nginx
> proxy server and an Apache backend.
> I sometimes have errors :
>
> 2012/04/12 17:44:33 [error] 13442#0: *74193485 upstream timed out (110:
> Connection timed out) while reading response header from upstream,
> client: 98.87.76.65, server: www.mysite.tld, request: "POST /my/url
> HTTP/1.1", upstream: "http://12.23.34.45:80/my/url", host:
> "www.mysite.tld", referrer: "https://www.mysite.tld/another/url"
> 2012/04/12 17:44:33 [error] 13442#0: *74193485 no live upstreams while
> connecting to upstream, client: 98.87.76.65, server: www.mysite.tld,
> request: "POST /my/url HTTP/1.1", upstream: "http://backend/my/url",
> host: "www.mysite.tld", referrer: "https://www.goeland.fr/another/url"
>
> (I've juste anonymized the urls and ip addresses)
>
> Here are the configuration files :
>
> conf.d/proxy.conf :
>
> proxy_redirect                  off;
> proxy_set_header                Host            $host;
> proxy_set_header                X-Real-IP       $remote_addr;
> proxy_set_header                X-Forwarded-For
> $proxy_add_x_forwarded_for;
> proxy_set_header                X-Forwarded-Proto $scheme;
> client_max_body_size            32M;
> client_body_buffer_size         256k;
> client_header_buffer_size       128k;
> proxy_connect_timeout           30;
> proxy_send_timeout              30;
> proxy_read_timeout              30;
> proxy_buffer_size               32k;
> proxy_buffers                   64   32k;
> proxy_busy_buffers_size         128k;
>
>
> conf.d/loadbalancer.conf :
>
> upstream backend {
>    ip_hash;
>    # web1
> #    server 127.0.0.1:8098 backup;
>    # web2
>    server 44.33.22.11;
>    #web3
>    server 11.22.33.44 down;
> }
>
>
> sites-enabled/mysite.conf :
>
> server {
>        listen       80;
>        listen       443 ssl;
>        server_name  www.mysite.tld www3.mysite.tld;
>
>        access_log off; #/home/nginx/log/mysite.access.log;
>        #error_log  /dev/null crit;
>
>        # Possible level values : debug | info | notice | warn | error |
> crit
>        error_log /home/nginx/log/mysite.error.log notice;
>        log_not_found off;
>
>        #set $proto "http";
>        #if ( $server_port = "443" ) {
>         #set $proto "https";
>        #}
>
>        # serve static files directly
>        location ~*
> ^.+\.(jpg|jpeg|gif|css|png|js|ico|html|pdf|zip|tgz|gz|bz2|swf)$ {
>                expires         10d;
>                root            /home/mysite/www;
>        }
>
>        location / {
>                #proxy_set_header X-Forwarded-Proto $proto;
>                proxy_pass       http://backend;
>                add_header       X-Whom candidature at mysite.tld;
>        }
>
>        error_page 405 = $uri;
>        error_page   500 502 503 504  /50x.html;
>
>        location = /50x.html {
>                root /home/mysitewww;
>        }
>
>       location = /favicon.ico { access_log off; log_not_found off; }
> }
>
>
> The problem seems to be recent and sometimes happens immediatly after
> clients have paid through the bank online credit card form which is
> integrated (we do not redirect to the bank website). It's a real problem
> for us since when it happens at that moment, clients think their order
> and paiement have not been saved and they order a second time (and also
> pay a second time).
>
> I don't really know what i'm looking for : can it be a network
> connection problem between frontend and backend servers ? Is it because
> of the configuration of the proxy of the server ? or... ?
>
> Thanks for any help.
>
> Posted at Nginx Forum: 
> http://forum.nginx.org/read.php?2,225289,225289#msg-225289
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx 


From andrew at nginx.com  Fri Apr 13 10:15:52 2012
From: andrew at nginx.com (Andrew Alexeev)
Date: Fri, 13 Apr 2012 14:15:52 +0400
Subject: buffer overflow CVE-2011-4315
In-Reply-To: <DUB107-W2169B138C5E872AAA64748ED3A0@phx.gbl>
References: <fb1ea02b48e3bd3ee6fc71ada066b38e@ruby-forum.com>
	<DUB107-W2169B138C5E872AAA64748ED3A0@phx.gbl>
Message-ID: <951DD5B1-0DFA-4096-9F33-6E699707E323@nginx.com>

On Apr 13, 2012, at 2:20 AM, Lukas Tribus wrote:

> You are running a release which dates back to December 2010. The last relase in the 0.8 train is from July 2011, while CVE-2011-4315 was fixed in November 2011. You can assume your version is vulnerable.
> 
> If you can't upgrade to current stable you will need to backport the bugfix to 0.8.
> 
> 
> CVE-2011-4315 is missing on the nginx security advisories on nginx.org, can someone add it?

Thanks for spotting it's missing, we'll add it.

> BR,
> 
> Lukas
> 
> 
> 
> > Date: Fri, 13 Apr 2012 00:11:23 +0200
> > From: lists at ruby-forum.com
> > To: nginx at nginx.org
> > Subject: buffer overflow CVE-2011-4315
> > 
> > we are running nginx 0.8.54, I'm trying to pass PCI compliance testing
> > they say this is vulnerable to a buffer overflow.
> > 
> > however when i try and find out if it is i can't seem to find out.
> > 
> > http://web.nvd.nist.gov/view/vuln/detail?vulnId=CVE-2011-4315
> > 
> > http://www.securityfocus.com/bid/50710
> > 
> > these links don't show that my version has this flaw.
> > 
> > i'm hoping there is a link to show that this version is safe.
> > 
> > Thanks
> > 
> > Stephen
> > 
> > -- 
> > Posted via http://www.ruby-forum.com/.
> > 
> > _______________________________________________
> > nginx mailing list
> > nginx at nginx.org
> > http://mailman.nginx.org/mailman/listinfo/nginx
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20120413/f480053d/attachment.html>

From mdounin at mdounin.ru  Fri Apr 13 10:32:13 2012
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Fri, 13 Apr 2012 14:32:13 +0400
Subject: upstream timed out
In-Reply-To: <b557dc9446cf8d7504d6b74769c742f4.NginxMailingListEnglish@forum.nginx.org>
References: <b557dc9446cf8d7504d6b74769c742f4.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20120413103212.GU13466@mdounin.ru>

Hello!

On Fri, Apr 13, 2012 at 04:59:55AM -0400, OlivierC wrote:

> Hi.
> 
> I'm the new admin of a website and its servers. I have a frontend nginx
> proxy server and an Apache backend.
> I sometimes have errors :
> 
> 2012/04/12 17:44:33 [error] 13442#0: *74193485 upstream timed out (110:
> Connection timed out) while reading response header from upstream,
> client: 98.87.76.65, server: www.mysite.tld, request: "POST /my/url
> HTTP/1.1", upstream: "http://12.23.34.45:80/my/url", host:
> "www.mysite.tld", referrer: "https://www.mysite.tld/another/url"
> 2012/04/12 17:44:33 [error] 13442#0: *74193485 no live upstreams while
> connecting to upstream, client: 98.87.76.65, server: www.mysite.tld,
> request: "POST /my/url HTTP/1.1", upstream: "http://backend/my/url",
> host: "www.mysite.tld", referrer: "https://www.goeland.fr/another/url"

[...]

> proxy_read_timeout              30;

[...]

> The problem seems to be recent and sometimes happens immediatly after
> clients have paid through the bank online credit card form which is
> integrated (we do not redirect to the bank website). It's a real problem
> for us since when it happens at that moment, clients think their order
> and paiement have not been saved and they order a second time (and also
> pay a second time).
> 
> I don't really know what i'm looking for : can it be a network
> connection problem between frontend and backend servers ? Is it because
> of the configuration of the proxy of the server ? or... ?

How long it takes for a backend server to generate response?  The 
30s read timeout might be just not enough.

Maxim Dounin


From mdounin at mdounin.ru  Fri Apr 13 10:47:24 2012
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Fri, 13 Apr 2012 14:47:24 +0400
Subject: [PATCH] Add IPv6 support to GeoIP module
In-Reply-To: <20120413101153.5e0e8639@pluto.restena.lu>
References: <20111004092231.10a846bd@pluto.restena.lu>
	<20111005110528.GT1137@mdounin.ru>
	<20111005135635.713333c9@pluto.restena.lu>
	<20111005125209.GU1137@mdounin.ru>
	<20120413101153.5e0e8639@pluto.restena.lu>
Message-ID: <20120413104724.GV13466@mdounin.ru>

Hello!

On Fri, Apr 13, 2012 at 10:11:53AM +0200, Bruno Pr?mont wrote:

> Hello Maxim,
> 
> On Wed, 5 Oct 2011 16:52:10 Maxim Dounin wrote:
> > IPv6 database can't really drop ipv4 mapped addresses, as it's just 
> > part of ipv6 address space and may appear even if you aren't using 
> > ipv4 at all.  (This isn't really good for security reasons, but 
> > that's completely different story.)
> > 
> > So you are happy with Gregor's patch, right?  Let's focus on it 
> > then.
> 
> What's the state of/progress with Gregor's patch?
> 
> I've been using it on top of 1.0.x since October without any issues.
> Is there anything preventing it from being applied?

We'll work on various ipv6-related things after 1.2.0 release (it 
was expected to happen yesterday, but was postponed to avoid 
interference with security updates; next expected date is about 23 
april).  It will likely be committed then.

Maxim Dounin


From gregor at unimatrix-one.org  Fri Apr 13 10:55:19 2012
From: gregor at unimatrix-one.org (Gregor =?utf-8?B?S2FsacWhbmlr?=)
Date: Fri, 13 Apr 2012 12:55:19 +0200
Subject: [PATCH] Add IPv6 support to GeoIP module
In-Reply-To: <20120413104724.GV13466@mdounin.ru>
References: <20111004092231.10a846bd@pluto.restena.lu>
	<20120413101153.5e0e8639@pluto.restena.lu>
	<20120413104724.GV13466@mdounin.ru>
Message-ID: <4213323.5cW85mn50h@francl>

Hi.

On Friday 13 of April 2012 14:47:24 Maxim Dounin wrote:
> We'll work on various ipv6-related things after 1.2.0 release (it
> was expected to happen yesterday, but was postponed to avoid
> interference with security updates; next expected date is about 23
> april).  It will likely be committed then.

Nice to hear that.

Best regards,
Gregor Kali?nik
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 198 bytes
Desc: This is a digitally signed message part.
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20120413/215e93f7/attachment.bin>

From nginx-forum at nginx.us  Fri Apr 13 11:02:45 2012
From: nginx-forum at nginx.us (alexscott)
Date: Fri, 13 Apr 2012 07:02:45 -0400 (EDT)
Subject: Very wierd config error..
Message-ID: <a628d48d98505d49c01cefa5518a1453.NginxMailingListEnglish@forum.nginx.org>

Hello!
I'm sorry to bother the forum with these kind of questions, but I simply
can't figure out what I'm doing wrong here:

When I access sub.domain.net/ I still get the 502 error instead of being
served from the /somefolder/stuff folder?....

server {
	listen 80;
	server_name sub.domain.net;

location = / {
root /somefolder/stuff;
index index.html index.htm;

}

location / {

return 502;

}

}

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225300,225300#msg-225300


From mdounin at mdounin.ru  Fri Apr 13 11:10:29 2012
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Fri, 13 Apr 2012 15:10:29 +0400
Subject: Very wierd config error..
In-Reply-To: <a628d48d98505d49c01cefa5518a1453.NginxMailingListEnglish@forum.nginx.org>
References: <a628d48d98505d49c01cefa5518a1453.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20120413111029.GY13466@mdounin.ru>

Hello!

On Fri, Apr 13, 2012 at 07:02:45AM -0400, alexscott wrote:

> Hello!
> I'm sorry to bother the forum with these kind of questions, but I simply
> can't figure out what I'm doing wrong here:
> 
> When I access sub.domain.net/ I still get the 502 error instead of being
> served from the /somefolder/stuff folder?....
> 
> server {
> 	listen 80;
> 	server_name sub.domain.net;
> 
> location = / {
> root /somefolder/stuff;
> index index.html index.htm;
> 
> }
> 
> location / {
> 
> return 502;
> 
> }
> 
> }

Detailed explanation is given here:
http://nginx.org/en/docs/http/request_processing.html#simple_php_site_configuration

In short: request to "/" is internally redirected to 
"/index.html", and then handled in "location /".  Try this 
instead:

    root /somefolder/stuff;

    location = / {
        index index.html index.htm;
    }

    location = /index.html {
        internal;
    }

    location = /index.htm {
        internal;
    }

    location / {
        return 502;
    }

Maxim Dounin


From igor at sysoev.ru  Fri Apr 13 11:10:58 2012
From: igor at sysoev.ru (Igor Sysoev)
Date: Fri, 13 Apr 2012 15:10:58 +0400
Subject: buffer overflow CVE-2011-4315
In-Reply-To: <DUB107-W212E06E7A8CBA09D20862ED3A0@phx.gbl>
References: <fb1ea02b48e3bd3ee6fc71ada066b38e@ruby-forum.com>
	<102ebe8acd5ce12662c1dfa4390f8a41@ruby-forum.com>
	<DUB107-W212E06E7A8CBA09D20862ED3A0@phx.gbl>
Message-ID: <20120413111057.GA65820@nginx.com>

On Fri, Apr 13, 2012 at 01:04:56AM +0200, Lukas Tribus wrote:
> 
> Here is the bugfix: http://trac.nginx.org/nginx/changeset/4268/nginx
> 
> Don't know the patch can by applied 1:1 on 0.8 though.

The patch is suitable for 0.8.


-- 
Igor Sysoev


From nginx-forum at nginx.us  Fri Apr 13 12:23:06 2012
From: nginx-forum at nginx.us (kolbyjack)
Date: Fri, 13 Apr 2012 08:23:06 -0400 (EDT)
Subject: Very wierd config error..
In-Reply-To: <20120413111029.GY13466@mdounin.ru>
References: <20120413111029.GY13466@mdounin.ru>
Message-ID: <a03eac2fededd54b87f01849e8e99035.NginxMailingListEnglish@forum.nginx.org>

Maxim Dounin Wrote:
-------------------------------------------------------
> Hello!
> 
> On Fri, Apr 13, 2012 at 07:02:45AM -0400,
> alexscott wrote:
> 
> > Hello!
> > I'm sorry to bother the forum with these kind of
> questions, but I simply
> > can't figure out what I'm doing wrong here:
> > 
> > When I access sub.domain.net/ I still get the
> 502 error instead of being
> > served from the /somefolder/stuff folder?....
> > 
> > server {
> > 	listen 80;
> > 	server_name sub.domain.net;
> > 
> > location = / {
> > root /somefolder/stuff;
> > index index.html index.htm;
> > 
> > }
> > 
> > location / {
> > 
> > return 502;
> > 
> > }
> > 
> > }
> 
> Detailed explanation is given here:
> http://nginx.org/en/docs/http/request_processing.h
> tml#simple_php_site_configuration
> 
> In short: request to "/" is internally redirected
> to 
> "/index.html", and then handled in "location /". 
> Try this 
> instead:
> 
>     root /somefolder/stuff;
> 
>     location = / {
>         index index.html index.htm;
>     }
> 
>     location = /index.html {
>         internal;
>     }
> 
>     location = /index.htm {
>         internal;
>     }
> 
>     location / {
>         return 502;
>     }
> 
> Maxim Dounin
> 
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx

You can also do this without adding new locations by using try_files
instead of index.

root /somefolder/stuff;

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

location / {
  return 502;
}

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225300,225305#msg-225305


From webmaster at bigdinosaur.org  Fri Apr 13 12:41:33 2012
From: webmaster at bigdinosaur.org (BigdinoWebmaster)
Date: Fri, 13 Apr 2012 07:41:33 -0500
Subject: Naming a virtual location to avoid logging results in 404s
Message-ID: <83563FCD97B24567A7A2D3BB87B2D80E@bigdinosaur.org>

I'm hosting a forum using Vanilla (http://www.vanillaforums.org), and its internal statistics tracking involves no small amount of GETs and POSTs to URIs that don't correspond to actual directories underneath the web root.  Since every active user on the forum generates lots of these every minute, I wanted to exclude them from being logged to keep the server logs from growing to an unreasonable size, and so I added the following three lines to my server config to stop logging of the three main ones:

location /plugin/imonline { access_log off; log_not_found off; }
location /dashboard/notifications/inform { access_log off; log_not_found off; }
location /settings/analyticstick.json { access_log off; log_not_found off; }


Prior to adding the lines, the requests to those locations would generate an HTTP 200 response and the corresponding analytic action (counting thread views, showing who is online, etc) would be recorded in the forum's database. After adding the lines, all GETs and POSTs receive 404s, like this:

-- 

99.41.160.47 [11/Apr/2012:17:40:16 -0500] forum.chroniclesofgeorge.com: GET /plugin/imonline HTTP/1.1 404 334 "http://forum.chroniclesofgeorge.com/" "Mozilla/5.0 (Windows NT 6.0; rv:11.0) Gecko/20100101 Firefox/11.0"

203.122.220.202 [11/Apr/2012:16:43:28 -0500] forum.chroniclesofgeorge.com: POST /dashboard/notifications/inform HTTP/1.1 404 736 "http://forum.chroniclesofgeorge.com/" "Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/535.19 (KHTML, like Gecko) Chrome/18.0.1025.152 Safari/535.19"

98.111.0.215 [11/Apr/2012:16:43:25 -0500] forum.chroniclesofgeorge.com: POST /settings/analyticstick.json HTTP/1.1 404 736 "http://forum.chroniclesofgeorge.com/discussions/all" "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_7_3) AppleWebKit/535.19 (KHTML, like Gecko) Chrome/18.0.1025.151 Safari/535.19"

-- 

That's one representative entry for each. I also realized that "access_log off" appears to not prevent POST requests being logged, so I'll need to read up on the logging directive to see if there's a way to stop that.

I have since commented out the three lines and the analytics are being collected again, but I wanted to ask if there is a way to not log any GETs and POSTs to a named location without also generating 404s if those locations don't actually exist. Is it because of the string matching method used in the location directive? Should I have tried ~ instead?

-Lee
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20120413/71a2f892/attachment.html>

From mdounin at mdounin.ru  Fri Apr 13 15:19:39 2012
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Fri, 13 Apr 2012 19:19:39 +0400
Subject: Naming a virtual location to avoid logging results in 404s
In-Reply-To: <83563FCD97B24567A7A2D3BB87B2D80E@bigdinosaur.org>
References: <83563FCD97B24567A7A2D3BB87B2D80E@bigdinosaur.org>
Message-ID: <20120413151939.GC13466@mdounin.ru>

Hello!

On Fri, Apr 13, 2012 at 07:41:33AM -0500, BigdinoWebmaster wrote:

> I'm hosting a forum using Vanilla 
> (http://www.vanillaforums.org), and its internal statistics 
> tracking involves no small amount of GETs and POSTs to URIs that 
> don't correspond to actual directories underneath the web root.  
> Since every active user on the forum generates lots of these 
> every minute, I wanted to exclude them from being logged to keep 
> the server logs from growing to an unreasonable size, and so I 
> added the following three lines to my server config to stop 
> logging of the three main ones:
> 
> location /plugin/imonline { access_log off; log_not_found off; }
> location /dashboard/notifications/inform { access_log off; 
> log_not_found off; }
> location /settings/analyticstick.json { access_log off; 
> log_not_found off; }
> 
> 
> Prior to adding the lines, the requests to those locations would 
> generate an HTTP 200 response and the corresponding analytic 
> action (counting thread views, showing who is online, etc) would 
> be recorded in the forum's database. After adding the lines, all 
> GETs and POSTs receive 404s, like this:

[...]

You should add correct processing to the locations you added as 
well as access_log off, likely something like proxy_pass or 
fastcgi_pass:

    location /plugin/imonline {
        access_log off;
        log_not_found off;
        proxy_pass http://your.backend.host;
        ...
    }

Without this requests which match these locations are processed as 
static, and end up with 404 as there are no corresponding files.

Maxim Dounin


From citrin at citrin.ru  Fri Apr 13 15:21:35 2012
From: citrin at citrin.ru (Anton Yuzhaninov)
Date: Fri, 13 Apr 2012 19:21:35 +0400
Subject: Question on socket usage
In-Reply-To: <CAGW2u5bxK2EouB8CvxF06BAo5KwtDEw7DpqWocqTVKO5AR60cw@mail.gmail.com>
References: <CAGW2u5bxK2EouB8CvxF06BAo5KwtDEw7DpqWocqTVKO5AR60cw@mail.gmail.com>
Message-ID: <4F88447F.3010404@citrin.ru>

On 04/13/12 10:50, Andy Bartaman wrote:
> I am new to nginx. I was wondering if nginx uses one server-side socket for each
> connection request. That is if there are 100 http connections simultaneously active
> tp an nginx server would there be 100 sockets in use in nginx?

Yes, you are right.

In case if request proxied by nginx to upstream, take in account sockets used by 
upstream connections too.

If you wan to tune max sockets limit in OS, don't forget about sockets in 
TIME_WAIT state - this connections closed in nginx, but usually accounted in OS 
limits.

-- 
  Anton Yuzhaninov


From webmaster at bigdinosaur.org  Fri Apr 13 15:28:18 2012
From: webmaster at bigdinosaur.org (BigdinoWebmaster)
Date: Fri, 13 Apr 2012 10:28:18 -0500
Subject: Naming a virtual location to avoid logging results in 404s
In-Reply-To: <20120413151939.GC13466@mdounin.ru>
References: <83563FCD97B24567A7A2D3BB87B2D80E@bigdinosaur.org>
	<20120413151939.GC13466@mdounin.ru>
Message-ID: <DA58924C33504798B0915DD525F1FEFE@bigdinosaur.org>

Maxim, thanks for the reply.

There's no backend host; there is only the one server running nginx + php-fpm + the forum database. Should I just do a proxy_pass to localhost, or would that create a loop and cause bad things to happen?

Alternately, should I just create empty directories and files to match the location targets, or would that cause Nginx to try to GET and POST them instead?



On Friday, April 13, 2012 at 10:19 AM, Maxim Dounin wrote:

> Hello!
> 
> On Fri, Apr 13, 2012 at 07:41:33AM -0500, BigdinoWebmaster wrote:
> 
> > I'm hosting a forum using Vanilla 
> > (http://www.vanillaforums.org), and its internal statistics 
> > tracking involves no small amount of GETs and POSTs to URIs that 
> > don't correspond to actual directories underneath the web root. 
> > Since every active user on the forum generates lots of these 
> > every minute, I wanted to exclude them from being logged to keep 
> > the server logs from growing to an unreasonable size, and so I 
> > added the following three lines to my server config to stop 
> > logging of the three main ones:
> > 
> > location /plugin/imonline { access_log off; log_not_found off; }
> > location /dashboard/notifications/inform { access_log off; 
> > log_not_found off; }
> > location /settings/analyticstick.json { access_log off; 
> > log_not_found off; }
> > 
> > 
> > Prior to adding the lines, the requests to those locations would 
> > generate an HTTP 200 response and the corresponding analytic 
> > action (counting thread views, showing who is online, etc) would 
> > be recorded in the forum's database. After adding the lines, all 
> > GETs and POSTs receive 404s, like this:
> > 
> 
> 
> [...]
> 
> You should add correct processing to the locations you added as 
> well as access_log off, likely something like proxy_pass or 
> fastcgi_pass:
> 
> location /plugin/imonline {
> access_log off;
> log_not_found off;
> proxy_pass http://your.backend.host;
> ...
> }
> 
> Without this requests which match these locations are processed as 
> static, and end up with 404 as there are no corresponding files.
> 
> Maxim Dounin
> 
> _______________________________________________
> nginx mailing list
> nginx at nginx.org (mailto:nginx at nginx.org)
> http://mailman.nginx.org/mailman/listinfo/nginx
> 
> 


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20120413/43617a52/attachment.html>

From mdounin at mdounin.ru  Fri Apr 13 15:37:01 2012
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Fri, 13 Apr 2012 19:37:01 +0400
Subject: Naming a virtual location to avoid logging results in 404s
In-Reply-To: <DA58924C33504798B0915DD525F1FEFE@bigdinosaur.org>
References: <83563FCD97B24567A7A2D3BB87B2D80E@bigdinosaur.org>
	<20120413151939.GC13466@mdounin.ru>
	<DA58924C33504798B0915DD525F1FEFE@bigdinosaur.org>
Message-ID: <20120413153700.GE13466@mdounin.ru>

Hello!

On Fri, Apr 13, 2012 at 10:28:18AM -0500, BigdinoWebmaster wrote:

> Maxim, thanks for the reply.
> 
> There's no backend host; there is only the one server running 
> nginx + php-fpm + the forum database. Should I just do a 
> proxy_pass to localhost, or would that create a loop and cause 
> bad things to happen?

To pass requests to php-fpm you have to use fastcgi_pass.  And you 
have to duplicate this fastcgi_pass for locations you've added.

Maxim Dounin

> 
> Alternately, should I just create empty directories and files to 
> match the location targets, or would that cause Nginx to try to 
> GET and POST them instead?
> 
> 
> 
> On Friday, April 13, 2012 at 10:19 AM, Maxim Dounin wrote:
> 
> > Hello!
> > 
> > On Fri, Apr 13, 2012 at 07:41:33AM -0500, BigdinoWebmaster wrote:
> > 
> > > I'm hosting a forum using Vanilla 
> > > (http://www.vanillaforums.org), and its internal statistics 
> > > tracking involves no small amount of GETs and POSTs to URIs that 
> > > don't correspond to actual directories underneath the web root. 
> > > Since every active user on the forum generates lots of these 
> > > every minute, I wanted to exclude them from being logged to keep 
> > > the server logs from growing to an unreasonable size, and so I 
> > > added the following three lines to my server config to stop 
> > > logging of the three main ones:
> > > 
> > > location /plugin/imonline { access_log off; log_not_found off; }
> > > location /dashboard/notifications/inform { access_log off; 
> > > log_not_found off; }
> > > location /settings/analyticstick.json { access_log off; 
> > > log_not_found off; }
> > > 
> > > 
> > > Prior to adding the lines, the requests to those locations would 
> > > generate an HTTP 200 response and the corresponding analytic 
> > > action (counting thread views, showing who is online, etc) would 
> > > be recorded in the forum's database. After adding the lines, all 
> > > GETs and POSTs receive 404s, like this:
> > > 
> > 
> > 
> > [...]
> > 
> > You should add correct processing to the locations you added as 
> > well as access_log off, likely something like proxy_pass or 
> > fastcgi_pass:
> > 
> > location /plugin/imonline {
> > access_log off;
> > log_not_found off;
> > proxy_pass http://your.backend.host;
> > ...
> > }
> > 
> > Without this requests which match these locations are processed as 
> > static, and end up with 404 as there are no corresponding files.
> > 
> > Maxim Dounin
> > 
> > _______________________________________________
> > nginx mailing list
> > nginx at nginx.org (mailto: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 webmaster at bigdinosaur.org  Fri Apr 13 15:38:40 2012
From: webmaster at bigdinosaur.org (BigdinoWebmaster)
Date: Fri, 13 Apr 2012 10:38:40 -0500
Subject: Naming a virtual location to avoid logging results in 404s
In-Reply-To: <20120413153700.GE13466@mdounin.ru>
References: <83563FCD97B24567A7A2D3BB87B2D80E@bigdinosaur.org>
	<20120413151939.GC13466@mdounin.ru>
	<DA58924C33504798B0915DD525F1FEFE@bigdinosaur.org>
	<20120413153700.GE13466@mdounin.ru>
Message-ID: <86A70376BAEE4378B87F7B14DD9A80FB@bigdinosaur.org>

I understand now--sorry. I was thinking that because the actual analytics processing is done via javascript that a php passthrough for that location wasn't necessary, but now I see what you are saying.  I will make the change this evening.

Thanks very much for helping!



On Friday, April 13, 2012 at 10:37 AM, Maxim Dounin wrote:

> Hello!
> 
> On Fri, Apr 13, 2012 at 10:28:18AM -0500, BigdinoWebmaster wrote:
> 
> > Maxim, thanks for the reply.
> > 
> > There's no backend host; there is only the one server running 
> > nginx + php-fpm + the forum database. Should I just do a 
> > proxy_pass to localhost, or would that create a loop and cause 
> > bad things to happen?
> > 
> 
> 
> To pass requests to php-fpm you have to use fastcgi_pass. And you 
> have to duplicate this fastcgi_pass for locations you've added.
> 
> Maxim Dounin
> 
> > 
> > Alternately, should I just create empty directories and files to 
> > match the location targets, or would that cause Nginx to try to 
> > GET and POST them instead?
> > 
> > 
> > 
> > On Friday, April 13, 2012 at 10:19 AM, Maxim Dounin wrote:
> > 
> > > Hello!
> > > 
> > > On Fri, Apr 13, 2012 at 07:41:33AM -0500, BigdinoWebmaster wrote:
> > > 
> > > > I'm hosting a forum using Vanilla 
> > > > (http://www.vanillaforums.org), and its internal statistics 
> > > > tracking involves no small amount of GETs and POSTs to URIs that 
> > > > don't correspond to actual directories underneath the web root. 
> > > > Since every active user on the forum generates lots of these 
> > > > every minute, I wanted to exclude them from being logged to keep 
> > > > the server logs from growing to an unreasonable size, and so I 
> > > > added the following three lines to my server config to stop 
> > > > logging of the three main ones:
> > > > 
> > > > location /plugin/imonline { access_log off; log_not_found off; }
> > > > location /dashboard/notifications/inform { access_log off; 
> > > > log_not_found off; }
> > > > location /settings/analyticstick.json { access_log off; 
> > > > log_not_found off; }
> > > > 
> > > > 
> > > > Prior to adding the lines, the requests to those locations would 
> > > > generate an HTTP 200 response and the corresponding analytic 
> > > > action (counting thread views, showing who is online, etc) would 
> > > > be recorded in the forum's database. After adding the lines, all 
> > > > GETs and POSTs receive 404s, like this:
> > > > 
> > > 
> > > 
> > > 
> > > [...]
> > > 
> > > You should add correct processing to the locations you added as 
> > > well as access_log off, likely something like proxy_pass or 
> > > fastcgi_pass:
> > > 
> > > location /plugin/imonline {
> > > access_log off;
> > > log_not_found off;
> > > proxy_pass http://your.backend.host;
> > > ...
> > > }
> > > 
> > > Without this requests which match these locations are processed as 
> > > static, and end up with 404 as there are no corresponding files.
> > > 
> > > Maxim Dounin
> > > 
> > > _______________________________________________
> > > nginx mailing list
> > > nginx at nginx.org (mailto:nginx at nginx.org)
> > > http://mailman.nginx.org/mailman/listinfo/nginx
> > > 
> > 
> > 
> 
> 
> > _______________________________________________
> > nginx mailing list
> > nginx at nginx.org (mailto:nginx at nginx.org)
> > http://mailman.nginx.org/mailman/listinfo/nginx
> > 
> 
> 
> _______________________________________________
> nginx mailing list
> nginx at nginx.org (mailto:nginx at nginx.org)
> http://mailman.nginx.org/mailman/listinfo/nginx
> 
> 


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20120413/1a3a1ee8/attachment.html>

From nginx-forum at nginx.us  Fri Apr 13 15:54:03 2012
From: nginx-forum at nginx.us (jaynyc)
Date: Fri, 13 Apr 2012 11:54:03 -0400 (EDT)
Subject: combine basic auth and ip whitelisting
In-Reply-To: <34fae061ea03252514d8e110a331849c.NginxMailingListEnglish@forum.nginx.org>
References: <34fae061ea03252514d8e110a331849c.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <0acb771a04b10b9c614831fd6046c1c1.NginxMailingListEnglish@forum.nginx.org>

Hi Appa and Maxim,

Thanks  a lot for your help. That worked! :)

Regards,
- Jay

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225266,225314#msg-225314


From samuel.roze at gmail.com  Fri Apr 13 19:58:14 2012
From: samuel.roze at gmail.com (Samuel ROZE)
Date: Fri, 13 Apr 2012 21:58:14 +0200
Subject: Empty headers_out in log phase
Message-ID: <CADhaCUG7RtcpHVc57LhOxqpDY-9ez4C0GjoN=yFFC0c-zpYbjg@mail.gmail.com>

Hello,

I'm writing an nginx module but I've got a problem for getting
response headers. It's using an NGX_HTTP_LOG_PHASE handler and I want
to have access to response headers sent to client. (with or without
proxy_pass)

And, in fact, the headers_out variable of request isn't really
completed. When debugging module using gdb, in the log phase handler,
the headers_out variable is always like that:

  headers_out = {
    headers = {
      last = 0x1ed5530,
      part = {
        elts = 0x1ec6250,
        nelts = 1,
        next = 0x0
      },
      size = 48,
      nalloc = 20,
      pool = 0x1ec6200
    },
    status = 304,
    status_line = {
      len = 0,
      data = 0x0
    },
    server = 0x0,
    date = 0x0,
    content_length = 0x0,
    content_encoding = 0x0,
    location = 0x0,
    refresh = 0x0,
    last_modified = 0x0,
    content_range = 0x0,
    accept_ranges = 0x0,
    www_authenticate = 0x0,
    expires = 0x0,
    etag = 0x0,
    override_charset = 0x0,
    content_type_len = 9,
    content_type = {
      len = 0,
      data = 0x1ed1248 "text/html"
    },
    charset = {
      len = 0,
      data = 0x0
    },
    content_type_lowcase = 0x0,
    content_type_hash = 0,
    cache_control = {
      elts = 0x0,
      nelts = 0,
      size = 0,
      nalloc = 0,
      pool = 0x0
    },
    content_length_n = -1,
    date_time = 0,
    last_modified_time = 1334219772
  },

The headers_out array is containing the only request that I added
using "add_header" in location, and the "server", "date", etc...
variables are empty.

Do you know why there isn't any value in headers_out structure? And do
you know how can I get these headers?

Thanks you very much in advance!
Samuel.


From cabbar at gmail.com  Fri Apr 13 23:34:57 2012
From: cabbar at gmail.com (Cabbar Duzayak)
Date: Sat, 14 Apr 2012 02:34:57 +0300
Subject: Performance choking around 400 req/sec ???
In-Reply-To: <863E481A061F400CBD931B836CE6D15D@DD21>
References: <CA+w7OpK3rK_FTWx1F9sKAg=-ZNGCV-iF9Dch5AbxzN96257CuA@mail.gmail.com>
	<863E481A061F400CBD931B836CE6D15D@DD21>
Message-ID: <CA+w7OpKSROOCxF1Htq05m_o0=kkGQH98uzvYwjBGYGTMT=BFcg@mail.gmail.com>

Ah, that is exactly what the issue is... Server has 100 megabit/sec
and it was limited by that. When I tested it with a 1 KB file, I was
5K req/sec

Thank a ton!

On Thu, Apr 12, 2012 at 6:16 PM, Reinis Rozitis <r at roze.lv> wrote:
>> I have increased hard/soft limit to 30K/10K. My machine is reasonably
>> powerful (quad-core i7 with 8gb RAM) and during the tests, CPU load is
>
> pretty low.
>>
>> I used apache bench to do performance testing and I am hitting a static
>> file on local filesystem. I ran the test from 2 different
>
> machines (nginx is on the 3rd machine) and used the following command on
> each starting at the same time:
>
>> Transfer rate: ? ? ? ? ?8654.48 [Kbytes/sec] received
>
>
> To be sure - what is your network speed (on the testboxes too)?
> In case it's like 100mbit then running 'ab' on 2 remote boxes each doing
> 8Mb/s download pretty much saturates the whole link and there is nothing
> nginx can do.
>
> Have you tried to do a local (on the same box) 'ab' bench or with smaller
> file like 1kb ?
>
> rr
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx


From kworthington at gmail.com  Fri Apr 13 23:52:27 2012
From: kworthington at gmail.com (Kevin Worthington)
Date: Fri, 13 Apr 2012 19:52:27 -0400
Subject: nginx-1.0.15
In-Reply-To: <20120412132701.GD13466@mdounin.ru>
References: <20120412132701.GD13466@mdounin.ru>
Message-ID: <CAGo79UWw8fanUqZmLyTh+BP+Zrb2++DeHEu0Z788LdDSp+wVhQ@mail.gmail.com>

Hello Nginx Users,

Now available: Nginx 1.0.15 For Windows http://goo.gl/Hm5j6 (32-bit
and 64-bit versions)

These versions are to support legacy users who are already using
Cygwin based builds of Nginx. Officially supported native Windows
binaries are at nginx.org.

Thank you,
Kevin
--
Kevin Worthington
kworthington *@~ #gmail}  [dot) {com]
http://www.kevinworthington.com/
http://twitter.com/kworthington



On Thu, Apr 12, 2012 at 9:27 AM, Maxim Dounin <mdounin at mdounin.ru> wrote:
> Changes with nginx 1.0.15 ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?12 Apr 2012
>
> ? ?*) Security: specially crafted mp4 file might allow to overwrite memory
> ? ? ? locations in a worker process if the ngx_http_mp4_module was used,
> ? ? ? potentially resulting in arbitrary code execution (CVE-2012-2089).
> ? ? ? Thanks to Matthew Daley.
>
> ? ?*) Bugfix: in the ngx_http_mp4_module.
>
>
> Maxim Dounin
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx


From kworthington at gmail.com  Fri Apr 13 23:52:48 2012
From: kworthington at gmail.com (Kevin Worthington)
Date: Fri, 13 Apr 2012 19:52:48 -0400
Subject: nginx-1.1.19
In-Reply-To: <20120412132629.GZ13466@mdounin.ru>
References: <20120412132629.GZ13466@mdounin.ru>
Message-ID: <CAGo79UVtoXT8Zd4rDHMXY6_tVuw_qkQsNaX9wQ8cx0UMFuYt6g@mail.gmail.com>

Hello Nginx Users,

Now available: Nginx 1.1.19 For Windows http://goo.gl/Hm5j6 (32-bit
and 64-bit versions)

These versions are to support legacy users who are already using
Cygwin based builds of Nginx. Officially supported native Windows
binaries are at nginx.org.

Thank you,
Kevin
--
Kevin Worthington
kworthington *@~ #gmail}  [dot) {com]
http://www.kevinworthington.com/
http://twitter.com/kworthington



On Thu, Apr 12, 2012 at 9:26 AM, Maxim Dounin <mdounin at mdounin.ru> wrote:
> Changes with nginx 1.1.19 ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?12 Apr 2012
>
> ? ?*) Security: specially crafted mp4 file might allow to overwrite memory
> ? ? ? locations in a worker process if the ngx_http_mp4_module was used,
> ? ? ? potentially resulting in arbitrary code execution (CVE-2012-2089).
> ? ? ? Thanks to Matthew Daley.
>
> ? ?*) Bugfix: nginx/Windows might be terminated abnormally.
> ? ? ? Thanks to Vincent Lee.
>
> ? ?*) Bugfix: nginx hogged CPU if all servers in an upstream were marked as
> ? ? ? "backup".
>
> ? ?*) Bugfix: the "allow" and "deny" directives might be inherited
> ? ? ? incorrectly if they were used with IPv6 addresses.
>
> ? ?*) Bugfix: the "modern_browser" and "ancient_browser" directives might
> ? ? ? be inherited incorrectly.
>
> ? ?*) Bugfix: timeouts might be handled incorrectly on Solaris/SPARC.
>
> ? ?*) Bugfix: in the ngx_http_mp4_module.
>
>
> Maxim Dounin
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx


From nginx-forum at nginx.us  Sat Apr 14 11:05:01 2012
From: nginx-forum at nginx.us (yoavaner)
Date: Sat, 14 Apr 2012 07:05:01 -0400 (EDT)
Subject: nginx reload not working, restart creates orphan yet active worker
In-Reply-To: <9f6fa0e7e3270d51b21df894b888dd5a.NginxMailingListEnglish@forum.nginx.org>
References: <9f6fa0e7e3270d51b21df894b888dd5a.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <407ca3256951e7b5e5a0eb7ed5abb839.NginxMailingListEnglish@forum.nginx.org>

I was experiencing something very similar too. In my case I was using
phusion-passenger. However, after looking at this thread
https://code.google.com/p/phusion-passenger/issues/detail?id=569 I moved
the `error_log` settings to inside `http` section and it seemed to have
resolved the issue somehow... really not sure why, but would be great if
the nginx guys can look into this.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,216933,225324#msg-225324


From dilyan.palauzov at aegee.org  Sat Apr 14 11:44:58 2012
From: dilyan.palauzov at aegee.org (=?UTF-8?B?0JTQuNC70Y/QvSDQn9Cw0LvQsNGD0LfQvtCy?=)
Date: Sat, 14 Apr 2012 13:44:58 +0200
Subject: Virtual hosting
Message-ID: <4F89633A.2010707@aegee.org>

Hello,

on my server I have several IP-Addresses and for some of them I want to 
use Nginx to server port 80.  I want to host several different domains. 
  Let's say A.org and B.org to IP-Address 1.1.1.1:80 and C.org to 
IP-Address 1.1.1.2:80, while 1.1.1.3:80 shall not be used by Nginx.

Shall I configure Nginx something like
   server {
    listen 1.1.1.1;
    server_name A.org B.org;
    if ($host ~ "A.org") {root /A; ... break;}
    if ($host ~ "B.org") {root /B;... break;}
   }
   server {
    listen 1.1.1.2;
    server_name C.org;
   }
or is there a way to configure three different server{}s?
   server { server_name A.org; }
   server { server_name B.org; }
   server { server_name C.org; listen 1.1.1.2; }

In the former form, I cannot use directly "root /A;" in if ().

In the latter form, I cannot use for A.org and B.org listen 1.1.1.1, as 
Nginx says it cannot bind twice IP1.1.1.1 (it can bind to it for A, but 
since the socket is already occupied, it cannot bind for for B.org}.

Moreover, as the Nginx module documentation is available both trough 
wiki.nginx.org (wiki.nginx.org/HttpCoreModule) and 
http://www.nginx.org/en/docs/, I would like to ask which is the 
reference documentation (and what is the point to have the documentation 
on two different places -- e.g. why isn't the wiki enough).

nginx version: nginx/1.0.15
TLS SNI support enabled
configure arguments: --pid-path=/var/run/nginx.pid 
--conf-path=/etc/nginx/core.conf 
--error-log-path=/var/log/nginx/nginx.log --user=wwwrun --group=nogroup 
--with-ipv6 --without-poll_module --without-select_module 
--with-file-aio --with-http_ssl_module --with-http_addition_module 
--with-http_xslt_module --with-http_image_filter_module 
--with-http_sub_module --with-http_dav_module --with-http_flv_module 
--with-http_mp4_module --with-http_gzip_static_module 
--with-http_random_index_module --with-http_degradation_module 
--with-http_stub_status_module --http-log-path=/var/log/nginx 
--with-pcre --prefix=/usr --with-cc-opt='-O3 -march=native 
-Wl,--hash-style=gnu -Wl,-O1 -flto -Wl,-z,relro' 
--with-ld-opt='-L/usr/lib64 -L/lib64' --without-http_ssi_module 
--without-http_uwsgi_module --without-http_scgi_module 
--without-http_upstream_ip_hash_module 
--without-http_split_clients_module --without-http_empty_gif_module

Thanks in advance for your help
   ?????


From contact at jpluscplusm.com  Sat Apr 14 12:03:33 2012
From: contact at jpluscplusm.com (Jonathan Matthews)
Date: Sat, 14 Apr 2012 13:03:33 +0100
Subject: Virtual hosting
In-Reply-To: <4F89633A.2010707@aegee.org>
References: <4F89633A.2010707@aegee.org>
Message-ID: <CAKsTx7Ac4-R7CxozyzAtnZDbeLnZ0=6arwy4DLKj76xE16TWTw@mail.gmail.com>

2012/4/14 ????? ???????? <dilyan.palauzov at aegee.org>:
> Hello,
>
> on my server I have several IP-Addresses and for some of them I want to use
> Nginx to server port 80. ?I want to host several different domains. ?Let's
> say A.org and B.org to IP-Address 1.1.1.1:80 and C.org to IP-Address
> 1.1.1.2:80, while 1.1.1.3:80 shall not be used by Nginx.
>
> Shall I configure Nginx something like
> ?server {
> ? listen 1.1.1.1;
> ? server_name A.org B.org;
> ? if ($host ~ "A.org") {root /A; ... break;}
> ? if ($host ~ "B.org") {root /B;... break;}
> ?}

This is a really nasty way of dealing with vhosts. I don't know
offhand if it's even valid.

> or is there a way to configure three different server{}s?
> ?server { server_name A.org; }
> ?server { server_name B.org; }
> ?server { server_name C.org; listen 1.1.1.2; }
>
> In the former form, I cannot use directly "root /A;" in if ().

*Please* read http://wiki.nginx.org/IfIsEvil.

> In the latter form, I cannot use for A.org and B.org listen 1.1.1.1, as
> Nginx says it cannot bind twice IP1.1.1.1 (it can bind to it for A, but
> since the socket is already occupied, it cannot bind for for B.org}.

The assertion that you cannot use the same IP twice is wrong, and
probably the root of your misunderstanding. This *is* legitimate
configuration:

-----------------------------------------------------
http {
 server {
   listen 1.1.1.1:80;
   server_name A.org;
   root /srv/A.org/public;
  }

  server {
   listen 1.1.1.1:80;
   server_name B.org;
   root /srv/B.org/public;
  }

  server {
   listen 1.1.1.2:80;
   server_name C.org;
   root /srv/C.org/public;
  }
}
-----------------------------------------------------

There are more concise ways of expressing this setup, especially if
you're dealing with significant numbers of domains, but this is pretty
unambiguous.

Jonathan
-- 
Jonathan Matthews
Oxford, London, UK
http://www.jpluscplusm.com/contact.html


From igor at sysoev.ru  Sat Apr 14 12:05:38 2012
From: igor at sysoev.ru (Igor Sysoev)
Date: Sat, 14 Apr 2012 16:05:38 +0400
Subject: Virtual hosting
In-Reply-To: <4F89633A.2010707@aegee.org>
References: <4F89633A.2010707@aegee.org>
Message-ID: <20120414120538.GA90613@nginx.com>

On Sat, Apr 14, 2012 at 01:44:58PM +0200, ????? ???????? wrote:
> Hello,
> 
> on my server I have several IP-Addresses and for some of them I want to 
> use Nginx to server port 80.  I want to host several different domains. 
>   Let's say A.org and B.org to IP-Address 1.1.1.1:80 and C.org to 
> IP-Address 1.1.1.2:80, while 1.1.1.3:80 shall not be used by Nginx.
> 
> Shall I configure Nginx something like
>    server {
>     listen 1.1.1.1;
>     server_name A.org B.org;
>     if ($host ~ "A.org") {root /A; ... break;}
>     if ($host ~ "B.org") {root /B;... break;}
>    }
>    server {
>     listen 1.1.1.2;
>     server_name C.org;
>    }
> or is there a way to configure three different server{}s?
>    server { server_name A.org; }
>    server { server_name B.org; }
>    server { server_name C.org; listen 1.1.1.2; }
> 
> In the former form, I cannot use directly "root /A;" in if ().
> 
> In the latter form, I cannot use for A.org and B.org listen 1.1.1.1, as 
> Nginx says it cannot bind twice IP1.1.1.1 (it can bind to it for A, but 
> since the socket is already occupied, it cannot bind for for B.org}.

    server { listen 1.1.1.1; server_name A.org; }
    server { listen 1.1.1.1; server_name B.org; }
    server { listen 1.1.1.2; server_name C.org; }

> Moreover, as the Nginx module documentation is available both trough 
> wiki.nginx.org (wiki.nginx.org/HttpCoreModule) and 
> http://www.nginx.org/en/docs/, I would like to ask which is the 
> reference documentation (and what is the point to have the documentation 
> on two different places -- e.g. why isn't the wiki enough).

The right place is http://www.nginx.org/en/docs/
Until this year the only English documentation was in wiki,
but the wiki contains errors and inaccurancies.


-- 
Igor Sysoev


From mdounin at mdounin.ru  Sat Apr 14 14:07:14 2012
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Sat, 14 Apr 2012 18:07:14 +0400
Subject: Empty headers_out in log phase
In-Reply-To: <CADhaCUG7RtcpHVc57LhOxqpDY-9ez4C0GjoN=yFFC0c-zpYbjg@mail.gmail.com>
References: <CADhaCUG7RtcpHVc57LhOxqpDY-9ez4C0GjoN=yFFC0c-zpYbjg@mail.gmail.com>
Message-ID: <20120414140713.GF13466@mdounin.ru>

Hello!

On Fri, Apr 13, 2012 at 09:58:14PM +0200, Samuel ROZE wrote:

> Hello,
> 
> I'm writing an nginx module but I've got a problem for getting
> response headers. It's using an NGX_HTTP_LOG_PHASE handler and I want
> to have access to response headers sent to client. (with or without
> proxy_pass)
> 
> And, in fact, the headers_out variable of request isn't really
> completed. When debugging module using gdb, in the log phase handler,
> the headers_out variable is always like that:
> 
>   headers_out = {
>     headers = {
>       last = 0x1ed5530,
>       part = {
>         elts = 0x1ec6250,
>         nelts = 1,
>         next = 0x0
>       },
>       size = 48,
>       nalloc = 20,
>       pool = 0x1ec6200
>     },
>     status = 304,
>     status_line = {
>       len = 0,
>       data = 0x0
>     },
>     server = 0x0,
>     date = 0x0,
>     content_length = 0x0,
>     content_encoding = 0x0,
>     location = 0x0,
>     refresh = 0x0,
>     last_modified = 0x0,
>     content_range = 0x0,
>     accept_ranges = 0x0,
>     www_authenticate = 0x0,
>     expires = 0x0,
>     etag = 0x0,
>     override_charset = 0x0,
>     content_type_len = 9,
>     content_type = {
>       len = 0,
>       data = 0x1ed1248 "text/html"
>     },
>     charset = {
>       len = 0,
>       data = 0x0
>     },
>     content_type_lowcase = 0x0,
>     content_type_hash = 0,
>     cache_control = {
>       elts = 0x0,
>       nelts = 0,
>       size = 0,
>       nalloc = 0,
>       pool = 0x0
>     },
>     content_length_n = -1,
>     date_time = 0,
>     last_modified_time = 1334219772
>   },
> 
> The headers_out array is containing the only request that I added
> using "add_header" in location, and the "server", "date", etc...
> variables are empty.
> 
> Do you know why there isn't any value in headers_out structure? And do
> you know how can I get these headers?

The "Server", "Date", "Content-Length", "Last-Modified" and other 
standard headers aren't normally exists in r->headers_out for 
nginx's own responses, they are instead constructed on the fly in 
ngx_http_header_filter.c.

The r->headers_out structure either hold needed data (as 
last_modified_time for the "Last-Modified" header) or just nothing 
(as for the "Date" header, as it just uses current time).

Maxim Dounin


From adnan at rihan.fr  Sun Apr 15 15:22:03 2012
From: adnan at rihan.fr (Adnan RIHAN)
Date: Sun, 15 Apr 2012 17:22:03 +0200
Subject: Rewrite non-ssl to ssl except for given location
Message-ID: <EE20F18F5E0247F8B7E6BCA007F7A98B@gmail.com>

Hi everybody,

I'm using an internal proxy for one of my websites (just a PHP script which does a curl request for a given script).

I've forced every non-ssl location to ssl with this:
> rewrite ^/(.*) https://$host/$1 permanent;


The thing is that my scripts doesn't need SSL as it's internal, so I want to force SSL except for a given location:
> location = /includes/api.php {
> }  
> rewrite ^/(.*) https://$host/$1 permanent;



Unfortunately, this isn't enough and queries to /includes/api.php?foo=bar still sends me a 301 redirect.
Am I missing something ?
Thanks for your help.
---
Cordialement, Adnan RIHAN.
Pr?sident-Fondateur de l'association (de loi 1901) Virtual-Info (http://www.virtual-info.info/), h?bergeur Web et Serveurs de Jeux.
Consultant (http://rihan.fr/)-Technicien Sup?rieur en Informatique de Gestion.
Ambassadeur Qt (http://lyt.me/7E) (Projet Tag-PG (http://rihan.fr/fr/projets/tagpg)).

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20120415/fd4fd27d/attachment.html>

From edho at myconan.net  Sun Apr 15 15:28:18 2012
From: edho at myconan.net (Edho Arief)
Date: Sun, 15 Apr 2012 22:28:18 +0700
Subject: Rewrite non-ssl to ssl except for given location
In-Reply-To: <EE20F18F5E0247F8B7E6BCA007F7A98B@gmail.com>
References: <EE20F18F5E0247F8B7E6BCA007F7A98B@gmail.com>
Message-ID: <CAAtReC=BN+A2=0_fk=uDS9oT3Hn_zLG2=yC5Ku2N663Maegtcw@mail.gmail.com>

2012/4/15 Adnan RIHAN <adnan at rihan.fr>:
> Hi everybody,
>
> I'm using an internal proxy for one of my websites (just a PHP script which
> does a curl request for a given script).
>
> I've forced every non-ssl location to ssl with this:
>
> rewrite ^/(.*) https://$host/$1 permanent;
>
>
> The thing is that my scripts doesn't need SSL as it's internal, so I want to
> force SSL except for a given location:
>
> location = /includes/api.php {
> }
> rewrite ^/(.*) https://$host/$1 permanent;
>
>
> Unfortunately, this isn't enough and queries to /includes/api.php?foo=bar
> still sends me a 301 redirect.
> Am I missing something ?

location / {
  return 301 https://$host$request_uri;
}
location = /includes/api.php {
  ...
}


From adnan at rihan.fr  Sun Apr 15 15:32:42 2012
From: adnan at rihan.fr (Adnan RIHAN)
Date: Sun, 15 Apr 2012 17:32:42 +0200
Subject: Rewrite non-ssl to ssl except for given location
In-Reply-To: <CAAtReC=BN+A2=0_fk=uDS9oT3Hn_zLG2=yC5Ku2N663Maegtcw@mail.gmail.com>
References: <EE20F18F5E0247F8B7E6BCA007F7A98B@gmail.com>
	<CAAtReC=BN+A2=0_fk=uDS9oT3Hn_zLG2=yC5Ku2N663Maegtcw@mail.gmail.com>
Message-ID: <45625DD6CC8B4612A27C9274CB1826AE@gmail.com>

> location / {
> return 301 https://$host$request_uri;
> }
> location = /includes/api.php {
> ...
> }
>  
>  

The problem is that I'm on a web hosting service which allow s customers to edit an included config file.
 So, "location /" is a duplicate :/

Isn't there a way to write "location [not matching] /includes/api.php {}" ?  

---
Cordialement, Adnan RIHAN.
Pr?sident-Fondateur de l'association (de loi 1901) Virtual-Info (http://www.virtual-info.info/), h?bergeur Web et Serveurs de Jeux.
Consultant (http://rihan.fr/)-Technicien Sup?rieur en Informatique de Gestion.
Ambassadeur Qt (http://lyt.me/7E) (Projet Tag-PG (http://rihan.fr/fr/projets/tagpg)).


Le dimanche 15 avril 2012 ? 17:28, Edho Arief a ?crit :

> 2012/4/15 Adnan RIHAN <adnan at rihan.fr (mailto:adnan at rihan.fr)>:
> > Hi everybody,
> >  
> > I'm using an internal proxy for one of my websites (just a PHP script which
> > does a curl request for a given script).
> >  
> > I've forced every non-ssl location to ssl with this:
> >  
> > rewrite ^/(.*) https://$host/$1 permanent;
> >  
> >  
> > The thing is that my scripts doesn't need SSL as it's internal, so I want to
> > force SSL except for a given location:
> >  
> > location = /includes/api.php {
> > }
> > rewrite ^/(.*) https://$host/$1 permanent;
> >  
> >  
> > Unfortunately, this isn't enough and queries to /includes/api.php?foo=bar
> > still sends me a 301 redirect.
> > Am I missing something ?
> >  
>  
>  
> location / {
> return 301 https://$host$request_uri;
> }
> location = /includes/api.php {
> ...
> }
>  
> _______________________________________________
> nginx mailing list
> nginx at nginx.org (mailto:nginx at nginx.org)
> http://mailman.nginx.org/mailman/listinfo/nginx
>  
>  


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20120415/0be9b6af/attachment.html>

From contact at jpluscplusm.com  Sun Apr 15 15:42:54 2012
From: contact at jpluscplusm.com (Jonathan Matthews)
Date: Sun, 15 Apr 2012 16:42:54 +0100
Subject: Rewrite non-ssl to ssl except for given location
In-Reply-To: <45625DD6CC8B4612A27C9274CB1826AE@gmail.com>
References: <EE20F18F5E0247F8B7E6BCA007F7A98B@gmail.com>
	<CAAtReC=BN+A2=0_fk=uDS9oT3Hn_zLG2=yC5Ku2N663Maegtcw@mail.gmail.com>
	<45625DD6CC8B4612A27C9274CB1826AE@gmail.com>
Message-ID: <CAKsTx7D23FvnMpW_ssDQX38EKngoDZ-cbWcuwbZ2L3Nk+EGR=Q@mail.gmail.com>

On 15 April 2012 16:32, Adnan RIHAN <adnan at rihan.fr> wrote:
> location / {
> return 301 https://$host$request_uri;
> }
> location = /includes/api.php {
> ...
> }
>
> The problem is that I'm on a web hosting service which allow s customers to
> edit an included config file.
> ?So, "location /" is a duplicate :/
>
> Isn't there a way to write "location [not matching] /includes/api.php {}" ?

The method mentioned above *is* the canonical way to do that. It's in
the docs somewhere.

Anything else will be a hack of some kind, strongly influenced by your
specific nginx setup.  You'll need to give more complete information
about the nuances of the shared hosting setup in order for someone to
work out if there's a way to achieve what you want to.

> ---
> Cordialement, Adnan RIHAN.
> Pr?sident-Fondateur de l'association (de loi 1901) Virtual-Info, h?bergeur

You have the wrong signature separator here. It's not "---"
(dash-dash-dash), it's "-- " (dash-dash-space).

Jonathan
-- 
Jonathan Matthews
Oxford, London, UK
http://www.jpluscplusm.com/contact.html


From edho at myconan.net  Sun Apr 15 15:43:46 2012
From: edho at myconan.net (Edho Arief)
Date: Sun, 15 Apr 2012 22:43:46 +0700
Subject: Rewrite non-ssl to ssl except for given location
In-Reply-To: <45625DD6CC8B4612A27C9274CB1826AE@gmail.com>
References: <EE20F18F5E0247F8B7E6BCA007F7A98B@gmail.com>
	<CAAtReC=BN+A2=0_fk=uDS9oT3Hn_zLG2=yC5Ku2N663Maegtcw@mail.gmail.com>
	<45625DD6CC8B4612A27C9274CB1826AE@gmail.com>
Message-ID: <CAAtReCnnxjk1w012Qy7LvhnuLrQmJkh1UxOHEk2evsjUSTPrOA@mail.gmail.com>

2012/4/15 Adnan RIHAN <adnan at rihan.fr>:
> location / {
> return 301 https://$host$request_uri;
> }
> location = /includes/api.php {
> ...
> }
>
> The problem is that I'm on a web hosting service which allow s customers to
> edit an included config file.
> ?So, "location /" is a duplicate :/
>

This is valid:

    location / {
      location / {
      }
    }


From adnan at rihan.fr  Sun Apr 15 16:16:19 2012
From: adnan at rihan.fr (Adnan RIHAN)
Date: Sun, 15 Apr 2012 18:16:19 +0200
Subject: Rewrite non-ssl to ssl except for given location
In-Reply-To: <CAKsTx7D23FvnMpW_ssDQX38EKngoDZ-cbWcuwbZ2L3Nk+EGR=Q@mail.gmail.com>
References: <EE20F18F5E0247F8B7E6BCA007F7A98B@gmail.com>
	<CAAtReC=BN+A2=0_fk=uDS9oT3Hn_zLG2=yC5Ku2N663Maegtcw@mail.gmail.com>
	<45625DD6CC8B4612A27C9274CB1826AE@gmail.com>
	<CAKsTx7D23FvnMpW_ssDQX38EKngoDZ-cbWcuwbZ2L3Nk+EGR=Q@mail.gmail.com>
Message-ID: <BB221AC1EB26420193E1FFF90FE72350@gmail.com>

@Jonathan: okok

Here is the known but can't be touched vhost file:
> #################################################################
> ########### Vhost configuration of  example.com
> #################################################################
> server {
> access_log off;
> error_log  logs/vhost-error_log warn;
> listen    80;
> server_name  example.com www.example.com;
> location ~* ^.+.(jpg|jpeg|gif|png|ico|zip|tgz|gz|rar|bz2|iso|doc|xls|exe|pdf|ppt|txt|tar|mid|midi|wav|bmp|rtf|mp3|ogv|ogg|flv|swf|mpeg|mpg|mpeg4|mp4|avi|wmv|js|css)$ {
> expires 24h;
> root /home/example/public_html;
> error_page  404 = @apache;
> access_log /usr/local/apache/domlogs/example.com;
> log_not_found  off;  
> }  
> location ~ /\.ht {
>           deny all;
> }
>  
> location / {
> log_not_found  off;
>  
>   client_max_body_size    2000m;
>   client_body_buffer_size 512k;
>   proxy_send_timeout   90;
>   proxy_read_timeout   90;
>   proxy_buffer_size    32k;
>   proxy_buffers     16 32k;
>   proxy_busy_buffers_size 64k;
>   proxy_temp_file_write_size 64k;
>   proxy_connect_timeout 30s;
>  
>         proxy_redirect  http://www.example.com:8888   http://www.example.com;
>         proxy_redirect  http://example.com:8888   http://example.com;
>         proxy_pass   http://87.98.132.48:8888/;
>         proxy_set_header   Host   $host;
>         proxy_set_header   X-Real-IP  $remote_addr;
>         proxy_set_header   X-Forwarded-For $proxy_add_x_forwarded_for;
>         }
> location @apache {
> internal;
>  
>   client_max_body_size    2000m;
>   client_body_buffer_size 512k;
>   proxy_send_timeout   90;
>   proxy_read_timeout   90;
>   proxy_buffer_size    32k;
>   proxy_buffers     16 32k;
>   proxy_busy_buffers_size 64k;
>   proxy_temp_file_write_size 64k;
>   proxy_connect_timeout 30s;
>  
>         proxy_redirect  http://example.com:8888   http://example.com;
>         proxy_pass   http://87.98.132.48:8888;
>         proxy_set_header   Host   $host;
>         proxy_set_header   X-Real-IP  $remote_addr;
>         proxy_set_header   X-Forwarded-For $proxy_add_x_forwarded_for;
>         }
> include "/etc/cpnginx/custom/example.com";
> }


The only file I can modify is this one: /etc/cpnginx/custom/example.com

For forcing SSL, here is what I wrote in the custom file:

> root at shadow [/etc/cpnginx/custom]# cat example.com  
> rewrite ^/(.*) https://$host/$1 permanent;

I've also changed my signature ;)
--  
Cordialement, Adnan RIHAN.
Pr?sident-Fondateur de l'association (de loi 1901) Virtual-Info (http://www.virtual-info.info/), h?bergeur Web et Serveurs de Jeux.
Consultant (http://rihan.fr/)-Technicien Sup?rieur en Informatique de Gestion.
Ambassadeur Qt (http://lyt.me/7E) (Projet Tag-PG (http://rihan.fr/fr/projets/tagpg)).


Le dimanche 15 avril 2012 ? 17:42, Jonathan Matthews a ?crit :

> On 15 April 2012 16:32, Adnan RIHAN <adnan at rihan.fr (mailto:adnan at rihan.fr)> wrote:
> > location / {
> > return 301 https://$host$request_uri;
> > }
> > location = /includes/api.php {
> > ...
> > }
> >  
> > The problem is that I'm on a web hosting service which allow s customers to
> > edit an included config file.
> >  So, "location /" is a duplicate :/
> >  
> > Isn't there a way to write "location [not matching] /includes/api.php {}" ?
>  
> The method mentioned above *is* the canonical way to do that. It's in
> the docs somewhere.
>  
> Anything else will be a hack of some kind, strongly influenced by your
> specific nginx setup. You'll need to give more complete information
> about the nuances of the shared hosting setup in order for someone to
> work out if there's a way to achieve what you want to.
>  
> > ---
> > Cordialement, Adnan RIHAN.
> > Pr?sident-Fondateur de l'association (de loi 1901) Virtual-Info, h?bergeur
> >  
>  
>  
> You have the wrong signature separator here. It's not "---"
> (dash-dash-dash), it's "-- " (dash-dash-space).
>  
> Jonathan
> --  
> Jonathan Matthews
> Oxford, London, UK
> http://www.jpluscplusm.com/contact.html
>  
> _______________________________________________
> nginx mailing list
> nginx at nginx.org (mailto:nginx at nginx.org)
> http://mailman.nginx.org/mailman/listinfo/nginx
>  
>  


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20120415/20705939/attachment-0001.html>

From contact at jpluscplusm.com  Sun Apr 15 16:50:53 2012
From: contact at jpluscplusm.com (Jonathan Matthews)
Date: Sun, 15 Apr 2012 17:50:53 +0100
Subject: Rewrite non-ssl to ssl except for given location
In-Reply-To: <BB221AC1EB26420193E1FFF90FE72350@gmail.com>
References: <EE20F18F5E0247F8B7E6BCA007F7A98B@gmail.com>
	<CAAtReC=BN+A2=0_fk=uDS9oT3Hn_zLG2=yC5Ku2N663Maegtcw@mail.gmail.com>
	<45625DD6CC8B4612A27C9274CB1826AE@gmail.com>
	<CAKsTx7D23FvnMpW_ssDQX38EKngoDZ-cbWcuwbZ2L3Nk+EGR=Q@mail.gmail.com>
	<BB221AC1EB26420193E1FFF90FE72350@gmail.com>
Message-ID: <CAKsTx7AC-Y+k1nWkcXg=iWah-x92k655_cZrqFVYYVtQ-xy6Ag@mail.gmail.com>

On 15 April 2012 17:16, Adnan RIHAN <adnan at rihan.fr> wrote:
> @Jonathan: okok
>
> Here is the known but can't be touched vhost file:
>
> #################################################################
> ########### Vhost configuration of ?example.com
> #################################################################
> server {
> access_log off;
> error_log ?logs/vhost-error_log warn;
> listen ? ?80;
> server_name ?example.com www.example.com;
> location ~*
> ^.+.(jpg|jpeg|gif|png|ico|zip|tgz|gz|rar|bz2|iso|doc|xls|exe|pdf|ppt|txt|tar|mid|midi|wav|bmp|rtf|mp3|ogv|ogg|flv|swf|mpeg|mpg|mpeg4|mp4|avi|wmv|js|css)$
> {
> expires 24h;
> root /home/example/public_html;
> error_page ?404 = @apache;
> access_log /usr/local/apache/domlogs/example.com;
> log_not_found ?off;
> }
> location ~ /\.ht {
> ? ? ? ? ? deny all;
> }
>
> location / {
> log_not_found ?off;
>
> ? client_max_body_size ? ?2000m;
> ? client_body_buffer_size 512k;
> ? proxy_send_timeout ? 90;
> ? proxy_read_timeout ? 90;
> ? proxy_buffer_size ? ?32k;
> ? proxy_buffers ? ? 16 32k;
> ? proxy_busy_buffers_size 64k;
> ? proxy_temp_file_write_size 64k;
> ? proxy_connect_timeout 30s;
>
> ? ? ? ? proxy_redirect ?http://www.example.com:8888
> http://www.example.com;
> ? ? ? ? proxy_redirect ?http://example.com:8888 ? http://example.com;
> ? ? ? ? proxy_pass ? http://87.98.132.48:8888/;
> ? ? ? ? proxy_set_header ? Host ? $host;
> ? ? ? ? proxy_set_header ? X-Real-IP ?$remote_addr;
> ? ? ? ? proxy_set_header ? X-Forwarded-For $proxy_add_x_forwarded_for;
> ? ? ? ? }
> location @apache {
> internal;
>
> ? client_max_body_size ? ?2000m;
> ? client_body_buffer_size 512k;
> ? proxy_send_timeout ? 90;
> ? proxy_read_timeout ? 90;
> ? proxy_buffer_size ? ?32k;
> ? proxy_buffers ? ? 16 32k;
> ? proxy_busy_buffers_size 64k;
> ? proxy_temp_file_write_size 64k;
> ? proxy_connect_timeout 30s;
>
> ? ? ? ? proxy_redirect ?http://example.com:8888 ? http://example.com;
> ? ? ? ? proxy_pass ? http://87.98.132.48:8888;
> ? ? ? ? proxy_set_header ? Host ? $host;
> ? ? ? ? proxy_set_header ? X-Real-IP ?$remote_addr;
> ? ? ? ? proxy_set_header ? X-Forwarded-For $proxy_add_x_forwarded_for;
> ? ? ? ? }
> include "/etc/cpnginx/custom/example.com";
> }
>
> The only file I can modify is this one:?/etc/cpnginx/custom/example.com

I don't understand how this could ever work. The only entry point
you've shown to the @apache location is if the URI's path ends in
".(jpg|jpeg|gif|png|ico|zip|tgz|gz|rar|bz2|iso|doc|xls|exe|pdf|ppt|txt|tar|mid|midi|wav|bmp|rtf|mp3|ogv|ogg|flv|swf|mpeg|mpg|mpeg4|mp4|avi|wmv|js|css)"

That doesn't include ".php", so I don't see how control of
"/includes/api.php" ever hits the file you can amend.

Assuming there's something I'm missing here and .php does in fact
reach that file, here's how you might achieve what you want. It uses
PCRE negative lookaheads, which I don't know definitely work inside
nginx.

    rewrite ^/(?!includes/api\.php) https://$http_host$uri?$args;

For what it's worth, I personally wouldn't use a host that enforced
these restrictions on my configuration, like OVH appear to be doing
here. It's *extremely* unpleasant.

HTH,
Jonathan
-- 
Jonathan Matthews
Oxford, London, UK
http://www.jpluscplusm.com/contact.html


From adnan at rihan.fr  Sun Apr 15 16:57:11 2012
From: adnan at rihan.fr (Adnan RIHAN)
Date: Sun, 15 Apr 2012 18:57:11 +0200
Subject: Rewrite non-ssl to ssl except for given location
In-Reply-To: <CAKsTx7AC-Y+k1nWkcXg=iWah-x92k655_cZrqFVYYVtQ-xy6Ag@mail.gmail.com>
References: <EE20F18F5E0247F8B7E6BCA007F7A98B@gmail.com>
	<CAAtReC=BN+A2=0_fk=uDS9oT3Hn_zLG2=yC5Ku2N663Maegtcw@mail.gmail.com>
	<45625DD6CC8B4612A27C9274CB1826AE@gmail.com>
	<CAKsTx7D23FvnMpW_ssDQX38EKngoDZ-cbWcuwbZ2L3Nk+EGR=Q@mail.gmail.com>
	<BB221AC1EB26420193E1FFF90FE72350@gmail.com>
	<CAKsTx7AC-Y+k1nWkcXg=iWah-x92k655_cZrqFVYYVtQ-xy6Ag@mail.gmail.com>
Message-ID: <22D04174D3ED4139ABBD80B2BBD30405@gmail.com>

Actually, I'm the CEO of that host.  

The thing is that we're using CPanel, and a module called CPNginx.
I've send some features requests like moving the include and/or changing some positions. I've also asked them to add a "custom field" in the customer's panel to allow customers to edit their config (which will only be the include file).

Currently, I'm editing by hands included files, and I can't change the way cpnginx rebuild vhosts because it's scripts are ioncubed T_T

So, I'm hacking as I can.
-----
I don't get what you're "missing". PHP Files are never matched because there is no location block including them.
I've put my rewrite outside any blocks actually.

Thank your for your help, I'll check this soon.
--  
Cordialement, Adnan RIHAN.
Pr?sident-Fondateur de l'association (de loi 1901) Virtual-Info (http://www.virtual-info.info/), h?bergeur Web et Serveurs de Jeux.
Consultant (http://rihan.fr/)-Technicien Sup?rieur en Informatique de Gestion.
Ambassadeur Qt (http://lyt.me/7E) (Projet Tag-PG (http://rihan.fr/fr/projets/tagpg)).


Le dimanche 15 avril 2012 ? 18:50, Jonathan Matthews a ?crit :

> On 15 April 2012 17:16, Adnan RIHAN <adnan at rihan.fr (mailto:adnan at rihan.fr)> wrote:
> > @Jonathan: okok
> >  
> > Here is the known but can't be touched vhost file:
> >  
> > #################################################################
> > ########### Vhost configuration of  example.com (http://example.com)
> > #################################################################
> > server {
> > access_log off;
> > error_log  logs/vhost-error_log warn;
> > listen    80;
> > server_name  example.com www.example.com (http://www.example.com);
> > location ~*
> > ^.+.(jpg|jpeg|gif|png|ico|zip|tgz|gz|rar|bz2|iso|doc|xls|exe|pdf|ppt|txt|tar|mid|midi|wav|bmp|rtf|mp3|ogv|ogg|flv|swf|mpeg|mpg|mpeg4|mp4|avi|wmv|js|css)$
> > {
> > expires 24h;
> > root /home/example/public_html;
> > error_page  404 = @apache;
> > access_log /usr/local/apache/domlogs/example.com (http://example.com);
> > log_not_found  off;
> > }
> > location ~ /\.ht {
> >           deny all;
> > }
> >  
> > location / {
> > log_not_found  off;
> >  
> >   client_max_body_size    2000m;
> >   client_body_buffer_size 512k;
> >   proxy_send_timeout   90;
> >   proxy_read_timeout   90;
> >   proxy_buffer_size    32k;
> >   proxy_buffers     16 32k;
> >   proxy_busy_buffers_size 64k;
> >   proxy_temp_file_write_size 64k;
> >   proxy_connect_timeout 30s;
> >  
> >         proxy_redirect  http://www.example.com:8888
> > http://www.example.com;
> >         proxy_redirect  http://example.com:8888   http://example.com;
> >         proxy_pass   http://87.98.132.48:8888/;
> >         proxy_set_header   Host   $host;
> >         proxy_set_header   X-Real-IP  $remote_addr;
> >         proxy_set_header   X-Forwarded-For $proxy_add_x_forwarded_for;
> >         }
> > location @apache {
> > internal;
> >  
> >   client_max_body_size    2000m;
> >   client_body_buffer_size 512k;
> >   proxy_send_timeout   90;
> >   proxy_read_timeout   90;
> >   proxy_buffer_size    32k;
> >   proxy_buffers     16 32k;
> >   proxy_busy_buffers_size 64k;
> >   proxy_temp_file_write_size 64k;
> >   proxy_connect_timeout 30s;
> >  
> >         proxy_redirect  http://example.com:8888   http://example.com;
> >         proxy_pass   http://87.98.132.48:8888;
> >         proxy_set_header   Host   $host;
> >         proxy_set_header   X-Real-IP  $remote_addr;
> >         proxy_set_header   X-Forwarded-For $proxy_add_x_forwarded_for;
> >         }
> > include "/etc/cpnginx/custom/example.com (http://example.com)";
> > }
> >  
> > The only file I can modify is this one: /etc/cpnginx/custom/example.com (http://example.com)
>  
> I don't understand how this could ever work. The only entry point
> you've shown to the @apache location is if the URI's path ends in
> ".(jpg|jpeg|gif|png|ico|zip|tgz|gz|rar|bz2|iso|doc|xls|exe|pdf|ppt|txt|tar|mid|midi|wav|bmp|rtf|mp3|ogv|ogg|flv|swf|mpeg|mpg|mpeg4|mp4|avi|wmv|js|css)"
>  
> That doesn't include ".php", so I don't see how control of
> "/includes/api.php" ever hits the file you can amend.
>  
> Assuming there's something I'm missing here and .php does in fact
> reach that file, here's how you might achieve what you want. It uses
> PCRE negative lookaheads, which I don't know definitely work inside
> nginx.
>  
> rewrite ^/(?!includes/api\.php) https://$http_host$uri?$args;
>  
> For what it's worth, I personally wouldn't use a host that enforced
> these restrictions on my configuration, like OVH appear to be doing
> here. It's *extremely* unpleasant.
>  
> HTH,
> Jonathan
> --  
> Jonathan Matthews
> Oxford, London, UK
> http://www.jpluscplusm.com/contact.html
>  
> _______________________________________________
> nginx mailing list
> nginx at nginx.org (mailto:nginx at nginx.org)
> http://mailman.nginx.org/mailman/listinfo/nginx
>  
>  


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20120415/d6e860e2/attachment.html>

From adnan at rihan.fr  Sun Apr 15 17:16:04 2012
From: adnan at rihan.fr (Adnan RIHAN)
Date: Sun, 15 Apr 2012 19:16:04 +0200
Subject: Rewrite non-ssl to ssl except for given location
In-Reply-To: <CAKsTx7AC-Y+k1nWkcXg=iWah-x92k655_cZrqFVYYVtQ-xy6Ag@mail.gmail.com>
References: <EE20F18F5E0247F8B7E6BCA007F7A98B@gmail.com>
	<CAAtReC=BN+A2=0_fk=uDS9oT3Hn_zLG2=yC5Ku2N663Maegtcw@mail.gmail.com>
	<45625DD6CC8B4612A27C9274CB1826AE@gmail.com>
	<CAKsTx7D23FvnMpW_ssDQX38EKngoDZ-cbWcuwbZ2L3Nk+EGR=Q@mail.gmail.com>
	<BB221AC1EB26420193E1FFF90FE72350@gmail.com>
	<CAKsTx7AC-Y+k1nWkcXg=iWah-x92k655_cZrqFVYYVtQ-xy6Ag@mail.gmail.com>
Message-ID: <78B3CA16246A478F8AF2D4D1176379C8@gmail.com>

By the way,
rewrite ^/(?!includes/api\.php) https://$http_host$uri?$args;

Is working weeeell, thank you :)  

--  
Cordialement, Adnan RIHAN.
Pr?sident-Fondateur de l'association (de loi 1901) Virtual-Info (http://www.virtual-info.info/), h?bergeur Web et Serveurs de Jeux.
Consultant (http://rihan.fr/)-Technicien Sup?rieur en Informatique de Gestion.
Ambassadeur Qt (http://lyt.me/7E) (Projet Tag-PG (http://rihan.fr/fr/projets/tagpg)).


Le dimanche 15 avril 2012 ? 18:50, Jonathan Matthews a ?crit :

> On 15 April 2012 17:16, Adnan RIHAN <adnan at rihan.fr (mailto:adnan at rihan.fr)> wrote:
> > @Jonathan: okok
> >  
> > Here is the known but can't be touched vhost file:
> >  
> > #################################################################
> > ########### Vhost configuration of  example.com (http://example.com)
> > #################################################################
> > server {
> > access_log off;
> > error_log  logs/vhost-error_log warn;
> > listen    80;
> > server_name  example.com www.example.com (http://www.example.com);
> > location ~*
> > ^.+.(jpg|jpeg|gif|png|ico|zip|tgz|gz|rar|bz2|iso|doc|xls|exe|pdf|ppt|txt|tar|mid|midi|wav|bmp|rtf|mp3|ogv|ogg|flv|swf|mpeg|mpg|mpeg4|mp4|avi|wmv|js|css)$
> > {
> > expires 24h;
> > root /home/example/public_html;
> > error_page  404 = @apache;
> > access_log /usr/local/apache/domlogs/example.com (http://example.com);
> > log_not_found  off;
> > }
> > location ~ /\.ht {
> >           deny all;
> > }
> >  
> > location / {
> > log_not_found  off;
> >  
> >   client_max_body_size    2000m;
> >   client_body_buffer_size 512k;
> >   proxy_send_timeout   90;
> >   proxy_read_timeout   90;
> >   proxy_buffer_size    32k;
> >   proxy_buffers     16 32k;
> >   proxy_busy_buffers_size 64k;
> >   proxy_temp_file_write_size 64k;
> >   proxy_connect_timeout 30s;
> >  
> >         proxy_redirect  http://www.example.com:8888
> > http://www.example.com;
> >         proxy_redirect  http://example.com:8888   http://example.com;
> >         proxy_pass   http://87.98.132.48:8888/;
> >         proxy_set_header   Host   $host;
> >         proxy_set_header   X-Real-IP  $remote_addr;
> >         proxy_set_header   X-Forwarded-For $proxy_add_x_forwarded_for;
> >         }
> > location @apache {
> > internal;
> >  
> >   client_max_body_size    2000m;
> >   client_body_buffer_size 512k;
> >   proxy_send_timeout   90;
> >   proxy_read_timeout   90;
> >   proxy_buffer_size    32k;
> >   proxy_buffers     16 32k;
> >   proxy_busy_buffers_size 64k;
> >   proxy_temp_file_write_size 64k;
> >   proxy_connect_timeout 30s;
> >  
> >         proxy_redirect  http://example.com:8888   http://example.com;
> >         proxy_pass   http://87.98.132.48:8888;
> >         proxy_set_header   Host   $host;
> >         proxy_set_header   X-Real-IP  $remote_addr;
> >         proxy_set_header   X-Forwarded-For $proxy_add_x_forwarded_for;
> >         }
> > include "/etc/cpnginx/custom/example.com (http://example.com)";
> > }
> >  
> > The only file I can modify is this one: /etc/cpnginx/custom/example.com (http://example.com)
>  
> I don't understand how this could ever work. The only entry point
> you've shown to the @apache location is if the URI's path ends in
> ".(jpg|jpeg|gif|png|ico|zip|tgz|gz|rar|bz2|iso|doc|xls|exe|pdf|ppt|txt|tar|mid|midi|wav|bmp|rtf|mp3|ogv|ogg|flv|swf|mpeg|mpg|mpeg4|mp4|avi|wmv|js|css)"
>  
> That doesn't include ".php", so I don't see how control of
> "/includes/api.php" ever hits the file you can amend.
>  
> Assuming there's something I'm missing here and .php does in fact
> reach that file, here's how you might achieve what you want. It uses
> PCRE negative lookaheads, which I don't know definitely work inside
> nginx.
>  
> rewrite ^/(?!includes/api\.php) https://$http_host$uri?$args;
>  
> For what it's worth, I personally wouldn't use a host that enforced
> these restrictions on my configuration, like OVH appear to be doing
> here. It's *extremely* unpleasant.
>  
> HTH,
> Jonathan
> --  
> Jonathan Matthews
> Oxford, London, UK
> http://www.jpluscplusm.com/contact.html
>  
> _______________________________________________
> nginx mailing list
> nginx at nginx.org (mailto:nginx at nginx.org)
> http://mailman.nginx.org/mailman/listinfo/nginx
>  
>  


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20120415/6c7d09e3/attachment-0001.html>

From contact at jpluscplusm.com  Sun Apr 15 17:20:09 2012
From: contact at jpluscplusm.com (Jonathan Matthews)
Date: Sun, 15 Apr 2012 18:20:09 +0100
Subject: Rewrite non-ssl to ssl except for given location
In-Reply-To: <22D04174D3ED4139ABBD80B2BBD30405@gmail.com>
References: <EE20F18F5E0247F8B7E6BCA007F7A98B@gmail.com>
	<CAAtReC=BN+A2=0_fk=uDS9oT3Hn_zLG2=yC5Ku2N663Maegtcw@mail.gmail.com>
	<45625DD6CC8B4612A27C9274CB1826AE@gmail.com>
	<CAKsTx7D23FvnMpW_ssDQX38EKngoDZ-cbWcuwbZ2L3Nk+EGR=Q@mail.gmail.com>
	<BB221AC1EB26420193E1FFF90FE72350@gmail.com>
	<CAKsTx7AC-Y+k1nWkcXg=iWah-x92k655_cZrqFVYYVtQ-xy6Ag@mail.gmail.com>
	<22D04174D3ED4139ABBD80B2BBD30405@gmail.com>
Message-ID: <CAKsTx7BwO5XDOSW2v2sOxE580FPaXxH2OYtDPQsVSH4UJ=f_fg@mail.gmail.com>

On 15 April 2012 17:57, Adnan RIHAN <adnan at rihan.fr> wrote:
[snip]
> I don't get what you're "missing". PHP Files are never matched because there
> is no location block including them.
> I've put my rewrite outside any blocks actually.

Ah; your indenting fooled me into thinking your include was the last
line of the @apache location. Now that I've noticed it's /outside/ any
location, I can see how it works. The negative lookahead should work
fine here :-)

Jonathan
-- 
Jonathan Matthews
Oxford, London, UK
http://www.jpluscplusm.com/contact.html


From adnan at rihan.fr  Sun Apr 15 17:22:50 2012
From: adnan at rihan.fr (Adnan RIHAN)
Date: Sun, 15 Apr 2012 19:22:50 +0200
Subject: Rewrite non-ssl to ssl except for given location
In-Reply-To: <CAKsTx7BwO5XDOSW2v2sOxE580FPaXxH2OYtDPQsVSH4UJ=f_fg@mail.gmail.com>
References: <EE20F18F5E0247F8B7E6BCA007F7A98B@gmail.com>
	<CAAtReC=BN+A2=0_fk=uDS9oT3Hn_zLG2=yC5Ku2N663Maegtcw@mail.gmail.com>
	<45625DD6CC8B4612A27C9274CB1826AE@gmail.com>
	<CAKsTx7D23FvnMpW_ssDQX38EKngoDZ-cbWcuwbZ2L3Nk+EGR=Q@mail.gmail.com>
	<BB221AC1EB26420193E1FFF90FE72350@gmail.com>
	<CAKsTx7AC-Y+k1nWkcXg=iWah-x92k655_cZrqFVYYVtQ-xy6Ag@mail.gmail.com>
	<22D04174D3ED4139ABBD80B2BBD30405@gmail.com>
	<CAKsTx7BwO5XDOSW2v2sOxE580FPaXxH2OYtDPQsVSH4UJ=f_fg@mail.gmail.com>
Message-ID: <DE72A6FC6F174F0FB076A8449761DCBC@gmail.com>

That's what I answered, it's working and thank you.
And sorry for being a host which enforce strange configs, but I'm currently trying to  do better, and it's not easy (as the dev of the softwares I'm using are a bit short-minded...).

Have a nice day !  

--  
Cordialement, Adnan RIHAN.
Pr?sident-Fondateur de l'association (de loi 1901) Virtual-Info (http://www.virtual-info.info/), h?bergeur Web et Serveurs de Jeux.
Consultant (http://rihan.fr/)-Technicien Sup?rieur en Informatique de Gestion.
Ambassadeur Qt (http://lyt.me/7E) (Projet Tag-PG (http://rihan.fr/fr/projets/tagpg)).


Le dimanche 15 avril 2012 ? 19:20, Jonathan Matthews a ?crit :

> On 15 April 2012 17:57, Adnan RIHAN <adnan at rihan.fr (mailto:adnan at rihan.fr)> wrote:
> [snip]
> > I don't get what you're "missing". PHP Files are never matched because there
> > is no location block including them.
> > I've put my rewrite outside any blocks actually.
> >  
>  
>  
> Ah; your indenting fooled me into thinking your include was the last
> line of the @apache location. Now that I've noticed it's /outside/ any
> location, I can see how it works. The negative lookahead should work
> fine here :-)
>  
> Jonathan
> --  
> Jonathan Matthews
> Oxford, London, UK
> http://www.jpluscplusm.com/contact.html
>  
> _______________________________________________
> nginx mailing list
> nginx at nginx.org (mailto:nginx at nginx.org)
> http://mailman.nginx.org/mailman/listinfo/nginx
>  
>  


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20120415/c1b39db6/attachment.html>

From tdgh2323 at hotmail.com  Sun Apr 15 20:18:27 2012
From: tdgh2323 at hotmail.com (Micheal Wolfskill)
Date: Sun, 15 Apr 2012 20:18:27 +0000
Subject: How make nginx write cache files in 770 permissions instead of 700?
Message-ID: <BLU144-W1426197A197E2A32AB663FD0390@phx.gbl>


Hello all,

I included a special user into the nginx group with the purpose of giving privilege to be able to delete or read nginx cache files. The problem is nginx writes these in 700 mode. What is the simplest way I can achieve what I want? Perhaps forcing nginx to write cache files in 770 ?

Regards,
Mike
 		 	   		  
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20120415/47a43aac/attachment.html>

From appa at perusio.net  Sun Apr 15 22:13:04 2012
From: appa at perusio.net (=?UTF-8?B?QW50w7NuaW8=?= P. P. Almeida)
Date: Mon, 16 Apr 2012 00:13:04 +0200
Subject: How make nginx write cache files in 770 permissions instead of
	700?
In-Reply-To: <BLU144-W1426197A197E2A32AB663FD0390@phx.gbl>
References: <BLU144-W1426197A197E2A32AB663FD0390@phx.gbl>
Message-ID: <87sjg43aq7.wl%appa@perusio.net>

On 15 Abr 2012 22h18 CEST, tdgh2323 at hotmail.com wrote:

>
> Hello all,
>
> I included a special user into the nginx group with the purpose of
> giving privilege to be able to delete or read nginx cache files. The
> problem is nginx writes these in 700 mode. What is the simplest way
> I can achieve what I want? Perhaps forcing nginx to write cache
> files in 770 ?

Isn't it simpler and potentially safer to use a setuid wrapper?

I like super: http://www.ucolick.org/~will/RUE/super/super.1.html

You can create a wrapper command that grants the right to delete the
cache files to a specific user.

--- appa


From nginx-forum at nginx.us  Mon Apr 16 03:34:48 2012
From: nginx-forum at nginx.us (bpiraeus)
Date: Sun, 15 Apr 2012 23:34:48 -0400 (EDT)
Subject: SSL negotiation failures causing 408 error to be returned
Message-ID: <3f7ca0ac8ba89cce43fad095d7821c7a.NginxMailingListEnglish@forum.nginx.org>

While debugging for a client, we found that we're seeing significant
numbers of 408s being generated for SSL connections. It seems to be
MOSTLY Chrome on a Windows 7 base with some MSIE 9.0 (also on Win 7) as
user agents go, and at this point I'm completely stumped. We've tried
disabling session cache, upping timeout values, enabling/disabling
keepalives, nothing seems to stem the steady flow of these. Below is a
sample debug log (I've redacted specific pieces for business reasons).
What I see a lot of is 6 simultaneous connections being setup, 2 being
used, and then 4 timing out 60s later. Invariably, if I enable full
debug, I see 'client timed out (60: Operation timed out) while SSL
handshaking' which makes no sense. If I turn on logging for session IDs,
I can clearly see the clients being sent 408s are being assigned session
IDs which says to me they're completing the SSL negotiation and then
just.. sitting idle.

2012/04/14 14:51:41 [debug] 99433#0: *3313 accept: <CLIENT IP> fd:100
2012/04/14 14:51:41 [debug] 99433#0: *3313 post event 0000000801A026F0
2012/04/14 14:51:41 [debug] 99433#0: *3313 delete posted event
0000000801A026F0
2012/04/14 14:51:41 [debug] 99433#0: *3313 malloc:
0000000801EEF000:1256
2012/04/14 14:51:41 [debug] 99433#0: *3313 posix_memalign:
0000000801EE8300:256 @16
2012/04/14 14:51:41 [debug] 99433#0: *3313 malloc:
0000000800DE5000:4096
2012/04/14 14:51:41 [debug] 99433#0: *3313 posix_memalign:
0000000800DFE000:4096 @16
2012/04/14 14:51:41 [debug] 99433#0: *3313 http check ssl handshake
2012/04/14 14:51:41 [debug] 99433#0: *3313 https ssl handshake: 0x16
2012/04/14 14:51:41 [debug] 99433#0: *3313 SSL server name: "<SERVER
NAME>"
2012/04/14 14:51:41 [debug] 99433#0: *3313 SSL_do_handshake: -1
2012/04/14 14:51:41 [debug] 99433#0: *3313 SSL_get_error: 2
2012/04/14 14:51:41 [debug] 99433#0: *3313 kevent set event: 100: ft:-1
fl:0025
2012/04/14 14:51:41 [debug] 99433#0: *3313 event timer add: 100:
60000:1334440361120
2012/04/14 14:51:41 [debug] 99433#0: *3313 post event 0000000801A026F0
2012/04/14 14:51:41 [debug] 99433#0: *3313 delete posted event
0000000801A026F0
2012/04/14 14:51:41 [debug] 99433#0: *3313 SSL handshake handler: 0
2012/04/14 14:51:41 [debug] 99433#0: *3313 SSL_do_handshake: 1
2012/04/14 14:51:41 [debug] 99433#0: *3313 SSL: TLSv1, cipher: "RC4-SHA
SSLv3 Kx=RSA Au=RSA Enc=RC4(128) Mac=SHA1"
2012/04/14 14:51:41 [debug] 99433#0: *3313 SSL reused session
2012/04/14 14:51:41 [debug] 99433#0: *3313 http process request line
2012/04/14 14:51:41 [debug] 99433#0: *3313 SSL_read: -1
2012/04/14 14:51:41 [debug] 99433#0: *3313 SSL_get_error: 2
2012/04/14 14:52:41 [debug] 99433#0: *3313 event timer del: 100:
1334440361120
2012/04/14 14:52:41 [debug] 99433#0: *3313 http process request line
2012/04/14 14:52:41 [info] 99433#0: *3313 client timed out (60:
Operation timed out) while SSL handshaking, client: <CLIENT IP>,
server:
2012/04/14 14:52:41 [debug] 99433#0: *3313 http request count:1 blk:0
2012/04/14 14:52:41 [debug] 99433#0: *3313 http close request
2012/04/14 14:52:41 [debug] 99433#0: *3313 http log handler
2012/04/14 14:52:41 [debug] 99433#0: *3313 free: 0000000800DFE000,
unused: 2426
2012/04/14 14:52:41 [debug] 99433#0: *3313 close http connection: 100
2012/04/14 14:52:41 [debug] 99433#0: *3313 SSL_shutdown: 1
2012/04/14 14:52:41 [debug] 99433#0: *3313 reusable connection: 0
2012/04/14 14:52:41 [debug] 99433#0: *3313 free: 0000000800DE5000
2012/04/14 14:52:41 [debug] 99433#0: *3313 free: 0000000801EEF000
2012/04/14 14:52:41 [debug] 99433#0: *3313 free: 0000000801EE8200,
unused: 8
2012/04/14 14:52:41 [debug] 99433#0: *3313 free: 0000000801EE8300,
unused: 72

The following 5 UAs are by far the most common

Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/535.19 (KHTML, like
Gecko) Chrome/18.0.1025.152 Safari/535.19
Mozilla/5.0 (Windows NT 6.1) AppleWebKit/535.19 (KHTML, like Gecko)
Chrome/18.0.1025.162 Safari/535.19
Mozilla/5.0 (Windows NT 5.1) AppleWebKit/535.19 (KHTML, like Gecko)
Chrome/18.0.1025.162 Safari/535.19
Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/535.19 (KHTML, like
Gecko) Chrome/18.0.1025.162 Safari/535.19
Mozilla/5.0 (compatible; MSIE 9.0; Windows NT 6.1; WOW64; Trident/5.0)

This appears to implicate Windows in some way, but I've yet to figure it
out and am, quite frankly, stumped.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225349,225349#msg-225349


From nginx-forum at nginx.us  Mon Apr 16 03:38:29 2012
From: nginx-forum at nginx.us (bpiraeus)
Date: Sun, 15 Apr 2012 23:38:29 -0400 (EDT)
Subject: SSL negotiation failures causing 408 error to be returned
In-Reply-To: <3f7ca0ac8ba89cce43fad095d7821c7a.NginxMailingListEnglish@forum.nginx.org>
References: <3f7ca0ac8ba89cce43fad095d7821c7a.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <989062e0f76afb1d746dee0e6170ab3c.NginxMailingListEnglish@forum.nginx.org>

I should note, I've also tried various revisions of nginx, the debug was
taken using 1.1.19 w/ the following configure options

--with-http_ssl_module --with-http_stub_status_module --with-ipv6
--with-pcre --with-debug --with-openssl=/usr/local/src/openssl-1.0.1

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225349,225350#msg-225350


From list at oxdrove.co.uk  Mon Apr 16 08:37:27 2012
From: list at oxdrove.co.uk (James Lee)
Date: Mon, 16 Apr 2012 09:37:27 +0100
Subject: truncated output with gzip, large files and slow rate
Message-ID: <op.wcunkpg53ig1dk@sessia>

Hello,

nginx is truncating output with the following:
  + gzip on
  + compressed size greater than 100k (approx)
  + client downloads at a slow rate


*** Steps to reproduce ***

nginx.conf:

events {
}
http {
     types {
         text/plain txt;
     }
     gzip on;
     gzip_types text/plain;
     server {
         listen 80 default_server;
         server_name _;
         root /path/to/root;
     }
}


Create a test file that is larger than 100k compressed:

$ head -1000 /dev/random > /path/to/root/test.txt


Request exhibiting failure:

$ curl -H "Accept-Encoding: gzip" --limit-rate 50k localhost/test.txt |  
gzcat | cksum
   % Total    % Received % Xferd  Average Speed   Time    Time     Time   
Current
                                  Dload  Upload   Total   Spent    Left   
Speed
100  180k    0  180k    0     0   2966      0 --:--:--  0:01:02  
--:--:--     0
curl: (18) transfer closed with outstanding read data remaining

gzcat: stdin: unexpected end of file
649550918       184260


and with wget:

$ wget --header "Accept-Encoding: gzip" --limit-rate=50k -O /dev/stdout  
localhost/test.txt | gzcat | cksum

--2012-04-15 11:27:04--  http://localhost/test.txt
Resolving localhost (localhost)... 127.0.0.1, ::1
Connecting to localhost (localhost)|127.0.0.1|:80... connected.
HTTP request sent, awaiting response... 200 OK
Length: unspecified [text/plain]
Saving to: `/dev/stdout'

     [                                <=>     ] 184,330     --.-K/s   in 61s

2012-04-15 11:28:06 (2.96 KB/s) - `/dev/stdout' saved [184330]


gzcat: stdin: unexpected end of file
649550918       184260




Some variants showing success (correct size and cksum) by avoiding one of  
compression or limit-rate:

$ curl -H "Accept-Encoding: gzip" localhost/test.txt | gzcat | cksum
   % Total    % Received % Xferd  Average Speed   Time    Time     Time   
Current
                                  Dload  Upload   Total   Spent    Left   
Speed
100  235k    0  235k    0     0  5970k      0 --:--:-- --:--:-- --:--:--  
6370k
2871747365      241269


$ curl --limit-rate 50k localhost/test.txt | cksum
   % Total    % Received % Xferd  Average Speed   Time    Time     Time   
Current
                                  Dload  Upload   Total   Spent    Left   
Speed
100  235k  100  235k    0     0  47700      0  0:00:05  0:00:05 --:--:--  
55650
2871747365      241269



There are also failures with major web browsers including Firefox 11.0,  
MSIE 9.0 and Safari.

No error is reported by nginx.  The http return code and logged code is  
200.  The only clue in the server is the sent size reported is smaller  
than the expected compressed size.



*** Other information ***

nginx 1.1.19
Solaris 10 u10




James.


From nginx-forum at nginx.us  Mon Apr 16 08:43:46 2012
From: nginx-forum at nginx.us (lockev3.0)
Date: Mon, 16 Apr 2012 04:43:46 -0400 (EDT)
Subject: Default_server catch all block not working
In-Reply-To: <CAKsTx7A+HKp65M0meMn5v1XQ2emLRGYj8Htj0+akL9=ZY_ZmMA@mail.gmail.com>
References: <CAKsTx7A+HKp65M0meMn5v1XQ2emLRGYj8Htj0+akL9=ZY_ZmMA@mail.gmail.com>
Message-ID: <5c1276b46f9cd1e19931715fee0069ad.NginxMailingListEnglish@forum.nginx.org>

Well, probably I'm, still novice at Nginx but, that default server block
is meant to fit when no other block server (server_name)  fits. So my
current problem is that 444 behaviour is arising all thru my nginx
server, not just that default_server block/vhost 
.....not quit sure If I explained my self better now ? or even
understood you right ....

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,221572,225354#msg-225354


From hagaizzz at yahoo.com  Mon Apr 16 08:53:39 2012
From: hagaizzz at yahoo.com (hagai avrahami)
Date: Mon, 16 Apr 2012 01:53:39 -0700 (PDT)
Subject: ngx_http_send_special_response
Message-ID: <1334566419.86321.YahooMailNeo@web125506.mail.ne1.yahoo.com>



Hi

I am appreciate any help I can get on the following issues.

I am writing my own module.

1. I can see in function ngx_http_send_special_response
??? that it calculates the content length of the error page even if the request signed as header only (r->header_only = 1)
??? and set the content length header with value, after sending the header it does not send the body ( because? r->header_only = 1)
??? so the response arrive with content length different than 0 but without body 

??? Is it a BUG?


2.In the module I am trying to redirect the request with status code 302 (NGX_HTTP_MOVED_TEMPORARILY)
?? Trying to user->headers_out.location for the Alternate URL failed, I looked in the code and saw the Location must start with "/" - why?

?? In the end I add it to r->headers_out.headers
?? using header = ngx_list_push(&r->headers_out.headers);
?? Is it the write way to do it?

3. Can you explain please the different between
??? a. r->headers_out.content_length_n
??? b. r->headers_out.content_length

4. Can you explain the usage of the hash field in ngx_table_elt_t


Thanks
Hagai Avrahami
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20120416/a7eb11ce/attachment.html>

From mdounin at mdounin.ru  Mon Apr 16 10:17:41 2012
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Mon, 16 Apr 2012 14:17:41 +0400
Subject: truncated output with gzip, large files and slow rate
In-Reply-To: <op.wcunkpg53ig1dk@sessia>
References: <op.wcunkpg53ig1dk@sessia>
Message-ID: <20120416101741.GH13466@mdounin.ru>

Hello!

On Mon, Apr 16, 2012 at 09:37:27AM +0100, James Lee wrote:

> Hello,
> 
> nginx is truncating output with the following:
>  + gzip on
>  + compressed size greater than 100k (approx)
>  + client downloads at a slow rate
> 
> 
> *** Steps to reproduce ***
> 
> nginx.conf:
> 
> events {
> }
> http {
>     types {
>         text/plain txt;
>     }
>     gzip on;
>     gzip_types text/plain;
>     server {
>         listen 80 default_server;
>         server_name _;
>         root /path/to/root;
>     }
> }
> 
> 
> Create a test file that is larger than 100k compressed:
> 
> $ head -1000 /dev/random > /path/to/root/test.txt
> 
> 
> Request exhibiting failure:
> 
> $ curl -H "Accept-Encoding: gzip" --limit-rate 50k
> localhost/test.txt | gzcat | cksum
>   % Total    % Received % Xferd  Average Speed   Time    Time
> Time  Current
>                                  Dload  Upload   Total   Spent
> Left  Speed
> 100  180k    0  180k    0     0   2966      0 --:--:--  0:01:02
> --:--:--     0
> curl: (18) transfer closed with outstanding read data remaining
> 
> gzcat: stdin: unexpected end of file
> 649550918       184260
> 
> 
> and with wget:
> 
> $ wget --header "Accept-Encoding: gzip" --limit-rate=50k -O
> /dev/stdout localhost/test.txt | gzcat | cksum
> 
> --2012-04-15 11:27:04--  http://localhost/test.txt
> Resolving localhost (localhost)... 127.0.0.1, ::1
> Connecting to localhost (localhost)|127.0.0.1|:80... connected.
> HTTP request sent, awaiting response... 200 OK
> Length: unspecified [text/plain]
> Saving to: `/dev/stdout'
> 
>     [                                <=>     ] 184,330     --.-K/s   in 61s
> 
> 2012-04-15 11:28:06 (2.96 KB/s) - `/dev/stdout' saved [184330]
> 
> 
> gzcat: stdin: unexpected end of file
> 649550918       184260
> 
> 
> 
> 
> Some variants showing success (correct size and cksum) by avoiding
> one of compression or limit-rate:
> 
> $ curl -H "Accept-Encoding: gzip" localhost/test.txt | gzcat | cksum
>   % Total    % Received % Xferd  Average Speed   Time    Time
> Time  Current
>                                  Dload  Upload   Total   Spent
> Left  Speed
> 100  235k    0  235k    0     0  5970k      0 --:--:-- --:--:--
> --:--:-- 6370k
> 2871747365      241269
> 
> 
> $ curl --limit-rate 50k localhost/test.txt | cksum
>   % Total    % Received % Xferd  Average Speed   Time    Time
> Time  Current
>                                  Dload  Upload   Total   Spent
> Left  Speed
> 100  235k  100  235k    0     0  47700      0  0:00:05  0:00:05
> --:--:-- 55650
> 2871747365      241269
> 
> 
> 
> There are also failures with major web browsers including Firefox
> 11.0, MSIE 9.0 and Safari.
> 
> No error is reported by nginx.  The http return code and logged code
> is 200.  The only clue in the server is the sent size reported is
> smaller than the expected compressed size.
> 
> 
> 
> *** Other information ***
> 
> nginx 1.1.19
> Solaris 10 u10

Could you please provide debug log and nginx -V output?
See http://wiki.nginx.org/Debugging for details.

Maxim Dounin


From mdounin at mdounin.ru  Mon Apr 16 10:37:45 2012
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Mon, 16 Apr 2012 14:37:45 +0400
Subject: ngx_http_send_special_response
In-Reply-To: <1334566419.86321.YahooMailNeo@web125506.mail.ne1.yahoo.com>
References: <1334566419.86321.YahooMailNeo@web125506.mail.ne1.yahoo.com>
Message-ID: <20120416103744.GI13466@mdounin.ru>

Hello!

On Mon, Apr 16, 2012 at 01:53:39AM -0700, hagai avrahami wrote:

> 
> 
> Hi
> 
> I am appreciate any help I can get on the following issues.
> 
> I am writing my own module.
> 
> 1. I can see in function ngx_http_send_special_response
> ??? that it calculates the content length of the error page even if the request signed as header only (r->header_only = 1)
> ??? and set the content length header with value, after sending the header it does not send the body ( because? r->header_only = 1)
> ??? so the response arrive with content length different than 0 but without body 
> 
> ??? Is it a BUG?

No.  The r->header_only flag should is set when body isn't 
expected per protocol (HEAD requests, 304 responses).  Having 
Content-Length present in such responses is ok.

> 2.In the module I am trying to redirect the request with status code 302 (NGX_HTTP_MOVED_TEMPORARILY)
> ?? Trying to user->headers_out.location for the Alternate URL failed, I looked in the code and saw the Location must start with "/" - why?

You have to set it correctly, i.e. add to r->headers_out.headers 
and link to r->headers_out.location.  If it starts with "/" it 
works even if set incorrectly due to adding of a server_name, 
that's probably what confused you.

> ?? In the end I add it to r->headers_out.headers
> ?? using header = ngx_list_push(&r->headers_out.headers);
> ?? Is it the write way to do it?
> 
> 3. Can you explain please the different between
> ??? a. r->headers_out.content_length_n
> ??? b. r->headers_out.content_length

The content_length_n is a numeric length, this what you normaly 
should set when generating response in nginx.  The content_length 
is a pointer to a headers array element, it might be present if 
there is string representation of a Content-Length header 
available for some reason (e.g. from a backend).

> 4. Can you explain the usage of the hash field in ngx_table_elt_t

Normally the hash field is used for fast lookups in hash 
structures.  In case of r->headers_out it's usually just set to 1 
indicate the header is valid (or reset to 0 to indicate the 
header should not be sent as it was superseeded by some other 
header).

Maxim Dounin


From mdounin at mdounin.ru  Mon Apr 16 10:59:30 2012
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Mon, 16 Apr 2012 14:59:30 +0400
Subject: SSL negotiation failures causing 408 error to be returned
In-Reply-To: <3f7ca0ac8ba89cce43fad095d7821c7a.NginxMailingListEnglish@forum.nginx.org>
References: <3f7ca0ac8ba89cce43fad095d7821c7a.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20120416105930.GJ13466@mdounin.ru>

Hello!

On Sun, Apr 15, 2012 at 11:34:48PM -0400, bpiraeus wrote:

> While debugging for a client, we found that we're seeing significant
> numbers of 408s being generated for SSL connections. It seems to be
> MOSTLY Chrome on a Windows 7 base with some MSIE 9.0 (also on Win 7) as
> user agents go, and at this point I'm completely stumped. We've tried
> disabling session cache, upping timeout values, enabling/disabling
> keepalives, nothing seems to stem the steady flow of these. Below is a
> sample debug log (I've redacted specific pieces for business reasons).
> What I see a lot of is 6 simultaneous connections being setup, 2 being
> used, and then 4 timing out 60s later. Invariably, if I enable full
> debug, I see 'client timed out (60: Operation timed out) while SSL
> handshaking' which makes no sense. If I turn on logging for session IDs,
> I can clearly see the clients being sent 408s are being assigned session
> IDs which says to me they're completing the SSL negotiation and then
> just.. sitting idle.

This is because Chrome opens many connections to a server from 
very start, trying to reduce latence in case of these connections 
will be useful.  If they aren't - they are just kept idle and 
either closed by Chrome (in this case you'll likely see 400 in 
logs) or timed out (in this case you'll see 408).

[...]

> 2012/04/14 14:52:41 [debug] 99433#0: *3313 http process request line
> 2012/04/14 14:52:41 [info] 99433#0: *3313 client timed out (60:
> Operation timed out) while SSL handshaking, client: <CLIENT IP>,
> server:

The "while SSL handshaking" part of the message is a bit 
misleading.  Probably something like this will make things a bit 
better:

--- a/src/http/ngx_http_request.c
+++ b/src/http/ngx_http_request.c
@@ -612,6 +612,8 @@ ngx_http_ssl_handshake_handler(ngx_conne
 
         c->ssl->no_wait_shutdown = 1;
 
+        c->log->action = "reading client request line";
+
         c->read->handler = ngx_http_process_request_line;
         /* STUB: epoll edge */ c->write->handler = ngx_http_empty_handler;
 


Maxim Dounin


From hagaizzz at yahoo.com  Mon Apr 16 11:29:18 2012
From: hagaizzz at yahoo.com (hagai avrahami)
Date: Mon, 16 Apr 2012 04:29:18 -0700 (PDT)
Subject: ngx_http_send_special_response
In-Reply-To: <20120416103744.GI13466@mdounin.ru>
References: <1334566419.86321.YahooMailNeo@web125506.mail.ne1.yahoo.com>
	<20120416103744.GI13466@mdounin.ru>
Message-ID: <1334575758.9686.YahooMailNeo@web125504.mail.ne1.yahoo.com>

WOW thanks for the quick and elaborate response

So what is the right way to send response with status code 302 on GET request but
with content length 0 and no body (I want to redirect the request to other Server)

Thanks 

Hagai




>________________________________
> From: Maxim Dounin <mdounin at mdounin.ru>
>To: nginx at nginx.org 
>Sent: Monday, April 16, 2012 12:37 PM
>Subject: Re: ngx_http_send_special_response
> 
>Hello!
>
>On Mon, Apr 16, 2012 at 01:53:39AM -0700, hagai avrahami wrote:
>
>> 
>> 
>> Hi
>> 
>> I am appreciate any help I can get on the following issues.
>> 
>> I am writing my own module.
>> 
>> 1. I can see in function ngx_http_send_special_response
>> ??? that it calculates the content length of the error page even if the request signed as header only (r->header_only = 1)
>> ??? and set the content length header with value, after sending the header it does not send the body ( because? r->header_only = 1)
>> ??? so the response arrive with content length different than 0 but without body 
>> 
>> ??? Is it a BUG?
>
>No.? The r->header_only flag should is set when body isn't 
>expected per protocol (HEAD requests, 304 responses).? Having 
>Content-Length present in such responses is ok.
>
>> 2.In the module I am trying to redirect the request with status code 302 (NGX_HTTP_MOVED_TEMPORARILY)
>> ?? Trying to user->headers_out.location for the Alternate URL failed, I looked in the code and saw the Location must start with "/" - why?
>
>You have to set it correctly, i.e. add to r->headers_out.headers 
>and link to r->headers_out.location.? If it starts with "/" it 
>works even if set incorrectly due to adding of a server_name, 
>that's probably what confused you.
>
>> ?? In the end I add it to r->headers_out.headers
>> ?? using header = ngx_list_push(&r->headers_out.headers);
>> ?? Is it the write way to do it?
>> 
>> 3. Can you explain please the different between
>> ??? a. r->headers_out.content_length_n
>> ??? b. r->headers_out.content_length
>
>The content_length_n is a numeric length, this what you normaly 
>should set when generating response in nginx.? The content_length 
>is a pointer to a headers array element, it might be present if 
>there is string representation of a Content-Length header 
>available for some reason (e.g. from a backend).
>
>> 4. Can you explain the usage of the hash field in ngx_table_elt_t
>
>Normally the hash field is used for fast lookups in hash 
>structures.? In case of r->headers_out it's usually just set to 1 
>indicate the header is valid (or reset to 0 to indicate the 
>header should not be sent as it was superseeded by some other 
>header).
>
>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/20120416/ddc44f5e/attachment-0001.html>

From list at oxdrove.co.uk  Mon Apr 16 13:27:08 2012
From: list at oxdrove.co.uk (James Lee)
Date: Mon, 16 Apr 2012 14:27:08 +0100
Subject: truncated output with gzip, large files and slow rate
In-Reply-To: <20120416101741.GH13466@mdounin.ru>
References: <op.wcunkpg53ig1dk@sessia> <20120416101741.GH13466@mdounin.ru>
Message-ID: <op.wcu0zino3ig1dk@sessia>

On Mon, 16 Apr 2012 11:17:41 +0100, Maxim Dounin <mdounin at mdounin.ru>  
wrote:

> Could you please provide debug log and nginx -V output?

debug.log.bz2 as attachment.


$ nginx -V
nginx version: nginx/1.1.19
TLS SNI support enabled
configure arguments: --prefix=/opt/prefix --prefix=/var/opt/prefix/nginx  
--http-client-body-temp-path=/var/opt/prefix/nginx/client  
--http-fastcgi-temp-path=/var/opt/prefix/nginx/fastcgi  
--http-scgi-temp-path=/var/opt/prefix/nginx/scgi  
--http-proxy-temp-path=/var/opt/prefix/nginx/proxy  
--conf-path=/etc/opt/prefix/nginx/nginx.conf  
--error-log-path=/var/opt/prefix/nginx/logs/error.log  
--http-log-path=/var/opt/prefix/nginx/logs/access.log  
--pid-path=/var/opt/prefix/nginx/nginx.pid  
--lock-path=/var/opt/prefix/nginx/nginx.lock --user=webservd  
--group=webservd --without-mail_pop3_module --without-mail_imap_module  
--without-mail_smtp_module --with-http_gzip_static_module  
--with-http_realip_module --with-http_ssl_module  
--with-http_stub_status_module --with-debug




James.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: debug.log.bz2
Type: application/bzip2
Size: 10955 bytes
Desc: not available
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20120416/824ef6ab/attachment.bin>

From mdounin at mdounin.ru  Mon Apr 16 13:38:23 2012
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Mon, 16 Apr 2012 17:38:23 +0400
Subject: ngx_http_send_special_response
In-Reply-To: <1334575758.9686.YahooMailNeo@web125504.mail.ne1.yahoo.com>
References: <1334566419.86321.YahooMailNeo@web125506.mail.ne1.yahoo.com>
	<20120416103744.GI13466@mdounin.ru>
	<1334575758.9686.YahooMailNeo@web125504.mail.ne1.yahoo.com>
Message-ID: <20120416133823.GK13466@mdounin.ru>

Hello!

On Mon, Apr 16, 2012 at 04:29:18AM -0700, hagai avrahami wrote:

> WOW thanks for the quick and elaborate response
> 
> So what is the right way to send response with status code 302 on GET request but
> with content length 0 and no body (I want to redirect the request to other Server)

Easiest solution is to just write

    return 302 $where;

in config. :)

>From the code point of view (assuming content handler context), 
something like this should work:

    ngx_table_elt_t  *h;

    ngx_http_clear_location(r);

    h = ngx_list_push(&r->headers_out.headers);
    if (h == NULL) {
        return NGX_HTTP_INTERNAL_SERVER_ERROR;
    }

    h->hash = 1;
    ngx_str_set(&h->key, "Location");

    h->value.data = ...
    h->value.len = ...

    r->headers_out.location = h;

    return NGX_HTTP_MOVED_TEMPORARILY;

Maxim Dounin


> 
> Thanks 
> 
> Hagai
> 
> 
> 
> 
> >________________________________
> > From: Maxim Dounin <mdounin at mdounin.ru>
> >To: nginx at nginx.org 
> >Sent: Monday, April 16, 2012 12:37 PM
> >Subject: Re: ngx_http_send_special_response
> > 
> >Hello!
> >
> >On Mon, Apr 16, 2012 at 01:53:39AM -0700, hagai avrahami wrote:
> >
> >> 
> >> 
> >> Hi
> >> 
> >> I am appreciate any help I can get on the following issues.
> >> 
> >> I am writing my own module.
> >> 
> >> 1. I can see in function ngx_http_send_special_response
> >> ??? that it calculates the content length of the error page even if the request signed as header only (r->header_only = 1)
> >> ??? and set the content length header with value, after sending the header it does not send the body ( because? r->header_only = 1)
> >> ??? so the response arrive with content length different than 0 but without body 
> >> 
> >> ??? Is it a BUG?
> >
> >No.? The r->header_only flag should is set when body isn't 
> >expected per protocol (HEAD requests, 304 responses).? Having 
> >Content-Length present in such responses is ok.
> >
> >> 2.In the module I am trying to redirect the request with status code 302 (NGX_HTTP_MOVED_TEMPORARILY)
> >> ?? Trying to user->headers_out.location for the Alternate URL failed, I looked in the code and saw the Location must start with "/" - why?
> >
> >You have to set it correctly, i.e. add to r->headers_out.headers 
> >and link to r->headers_out.location.? If it starts with "/" it 
> >works even if set incorrectly due to adding of a server_name, 
> >that's probably what confused you.
> >
> >> ?? In the end I add it to r->headers_out.headers
> >> ?? using header = ngx_list_push(&r->headers_out.headers);
> >> ?? Is it the write way to do it?
> >> 
> >> 3. Can you explain please the different between
> >> ??? a. r->headers_out.content_length_n
> >> ??? b. r->headers_out.content_length
> >
> >The content_length_n is a numeric length, this what you normaly 
> >should set when generating response in nginx.? The content_length 
> >is a pointer to a headers array element, it might be present if 
> >there is string representation of a Content-Length header 
> >available for some reason (e.g. from a backend).
> >
> >> 4. Can you explain the usage of the hash field in ngx_table_elt_t
> >
> >Normally the hash field is used for fast lookups in hash 
> >structures.? In case of r->headers_out it's usually just set to 1 
> >indicate the header is valid (or reset to 0 to indicate the 
> >header should not be sent as it was superseeded by some other 
> >header).
> >
> >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 Apr 16 14:43:03 2012
From: nginx-forum at nginx.us (bpiraeus)
Date: Mon, 16 Apr 2012 10:43:03 -0400 (EDT)
Subject: SSL negotiation failures causing 408 error to be returned
In-Reply-To: <20120416105930.GJ13466@mdounin.ru>
References: <20120416105930.GJ13466@mdounin.ru>
Message-ID: <23c6bb3114f978f9e223193522fab9f7.NginxMailingListEnglish@forum.nginx.org>

The problem with that diagnosis, is that in this case, it's actually
causing failures on the client side. The client in question has a flash
game which pulls some amount of data over https, when these sessions
time out and hand out a 408, it causes the game itself to barf up a
lung. Additionally after pulling 12 hours of logs and doing UA matching,
while Chrome on Win7 base seems to be the major player, we find these as
well:

Mozilla/4.0 (compatible; MSIE 7.0; Windows NT 6.0; Trident/5.0; SLCC1;
.NET CLR 2.0.50727; .NET CLR 3.5.30729; .NET CLR 3.0.30729; .NET4.0C)
Mozilla/5.0 (compatible; MSIE 9.0; Windows NT 6.0; Trident/5.0; yie9)
Mozilla/4.0 (compatible; MSIE 7.0b; Windows NT 6.0)
Mozilla/5.0 (compatible; MSIE 9.0; Windows NT 6.0; WOW64; Trident/5.0;
yie9)
Mozilla/5.0 (Windows NT 6.0; rv:11.0) Gecko/20100101 Firefox/11.0
Mozilla/4.0 (compatible; MSIE 7.0; Windows NT 6.1; WOW64; Trident/5.0;
SLCC2; .NET CLR 2.0.50727; .NET CLR 3.5.30729; .NET CLR 3.0.30729; Media
Center PC 6.0; CPNTDF; .NET4.0C)
Mozilla/4.0 (compatible; MSIE 7.0; Windows NT 6.0; GTB6; SLCC1; .NET CLR
2.0.50727; Media Center PC 5.0; .NET CLR 3.5.30729; .NET CLR 3.0.30618;
WinNT-PAI 26.09.2009; .NET4.0C)
Mozilla/5.0 (Windows NT 6.1; rv:11.0) Gecko/20100101 Firefox/11.0
Mozilla/5.0 (compatible; MSIE 9.0; Windows NT 6.1; WOW64; Trident/5.0;
NP06)
Mozilla/5.0 (compatible; MSIE 9.0; Windows NT 6.1; WOW64; Trident/5.0;
SlimBrowser)
Mozilla/4.0 (compatible; MSIE 7.0; Windows NT 6.0; Trident/5.0; SLCC1;
.NET CLR 2.0.50727; Media Center PC 5.0; MS-RTC EA 2; .NET CLR
3.5.30729; .NET CLR 3.0.30729; .NET4.0C; BRI/2)
Mozilla/5.0 (compatible; MSIE 9.0; Windows NT 6.1; Win64; x64;
Trident/5.0)
Mozilla/5.0 (Macintosh; Intel Mac OS X 10.6; rv:11.0) Gecko/20100101
Firefox/11.0
Mozilla/5.0 (Macintosh; U; Intel Mac OS X 10_4_11; en)
AppleWebKit/533.19.4 (KHTML, like Gecko) Version/4.1.3 Safari/533.19.4
Mozilla/5.0 (Windows NT 5.1; rv:11.0) Gecko/20100101 Firefox/11.0
Mozilla/5.0 (compatible; MSIE 9.0; Windows NT 6.1; WOW64; Trident/5.0;
FunWebProducts)
Mozilla/5.0 (compatible; MSIE 9.0; Windows NT 6.1; WOW64; Trident/5.0;
yie9)
Mozilla/5.0 (Windows NT 6.1; WOW64; rv:11.0) Gecko/20100101
Firefox/11.0
Mozilla/5.0 (Macintosh; Intel Mac OS X 10_5_8) AppleWebKit/534.50.2
(KHTML, like Gecko) Version/5.0.6 Safari/533.22.3
Mozilla/5.0 (compatible; MSIE 9.0; Windows NT 6.0; WOW64; Trident/5.0)
Mozilla/5.0 (compatible; MSIE 9.0; Windows NT 6.0; Trident/5.0)
Mozilla/5.0 (compatible; MSIE 9.0; Windows NT 6.1; Trident/5.0)
Mozilla/5.0 (compatible; MSIE 9.0; Windows NT 6.1; WOW64; Trident/5.0)


The interesting bit is that the significant amount of these (when we
strip out Chrome) are MSIE on a Win7 base, which makes me inclined to
suspect something funny with Win 7, but there's at least 2 instances of
OS X in there, and 2 instances (FF 10.0.02 and FF 11) on NT 5.1 (XP
base)

If we throw out the OS X and XP ones as outliers due to something else
(people monkeying with settings, who knows), that leaves us with a
pretty clear indication that -something- in Win 7 has got some kind of
naughty behavior with SSL, but I'll be damned if I can come up with
anything. I spent half of last night trawling google for Win 7+SSL
issues and just didn't find anything of relevance.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225349,225368#msg-225368


From nginx-forum at nginx.us  Mon Apr 16 15:14:13 2012
From: nginx-forum at nginx.us (bpiraeus)
Date: Mon, 16 Apr 2012 11:14:13 -0400 (EDT)
Subject: SSL negotiation failures causing 408 error to be returned
In-Reply-To: <20120416105930.GJ13466@mdounin.ru>
References: <20120416105930.GJ13466@mdounin.ru>
Message-ID: <63f68a99756125f01db7f1c33d06bd0e.NginxMailingListEnglish@forum.nginx.org>

I stumbled across this this morning after talking with some co-workers
about what we're seeing

http://code.google.com/p/chromium/issues/detail?id=85229

I'm wondering if someone @ MS decided IE9 would benefit from the same
speculative pre-connect sockets. If we strip out anything that's not
IE9/Chrome we're left with a pretty small subset of 'other browsers'
which I'm leaning towards considering outliers just by virtue of being <
.5% of the overall error count.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225349,225369#msg-225369


From hagaizzz at yahoo.com  Mon Apr 16 15:14:25 2012
From: hagaizzz at yahoo.com (hagai avrahami)
Date: Mon, 16 Apr 2012 08:14:25 -0700 (PDT)
Subject: ngx_http_send_special_response
In-Reply-To: <20120416133823.GK13466@mdounin.ru>
References: <1334566419.86321.YahooMailNeo@web125506.mail.ne1.yahoo.com>
	<20120416103744.GI13466@mdounin.ru>
	<1334575758.9686.YahooMailNeo@web125504.mail.ne1.yahoo.com>
	<20120416133823.GK13466@mdounin.ru>
Message-ID: <1334589265.45104.YahooMailNeo@web125504.mail.ne1.yahoo.com>

Hi
Thanks

The redirect response is from content handler context.

Using the code you added 

cause Nginx to send back the Error Page (302) with Content-Length and Content-Type ("test/html") (And I am trying avoid that) without Body


That's why I assumed using r->header_only = 1? will solve my problem but as you explained it is used for other cases.
please see that in ngx_http_send_special_response when this flag set to 1 the Header is set with this fields but no body in response (I think this is not a good result)

if adding in the beginning of the function (ngx_http_send_special_response) the following code it solves the problem:

? if (r->header_only == 1)
? {
?????? ngx_http_clear_accept_ranges(r);
?????? ngx_http_clear_last_modified(r);

??????? rc = ngx_http_send_header(r);

??????? return rc;
?? }

Is there any other way I can avoid content length and content-type in 302 response
Thanks
Hagai





>________________________________
> From: Maxim Dounin <mdounin at mdounin.ru>
>To: nginx at nginx.org 
>Sent: Monday, April 16, 2012 3:38 PM
>Subject: Re: ngx_http_send_special_response
> 
>Hello!
>
>On Mon, Apr 16, 2012 at 04:29:18AM -0700, hagai avrahami wrote:
>
>> WOW thanks for the quick and elaborate response
>> 
>> So what is the right way to send response with status code 302 on GET request but
>> with content length 0 and no body (I want to redirect the request to other Server)
>
>Easiest solution is to just write
>
>? ? return 302 $where;
>
>in config. :)
>
>From the code point of view (assuming content handler context), 
>something like this should work:
>
>? ? ngx_table_elt_t? *h;
>
>? ? ngx_http_clear_location(r);
>
>? ? h = ngx_list_push(&r->headers_out.headers);
>? ? if (h == NULL) {
>? ? ? ? return NGX_HTTP_INTERNAL_SERVER_ERROR;
>? ? }
>
>? ? h->hash = 1;
>? ? ngx_str_set(&h->key, "Location");
>
>? ? h->value.data = ...
>? ? h->value.len = ...
>
>? ? r->headers_out.location = h;
>
>? ? return NGX_HTTP_MOVED_TEMPORARILY;
>
>Maxim Dounin
>
>
>> 
>> Thanks 
>> 
>> Hagai
>> 
>> 
>> 
>> 
>> >________________________________
>> > From: Maxim Dounin <mdounin at mdounin.ru>
>> >To: nginx at nginx.org 
>> >Sent: Monday, April 16, 2012 12:37 PM
>> >Subject: Re: ngx_http_send_special_response
>> > 
>> >Hello!
>> >
>> >On Mon, Apr 16, 2012 at 01:53:39AM -0700, hagai avrahami wrote:
>> >
>> >> 
>> >> 
>> >> Hi
>> >> 
>> >> I am appreciate any help I can get on the following issues.
>> >> 
>> >> I am writing my own module.
>> >> 
>> >> 1. I can see in function ngx_http_send_special_response
>> >> ??? that it calculates the content length of the error page even if the request signed as header only (r->header_only = 1)
>> >> ??? and set the content length header with value, after sending the header it does not send the body ( because? r->header_only = 1)
>> >> ??? so the response arrive with content length different than 0 but without body 
>> >> 
>> >> ??? Is it a BUG?
>> >
>> >No.? The r->header_only flag should is set when body isn't 
>> >expected per protocol (HEAD requests, 304 responses).? Having 
>> >Content-Length present in such responses is ok.
>> >
>> >> 2.In the module I am trying to redirect the request with status code 302 (NGX_HTTP_MOVED_TEMPORARILY)
>> >> ?? Trying to user->headers_out.location for the Alternate URL failed, I looked in the code and saw the Location must start with "/" - why?
>> >
>> >You have to set it correctly, i.e. add to r->headers_out.headers 
>> >and link to r->headers_out.location.? If it starts with "/" it 
>> >works even if set incorrectly due to adding of a server_name, 
>> >that's probably what confused you.
>> >
>> >> ?? In the end I add it to r->headers_out.headers
>> >> ?? using header = ngx_list_push(&r->headers_out.headers);
>> >> ?? Is it the write way to do it?
>> >> 
>> >> 3. Can you explain please the different between
>> >> ??? a. r->headers_out.content_length_n
>> >> ??? b. r->headers_out.content_length
>> >
>> >The content_length_n is a numeric length, this what you normaly 
>> >should set when generating response in nginx.? The content_length 
>> >is a pointer to a headers array element, it might be present if 
>> >there is string representation of a Content-Length header 
>> >available for some reason (e.g. from a backend).
>> >
>> >> 4. Can you explain the usage of the hash field in ngx_table_elt_t
>> >
>> >Normally the hash field is used for fast lookups in hash 
>> >structures.? In case of r->headers_out it's usually just set to 1 
>> >indicate the header is valid (or reset to 0 to indicate the 
>> >header should not be sent as it was superseeded by some other 
>> >header).
>> >
>> >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/20120416/f7e06dd5/attachment.html>

From mdounin at mdounin.ru  Mon Apr 16 15:25:03 2012
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Mon, 16 Apr 2012 19:25:03 +0400
Subject: SSL negotiation failures causing 408 error to be returned
In-Reply-To: <23c6bb3114f978f9e223193522fab9f7.NginxMailingListEnglish@forum.nginx.org>
References: <20120416105930.GJ13466@mdounin.ru>
	<23c6bb3114f978f9e223193522fab9f7.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20120416152503.GM13466@mdounin.ru>

Hello!

On Mon, Apr 16, 2012 at 10:43:03AM -0400, bpiraeus wrote:

> The problem with that diagnosis, is that in this case, it's actually
> causing failures on the client side. The client in question has a flash
> game which pulls some amount of data over https, when these sessions
> time out and hand out a 408, it causes the game itself to barf up a
> lung. Additionally after pulling 12 hours of logs and doing UA matching,
> while Chrome on Win7 base seems to be the major player, we find these as
> well:

Similar errors might as well be a result of e.g. network connectivity 
problems.  It's hard to say anything without detailed inspection 
of a failure on the client side.

[...]

> The interesting bit is that the significant amount of these (when we
> strip out Chrome) are MSIE on a Win7 base, which makes me inclined to
> suspect something funny with Win 7, but there's at least 2 instances of
> OS X in there, and 2 instances (FF 10.0.02 and FF 11) on NT 5.1 (XP
> base)

There is at least several Vista user-agents in your list (Windows 
NT 6.0), and I would rather suggest to compare the distribution to 
your site's OS/UA stats before concluding anything.  And you need 
to trace real client failures, not 408 errors in general.

Maxim Dounin


From mdounin at mdounin.ru  Mon Apr 16 15:57:16 2012
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Mon, 16 Apr 2012 19:57:16 +0400
Subject: ngx_http_send_special_response
In-Reply-To: <1334589265.45104.YahooMailNeo@web125504.mail.ne1.yahoo.com>
References: <1334566419.86321.YahooMailNeo@web125506.mail.ne1.yahoo.com>
	<20120416103744.GI13466@mdounin.ru>
	<1334575758.9686.YahooMailNeo@web125504.mail.ne1.yahoo.com>
	<20120416133823.GK13466@mdounin.ru>
	<1334589265.45104.YahooMailNeo@web125504.mail.ne1.yahoo.com>
Message-ID: <20120416155716.GN13466@mdounin.ru>

Hello!

On Mon, Apr 16, 2012 at 08:14:25AM -0700, hagai avrahami wrote:

> Hi
> Thanks
> 
> The redirect response is from content handler context.
> 
> Using the code you added 
> 
> cause Nginx to send back the Error Page (302) with 
> Content-Length and Content-Type ("test/html") (And I am trying 
> avoid that) without Body

Ah, sorry, I missed you want to send 302 with an empty body, not 
just 302.  Then instead of

    return NGX_HTTP_MOVED_TEMPORARY;

you have to send actual response youself, like this:

    r->headers_out.status = NGX_HTTP_MOVED_TEMPORARY;
    r->headers_out.content_length_n = 0;

    rc = ngx_http_send_header(r);

    if (rc == NGX_ERROR || rc > NGX_OK || r->header_only) {
        return rc;
    }

    return ngx_http_send_special(r, NGX_HTTP_LAST);

Maxim Dounin


From hagaizzz at yahoo.com  Mon Apr 16 16:28:01 2012
From: hagaizzz at yahoo.com (hagai avrahami)
Date: Mon, 16 Apr 2012 09:28:01 -0700 (PDT)
Subject: ngx_http_send_special_response
In-Reply-To: <20120416155716.GN13466@mdounin.ru>
References: <1334566419.86321.YahooMailNeo@web125506.mail.ne1.yahoo.com>
	<20120416103744.GI13466@mdounin.ru>
	<1334575758.9686.YahooMailNeo@web125504.mail.ne1.yahoo.com>
	<20120416133823.GK13466@mdounin.ru>
	<1334589265.45104.YahooMailNeo@web125504.mail.ne1.yahoo.com>
	<20120416155716.GN13466@mdounin.ru>
Message-ID: <1334593681.18904.YahooMailNeo@web125504.mail.ne1.yahoo.com>

Thanks!


>________________________________
> From: Maxim Dounin <mdounin at mdounin.ru>
>To: nginx at nginx.org 
>Sent: Monday, April 16, 2012 6:57 PM
>Subject: Re: ngx_http_send_special_response
> 
>Hello!
>
>On Mon, Apr 16, 2012 at 08:14:25AM -0700, hagai avrahami wrote:
>
>> Hi
>> Thanks
>> 
>> The redirect response is from content handler context.
>> 
>> Using the code you added 
>> 
>> cause Nginx to send back the Error Page (302) with 
>> Content-Length and Content-Type ("test/html") (And I am trying 
>> avoid that) without Body
>
>Ah, sorry, I missed you want to send 302 with an empty body, not 
>just 302.? Then instead of
>
>? ? return NGX_HTTP_MOVED_TEMPORARY;
>
>you have to send actual response youself, like this:
>
>? ? r->headers_out.status = NGX_HTTP_MOVED_TEMPORARY;
>? ? r->headers_out.content_length_n = 0;
>
>? ? rc = ngx_http_send_header(r);
>
>? ? if (rc == NGX_ERROR || rc > NGX_OK || r->header_only) {
>? ? ? ? return rc;
>? ? }
>
>? ? return ngx_http_send_special(r, NGX_HTTP_LAST);
>
>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/20120416/e8f17f9d/attachment-0001.html>

From nginx-forum at nginx.us  Mon Apr 16 17:24:17 2012
From: nginx-forum at nginx.us (bpiraeus)
Date: Mon, 16 Apr 2012 13:24:17 -0400 (EDT)
Subject: SSL negotiation failures causing 408 error to be returned
In-Reply-To: <20120416152503.GM13466@mdounin.ru>
References: <20120416152503.GM13466@mdounin.ru>
Message-ID: <42f0f0380cd3c7c65a4ed4707cea73b9.NginxMailingListEnglish@forum.nginx.org>

Maxim Dounin Wrote:
-------------------------------------------------------
> Hello!

> Similar errors might as well be a result of e.g.
> network connectivity 
> problems.  It's hard to say anything without
> detailed inspection 
> of a failure on the client side.
> 

We've been able to replicate once in a VM running Win 7 w/ Chrome.


> There is at least several Vista user-agents in
> your list (Windows 
> NT 6.0), and I would rather suggest to compare the
> distribution to 
> your site's OS/UA stats before concluding
> anything.  And you need 
> to trace real client failures, not 408 errors in
> general.

We've done numerous packet traces, at this point we're fairly well
stumped and are leaning towards it being some strange interaction with
SSL+Flash. The 408s, while not specifically problem, may indirectly
contribute to some behavioral issue that we're still trying to track
down. I would be hard pressed to put this on Nginx, I was, ultimately,
hoping someone had perhaps seen this behvior previously and had some
insight into the issue.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225349,225375#msg-225375


From mdounin at mdounin.ru  Mon Apr 16 17:56:03 2012
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Mon, 16 Apr 2012 21:56:03 +0400
Subject: truncated output with gzip, large files and slow rate
In-Reply-To: <op.wcu0zino3ig1dk@sessia>
References: <op.wcunkpg53ig1dk@sessia> <20120416101741.GH13466@mdounin.ru>
	<op.wcu0zino3ig1dk@sessia>
Message-ID: <20120416175603.GQ13466@mdounin.ru>

Hello!

On Mon, Apr 16, 2012 at 02:27:08PM +0100, James Lee wrote:

> On Mon, 16 Apr 2012 11:17:41 +0100, Maxim Dounin
> <mdounin at mdounin.ru> wrote:
> 
> >Could you please provide debug log and nginx -V output?
> 
> debug.log.bz2 as attachment.

Thanks, the problem looks clear.  Could you please test if the 
following patch fixes things for you?

# HG changeset patch
# User Maxim Dounin <mdounin at mdounin.ru>
# Date 1334597965 -14400
# Node ID f58e907f57e1136223c178ea81a18e52a136ec60
# Parent  56f926e54f3430f1f9ecc9ef154262de10c33005
Fixed loop in ngx_writev_chain() and ngx_solaris_sendfilev_chain().

diff --git a/src/os/unix/ngx_solaris_sendfilev_chain.c b/src/os/unix/ngx_solaris_sendfilev_chain.c
--- a/src/os/unix/ngx_solaris_sendfilev_chain.c
+++ b/src/os/unix/ngx_solaris_sendfilev_chain.c
@@ -74,7 +74,6 @@ ngx_solaris_sendfilev_chain(ngx_connecti
 
 
     send = 0;
-    complete = 0;
 
     vec.elts = sfvs;
     vec.size = sizeof(sendfilevec_t);
@@ -87,6 +86,7 @@ ngx_solaris_sendfilev_chain(ngx_connecti
         fprev = 0;
         sfv = NULL;
         eintr = 0;
+        complete = 0;
         sent = 0;
         prev_send = send;
 
diff --git a/src/os/unix/ngx_writev_chain.c b/src/os/unix/ngx_writev_chain.c
--- a/src/os/unix/ngx_writev_chain.c
+++ b/src/os/unix/ngx_writev_chain.c
@@ -54,7 +54,6 @@ ngx_writev_chain(ngx_connection_t *c, ng
     }
 
     send = 0;
-    complete = 0;
 
     vec.elts = iovs;
     vec.size = sizeof(struct iovec);
@@ -65,6 +64,7 @@ ngx_writev_chain(ngx_connection_t *c, ng
         prev = NULL;
         iov = NULL;
         eintr = 0;
+        complete = 0;
         prev_send = send;
 
         vec.nelts = 0;


Maxim Dounin


From list at oxdrove.co.uk  Mon Apr 16 18:22:16 2012
From: list at oxdrove.co.uk (James Lee)
Date: Mon, 16 Apr 2012 19:22:16 +0100
Subject: truncated output with gzip, large files and slow rate
In-Reply-To: <20120416175603.GQ13466@mdounin.ru>
References: <op.wcunkpg53ig1dk@sessia> <20120416101741.GH13466@mdounin.ru>
	<op.wcu0zino3ig1dk@sessia> <20120416175603.GQ13466@mdounin.ru>
Message-ID: <op.wcvenewd3ig1dk@sessia>

On Mon, 16 Apr 2012 18:56:03 +0100, Maxim Dounin <mdounin at mdounin.ru>  
wrote:

Hello,

> Thanks, the problem looks clear.  Could you please test if the
> following patch fixes things for you?

Glad it looked clear to you!  Quick test shows this fixes the problem.
Thank you for the fix, I'll check tomorrow on a working installation
and hopefully users will get full files at busy times.



James.





From lists at ruby-forum.com  Mon Apr 16 18:27:36 2012
From: lists at ruby-forum.com (maria m.)
Date: Mon, 16 Apr 2012 20:27:36 +0200
Subject: File Not Found logging in w/ nginx+phpMyAdmin in URL subdir
In-Reply-To: <2ade13bb167050f3e7310e42f175013c@ruby-forum.com>
References: <2ade13bb167050f3e7310e42f175013c@ruby-forum.com>
Message-ID: <c7279f13cea936af17b796fa036d98eb@ruby-forum.com>

server {
listen 80;
server_name support.example.com;
access_log /var/log/nginx/support.example.com.access.log;
error_log /var/log/nginx/support.example.com.error.log;

# this root dir below is to the thebuggenie htdocs folder
root /home/user/public_html/support.example.com/thebuggenie;
client_max_body_size 40M;
large_client_header_buffers 4 8k;
index index.php index.html;

location ^~ /files { }

#location ~ \..*/.*\.php$ { return 403; }

location ~ \.php$ {
fastcgi_pass 127.0.0.1:9000;
fastcgi_index index.php;
fastcgi_param SCRIPT_FILENAME $document_root$fastcgi_script_name;
include fastcgi_params;
}

location ~ /(?<suburi>.*){
set $suburi $1;
try_files $uri $uri/ /index.php?url=$suburi&$args;
}

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


From eltlims at yahoo.com.br  Mon Apr 16 19:34:30 2012
From: eltlims at yahoo.com.br (Elton Lima)
Date: Mon, 16 Apr 2012 12:34:30 -0700 (PDT)
Subject: File Not Found logging in w/ nginx+phpMyAdmin in URL subdir
In-Reply-To: <c7279f13cea936af17b796fa036d98eb@ruby-forum.com>
References: <2ade13bb167050f3e7310e42f175013c@ruby-forum.com>
	<c7279f13cea936af17b796fa036d98eb@ruby-forum.com>
Message-ID: <1334604870.68092.YahooMailNeo@web39304.mail.mud.yahoo.com>

Hi,

server {
? ? ? ? listen ?80;
? ? ? ? server_name ?localhost;
? ? ? ? #access_log ?/var/log/nginx/access.log ?main;

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

? ? ? ?? location /phpmyadmin {
? ? ? ? ? ? ? ?root /usr/share/;
? ? ? ? ? ? ? ?index index.php index.html index.htm;
? ? ? ? ? ? ? ?location ~ ^/phpmyadmin/(.+\.php)$ {
? ? ? ? ? ? ? ? ? ? ? ?try_files $uri =404;
? ? ? ? ? ? ? ? ? ? ? ?root /usr/share/;
? ? ? ? ? ? ? ? ? ? ? ?fastcgi_pass 127.0.0.1:9000;
? ? ? ? ? ? ? ? ? ? ? ?fastcgi_index index.php;
? ? ? ? ? ? ? ? ? ? ? ?fastcgi_param SCRIPT_FILENAME $document_root$fastcgi_script_name;
? ? ? ? ? ? ? ? ? ? ? ?include /etc/nginx/fastcgi_params;
? ? ? ? ? ? ? ?}
? ? ? ? ? ? ? ?location ~* ^/phpmyadmin/(.+\.(jpg|jpeg|gif|css|png|js|ico|html|xml|txt))$ {
? ? ? ? ? ? ? ? ? ? ? ?root /usr/share/;
? ? ? ? ? ? ? ?}
? ? ? ? }
? ? ? ? location /phpMyAdmin {
? ? ? ? ? ? ? ?rewrite ^/* /phpmyadmin last;
? ? ? ? }
}


?
Elton Lima
------------------------------------------------------
Analista de Suporte Linux
E-mail: eltonlim at gmail.com // eltlims at yahoo.com.br
Site: http://blog.eltonlima.com.br


________________________________
 De: maria m. <lists at ruby-forum.com>
Para: nginx at nginx.org 
Enviadas: Segunda-feira, 16 de Abril de 2012 15:27
Assunto: Re: File Not Found logging in w/ nginx+phpMyAdmin in URL subdir
 
server {
listen 80;
server_name support.example.com;
access_log /var/log/nginx/support.example.com.access.log;
error_log /var/log/nginx/support.example.com.error.log;

# this root dir below is to the thebuggenie htdocs folder
root /home/user/public_html/support.example.com/thebuggenie;
client_max_body_size 40M;
large_client_header_buffers 4 8k;
index index.php index.html;

location ^~ /files { }

#location ~ \..*/.*\.php$ { return 403; }

location ~ \.php$ {
fastcgi_pass 127.0.0.1:9000;
fastcgi_index index.php;
fastcgi_param SCRIPT_FILENAME $document_root$fastcgi_script_name;
include fastcgi_params;
}

location ~ /(?<suburi>.*){
set $suburi $1;
try_files $uri $uri/ /index.php?url=$suburi&$args;
}

-- 
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/20120416/93ba28a0/attachment.html>

From nginx-forum at nginx.us  Mon Apr 16 21:15:17 2012
From: nginx-forum at nginx.us (michael_teter)
Date: Mon, 16 Apr 2012 17:15:17 -0400 (EDT)
Subject: Display configuration at runtime?
Message-ID: <b005faa63d3aa42dc7945a624c0d1665.NginxMailingListEnglish@forum.nginx.org>

Howdy.

I'm working on setting up a maintenance page, and I can't figure out
where Nginx is looking for the maintenance.html that I have specified. 
I've tried placing the file in a lot of different locations, but I still
get the generic "503 Service Temporarily Unavailable" page.

My question is, is there a way I can get Nginx to display it's
configuration at run-time, so perhaps I can determine where it is
looking for the file?

Thanks in advance,
Michael

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225392,225392#msg-225392


From jim at ohlste.in  Mon Apr 16 21:38:38 2012
From: jim at ohlste.in (Jim Ohlstein)
Date: Mon, 16 Apr 2012 17:38:38 -0400
Subject: Display configuration at runtime?
In-Reply-To: <b005faa63d3aa42dc7945a624c0d1665.NginxMailingListEnglish@forum.nginx.org>
References: <b005faa63d3aa42dc7945a624c0d1665.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <CABxOBbTV9GqRBspZQj0wkv0gYeQfvUae+41Dz9hriOHZgY6Hqw@mail.gmail.com>

Jim Ohlstein
On Apr 16, 2012 5:15 PM, "michael_teter" <nginx-forum at nginx.us> wrote:
>
> Howdy.
>
> I'm working on setting up a maintenance page, and I can't figure out
> where Nginx is looking for the maintenance.html that I have specified.
> I've tried placing the file in a lot of different locations, but I still
> get the generic "503 Service Temporarily Unavailable" page.

What does the error log show?

If you use something like:

location / {
   root /some/path;
   try_files maintenance.html $uri $uri/;
   ...;
}

then nginx will look for maintenance.html in the root directory and serve
it if it's present.

So to put our sites into maintenance mode, we simply copy our
maintenance.html into the root directory. No need to reload nginx. When
we're ready to go back up, the file is removed.

>
> My question is, is there a way I can get Nginx to display it's
> configuration at run-time, so perhaps I can determine where it is
> looking for the file?
>
> Thanks in advance,
> Michael
>
> Posted at Nginx Forum:
http://forum.nginx.org/read.php?2,225392,225392#msg-225392
>

Jim
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20120416/a2cc1522/attachment-0001.html>

From ashishs.dev at gmail.com  Mon Apr 16 21:51:53 2012
From: ashishs.dev at gmail.com (Ashish S)
Date: Mon, 16 Apr 2012 14:51:53 -0700
Subject: Help with chaining of sub-requests
Message-ID: <CA+BpfUpZtvmetMJNp9-=rEq0cXQQR6hoCkY7UAXWQ4o8wRPVrQ@mail.gmail.com>

Hi,

I have 3 existing http backend web-services  S1, S2 and S3.  I am
trying to use nginx for a use-case where  i need to first get a "key"
from service S1, use this key along with some input data, to query S2,
and based on S2's response, query S3.    My question is, would
sub-request chaining be the best method to go about this?  I can use
the upstream method to connect to S1.   Am i doing something wrong, if
I then fire a sub-request (request to S2) from within the content
handler (process header) method registered for this upstream?.  How do
i then initiate the sub-request to S3 ?

Any pointers or example that i might be able to follow, for something
like the above?

Thanks,
Ashish


From nginx-forum at nginx.us  Mon Apr 16 22:02:45 2012
From: nginx-forum at nginx.us (michael_teter)
Date: Mon, 16 Apr 2012 18:02:45 -0400 (EDT)
Subject: Display configuration at runtime?
In-Reply-To: <CABxOBbTV9GqRBspZQj0wkv0gYeQfvUae+41Dz9hriOHZgY6Hqw@mail.gmail.com>
References: <CABxOBbTV9GqRBspZQj0wkv0gYeQfvUae+41Dz9hriOHZgY6Hqw@mail.gmail.com>
Message-ID: <34ad57c9bed43d3baf79a901d69b3d7b.NginxMailingListEnglish@forum.nginx.org>

Thanks for the quick response.  My error log shows nothing, but my
access log does show the visit.

Here's the relevant part of my config:


   server {
      listen      80;
      server_name www.ourname.com;
      rewrite     ^(.*)  https://$server_name$1 permanent;
   }
   server {
      listen 443;
      server_name www.ourname.com;
      #try_files $uri /maintenance.html @passenger;
      root /home/michael/ourname/public;

      # MT added 20120416
      error_page 503 /maintenance.html;
      if (-f $document_root/../tmp/stop.txt) {
        set $maintenance 1;
      }

      if ($maintenance) {
        return 503;
      }

      passenger_enabled on;
      #rails_env development;
      rails_env production;
      location @passenger {
        passenger_enabled on;
        root /home/michael/ourname/public;
      }

      ssl on;
      ssl_certificate /home/michael/ssl_keys/ourname.com_ssl.crt;
      ssl_certificate_key /home/michael/ssl_keys/Cert1.key;
      #return 403;
   }

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225392,225395#msg-225395


From nginx-forum at nginx.us  Mon Apr 16 22:05:23 2012
From: nginx-forum at nginx.us (michael_teter)
Date: Mon, 16 Apr 2012 18:05:23 -0400 (EDT)
Subject: Display configuration at runtime?
In-Reply-To: <34ad57c9bed43d3baf79a901d69b3d7b.NginxMailingListEnglish@forum.nginx.org>
References: <CABxOBbTV9GqRBspZQj0wkv0gYeQfvUae+41Dz9hriOHZgY6Hqw@mail.gmail.com>
	<34ad57c9bed43d3baf79a901d69b3d7b.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <36998251f3b4903876fa160e7d7d20c0.NginxMailingListEnglish@forum.nginx.org>

I should note that on this line:

error_page 503 /maintenance.html;

I have tried a variety of paths.
/maintenance.html
$document_root/maintenance.html
/some/other/path/maintenance.html

etc.

I don't understand how Nginx can see the stop.txt that I use to put it
in maint mode, but it can't see (or ignores)
$document_root/maintenance.html (when I had that set as the error_page)

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225392,225396#msg-225396


From ne at vbart.ru  Mon Apr 16 22:18:06 2012
From: ne at vbart.ru (Valentin V. Bartenev)
Date: Tue, 17 Apr 2012 02:18:06 +0400
Subject: Display configuration at runtime?
In-Reply-To: <b005faa63d3aa42dc7945a624c0d1665.NginxMailingListEnglish@forum.nginx.org>
References: <b005faa63d3aa42dc7945a624c0d1665.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <201204170218.06363.ne@vbart.ru>

On Tuesday 17 April 2012 01:15:17 michael_teter wrote:
[...]
> 
> My question is, is there a way I can get Nginx to display it's
> configuration at run-time, so perhaps I can determine where it is
> looking for the file?
> 

Debug log will help:
http://nginx.org/en/docs/debugging_log.html

 wbr, Valentin V. Bartenev


From jim at ohlste.in  Mon Apr 16 22:25:44 2012
From: jim at ohlste.in (Jim Ohlstein)
Date: Mon, 16 Apr 2012 18:25:44 -0400
Subject: Display configuration at runtime?
In-Reply-To: <34ad57c9bed43d3baf79a901d69b3d7b.NginxMailingListEnglish@forum.nginx.org>
References: <CABxOBbTV9GqRBspZQj0wkv0gYeQfvUae+41Dz9hriOHZgY6Hqw@mail.gmail.com>
	<34ad57c9bed43d3baf79a901d69b3d7b.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <CABxOBbTMTT0r=X77beEYmWXes-_Q0h0xMqv4PG=aoC5dQ3XQuw@mail.gmail.com>

On Apr 16, 2012 6:03 PM, "michael_teter" <nginx-forum at nginx.us> wrote:
>
> Thanks for the quick response.  My error log shows nothing, but my
> access log does show the visit.

Use higher level error logging. See
http://nginx.org/en/docs/ngx_core_module.html#error_log.

>
> Here's the relevant part of my config:
>
>
>   server {
>      listen      80;
>      server_name www.ourname.com;
>      rewrite     ^(.*)  https://$server_name$1 permanent;
>   }
>   server {
>      listen 443;
>      server_name www.ourname.com;
>      #try_files $uri /maintenance.html @passenger;

Why are you not using this line? It's the perfect use case for "try_files".
Note that I believe that you should remove the leading "/" from
"/maintenance.html".

>      root /home/michael/ourname/public;
>
>      # MT added 20120416
>      error_page 503 /maintenance.html;
>      if (-f $document_root/../tmp/stop.txt) {
>        set $maintenance 1;
>      }
>
>      if ($maintenance) {
>        return 503;
>      }

This is a lot more work than using the above "try_files". See also
http://wiki.nginx.org/IfIsEvil.

>
>      passenger_enabled on;
>      #rails_env development;
>      rails_env production;
>      location @passenger {
>        passenger_enabled on;
>        root /home/michael/ourname/public;
>      }
>
>      ssl on;
>      ssl_certificate /home/michael/ssl_keys/ourname.com_ssl.crt;
>      ssl_certificate_key /home/michael/ssl_keys/Cert1.key;
>      #return 403;
>   }
>
> Posted at Nginx Forum:
http://forum.nginx.org/read.php?2,225392,225395#msg-225395
>
>

Jim
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20120416/6ec30bb2/attachment.html>

From francis at daoine.org  Mon Apr 16 23:05:57 2012
From: francis at daoine.org (Francis Daly)
Date: Tue, 17 Apr 2012 00:05:57 +0100
Subject: Display configuration at runtime?
In-Reply-To: <34ad57c9bed43d3baf79a901d69b3d7b.NginxMailingListEnglish@forum.nginx.org>
References: <CABxOBbTV9GqRBspZQj0wkv0gYeQfvUae+41Dz9hriOHZgY6Hqw@mail.gmail.com>
	<34ad57c9bed43d3baf79a901d69b3d7b.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20120416230557.GB26729@craic.sysops.org>

On Mon, Apr 16, 2012 at 06:02:45PM -0400, michael_teter wrote:

Hi there,

> My error log shows nothing, but my
> access log does show the visit.

The error log doesn't seem immediately useful in this case; but with
the debug log you can work out what is happening.

The short answer is "internal redirect", or "subrequest".

> Here's the relevant part of my config:

>       error_page 503 /maintenance.html;
>       if (-f $document_root/../tmp/stop.txt) {
>         set $maintenance 1;
>       }
> 
>       if ($maintenance) {
>         return 503;
>       }

For any request, if the file exists then return 503, which does an
internal redirect to /maintenance.html. This subrequest goes through the
same process, and since the file still exists, you would end up in a loop.

With the following config, you can break out of the loop:

        error_page 503 /maintenance.html;
        if ($uri = "/maintenance.html") {
            break;
        }
        if (-f $document_root/../tmp/stop.txt) {
            return 503;
        }

Although I confess I don't know if there is a "best" way to implement
a 503 response to all requests in nginx based on the existence of a
flag file.

	f
-- 
Francis Daly        francis at daoine.org


From agentzh at gmail.com  Tue Apr 17 05:47:27 2012
From: agentzh at gmail.com (agentzh)
Date: Tue, 17 Apr 2012 13:47:27 +0800
Subject: Slides for My Nginx Talk at Percona Live MySQL Conference 2012
In-Reply-To: <CAB4Tn6NGVGx3rF064z-HQ9B5Gi90587uPS__g8+jr2iDUcVVVg@mail.gmail.com>
References: <CAB4Tn6NGVGx3rF064z-HQ9B5Gi90587uPS__g8+jr2iDUcVVVg@mail.gmail.com>
Message-ID: <CAB4Tn6NOdp1M83AKM8bXkrC3Qc690c-pY6x==XePRdBSBV52iw@mail.gmail.com>

Hi, folks!

Last week chaoslawful and I gave a talk at Percona Live MySQL Conference
2012:

?http://www.percona.com/live/mysql-conference-2012/sessions/scripting-mysql-lua-and-libdrizzle-inside-nginx

This was my first English presentation and I was glad that it went well :)

Here's the AJAX-based slides for my part:

? ?http://agentzh.org/misc/slides/libdrizzle-lua-nginx/#2

You can use the arrow keys on your keyboard to switch pages. The slides
contain a lot of benchmark results comparing various solutions like nginx +
php-fpm + php-mysql, ngx_drizzle + ngx_rds_json, ngx_drizzle + ngx_lua, and
nodejs + node-mysql + node-generic-pool.

There's also a PDF version for these slides:

? ?http://agentzh.org/misc/slides/libdrizzle-lua-nginx.pdf

Enjoy!
-agentzh


From agentzh at gmail.com  Tue Apr 17 07:19:56 2012
From: agentzh at gmail.com (agentzh)
Date: Tue, 17 Apr 2012 15:19:56 +0800
Subject: Help with chaining of sub-requests
In-Reply-To: <CA+BpfUpZtvmetMJNp9-=rEq0cXQQR6hoCkY7UAXWQ4o8wRPVrQ@mail.gmail.com>
References: <CA+BpfUpZtvmetMJNp9-=rEq0cXQQR6hoCkY7UAXWQ4o8wRPVrQ@mail.gmail.com>
Message-ID: <CAB4Tn6MPDfuwGouTB8f6_PRNTGuENQmxJgT_O+n8TrWFAfh4VQ@mail.gmail.com>

On Tue, Apr 17, 2012 at 5:51 AM, Ashish S <ashishs.dev at gmail.com> wrote:
>
> I have 3 existing http backend web-services ?S1, S2 and S3. ?I am
> trying to use nginx for a use-case where ?i need to first get a "key"
> from service S1, use this key along with some input data, to query S2,
> and based on S2's response, query S3. ? ?My question is, would
> sub-request chaining be the best method to go about this? ?I can use
> the upstream method to connect to S1. ? Am i doing something wrong, if
> I then fire a sub-request (request to S2) from within the content
> handler (process header) method registered for this upstream?. ?How do
> i then initiate the sub-request to S3 ?
>
> Any pointers or example that i might be able to follow, for something
> like the above?
>

Regarding chained subrequest implementation, see ngx_echo module's
echo_subrequest directive for an example:

?? http://wiki.nginx.org/HttpEchoModule#echo_subrequest

Also ngx_lua's ngx.location.capture() and ngx.location.capture_multi()
worth a look too:

??? http://wiki.nginx.org/HttpLuaModule#ngx.location.capture
??? http://wiki.nginx.org/HttpLuaModule#ngx.location.capture_multi

Regarding non-subrequest upstream communication, ngx_lua supports the
cosocket API which implements something similar to (but more general
than) ngx_http_upstream:

??? http://wiki.nginx.org/HttpLuaModule#ngx.socket.tcp

Hope these helps,
-agentzh


From ashishs.dev at gmail.com  Tue Apr 17 07:46:58 2012
From: ashishs.dev at gmail.com (Ashish S)
Date: Tue, 17 Apr 2012 00:46:58 -0700
Subject: Help with chaining of sub-requests
In-Reply-To: <CAB4Tn6MPDfuwGouTB8f6_PRNTGuENQmxJgT_O+n8TrWFAfh4VQ@mail.gmail.com>
References: <CA+BpfUpZtvmetMJNp9-=rEq0cXQQR6hoCkY7UAXWQ4o8wRPVrQ@mail.gmail.com>
	<CAB4Tn6MPDfuwGouTB8f6_PRNTGuENQmxJgT_O+n8TrWFAfh4VQ@mail.gmail.com>
Message-ID: <CA+BpfUq3WWd4zgSdGBMLmm9eanhMe-4YOnLeHTKa9wa=PTs-vQ@mail.gmail.com>

Thank you for the pointers, agentzh!

--Ashish

On Tue, Apr 17, 2012 at 12:19 AM, agentzh <agentzh at gmail.com> wrote:
> On Tue, Apr 17, 2012 at 5:51 AM, Ashish S <ashishs.dev at gmail.com> wrote:
>>
>> I have 3 existing http backend web-services ?S1, S2 and S3. ?I am
>> trying to use nginx for a use-case where ?i need to first get a "key"
>> from service S1, use this key along with some input data, to query S2,
>> and based on S2's response, query S3. ? ?My question is, would
>> sub-request chaining be the best method to go about this? ?I can use
>> the upstream method to connect to S1. ? Am i doing something wrong, if
>> I then fire a sub-request (request to S2) from within the content
>> handler (process header) method registered for this upstream?. ?How do
>> i then initiate the sub-request to S3 ?
>>
>> Any pointers or example that i might be able to follow, for something
>> like the above?
>>
>
> Regarding chained subrequest implementation, see ngx_echo module's
> echo_subrequest directive for an example:
>
> ?? http://wiki.nginx.org/HttpEchoModule#echo_subrequest
>
> Also ngx_lua's ngx.location.capture() and ngx.location.capture_multi()
> worth a look too:
>
> ??? http://wiki.nginx.org/HttpLuaModule#ngx.location.capture
> ??? http://wiki.nginx.org/HttpLuaModule#ngx.location.capture_multi
>
> Regarding non-subrequest upstream communication, ngx_lua supports the
> cosocket API which implements something similar to (but more general
> than) ngx_http_upstream:
>
> ??? http://wiki.nginx.org/HttpLuaModule#ngx.socket.tcp
>
> Hope these helps,
> -agentzh
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx


From nginx-forum at nginx.us  Tue Apr 17 09:09:54 2012
From: nginx-forum at nginx.us (bigplum)
Date: Tue, 17 Apr 2012 05:09:54 -0400 (EDT)
Subject: Why the benchmark results are different?
Message-ID: <c0fcdfa70cee82a9d8d23a502359d23b.NginxMailingListEnglish@forum.nginx.org>

Hi, 

I install the newest nginx, and run 3 benchmark tools:
ab/http_load/weighttp. I get 
three rps results 9129.48/2594.08 /1107. So which tool give the right
result? 

# ab2 -n 10000 -c 1000 http://10.6.2.213/
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 10.6.2.213 (be patient)
Completed 1000 requests
Completed 2000 requests
Completed 3000 requests
Completed 4000 requests
Completed 5000 requests
Completed 6000 requests
Completed 7000 requests
Completed 8000 requests
Completed 9000 requests
Completed 10000 requests
Finished 10000 requests


Server Software:        nginx/1.1.19
Server Hostname:        10.6.2.213
Server Port:            80

Document Path:          /
Document Length:        151 bytes

Concurrency Level:      1000
Time taken for tests:   1.095 seconds
Complete requests:      10000
Failed requests:        0
Write errors:           0
Total transferred:      3620000 bytes
HTML transferred:       1510000 bytes
Requests per second:    9129.48 [#/sec] (mean)
Time per request:       109.535 [ms] (mean)
Time per request:       0.110 [ms] (mean, across all concurrent
requests)
Transfer rate:          3227.41 [Kbytes/sec] received

Connection Times (ms)
              min  mean[+/-sd] median   max
Connect:        0    9  11.4      2      34
Processing:     2   26  30.3     16     638
Waiting:        1   24  29.3     16     638
Total:          3   35  36.9     16     641

Percentage of the requests served within a certain time (ms)
  50%     16
  66%     31
  75%     53
  80%     56
  90%     67
  95%    109
  98%    164
  99%    178
 100%    641 (longest request)

# ./http_load -parallel 1000 -fetches 10000 ./htl.url
10000 fetches, 1000 max parallel, 1.51e+06 bytes, in 3.85493 seconds
151 mean bytes/connection
2594.08 fetches/sec, 391706 bytes/sec
msecs/connect: 211.437 mean, 3001.19 max, 0.102 min
msecs/first-response: 38.354 mean, 1456.25 max, 0.189 min
HTTP response codes:
  code 200 -- 10000

# weighttp -n 10000 -t 10 -c 1000 http://10.6.2.213
weighttp - a lightweight and simple webserver benchmarking tool

starting benchmark...
spawning thread #1: 100 concurrent requests, 1000 total requests
spawning thread #2: 100 concurrent requests, 1000 total requests
spawning thread #3: 100 concurrent requests, 1000 total requests
spawning thread #4: 100 concurrent requests, 1000 total requests
spawning thread #5: 100 concurrent requests, 1000 total requests
spawning thread #6: 100 concurrent requests, 1000 total requests
spawning thread #7: 100 concurrent requests, 1000 total requests
spawning thread #8: 100 concurrent requests, 1000 total requests
spawning thread #9: 100 concurrent requests, 1000 total requests
progress:  10% done
spawning thread #10: 100 concurrent requests, 1000 total requests
progress:  20% done
progress:  30% done
progress:  40% done
progress:  50% done
progress:  60% done
progress:  70% done
progress:  80% done
progress:  90% done
progress: 100% done

finished in 9 sec, 26 millisec and 801 microsec, 1107 req/s, 391
kbyte/s
requests: 10000 total, 10000 started, 10000 done, 10000 succeeded, 0
failed, 0 errored
status codes: 10000 2xx, 0 3xx, 0 4xx, 0 5xx
traffic: 3620000 bytes total, 2110000 bytes http, 1510000 bytes data

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225415,225415#msg-225415


From igor at sysoev.ru  Tue Apr 17 09:18:29 2012
From: igor at sysoev.ru (Igor Sysoev)
Date: Tue, 17 Apr 2012 13:18:29 +0400
Subject: Why the benchmark results are different?
In-Reply-To: <c0fcdfa70cee82a9d8d23a502359d23b.NginxMailingListEnglish@forum.nginx.org>
References: <c0fcdfa70cee82a9d8d23a502359d23b.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20120417091829.GA95437@nginx.com>

On Tue, Apr 17, 2012 at 05:09:54AM -0400, bigplum wrote:

> I install the newest nginx, and run 3 benchmark tools:
> ab/http_load/weighttp. I get 
> three rps results 9129.48/2594.08 /1107. So which tool give the right
> result? 

I believe you benchmarked these tools themselves.
It seems the tools overhead is larger than nginx's overhead.


-- 
Igor Sysoev


From nginx-forum at nginx.us  Tue Apr 17 10:44:48 2012
From: nginx-forum at nginx.us (leontinus)
Date: Tue, 17 Apr 2012 06:44:48 -0400 (EDT)
Subject: some of my user cannot access my website
Message-ID: <d6335409b79bc643360c3e44f1a2b9b4.NginxMailingListEnglish@forum.nginx.org>

Hi guys, I need your help to help me identify a problem. I'm running an
online marketplace here in Indonesia. Right now I have 1 mod perl server
using apache and 1 nginx server as proxy. But some of my users complain
they cannot access my website, http://www.tokopedia.com. And I found out
a fact that if they use mikrotik HttpProxy, they cannot access my
website, but they can access other website.

If they use firefox, they found err message : Content Encoding Error
If they use google chrome, they found err message :
ERR_INVALID_CHUNKED_ENCODING

Can you help me with this?

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225421,225421#msg-225421


From nginx-forum at nginx.us  Tue Apr 17 10:51:54 2012
From: nginx-forum at nginx.us (leontinus)
Date: Tue, 17 Apr 2012 06:51:54 -0400 (EDT)
Subject: some of my user cannot access my website
In-Reply-To: <d6335409b79bc643360c3e44f1a2b9b4.NginxMailingListEnglish@forum.nginx.org>
References: <d6335409b79bc643360c3e44f1a2b9b4.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <953e03108e130f68986766f1c2dc6351.NginxMailingListEnglish@forum.nginx.org>

by the way here i put my website proxy configuration


#-- notice info
error_log  /data/share/project/www-conf/logs/nginx.err.proxy.log;
pid	   /data/share/project/www-conf/logs/nginx.proxy.pid;

user ngproxy tokopedia;
worker_processes  4;


events {
	worker_connections  1024;
}




http {
	include /share/project/www-conf/nginx/tkpmodperl1/mime.types;
	include /share/project/www-conf/nginx/deny.conf;

	default_type  text/plain;
	sendfile	   on;
	keepalive_timeout  5;

        set_real_ip_from   192.168.1.0/24;
        real_ip_header     X-Forwarded-For;

	gzip  on;
	#gzip off;
	gzip_types text/plain text/css application/x-javascript;
	gzip_comp_level 9;
	gzip_proxied any;
	#gzip_vary off;	
	#charset  ISO-8859-1;
	
	charset UTF-8;
	source_charset UTF-8;

	#-------------------
	server {
		server_name  tokopedia;
		listen	   80;
		proxy_set_header X-Forwarded-Host $host:$server_port;
		proxy_set_header X-Forwarded-Server www.tokopedia.com;
		proxy_set_header X-Forwarded-For $remote_addr;
		proxy_set_header Host $host;
		proxy_set_header Accept-Encoding "";
		proxy_buffering on;
		proxy_buffers 8 64k;

		if ($http_user_agent ~* big.brother|crawl ){return 403;}
		if ($http_user_agent = "" ){return 403;}

		location /___pstatus {
			stub_status on;
			access_log   off;
		}
		
		location /img/ {
			proxy_pass  http://xxx.xxx.xxx.xxx/img/;
		}
				
		location ~* ^/(.*)\.(js|css)$ {
			proxy_pass  http://xxx.xxx.xxx.xxx/$1.$2;
		}


		location / { proxy_pass http://xxx.xxx.xxx.xxx:888/;	}

	}

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225421,225422#msg-225422


From svoop at delirium.ch  Tue Apr 17 11:24:07 2012
From: svoop at delirium.ch (Sven Schwyn)
Date: Tue, 17 Apr 2012 13:24:07 +0200
Subject: Legitimate access with certificate auth or basic auth
Message-ID: <80A15EAE-1997-4A5B-ADEA-AA281787DCA4@delirium.ch>

Hi

Is there a way to legitimate access to a location (in my case an entire vhost) with certificate auth and - if no certificate is sent - a fallback to basic auth?

On Apache (from where I'm currently migrating) this is done with "FakeBasicAuth" as follows:

SSLVerifyClient require
SSLVerifyDepth 3
SSLCADNRequestFile /path/to/certs.pem
SSLOptions +FakeBasicAuth

<Location />
  Allow from all
  AuthName 'MySite'
  AuthType Basic
  AuthUserFile /path/to/htpasswd
  Require valid-user
</Location>

Thanks for your hints!

From Pekka.Panula at sofor.fi  Tue Apr 17 11:43:27 2012
From: Pekka.Panula at sofor.fi (Pekka.Panula at sofor.fi)
Date: Tue, 17 Apr 2012 14:43:27 +0300
Subject: I am getting wrong response from nginx
Message-ID: <OFAE5C9F69.4C4F8C40-ONC22579E3.003DBE70-C22579E3.004067BB@sofor.fi>

Hi

I am wondering how does nginx returns wrong content, atleast it seems so. 
I am terminating https addresses to nginx thru my firewall, what does nat 
public ips to nginx to ports 901 and 902 in this case, where nginx is 
configured to wait https connections. 

So when end user hits eg address: https://a.host.com/, firewall does nat 
and sends it to nginx private ip port 901, and https://b.host.com/ goes to 
nginx private ip port 902. 

Some config lines:
http {
    proxy_buffer_size           128k;
    proxy_buffers             4 256k;
    proxy_busy_buffers_size     256k;
    client_max_body_size 50M;
    # DNS resolver
    resolver 213.250.93.67;

   # nginx oma http proxy cache
    proxy_cache_path  /usr/share/nginx/cache  levels=1:2 keys_zone=one:10m 
max_size=1G;
    proxy_temp_path   /usr/share/nginx/tmp;
    proxy_cache_key   "$scheme$host$request_uri$args";
    proxy_cache_use_stale updating error timeout invalid_header http_500 
http_502 http_503 http_504;

   # ---- cache times ---- 
    proxy_cache_valid 200 302 5m;
    proxy_cache_valid 301 1h;
    proxy_cache_valid any 1m;

    proxy_cache one;
    proxy_cache_bypass $http_pragma     $http_authorization 
$cookie_DomAuthSessId   $args;
    proxy_no_cache     $http_pragma     $http_authorization 
$cookie_DomAuthSessId   $args;

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

   gzip_http_version 1.1;
   gzip_vary          on;
   gzip_min_length  1100;
   gzip_buffers     16 8k;
   gzip_disable "MSIE [1-6]\.(?!.*SV1)";
   gzip_proxied     any;
   gzip_types       text/css text/plain application/atom+xml 
application/x-javascript application/xml text/javascript 
application/xml+rss;
   gzip on;

   server_tokens                 off;
   client_header_timeout          3m;
   send_timeout                   3m;
   client_header_buffer_size      8k;
   large_client_header_buffers 4  8k;
   output_buffers              1 32k;
   postpone_output              1460;
   sendfile                       on;
   tcp_nopush                     on;
   tcp_nodelay                    on;
   keepalive_timeout           75 20;
   server_names_hash_bucket_size 256; # this seems to be required for some 
vh

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

   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_connect_timeout 120;
   proxy_send_timeout    120;
   proxy_read_timeout    120;

   proxy_set_header      Accept-Encoding   "";
   proxy_pass_header     Set-Cookie;

  # ---- geoip ----- #
   geo $country {
      default no;
      include conf.d/geo.data;
   }

    # catchall
    server {
        listen          80 default;
        server_name     _;

        access_log  /var/log/nginx/default-access.log  main;

        server_name_in_redirect  off;

        location / {
            index index.html;
            root  /var/www/default/htdocs;
        }
   }

   include /etc/nginx/virtual-hosts/*;

}

then here is config file for a.host.com, file: 
/etc/nginx/virtual-hosts/a.host.com

server {
   listen               81.x.y.x:901;
   ssl                  on;
   ssl_certificate      /etc/nginx/ssl/a.crt;
   ssl_certificate_key  /etc/nginx/ssl/a.key;
   ssl_session_timeout  5m;
   ssl_protocols        SSLv3 TLSv1;
   ssl_ciphers          RC4:HIGH:!aNULL:!MD5;
   ssl_prefer_server_ciphers   on;

   # HTTP Strict Transport Security
   add_header Strict-Transport-Security max-age=500;

   proxy_connect_timeout 300;
   proxy_send_timeout    300;
   proxy_read_timeout    300;

   server_name  a.host.com atest.host.com;
   access_log   /logfiles/a.host.com/a.host.com-access_log combined;
   default_type text/html;
   root         /usr/share/nginx/huolto;

   location / {
      proxy_cache off;
      proxy_cache_valid 200 302 0;
      proxy_cache_valid 301 0;
      proxy_cache_valid any 0;
      proxy_set_header ClientProtocol HTTPS;
      proxy_set_header  Accept-Encoding  "";
      proxy_pass http://1.1.1.131;

      # Huoltoikkuna (manuaalinen)
      include /etc/nginx/maintenance.conf;
   }
}

then file: /etc/nginx/virtual-hosts/b.host.com

server {
   listen               81.x.y.x:902;
   ssl                  on;
   ssl_certificate      /etc/nginx/ssl/b.crt;
   ssl_certificate_key  /etc/nginx/ssl/b.key;
   ssl_session_timeout  5m;
   ssl_protocols        SSLv3 TLSv1;
   ssl_ciphers          RC4:HIGH:!aNULL:!MD5;
   ssl_prefer_server_ciphers   on;

   proxy_cache_use_stale off;

   # HTTP Strict Transport Security
   add_header Strict-Transport-Security max-age=500;

   proxy_connect_timeout 300;
   proxy_send_timeout    300;
   proxy_read_timeout    300;

   server_name  b.host.com btest.host.com;
   access_log   /logfiles/b.host.com/b.host.com-access_log combined;
   default_type text/html;
   root         /usr/share/nginx/huolto;

   location / {
      proxy_cache off;
      proxy_cache_valid 200 302 0;
      proxy_cache_valid 301 0;
      proxy_cache_valid any 0;
      proxy_set_header ClientProtocol HTTPS;
      proxy_set_header  Accept-Encoding  "";
      proxy_pass http://1.1.1.131;

   }
}

I have also a wildcard ssl cert, file 
/etc/nginx/virtual-hosts/wildcard.host.com:
ssl_certificate             /etc/nginx/ssl/wildcard.crt;
ssl_certificate_key         /etc/nginx/ssl/wildcard.key;
ssl_session_timeout         5m;
ssl_protocols                   SSLv3 TLSv1;
ssl_ciphers                     RC4:HIGH:!aNULL:!MD5;
ssl_prefer_server_ciphers   on;
ssl_session_cache           shared:SSL:20m;
server {
        listen       443 default ssl;
        ssl          on;
        server_name  my.default.hostname;
        access_log 
/logfiles/my.default.hostname/my.default.hostname-access_log combined;

        # HTTP Strict Transport Security
        add_header Strict-Transport-Security max-age=500;
 
   location / {

      proxy_cache off;
      proxy_cache_valid 200 302 0;
      proxy_cache_valid 301 0;
      proxy_cache_valid any 0;

      proxy_pass http://1.2.3.4;

      # Huoltoikkuna (manuaalinen)
      include /etc/nginx/maintenance.conf;
   }
}


Now when i do go to address https://b.host.com/, i am getting response 
from a.host.com.

 I am pretty if correct Host header goes to 1.1.1.131 server, it should 
return different content, i have tested it manually using curl, eg:
" curl --verbose --header 'Host: b.host.com' http://1.1.1.131" from nginx 
machine, i will get correct content, but not thru nginx. 
So i am assuming that somehow i am getting wrong content from a.host.com 
or does nginx somehow leave Host header off? 

Notice that i have same proxy_pass in both a.host.com and in b.host.com, 
they resides on same ip and same port (named virtual hosts).
What i am debugged, i can see it goes to b.host.com config block, atleast 
nginx writes to /logfiles/b.host.com/b.host.com-access_log when i do test 
it.

Pekka Panula | Jatkuvat palvelut | Sofor Oy | www.sofor.fi
Takakaarre 3 | PL 51 |FIN-62201 KAUHAVA | tel. +358 6 432 3111 | fax. +358 
6 432 3555 
Mob. + 358 50 384 3232 | pekka.panula at sofor.fi


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20120417/069885df/attachment-0001.html>

From mdounin at mdounin.ru  Tue Apr 17 12:17:57 2012
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Tue, 17 Apr 2012 16:17:57 +0400
Subject: I am getting wrong response from nginx
In-Reply-To: <OFAE5C9F69.4C4F8C40-ONC22579E3.003DBE70-C22579E3.004067BB@sofor.fi>
References: <OFAE5C9F69.4C4F8C40-ONC22579E3.003DBE70-C22579E3.004067BB@sofor.fi>
Message-ID: <20120417121756.GX13466@mdounin.ru>

Hello!

On Tue, Apr 17, 2012 at 02:43:27PM +0300, Pekka.Panula at sofor.fi wrote:

> I am wondering how does nginx returns wrong content, atleast it seems so. 
> I am terminating https addresses to nginx thru my firewall, what does nat 
> public ips to nginx to ports 901 and 902 in this case, where nginx is 
> configured to wait https connections. 
> 
> So when end user hits eg address: https://a.host.com/, firewall does nat 
> and sends it to nginx private ip port 901, and https://b.host.com/ goes to 
> nginx private ip port 902. 

[...]

> server {
>    listen               81.x.y.x:901;

[...]

>    server_name  a.host.com atest.host.com;

[...]

>    location / {
>       proxy_cache off;
>       proxy_cache_valid 200 302 0;
>       proxy_cache_valid 301 0;
>       proxy_cache_valid any 0;
>       proxy_set_header ClientProtocol HTTPS;
>       proxy_set_header  Accept-Encoding  "";
>       proxy_pass http://1.1.1.131;

This will result in a request with "Host: 1.1.1.131" to a backend 
server, see http://nginx.org/r/proxy_set_header.

[...]

> server {
>    listen               81.x.y.x:902;

[...]

>    server_name  b.host.com btest.host.com;

[...]

>    location / {
>       proxy_cache off;
>       proxy_cache_valid 200 302 0;
>       proxy_cache_valid 301 0;
>       proxy_cache_valid any 0;
>       proxy_set_header ClientProtocol HTTPS;
>       proxy_set_header  Accept-Encoding  "";
>       proxy_pass http://1.1.1.131;

Same here.

[...]

> Now when i do go to address https://b.host.com/, i am getting response 
> from a.host.com.
> 
>  I am pretty if correct Host header goes to 1.1.1.131 server, it should 
> return different content, i have tested it manually using curl, eg:
> " curl --verbose --header 'Host: b.host.com' http://1.1.1.131" from nginx 
> machine, i will get correct content, but not thru nginx. 
> So i am assuming that somehow i am getting wrong content from a.host.com 
> or does nginx somehow leave Host header off? 

By default nginx uses the name you wrote in a proxy_pass directive 
for the Host header.  You probably want to use

    proxy_set_header Host $host;

or something similar in the locations above.

[...]

Maxim Dounin


From Pekka.Panula at sofor.fi  Tue Apr 17 12:29:26 2012
From: Pekka.Panula at sofor.fi (Pekka.Panula at sofor.fi)
Date: Tue, 17 Apr 2012 15:29:26 +0300
Subject: I am getting wrong response from nginx
In-Reply-To: <20120417121756.GX13466@mdounin.ru>
References: <OFAE5C9F69.4C4F8C40-ONC22579E3.003DBE70-C22579E3.004067BB@sofor.fi>
	<20120417121756.GX13466@mdounin.ru>
Message-ID: <OF07E8D6A3.C78C03B6-ONC22579E3.004475C3-C22579E3.00449D5E@sofor.fi>

> By default nginx uses the name you wrote in a proxy_pass directive 
> for the Host header.  You probably want to use
> 
>     proxy_set_header Host $host;
> 
> or something similar in the locations above.
> 
> [...]
> 
> Maxim Dounin


Hi, look again, i do have a proxy_set_header line in my http block. It 
should inherit it to server blocks, afaik?


Pekka Panula | Jatkuvat palvelut | Sofor Oy | www.sofor.fi
Takakaarre 3 | PL 51 |FIN-62201 KAUHAVA | tel. +358 6 432 3111 | fax. +358 
6 432 3555 
Mob. + 358 50 384 3232 | pekka.panula at sofor.fi
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20120417/e89ec290/attachment.html>

From mdounin at mdounin.ru  Tue Apr 17 12:41:08 2012
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Tue, 17 Apr 2012 16:41:08 +0400
Subject: some of my user cannot access my website
In-Reply-To: <d6335409b79bc643360c3e44f1a2b9b4.NginxMailingListEnglish@forum.nginx.org>
References: <d6335409b79bc643360c3e44f1a2b9b4.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20120417124108.GY13466@mdounin.ru>

Hello!

On Tue, Apr 17, 2012 at 06:44:48AM -0400, leontinus wrote:

> Hi guys, I need your help to help me identify a problem. I'm running an
> online marketplace here in Indonesia. Right now I have 1 mod perl server
> using apache and 1 nginx server as proxy. But some of my users complain
> they cannot access my website, http://www.tokopedia.com. And I found out
> a fact that if they use mikrotik HttpProxy, they cannot access my
> website, but they can access other website.
> 
> If they use firefox, they found err message : Content Encoding Error
> If they use google chrome, they found err message :
> ERR_INVALID_CHUNKED_ENCODING
> 
> Can you help me with this?

The messages suggest there is something wrong with the proxy in 
question, most likely it fails to handle chunked encoding 
properly.  You may try to disable use of chunked transfer encoding 
with

    chunked_transfer_encoding off;

to see if it helps.  Note though that it will also make keepalive 
connections impossible if Content-Length isn't known.  You may 
also try to disabling keepalive connections completely with

    keepalive_timeout 0;

See here for details:

http://nginx.org/r/chunked_transfer_encoding
http://nginx.org/r/keepalive_timeout

Maxim Dounin


From mdounin at mdounin.ru  Tue Apr 17 12:43:03 2012
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Tue, 17 Apr 2012 16:43:03 +0400
Subject: I am getting wrong response from nginx
In-Reply-To: <OF07E8D6A3.C78C03B6-ONC22579E3.004475C3-C22579E3.00449D5E@sofor.fi>
References: <OFAE5C9F69.4C4F8C40-ONC22579E3.003DBE70-C22579E3.004067BB@sofor.fi>
	<20120417121756.GX13466@mdounin.ru>
	<OF07E8D6A3.C78C03B6-ONC22579E3.004475C3-C22579E3.00449D5E@sofor.fi>
Message-ID: <20120417124303.GZ13466@mdounin.ru>

Hello!

On Tue, Apr 17, 2012 at 03:29:26PM +0300, Pekka.Panula at sofor.fi wrote:

> > By default nginx uses the name you wrote in a proxy_pass directive 
> > for the Host header.  You probably want to use
> > 
> >     proxy_set_header Host $host;
> > 
> > or something similar in the locations above.
> > 
> > [...]
> > 
> > Maxim Dounin
> 
> 
> Hi, look again, i do have a proxy_set_header line in my http block. It 
> should inherit it to server blocks, afaik?

No.

Quote from http://nginx.org/r/proxy_set_header:

    These directives are inherited from the previous level if and only 
    if there are no proxy_set_header directives defined on the current 
    level.

Maxim Dounin


From Pekka.Panula at sofor.fi  Tue Apr 17 12:50:41 2012
From: Pekka.Panula at sofor.fi (Pekka.Panula at sofor.fi)
Date: Tue, 17 Apr 2012 15:50:41 +0300
Subject: I am getting wrong response from nginx
In-Reply-To: <20120417124303.GZ13466@mdounin.ru>
References: <OFAE5C9F69.4C4F8C40-ONC22579E3.003DBE70-C22579E3.004067BB@sofor.fi>	<20120417121756.GX13466@mdounin.ru>
	<OF07E8D6A3.C78C03B6-ONC22579E3.004475C3-C22579E3.00449D5E@sofor.fi>
	<20120417124303.GZ13466@mdounin.ru>
Message-ID: <OF87955D31.63407A26-ONC22579E3.00465BC4-C22579E3.00468F81@sofor.fi>

> > Hi, look again, i do have a proxy_set_header line in my http block. It 

> > should inherit it to server blocks, afaik?
> 
> No.
> 
> Quote from http://nginx.org/r/proxy_set_header:
> 
>     These directives are inherited from the previous level if and only 
>     if there are no proxy_set_header directives defined on the current 
>     level.
> 
> Maxim Dounin


Hi, that was my problem! Thx man, you saved my day. 
Now my configuration does do what it suppose todo.

Pekka Panula | Jatkuvat palvelut | Sofor Oy | www.sofor.fi
Takakaarre 3 | PL 51 |FIN-62201 KAUHAVA | tel. +358 6 432 3111 | fax. +358 
6 432 3555 
Mob. + 358 50 384 3232 | pekka.panula at sofor.fi
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20120417/006e63ad/attachment.html>

From nginx-forum at nginx.us  Tue Apr 17 13:44:13 2012
From: nginx-forum at nginx.us (mscheel)
Date: Tue, 17 Apr 2012 09:44:13 -0400 (EDT)
Subject: worker process exited on signal 11
Message-ID: <2bd9c9c86f68a802bb9544d26ad045bb.NginxMailingListEnglish@forum.nginx.org>

Dear all,

error logs show this error (worker process exited on signal 11). The
error appears after ~1 hour, then with the half of the worker processes.
No problems with 1.0-versions of nginx. Problem existed in 1.1.18
already.

I'm using a custom rpm build with the following additional modules:
--add-module=/root/rpm/SOURCES/agentzh-headers-more-nginx-module-3580526
\
--add-module=/root/rpm/SOURCES/FRiCKLE-ngx_cache_purge-e34bd2c \
--add-module=/root/rpm/SOURCES/substitutions4nginx-read-only \

Here is the backtrace, any ideas?

Thank you.

2012/04/17 10:16:36 [alert] 20802#0: worker process 20808 exited on
signal 11 (core dumped)
2012/04/17 10:16:45 [alert] 20802#0: worker process 20805 exited on
signal 11 (core dumped)
2012/04/17 10:17:33 [alert] 20802#0: worker process 23616 exited on
signal 11 (core dumped)
2012/04/17 10:18:14 [alert] 20802#0: worker process 20806 exited on
signal 11 (core dumped)
2012/04/17 10:22:41 [alert] 20802#0: worker process 23892 exited on
signal 11 (core dumped)

dmesg:
nginx[20808]: segfault at 0000000000000020 rip 000000000042a47b rsp
00007fff79073f80 error 4
nginx[20805]: segfault at 0000000000000020 rip 000000000042a47b rsp
00007fff79073f80 error 4
nginx[23616]: segfault at 0000000000000020 rip 000000000042a47b rsp
00007fff79073ff0 error 4
nginx[20806]: segfault at 0000000000000020 rip 000000000042a47b rsp
00007fff79073f80 error 4
nginx[23892]: segfault at 0000000000000008 rip 000000000042cbe8 rsp
00007fff79073f20 error 6



gdb /usr/sbin/nginx /var/local/dumps/core
GNU gdb 6.6
Copyright (C) 2006 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 "x86_64-suse-linux"...
Using host libthread_db library "/lib64/libthread_db.so.1".
Reading symbols from /lib64/libpthread.so.0...done.
Loaded symbols for /lib64/libpthread.so.0
Reading symbols from /lib64/libcrypt.so.1...done.
Loaded symbols for /lib64/libcrypt.so.1
Reading symbols from /usr/lib64/libpcre.so.0...done.
Loaded symbols for /usr/lib64/libpcre.so.0
Reading symbols from /usr/lib64/libssl.so.0.9.8...done.
Loaded symbols for /usr/lib64/libssl.so.0.9.8
Reading symbols from /usr/lib64/libcrypto.so.0.9.8...done.
Loaded symbols for /usr/lib64/libcrypto.so.0.9.8
Reading symbols from /lib64/libdl.so.2...done.
Loaded symbols for /lib64/libdl.so.2
Reading symbols from /lib64/libz.so.1...done.
Loaded symbols for /lib64/libz.so.1
Reading symbols from /usr/lib64/libGeoIP.so.1...done.
Loaded symbols for /usr/lib64/libGeoIP.so.1
Reading symbols from
/usr/lib/perl5/5.8.8/x86_64-linux-thread-multi/CORE/libperl.so...done.
Loaded symbols for
/usr/lib/perl5/5.8.8/x86_64-linux-thread-multi/CORE/libperl.so
Reading symbols from /lib64/libm.so.6...done.
Loaded symbols for /lib64/libm.so.6
Reading symbols from /lib64/libc.so.6...done.
Loaded symbols for /lib64/libc.so.6
Reading symbols from /lib64/ld-linux-x86-64.so.2...done.
Loaded symbols for /lib64/ld-linux-x86-64.so.2
Reading symbols from
/usr/lib/perl5/site_perl/5.8.8/x86_64-linux-thread-multi/auto/nginx/nginx.so...done.
Loaded symbols for
/usr/lib/perl5/site_perl/5.8.8/x86_64-linux-thread-multi/auto/nginx/nginx.so
Core was generated by `nginx: worker pr'.
Program terminated with signal 11, Segmentation fault.
#0  0x000000000042cbe8 in ngx_readv_chain (c=0x2b508a1d43d0,
chain=0x602348) at src/os/unix/ngx_readv_chain.c:195
195     src/os/unix/ngx_readv_chain.c: No such file or directory.
        in src/os/unix/ngx_readv_chain.c
(gdb) bt
#0  0x000000000042cbe8 in ngx_readv_chain (c=0x2b508a1d43d0,
chain=0x602348) at src/os/unix/ngx_readv_chain.c:195
#1  0x000000000042b133 in ngx_event_pipe (p=0x747170, do_write=<value
optimized out>) at src/event/ngx_event_pipe.c:271
#2  0x000000000045410e in ngx_http_upstream_process_upstream
(r=0x69ef00, u=0x746db8) at src/http/ngx_http_upstream.c:2641
#3  0x0000000000454269 in ngx_http_upstream_handler (ev=0x2b50cc7d9218)
at src/http/ngx_http_upstream.c:935
#4  0x0000000000428f64 in ngx_event_process_posted (cycle=0x5f6490,
posted=0x5d14a0) at src/event/ngx_event_posted.c:40
#5  0x000000000042fc36 in ngx_worker_process_cycle (cycle=0x5f6490,
data=<value optimized out>) at src/os/unix/ngx_process_cycle.c:806
#6  0x000000000042e14d in ngx_spawn_process (cycle=0x5f6490,
proc=0x42fb70 <ngx_worker_process_cycle>, data=0x0, name=0x49e05e
"worker process", respawn=5) at src/os/unix/ngx_process.c:198
#7  0x0000000000430c76 in ngx_master_process_cycle (cycle=0x5f6490) at
src/os/unix/ngx_process_cycle.c:622
#8  0x000000000041235d in main (argc=1, argv=<value optimized out>) at
src/core/nginx.c:410
(gdb) backtrace full
#0  0x000000000042cbe8 in ngx_readv_chain (c=0x2b508a1d43d0,
chain=0x602348) at src/os/unix/ngx_readv_chain.c:195
        prev = (u_char *) 0x0
        n = <value optimized out>
        size = 0
        err = <value optimized out>
        vec = {elts = 0x7fff79073f20, nelts = 0, size = 16, nalloc = 16,
pool = 0x5f57a0}
        rev = <value optimized out>
        iov = (struct iovec *) 0x0
        iovs = {{iov_base = 0x40000, iov_len = 140735223906344},
{iov_base = 0x69ef00, iov_len = 47624914552144}, {iov_base = 0x6024c0,
iov_len = 262144}, {iov_base = 0x1, iov_len = 4566241}, {
    iov_base = 0x40000, iov_len = 6942464}, {iov_base = 0x2b508a1d4550,
iov_len = 0}, {iov_base = 0x0, iov_len = 47624930312022}, {iov_base =
0x69b6e0, iov_len = 6942464}, {
    iov_base = 0x2b508a1d4550, iov_len = 6300864}, {iov_base = 0x40000,
iov_len = 1}, {iov_base = 0x602460, iov_len = 4367464}, {iov_base = 0x1,
iov_len = 2}, {iov_base = 0x0,
    iov_len = 47624930399933}, {iov_base = 0x40000, iov_len = 4309842},
{iov_base = 0x2b50cc80e2e8, iov_len = 4370439}, {iov_base = 0x2b,
iov_len = 2}}
#1  0x000000000042b133 in ngx_event_pipe (p=0x747170, do_write=<value
optimized out>) at src/event/ngx_event_pipe.c:271
        rc = <value optimized out>
        rev = <value optimized out>
        wev = <value optimized out>
#2  0x000000000045410e in ngx_http_upstream_process_upstream
(r=0x69ef00, u=0x746db8) at src/http/ngx_http_upstream.c:2641
        c = (ngx_connection_t *) 0x2b508a1d43d0
#3  0x0000000000454269 in ngx_http_upstream_handler (ev=0x2b50cc7d9218)
at src/http/ngx_http_upstream.c:935
        c = (ngx_connection_t *) 0x2b508a1d4550
        r = (ngx_http_request_t *) 0x69ef00
        u = (ngx_http_upstream_t *) 0x746db8
#4  0x0000000000428f64 in ngx_event_process_posted (cycle=0x5f6490,
posted=0x5d14a0) at src/event/ngx_event_posted.c:40
        ev = (ngx_event_t *) 0x2b50cc7d9218
#5  0x000000000042fc36 in ngx_worker_process_cycle (cycle=0x5f6490,
data=<value optimized out>) at src/os/unix/ngx_process_cycle.c:806
        i = 0
        c = <value optimized out>
#6  0x000000000042e14d in ngx_spawn_process (cycle=0x5f6490,
proc=0x42fb70 <ngx_worker_process_cycle>, data=0x0, name=0x49e05e
"worker process", respawn=5) at src/os/unix/ngx_process.c:198
        on = 1
        pid = 0
        s = 5
#7  0x0000000000430c76 in ngx_master_process_cycle (cycle=0x5f6490) at
src/os/unix/ngx_process_cycle.c:622
        title = <value optimized out>
        p = <value optimized out>
        size = 9
        i = <value optimized out>
        n = 432
        sigio = 0
        set = {__val = {0 <repeats 16 times>}}
        itv = {it_interval = {tv_sec = 0, tv_usec = 6251688}, it_value =
{tv_sec = 0, tv_usec = 47624916165168}}
        live = 1
        delay = 0
        ccf = (ngx_core_conf_t *) 0x5f74f0
#8  0x000000000041235d in main (argc=1, argv=<value optimized out>) at
src/core/nginx.c:410
        i = <value optimized out>
        log = (ngx_log_t *) 0x5ce6e0
        cycle = (ngx_cycle_t *) 0x5f6490
        init_cycle = {conf_ctx = 0x0, pool = 0x5f5440, log = 0x5ce6e0,
new_log = {log_level = 0, file = 0x0, connection = 0, handler = 0, data
= 0x0, action = 0x0}, files = 0x0,
  free_connections = 0x0, free_connection_n = 0,
reusable_connections_queue = {prev = 0x0, next = 0x0}, listening = {elts
= 0x0, nelts = 0, size = 0, nalloc = 0, pool = 0x0}, pathes = {
    elts = 0x0, nelts = 0, size = 0, nalloc = 0, pool = 0x0}, open_files
= {last = 0x0, part = {elts = 0x0, nelts = 0, next = 0x0}, size = 0,
nalloc = 0, pool = 0x0}, shared_memory = {
    last = 0x0, part = {elts = 0x0, nelts = 0, next = 0x0}, size = 0,
nalloc = 0, pool = 0x0}, connection_n = 0, files_n = 0, connections =
0x0, read_events = 0x0, write_events = 0x0,
  old_cycle = 0x0, conf_file = {len = 21, data = 0x4998b3
"/etc/nginx/nginx.conf"}, conf_param = {len = 0, data = 0x0},
conf_prefix = {len = 11, data = 0x4998b3 "/etc/nginx/nginx.conf"},
  prefix = {len = 6, data = 0x4998ac "/usr//"}, lock_file = {len = 0,
data = 0x0}, hostname = {len = 0, data = 0x0}}
        ccf = (ngx_core_conf_t *) 0x5f74f0

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225420,225420#msg-225420


From nginx-forum at nginx.us  Tue Apr 17 14:20:10 2012
From: nginx-forum at nginx.us (mscheel)
Date: Tue, 17 Apr 2012 10:20:10 -0400 (EDT)
Subject: worker process exited on signal 11
In-Reply-To: <2bd9c9c86f68a802bb9544d26ad045bb.NginxMailingListEnglish@forum.nginx.org>
References: <2bd9c9c86f68a802bb9544d26ad045bb.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <35b528922bf87b99900a8ada6b165ca8.NginxMailingListEnglish@forum.nginx.org>

sorry, forgot uname -a:

Linux hostname 2.6.16.60-0.42.10-smp #1 SMP Tue Apr 27 05:11:27 UTC 2010
x86_64 x86_64 x86_64 GNU/Linux

 nginx -V
nginx version: nginx/1.1.19
built by gcc 4.1.2 20070115 (SUSE Linux)
TLS SNI support disabled
configure arguments:
--add-module=/root/rpm/SOURCES/agentzh-headers-more-nginx-module-3580526
--add-module=/root/rpm/SOURCES/FRiCKLE-ngx_cache_purge-e34bd2c
--add-module=/root/rpm/SOURCES/substitutions4nginx-read-only
--with-debug --prefix=/usr/ --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/lib/nginx/tmp/
--http-proxy-temp-path=/var/lib/nginx/proxy/
--http-fastcgi-temp-path=/var/lib/nginx/fastcgi/
--http-uwsgi-temp-path=/var/lib/nginx/uwsgi/
--http-scgi-temp-path=/var/lib/nginx/scgi/ --user=wwwrun --group=www
--with-rtsig_module --with-select_module --with-poll_module --with-ipv6
--with-http_ssl_module --with-http_realip_module
--with-http_addition_module --with-http_geoip_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_degradation_module
--with-http_stub_status_module --with-http_perl_module
--with-perl=/usr/bin/perl --with-mail --with-mail_ssl_module --with-pcre
--with-md5=/usr --with-sha1=/usr --with-cc-opt='-O2 -g
-fmessage-length=0 -D_FORTIFY_SOURCE=2 -fstack-protector'

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225420,225433#msg-225433


From a_davy at bellsouth.net  Tue Apr 17 15:56:54 2012
From: a_davy at bellsouth.net (Allan Davy)
Date: Tue, 17 Apr 2012 10:56:54 -0500
Subject: Compiling nginx with zlib-1.2.6 error
Message-ID: <AB1F496419824766A1486A8A00253756@32PNBF1>

Hello,
I attempted to compile nginx-1.0.15 with zlib-1.2.6 on a Solaris 11 platform and received the following error.

gmake
.
.
.
cd ../zlib-1.2.6 \
    && gmake distclean \
    && CFLAGS="-O2 -fomit-frame-pointer -pipe " CC="gcc" \
        ./configure \
    && gmake libz.a
gmake[2]: Entering directory `/home/produser/Downloads/zlib-1.2.6'
make -f Makefile.in distclean
make: Fatal error in reader: Makefile.in, line 57: Unexpected end of line seen
Current working directory /home/produser/Downloads/zlib-1.2.6
gmake[2]: *** [distclean] Error 1
gmake[2]: Leaving directory `/home/produser/Downloads/zlib-1.2.6'
gmake[1]: *** [../zlib-1.2.6/libz.a] Error 2
gmake[1]: Leaving directory `/home/produser/Downloads/nginx-1.0.15'
gmake: *** [build] Error 2

Pleae advise what I should do.  There were many many bug fixes and enhancements to zlib 1.2.6 over the 1.2.5 and would like to know if its going to work or not.

Tank you,
Allan
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20120417/644821ce/attachment.html>

From sb at waeme.net  Tue Apr 17 16:21:15 2012
From: sb at waeme.net (Sergey Budnevitch)
Date: Tue, 17 Apr 2012 20:21:15 +0400
Subject: Compiling nginx with zlib-1.2.6 error
In-Reply-To: <AB1F496419824766A1486A8A00253756@32PNBF1>
References: <AB1F496419824766A1486A8A00253756@32PNBF1>
Message-ID: <6FA4E717-2E18-4C73-AA9E-73E2F55538C8@waeme.net>


On 17.04.2012, at 19:56, Allan Davy wrote:

> Hello,
> I attempted to compile nginx-1.0.15 with zlib-1.2.6 on a Solaris 11 platform and received the following error.

I doubt you will be able to build static library on solaris 10 or 11.
Try to build or install new dynamic zlib 1.2.6 and run nginx with LD_PRELOAD hack:
LD_PRELOAD=/path/to/new/libz.so nginx

>  
> gmake
> .
> .
> .
> cd ../zlib-1.2.6 \
>     && gmake distclean \
>     && CFLAGS="-O2 -fomit-frame-pointer -pipe " CC="gcc" \
>         ./configure \
>     && gmake libz.a
> gmake[2]: Entering directory `/home/produser/Downloads/zlib-1.2.6'
> make -f Makefile.in distclean
> make: Fatal error in reader: Makefile.in, line 57: Unexpected end of line seen
> Current working directory /home/produser/Downloads/zlib-1.2.6
> gmake[2]: *** [distclean] Error 1
> gmake[2]: Leaving directory `/home/produser/Downloads/zlib-1.2.6'
> gmake[1]: *** [../zlib-1.2.6/libz.a] Error 2
> gmake[1]: Leaving directory `/home/produser/Downloads/nginx-1.0.15'
> gmake: *** [build] Error 2
>  
> Pleae advise what I should do.  There were many many bug fixes and enhancements to zlib 1.2.6 over the 1.2.5 and would like to know if its going to work or not.
>  
> Tank you,
> Allan
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx


From mdounin at mdounin.ru  Tue Apr 17 16:48:24 2012
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Tue, 17 Apr 2012 20:48:24 +0400
Subject: Compiling nginx with zlib-1.2.6 error
In-Reply-To: <AB1F496419824766A1486A8A00253756@32PNBF1>
References: <AB1F496419824766A1486A8A00253756@32PNBF1>
Message-ID: <20120417164824.GB13466@mdounin.ru>

Hello!

On Tue, Apr 17, 2012 at 10:56:54AM -0500, Allan Davy wrote:

> Hello,
> I attempted to compile nginx-1.0.15 with zlib-1.2.6 on a Solaris 11 platform and received the following error.
> 
> gmake
> .
> .
> .
> cd ../zlib-1.2.6 \
>     && gmake distclean \
>     && CFLAGS="-O2 -fomit-frame-pointer -pipe " CC="gcc" \
>         ./configure \
>     && gmake libz.a
> gmake[2]: Entering directory `/home/produser/Downloads/zlib-1.2.6'
> make -f Makefile.in distclean
> make: Fatal error in reader: Makefile.in, line 57: Unexpected end of line seen
> Current working directory /home/produser/Downloads/zlib-1.2.6
> gmake[2]: *** [distclean] Error 1
> gmake[2]: Leaving directory `/home/produser/Downloads/zlib-1.2.6'
> gmake[1]: *** [../zlib-1.2.6/libz.a] Error 2
> gmake[1]: Leaving directory `/home/produser/Downloads/nginx-1.0.15'
> gmake: *** [build] Error 2
> 
> Pleae advise what I should do.  There were many many bug fixes 
> and enhancements to zlib 1.2.6 over the 1.2.5 and would like to 
> know if its going to work or not.

As usual, "many bug fixes" means "old bugs fixed, new bugs added".  
The zlib-1.2.6's Makefile.in now can't be parsed by Solaris's make, 
but shipped Makefile calls "make" as a hardcoded command for 
distclean target.  You may try the following patch (against 
zlib-1.2.6) to see if it helps (and report the problem to zlib 
folks):

--- Makefile.orig       Tue Apr 17 20:39:34 2012
+++ Makefile    Tue Apr 17 20:39:48 2012
@@ -2,4 +2,4 @@
        - at echo "Please use ./configure first.  Thank you."
 
 distclean:
-       make -f Makefile.in distclean
+       ${MAKE} -f Makefile.in distclean

If you want to just compile nginx, the fix should be as trivial as 
running ./configure in zlib's directory once.

Maxim Dounin


From mdounin at mdounin.ru  Tue Apr 17 16:56:46 2012
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Tue, 17 Apr 2012 20:56:46 +0400
Subject: Compiling nginx with zlib-1.2.6 error
In-Reply-To: <6FA4E717-2E18-4C73-AA9E-73E2F55538C8@waeme.net>
References: <AB1F496419824766A1486A8A00253756@32PNBF1>
	<6FA4E717-2E18-4C73-AA9E-73E2F55538C8@waeme.net>
Message-ID: <20120417165645.GC13466@mdounin.ru>

Hello!

On Tue, Apr 17, 2012 at 08:21:15PM +0400, Sergey Budnevitch wrote:

> 
> On 17.04.2012, at 19:56, Allan Davy wrote:
> 
> > Hello,
> > I attempted to compile nginx-1.0.15 with zlib-1.2.6 on a Solaris 11 platform and received the following error.
> 
> I doubt you will be able to build static library on solaris 10 or 11.
> Try to build or install new dynamic zlib 1.2.6 and run nginx with LD_PRELOAD hack:
> LD_PRELOAD=/path/to/new/libz.so nginx

It builds fine with zlib 1.2.5.  The problem is in zlib 1.2.6 
Makefile / Makefile.in, see my reply.

Maxim Dounin

> 
> >  
> > gmake
> > .
> > .
> > .
> > cd ../zlib-1.2.6 \
> >     && gmake distclean \
> >     && CFLAGS="-O2 -fomit-frame-pointer -pipe " CC="gcc" \
> >         ./configure \
> >     && gmake libz.a
> > gmake[2]: Entering directory `/home/produser/Downloads/zlib-1.2.6'
> > make -f Makefile.in distclean
> > make: Fatal error in reader: Makefile.in, line 57: Unexpected end of line seen
> > Current working directory /home/produser/Downloads/zlib-1.2.6
> > gmake[2]: *** [distclean] Error 1
> > gmake[2]: Leaving directory `/home/produser/Downloads/zlib-1.2.6'
> > gmake[1]: *** [../zlib-1.2.6/libz.a] Error 2
> > gmake[1]: Leaving directory `/home/produser/Downloads/nginx-1.0.15'
> > gmake: *** [build] Error 2
> >  
> > Pleae advise what I should do.  There were many many bug fixes and enhancements to zlib 1.2.6 over the 1.2.5 and would like to know if its going to work or not.
> >  
> > Tank you,
> > Allan
> > _______________________________________________
> > nginx mailing list
> > nginx at nginx.org
> > http://mailman.nginx.org/mailman/listinfo/nginx
> 
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx


From mdounin at mdounin.ru  Tue Apr 17 17:20:52 2012
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Tue, 17 Apr 2012 21:20:52 +0400
Subject: worker process exited on signal 11
In-Reply-To: <2bd9c9c86f68a802bb9544d26ad045bb.NginxMailingListEnglish@forum.nginx.org>
References: <2bd9c9c86f68a802bb9544d26ad045bb.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20120417172052.GE13466@mdounin.ru>

Hello!

On Tue, Apr 17, 2012 at 09:44:13AM -0400, mscheel wrote:

> Dear all,
> 
> error logs show this error (worker process exited on signal 11). The
> error appears after ~1 hour, then with the half of the worker processes.
> No problems with 1.0-versions of nginx. Problem existed in 1.1.18
> already.
> 
> I'm using a custom rpm build with the following additional modules:
> --add-module=/root/rpm/SOURCES/agentzh-headers-more-nginx-module-3580526
> \
> --add-module=/root/rpm/SOURCES/FRiCKLE-ngx_cache_purge-e34bd2c \
> --add-module=/root/rpm/SOURCES/substitutions4nginx-read-only \

Are you able to reproduce this without 3rd party modules?

[...]

> (gdb) bt
> #0  0x000000000042cbe8 in ngx_readv_chain (c=0x2b508a1d43d0,
> chain=0x602348) at src/os/unix/ngx_readv_chain.c:195

Please provide gdb output of the following commands:

fr 0
p chain
p *chain

> #1  0x000000000042b133 in ngx_event_pipe (p=0x747170, do_write=<value
> optimized out>) at src/event/ngx_event_pipe.c:271

fr 1
p *p

> #2  0x000000000045410e in ngx_http_upstream_process_upstream
> (r=0x69ef00, u=0x746db8) at src/http/ngx_http_upstream.c:2641

fr 2
p *r
p *u

[...]

It might be also helpful to see full configuration used.

Maxim Dounin


From nginx-forum at nginx.us  Tue Apr 17 17:40:05 2012
From: nginx-forum at nginx.us (leontinus)
Date: Tue, 17 Apr 2012 13:40:05 -0400 (EDT)
Subject: some of my user cannot access my website
In-Reply-To: <20120417124108.GY13466@mdounin.ru>
References: <20120417124108.GY13466@mdounin.ru>
Message-ID: <a7618092b1a28855940c85ce29b37f56.NginxMailingListEnglish@forum.nginx.org>

Maxim Dounin Wrote:
-------------------------------------------------------
> Hello!
> 
> On Tue, Apr 17, 2012 at 06:44:48AM -0400,
> leontinus wrote:
> 
> > Hi guys, I need your help to help me identify a
> problem. I'm running an
> > online marketplace here in Indonesia. Right now
> I have 1 mod perl server
> > using apache and 1 nginx server as proxy. But
> some of my users complain
> > they cannot access my website,
> http://www.tokopedia.com. And I found out
> > a fact that if they use mikrotik HttpProxy, they
> cannot access my
> > website, but they can access other website.
> > 
> > If they use firefox, they found err message :
> Content Encoding Error
> > If they use google chrome, they found err
> message :
> > ERR_INVALID_CHUNKED_ENCODING
> > 
> > Can you help me with this?
> 
> The messages suggest there is something wrong with
> the proxy in 
> question, most likely it fails to handle chunked
> encoding 
> properly.  You may try to disable use of chunked
> transfer encoding 
> with
> 
>     chunked_transfer_encoding off;
> 
> to see if it helps.  Note though that it will also
> make keepalive 
> connections impossible if Content-Length isn't
> known.  You may 
> also try to disabling keepalive connections
> completely with
> 
>     keepalive_timeout 0;
> 
> See here for details:
> 
> http://nginx.org/r/chunked_transfer_encoding
> http://nginx.org/r/keepalive_timeout
> 
> Maxim Dounin
> 
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx


Hi Maxim Dounin,

Thank you for these tips, I will try it and see if it help my user. I
will update again soon I get the result.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225421,225441#msg-225441


From siefke_listen at web.de  Tue Apr 17 22:17:49 2012
From: siefke_listen at web.de (Silvio Siefke)
Date: Wed, 18 Apr 2012 00:17:49 +0200
Subject: HTML as PHP
Message-ID: <20120418001749.8d31e902.siefke_listen@web.de>

Hello,

i want html files use as php file. So i have write the follow config:

/etc/nginx/nginx.conf
server {
listen	     80;
server_name  silviosiefke_de;
access_log /var/www/silviosiefke_de/log/access_log main;
error_log /var/www/silviosiefke_de/log/error_log info;


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

#location /article-(\d+)\.html$ {
#try_files $uri $uri/ /blogdetail.html?id=$1;
#}

# pass the PHP scripts to FastCGI 
location ~ \.(php|htm|html)$ {
#location ~ \.php$ {
fastcgi_pass   127.0.0.1:9000;
fastcgi_index  index.php;
fastcgi_param SCRIPT_FILENAME  /var/www/silviosiefke_de/htdocs$fastcgi_script_name;
include /etc/nginx/fastcgi_params; 
}

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

When i make restart nginx and go on the side i read only Access denied. 
In the log i have:

access.log
192.168.0.22 - - [18/Apr/2012:00:15:54 +0200] "GET /blog/manage.html 
HTTP/1.1" 403 25 "-" "Mozilla/5.0 (Macintosh; U; Intel Mac OS X; de-de)
AppleWebKit/535+ (KHTML, like Gecko) Version/5.0 Safari/535.4+ Midori/0.4" "-"

No entrys in the error.log. Has someone an idea?

Regards
Silvio


From cliff at develix.com  Tue Apr 17 22:56:34 2012
From: cliff at develix.com (Cliff Wells)
Date: Tue, 17 Apr 2012 15:56:34 -0700
Subject: HTML as PHP
In-Reply-To: <20120418001749.8d31e902.siefke_listen@web.de>
References: <20120418001749.8d31e902.siefke_listen@web.de>
Message-ID: <1334703394.2486.10.camel@portable-evil>

On Wed, 2012-04-18 at 00:17 +0200, Silvio Siefke wrote:

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

This also matches .html files.  Try making your regex more explicit.

Cliff



From siefke_listen at web.de  Tue Apr 17 23:02:23 2012
From: siefke_listen at web.de (siefke_listen at web.de)
Date: Wed, 18 Apr 2012 01:02:23 +0200
Subject: HTML as PHP
In-Reply-To: <1334703394.2486.10.camel@portable-evil>
References: <20120418001749.8d31e902.siefke_listen@web.de>
	<1334703394.2486.10.camel@portable-evil>
Message-ID: <20120418010223.d8bc2cd0.siefke_listen@web.de>

Hello,

On Tue, 17 Apr 2012 15:56:34 -0700
Cliff Wells <cliff at develix.com> wrote:

> On Wed, 2012-04-18 at 00:17 +0200, Silvio Siefke wrote:
> 
> > location ~ /\.ht {
> > deny  all;
> > }
> 
> This also matches .html files.  Try making your regex more explicit.


ok i have delete the part, but the error is the same. Access Denied. 
And in the logs the same message, 403. 

Regards
Silvio


From cliff at develix.com  Tue Apr 17 23:10:55 2012
From: cliff at develix.com (Cliff Wells)
Date: Tue, 17 Apr 2012 16:10:55 -0700
Subject: HTML as PHP
In-Reply-To: <1334703394.2486.10.camel@portable-evil>
References: <20120418001749.8d31e902.siefke_listen@web.de>
	<1334703394.2486.10.camel@portable-evil>
Message-ID: <1334704255.2486.18.camel@portable-evil>

On Tue, 2012-04-17 at 15:56 -0700, Cliff Wells wrote:
> On Wed, 2012-04-18 at 00:17 +0200, Silvio Siefke wrote:
> 
> > location ~ /\.ht {
> > deny  all;
> > }
> 
> This also matches .html files.  Try making your regex more explicit.

My mistake, it doesn't.  Check these:

1) file/directory permissions are correct
2) is the php app is setting the status to 403?

Regards,
Cliff


From cliff at develix.com  Tue Apr 17 23:12:33 2012
From: cliff at develix.com (Cliff Wells)
Date: Tue, 17 Apr 2012 16:12:33 -0700
Subject: HTML as PHP
In-Reply-To: <20120418010223.d8bc2cd0.siefke_listen@web.de>
References: <20120418001749.8d31e902.siefke_listen@web.de>
	<1334703394.2486.10.camel@portable-evil>
	<20120418010223.d8bc2cd0.siefke_listen@web.de>
Message-ID: <1334704353.2486.19.camel@portable-evil>

On Wed, 2012-04-18 at 01:02 +0200, siefke_listen at web.de wrote:

> ok i have delete the part, but the error is the same. Access Denied. 
> And in the logs the same message, 403. 

Can you also check the main nginx error log?
Usually /var/log/nginx/error.log.

Cliff


From siefke_listen at web.de  Tue Apr 17 23:19:46 2012
From: siefke_listen at web.de (siefke_listen at web.de)
Date: Wed, 18 Apr 2012 01:19:46 +0200
Subject: HTML as PHP
In-Reply-To: <1334704353.2486.19.camel@portable-evil>
References: <20120418001749.8d31e902.siefke_listen@web.de>
	<1334703394.2486.10.camel@portable-evil>
	<20120418010223.d8bc2cd0.siefke_listen@web.de>
	<1334704353.2486.19.camel@portable-evil>
Message-ID: <20120418011946.3328e5fb.siefke_listen@web.de>

On Tue, 17 Apr 2012 16:12:33 -0700
Cliff Wells <cliff at develix.com> wrote:

> Can you also check the main nginx error log?
> Usually /var/log/nginx/error.log.

Thats stand in the main error.log

2012/04/18 01:00:08 [info] 6802#0: Using 32768KiB of shared memory for push module in /etc/nginx/nginx.conf:85
2012/04/18 01:00:08 [notice] 6330#0: signal 15 (SIGTERM) received, exiting
2012/04/18 01:00:08 [notice] 6332#0: exiting
2012/04/18 01:00:08 [notice] 6332#0: exit
2012/04/18 01:00:08 [notice] 6330#0: signal 17 (SIGCHLD) received
2012/04/18 01:00:08 [notice] 6330#0: worker process 6332 exited with code 0
2012/04/18 01:00:08 [notice] 6330#0: exit
2012/04/18 01:00:08 [notice] 6817#0: using the "epoll" event method
2012/04/18 01:00:08 [notice] 6817#0: nginx/1.0.15
2012/04/18 01:00:08 [notice] 6817#0: OS: Linux 3.2.14
2012/04/18 01:00:08 [notice] 6817#0: getrlimit(RLIMIT_NOFILE): 1024:4096
2012/04/18 01:00:08 [notice] 6818#0: start worker processes
2012/04/18 01:00:08 [notice] 6818#0: start worker process 6820
2012/04/18 01:12:29 [notice] 6818#0: signal 15 (SIGTERM) received, exiting
2012/04/18 01:12:29 [notice] 6820#0: exiting
2012/04/18 01:12:29 [notice] 6820#0: exit
2012/04/18 01:12:29 [notice] 6818#0: signal 17 (SIGCHLD) received
2012/04/18 01:12:29 [notice] 6818#0: worker process 6820 exited with code 0
2012/04/18 01:12:29 [notice] 6818#0: exit
2012/04/18 01:12:32 [info] 7036#0: Using 32768KiB of shared memory for push module in /etc/nginx/nginx.conf:85
2012/04/18 01:12:32 [notice] 7040#0: using the "epoll" event method
2012/04/18 01:12:32 [notice] 7040#0: nginx/1.0.15
2012/04/18 01:12:32 [notice] 7040#0: OS: Linux 3.2.14
2012/04/18 01:12:32 [notice] 7040#0: getrlimit(RLIMIT_NOFILE): 1024:4096
2012/04/18 01:12:32 [notice] 7041#0: start worker processes
2012/04/18 01:12:32 [notice] 7041#0: start worker process 7043
2012/04/18 01:12:57 [info] 7058#0: Using 32768KiB of shared memory for push module in /etc/nginx/nginx.conf:85
2012/04/18 01:12:57 [notice] 7041#0: signal 15 (SIGTERM) received, exiting
2012/04/18 01:12:57 [notice] 7043#0: exiting
2012/04/18 01:12:57 [notice] 7043#0: exit
2012/04/18 01:12:57 [notice] 7041#0: signal 17 (SIGCHLD) received
2012/04/18 01:12:57 [notice] 7041#0: worker process 7043 exited with code 0
2012/04/18 01:12:57 [notice] 7041#0: exit
2012/04/18 01:12:58 [notice] 7073#0: using the "epoll" event method
2012/04/18 01:12:58 [notice] 7073#0: nginx/1.0.15
2012/04/18 01:12:58 [notice] 7073#0: OS: Linux 3.2.14
2012/04/18 01:12:58 [notice] 7073#0: getrlimit(RLIMIT_NOFILE): 1024:4096
2012/04/18 01:12:58 [notice] 7074#0: start worker processes
2012/04/18 01:12:58 [notice] 7074#0: start worker process 7076


From siefke_listen at web.de  Tue Apr 17 23:26:52 2012
From: siefke_listen at web.de (siefke_listen at web.de)
Date: Wed, 18 Apr 2012 01:26:52 +0200
Subject: HTML as PHP
In-Reply-To: <1334704255.2486.18.camel@portable-evil>
References: <20120418001749.8d31e902.siefke_listen@web.de>
	<1334703394.2486.10.camel@portable-evil>
	<1334704255.2486.18.camel@portable-evil>
Message-ID: <20120418012652.45339807.siefke_listen@web.de>

On Tue, 17 Apr 2012 16:10:55 -0700
Cliff Wells <cliff at develix.com> wrote:

> 1) file/directory permissions are correct

gentoo-desk htdocs # ls -l
insgesamt 40
drwxr-xr-x 2 siefke siefke 4096 18. Apr 00:08 blog
-rw-r--r-- 1 siefke siefke 1406 18. M?r 2011  favicon.ico
drwxr-xr-x 2 siefke siefke 4096 16. Apr 20:02 howto
drwxr-xr-x 4 siefke siefke 4096 16. Apr 13:38 img
drwxr-xr-x 2 siefke siefke 4096 18. Apr 00:08 inc
-rw-r--r-- 1 siefke siefke 1137 17. Apr 22:45 index.html
-rw-r--r-- 1 siefke siefke   17 14. Apr 00:09 info.php
-rw-r--r-- 1 siefke siefke   43 16. Apr 20:28 robots.txt
-rw-r--r-- 1 siefke siefke  508 16. Apr 20:28 sitemap.xml

> 2) is the php app is setting the status to 403?

The mistake come in blog directory, ok i try in root without any 
php inside html, but it come too Access denied.

The complete nginx.conf -> http://nopaste.info/d7665fafff.html
The php-php.conf -> http://nopaste.info/717a0a166f.html

Thank u for ur Help.


Regards
Silvio


From nginx-forum at nginx.us  Wed Apr 18 00:08:36 2012
From: nginx-forum at nginx.us (ragivan)
Date: Tue, 17 Apr 2012 20:08:36 -0400 (EDT)
Subject: MP4 pseudostreaming - seek delay
In-Reply-To: <20120305070815.GA28715@nginx.com>
References: <20120305070815.GA28715@nginx.com>
Message-ID: <a9e0ff79fbc7221000c17dcdf2641420.NginxMailingListEnglish@forum.nginx.org>

So how does YouTube manage to get 2 hour videos to start playing
instantly (1 second).. with nginx it takes me 10 seconds for a 2 hour
video to start playing.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,223375,225451#msg-225451


From jdorfman at netdna.com  Wed Apr 18 00:29:52 2012
From: jdorfman at netdna.com (Justin Dorfman)
Date: Tue, 17 Apr 2012 17:29:52 -0700
Subject: MP4 pseudostreaming - seek delay
In-Reply-To: <a9e0ff79fbc7221000c17dcdf2641420.NginxMailingListEnglish@forum.nginx.org>
References: <20120305070815.GA28715@nginx.com>
	<a9e0ff79fbc7221000c17dcdf2641420.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <CACDr0vFAzWMy6=cZTakZkpY6FN_mejb--yYn2LhnRAyj=jWRRQ@mail.gmail.com>

For video Youtube uses lighttp, not nginx -
http://highscalability.com/youtube-architecture

Regards,

Justin Dorfman <http://www.twitter.com/jdorfman>

NetDNA <http://www.netdna.com>?
The Science of Acceleration?

Email / gtalk: jdorfman at netdna.com
P: 1.323.301.1400 x 272
Skype: netdna-justind
Twitter: @jdorfman <http://www.twitter.com/jdorfman>

www.NetDNA.com <http://www.netdna.com/> | www.MaxCDN.com<http://www.maxcdn.com/>

@NetDNA <http://twitter.com/NetDNA> | @MaxCDN <http://twitter.com/MaxCDN>



On Tue, Apr 17, 2012 at 5:08 PM, ragivan <nginx-forum at nginx.us> wrote:

> So how does YouTube manage to get 2 hour videos to start playing
> instantly (1 second).. with nginx it takes me 10 seconds for a 2 hour
> video to start playing.
>
> Posted at Nginx Forum:
> http://forum.nginx.org/read.php?2,223375,225451#msg-225451
>
> _______________________________________________
> 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/20120417/4a658c7e/attachment.html>

From nginx-forum at nginx.us  Wed Apr 18 01:23:13 2012
From: nginx-forum at nginx.us (ragivan)
Date: Tue, 17 Apr 2012 21:23:13 -0400 (EDT)
Subject: MP4 pseudostreaming - seek delay
In-Reply-To: <CACDr0vFAzWMy6=cZTakZkpY6FN_mejb--yYn2LhnRAyj=jWRRQ@mail.gmail.com>
References: <CACDr0vFAzWMy6=cZTakZkpY6FN_mejb--yYn2LhnRAyj=jWRRQ@mail.gmail.com>
Message-ID: <d30c68eab7e112b3fd2ca0511801ccc2.NginxMailingListEnglish@forum.nginx.org>

That was back in 2008, they didn't have 2 hour long videos. But what I
was talking about what regarding the technology, lighttpd and nginx both
do pseudo streaming. The moov atom is about 5 MB so it takes about 10
seconds to download that before playing the video. Is there some
technique to start playing the video after partial download of the moov
atom? Or is it possible to cache the moov atom into memory for faster
retrieval?

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,223375,225453#msg-225453


From nginx-forum at nginx.us  Wed Apr 18 03:31:24 2012
From: nginx-forum at nginx.us (activa)
Date: Tue, 17 Apr 2012 23:31:24 -0400 (EDT)
Subject: nginx + FollowSymLinks owner verification
In-Reply-To: <d28b11bfc06a3323b9fab10b68e80d1f.NginxMailingListEnglish@forum.nginx.org>
References: <d28b11bfc06a3323b9fab10b68e80d1f.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <0c5d7c297e71276447c3ffd6b4705ef6.NginxMailingListEnglish@forum.nginx.org>

is this available in stable version 1.0.15 ? 

disable_symlinks if_not_owner;

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225152,225454#msg-225454


From edho at myconan.net  Wed Apr 18 03:37:43 2012
From: edho at myconan.net (Edho Arief)
Date: Wed, 18 Apr 2012 10:37:43 +0700
Subject: nginx + FollowSymLinks owner verification
In-Reply-To: <0c5d7c297e71276447c3ffd6b4705ef6.NginxMailingListEnglish@forum.nginx.org>
References: <d28b11bfc06a3323b9fab10b68e80d1f.NginxMailingListEnglish@forum.nginx.org>
	<0c5d7c297e71276447c3ffd6b4705ef6.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <CAAtReC=08cVxFbjSf86FYMKP8p2bkCyOHmmJWaCaPs4bM7YRbg@mail.gmail.com>

2012/4/18 activa <nginx-forum at nginx.us>:
> is this available in stable version 1.0.15 ?
>
> disable_symlinks if_not_owner;
>

Nope, it's new in 1.1.x (since 1.1.16)


From nginx-forum at nginx.us  Wed Apr 18 04:39:38 2012
From: nginx-forum at nginx.us (justin)
Date: Wed, 18 Apr 2012 00:39:38 -0400 (EDT)
Subject: Nginx Benchmark Serving Static Image Timeouts
Message-ID: <20d30f3c6faa2eaad43656025b0b0cc4.NginxMailingListEnglish@forum.nginx.org>

Hello, I am benchmarking nginx, serving a static image (png file which
is 6kb) with http://blitz.io. I am sending 250 requests a second for 60
seconds. When I run the benchmark approximately 15% of the requests are
taking longer than 2 seconds, which is completely unacceptable. Nginx
should be able to fly at this, even though its a lot of requests. At
first I thought that I must be maxing out the network interface, but I
looked and only pushing 10-12mbps. Any ideas what could be causing
this?

Here is the performance directives that I have modified in nginx:

worker_processes 4;

events {
  worker_connections  2048;
}

sendfile on;
tcp_nopush on;
tcp_nodelay on;
keepalive_timeout 30;
gzip on;
gzip_proxied any;
gzip_comp_level 3;
gzip_disable "MSIE [1-6].(?!.*SV1)";
gzip_types text/plain text/css application/x-javascript text/xml
application/xml application/xml+rss text/javascript;
client_max_body_size 16M;

location ~* \.(?:ico|css|js|gif|jpe?g|png)$ {
  expires 30d;
  add_header Pragma public;
  add_header Cache-Control "public, must-revalidate, proxy-revalidate";
}

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225456,225456#msg-225456


From nginx-forum at nginx.us  Wed Apr 18 05:06:25 2012
From: nginx-forum at nginx.us (mscheel)
Date: Wed, 18 Apr 2012 01:06:25 -0400 (EDT)
Subject: worker process exited on signal 11
In-Reply-To: <2bd9c9c86f68a802bb9544d26ad045bb.NginxMailingListEnglish@forum.nginx.org>
References: <2bd9c9c86f68a802bb9544d26ad045bb.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <02299bd23d1683f18815cac8d2f70f37.NginxMailingListEnglish@forum.nginx.org>

Hello Maxim,

it appears only on two productive servers. Compiling without these
modules is not possible at the moment but i've tried several times with
each module deactivated (and without any of these three).

Full config is huge. I'm using a php-fpm backend with keepalive on and
one upstream server with nginx and php-fpm.

  http {
	include mime.types;
	include gzip.conf;
	default_type  application/octet-stream;
	keepalive_timeout  240;
	keepalive_requests 2000;
	open_file_cache max=10000 inactive=3m; 
	open_file_cache_valid    3m; 
	open_file_cache_min_uses 2;
	open_file_cache_errors   off;
	index  wartung.html index.php index.html index.htm;
	recursive_error_pages on;
	
	large_client_header_buffers 8 256k;
	client_max_body_size 1024m;
	client_body_buffer_size 128K;
	client_header_buffer_size 10k;
	client_header_timeout 120;
	client_body_timeout 120;
	fastcgi_read_timeout 240;

	set_real_ip_from x.x.x.0/24;
	real_ip_header X-Forwarded-For;
	
	#nginx 1.0
		#limit_zone spiders $binary_remote_addr 10m;
	#nginx 1.1
		limit_conn_zone $binary_remote_addr zone=spiders:10m;
	
	limit_req_zone  $binary_remote_addr  zone=spider_req:10m   rate=5r/s;
	limit_conn_log_level info;
	
	upstream be_server_nginx {
		server IP:80;
		keepalive 300;
	}
	}
	
	...
	
	location @be_nginx {
	proxy_pass http://be_server_nginx;
	proxy_set_header Connection "";	proxy_http_version 1.1;
	include /etc/nginx/proxy.conf;


output of gdb:

(gdb) fr 0
#0  0x000000000042cbe8 in ngx_readv_chain (c=0x2b508a1d43d0,
chain=0x602348) at src/os/unix/ngx_readv_chain.c:195
195     in src/os/unix/ngx_readv_chain.c

(gdb) p chain
$1 = (ngx_chain_t *) 0x602348

(gdb) p *chain
$2 = {buf = 0x602460, next = 0x752628}

(gdb) fr 1
#1  0x000000000042b133 in ngx_event_pipe (p=0x747170, do_write=<value
optimized out>) at src/event/ngx_event_pipe.c:271
271     src/event/ngx_event_pipe.c: No such file or directory.
        in src/event/ngx_event_pipe.c

(gdb) p *p
$3 = {upstream = 0x2b508a1d43d0, downstream = 0x2b508a1d4550,
free_raw_bufs = 0x0, in = 0x6024d0, last_in = 0x7523a0, out = 0x0, free
= 0x602450, busy = 0x0,
  input_filter = 0x47bc30 <ngx_http_fastcgi_input_filter>, input_ctx =
0x69ef00, output_filter = 0x43ca40 <ngx_http_output_filter>, output_ctx
= 0x69ef00, read = 0, cacheable = 0,
  single_buf = 0, free_bufs = 1, upstream_done = 0, upstream_error = 0,
upstream_eof = 0, upstream_blocked = 0, downstream_done = 0,
downstream_error = 0, cyclic_temp_file = 0, allocated = 2,
  bufs = {num = 10, size = 262144}, tag = 0x5c8480, busy_size = 262144,
read_length = 378731, length = 1, max_temp_file_size = 0,
temp_file_write_size = 524288, read_timeout = 180000,
  send_timeout = 60000, send_lowat = 0, pool = 0x69b6e0, log = 0x6c5380,
preread_bufs = 0x0, preread_size = 129547, buf_to_file = 0x0, temp_file
= 0x602358, num = 6}

(gdb) fr 2
#2  0x000000000045410e in ngx_http_upstream_process_upstream
(r=0x69ef00, u=0x746db8) at src/http/ngx_http_upstream.c:2641
2641    src/http/ngx_http_upstream.c: No such file or directory.
        in src/http/ngx_http_upstream.c

(gdb) p *r
$4 = {signature = 1347703880, connection = 0x2b508a1d4550, ctx =
0x69baf0, main_conf = 0x5f7910, srv_conf = 0x617f48, loc_conf =
0x640e00,
  read_event_handler = 0x454510
<ngx_http_upstream_rd_check_broken_connection>, write_event_handler =
0x454b50 <ngx_http_upstream_process_downstream>, cache = 0x0, upstream =
0x746db8,
  upstream_states = 0x747890, pool = 0x69b6e0, header_in = 0x6c5450,
headers_in = {headers = {last = 0x69ef70, part = {elts = 0x69c008, nelts
= 11, next = 0x0}, size = 48, nalloc = 20,
      pool = 0x69b6e0}, host = 0x69c0f8, connection = 0x0,
if_modified_since = 0x0, if_unmodified_since = 0x0, user_agent =
0x69c098, referer = 0x69c038, content_length = 0x0,
    content_type = 0x0, range = 0x0, if_range = 0x0, transfer_encoding =
0x0, expect = 0x0, accept_encoding = 0x69c0c8, via = 0x69c188,
authorization = 0x0, keep_alive = 0x0,
    x_forwarded_for = 0x69c1b8, x_real_ip = 0x0, depth = 0x0,
destination = 0x0, overwrite = 0x0, date = 0x0, user = {len = 0, data =
0x0}, passwd = {len = 0, data = 0x0}, cookies = {
      elts = 0x69c3c8, nelts = 1, size = 8, nalloc = 2, pool =
0x69b6e0}, server = {len = 22, data = 0x74eadb "domain.com"},
content_length_n = -1, keep_alive_n = -1,
    connection_type = 2, msie = 1, msie6 = 0, opera = 0, gecko = 0,
chrome = 0, safari = 0, konqueror = 0}, headers_out = {headers = {last =
0x69f0c8, part = {elts = 0x69b730, nelts = 5,
        next = 0x0}, size = 48, nalloc = 20, pool = 0x69b6e0}, status =
200, status_line = {len = 6, data = 0x4a2662 "200 OK"}, server =
0x69b790, date = 0x0, content_length = 0x0,
    content_encoding = 0x69b7f0, location = 0x0, refresh = 0x0,
last_modified = 0x0, content_range = 0x0, accept_ranges = 0x0,
www_authenticate = 0x0, expires = 0x0, etag = 0x0,
    override_charset = 0x0, content_type_len = 9, content_type = {len =
24, data = 0x747a0d "text/html; charset=utf-8"}, charset = {len = 5,
data = 0x747a20 "utf-8"},
    content_type_lowcase = 0x602078 "text/html", content_type_hash =
101824002410317, cache_control = {elts = 0x6016c8, nelts = 1, size = 8,
nalloc = 2, pool = 0x69b6e0}, content_length_n = -1,
    date_time = 0, last_modified_time = -1}, request_body = 0x747270,
lingering_time = 0, start_sec = 1334650957, start_msec = 91, method = 2,
http_version = 1001, request_line = {len = 50,
    data = 0x74e8c0 "GET /nc/pdf-generation.html?pageID=776404
HTTP/1.1\r\nAccept"}, uri = {len = 10, data = 0x746d62 "/index.php"},
args = {len = 13, data = 0x746a8f "pageID=776404"}, exten = {
    len = 3, data = 0x746d69 "php"}, unparsed_uri = {len = 37, data =
0x74e8c4 "/nc/pdf-generation.html?pageID=776404 HTTP/1.1\r\nAccept"},
method_name = {len = 3,
    data = 0x74e8c0 "GET /nc/pdf-generation.html?pageID=776404
HTTP/1.1\r\nAccept"}, http_protocol = {len = 8, data = 0x74e8ea
"HTTP/1.1\r\nAccept"}, out = 0x7523b8, main = 0x69ef00,
  parent = 0x0, postponed = 0x0, post_subrequest = 0x0, posted_requests
= 0x0, virtual_names = 0x744098, phase_handler = 13, content_handler =
0x47b940 <ngx_http_fastcgi_handler>,
  access_code = 0, variables = 0x69bc98, ncaptures = 0, captures =
0x69c580, captures_data = 0x746a84 "/index.php?pageID=776404",
limit_rate = 0, header_size = 339, request_length = 785,
  err_status = 0, http_connection = 0x6c53d8, log_handler = 0x442e60
<ngx_http_log_error_handler>, cleanup = 0x7478f0, subrequests = 201,
count = 1, blocked = 0, aio = 0, http_state = 2,
  complex_uri = 0, quoted_uri = 0, plus_in_uri = 0, space_in_uri = 0,
invalid_header = 0, add_uri_to_alias = 0, valid_location = 1,
valid_unparsed_uri = 0, uri_changed = 0, uri_changes = 10,
  request_body_in_single_buf = 0, request_body_in_file_only = 0,
request_body_in_persistent_file = 0, request_body_in_clean_file = 0,
request_body_file_group_access = 0,
  request_body_file_log_level = 5, subrequest_in_memory = 0, waited = 0,
cached = 0, gzip_tested = 1, gzip_ok = 1, gzip_vary = 1, proxy = 0,
bypass_cache = 0, no_cache = 0, limit_conn_set = 1,
  limit_req_set = 1, pipeline = 0, plain_http = 0, chunked = 1,
header_only = 0, keepalive = 1, lingering_close = 0, discard_body = 0,
internal = 1, error_page = 0, ignore_content_encoding = 0,
  filter_finalize = 0, post_action = 0, request_complete = 0,
request_output = 1, header_sent = 1, expect_tested = 0, root_tested = 0,
done = 0, logged = 0, buffered = 0,
  main_filter_need_in_memory = 1, filter_need_in_memory = 1,
filter_need_temporary = 0, allow_ranges = 0, stat_reading = 0,
stat_writing = 1, state = 0, header_hash = 6740241951503570450,
  lowcase_index = 13, lowcase_header = "cache-controlsitiont", '\0'
<repeats 11 times>,
  header_name_start = 0x2b50cc8430b3
"^???????u?~??{???^???????u?~??{???^???????u?~??{???^???????u?~??{???^???????u?~??{???^???????u?~??{???^???????????^???????u?~??{???^???????u?~??{???^???????u?~??{???^???????u?~??{???^???????u?~??{???^"...,
  header_name_end = 0x2b50cc8430a8
"??u?~??{???^???????u?~??{???^???????u?~??{???^???????u?~??{???^???????u?~??{???^???????u?~??{???^???????u?~??{???^???????????^???????u?~??{???^???????u?~??{???^???????u?~??{???^???????u?~??{???^??????"...,
  header_start = 0x2b50cc8430aa
"u?~??{???^???????u?~??{???^???????u?~??{???^???????u?~??{???^???????u?~??{???^???????u?~??{???^???????u?~??{???^???????????^???????u?~??{???^???????u?~??{???^???????u?~??{???^???????u?~??{???^???????u"...,
  header_end = 0x2b50cc8430b3
"^???????u?~??{???^???????u?~??{???^???????u?~??{???^???????u?~??{???^???????u?~??{???^???????u?~??{???^???????????^???????u?~??{???^???????u?~??{???^???????u?~??{???^???????u?~??{???^???????u?~??{???^"...,
uri_start = 0x74e8c4 "/nc/pdf-generation.html?pageID=776404
HTTP/1.1\r\nAccept", uri_end = 0x74e8e9 " HTTP/1.1\r\nAccept",
  uri_ext = 0x74e8d7 "html?pageID=776404 HTTP/1.1\r\nAccept", args_start
= 0x74e8dc "pageID=776404 HTTP/1.1\r\nAccept",
  request_start = 0x74e8c0 "GET /nc/pdf-generation.html?pageID=776404
HTTP/1.1\r\nAccept", request_end = 0x74e8f2 "\r\nAccept",
  method_end = 0x74e8c2 "T /nc/pdf-generation.html?pageID=776404
HTTP/1.1\r\nAccept", schema_start = 0x0, schema_end = 0x0, host_start =
0x0, host_end = 0x0, port_start = 0x0, port_end = 0x0,
  http_minor = 1, http_major = 1}

(gdb) p *u
$5 = {read_event_handler = 0x454090
<ngx_http_upstream_process_upstream>, write_event_handler = 0x452980
<ngx_http_upstream_dummy_handler>, peer = {connection = 0x2b508a1d43d0,
    sockaddr = 0x617ea8, socklen = 110, name = 0x68a6b0, tries = 1, get
= 0x487370 <ngx_http_upstream_get_keepalive_peer>, free = 0x4875b0
<ngx_http_upstream_free_keepalive_peer>,
    data = 0x747908, set_session = 0x486e80
<ngx_http_upstream_keepalive_set_session>, save_session = 0x486e90
<ngx_http_upstream_keepalive_save_session>, local = 0x0, rcvbuf = 0,
    log = 0x6c5380, cached = 1, log_error = 1}, pipe = 0x747170,
request_bufs = 0x747880, output = {buf = 0x0, in = 0x0, free = 0x0, busy
= 0x0, sendfile = 0, directio = 0, unaligned = 0,
    need_in_memory = 0, need_in_temp = 0, alignment = 512, pool =
0x69b6e0, allocated = 0, bufs = {num = 1, size = 131072}, tag =
0x5c8480, output_filter = 0x414db0 <ngx_chain_writer>,
    filter_ctx = 0x746eb0}, writer = {out = 0x0, last = 0x746eb0,
connection = 0x2b508a1d43d0, pool = 0x69b6e0, limit = 0}, conf =
0x6416a8, headers_in = {headers = {last = 0x746ee8, part = {
        elts = 0x601520, nelts = 4, next = 0x0}, size = 48, nalloc = 8,
pool = 0x69b6e0}, status_n = 200, status_line = {len = 6, data =
0x4a2662 "200 OK"}, status = 0x0, date = 0x0,
    server = 0x0, connection = 0x0, expires = 0x0, etag = 0x0,
x_accel_expires = 0x0, x_accel_redirect = 0x0, x_accel_limit_rate = 0x0,
content_type = 0x601580, content_length = 0x601520,
    last_modified = 0x0, location = 0x0, accept_ranges = 0x0,
www_authenticate = 0x0, transfer_encoding = 0x0, content_encoding = 0x0,
content_length_n = 355230, cache_control = {
      elts = 0x747a38, nelts = 1, size = 8, nalloc = 2, pool =
0x69b6e0}, connection_close = 0, chunked = 0}, resolved = 0x0, buffer =
{
    pos = 0x2b50cc843010
"?u?~??{???^???????u?~??{???^???????u?~??{???^???????u?~??{???^???????u?~??{???^???????u?~??{???^???????u?~??{???^???????u?~??{???^???????u?~??{???^???????u?~??{???^???????u?~??{???^???????u?~??{???^??"...,
    last = 0x2b50cc843010
"?u?~??{???^???????u?~??{???^???????u?~??{???^???????u?~??{???^???????u?~??{???^???????u?~??{???^???????u?~??{???^???????u?~??{???^???????u?~??{???^???????u?~??{???^???????u?~??{???^???????u?~??{???^??"...,
file_pos = 0, file_last = 0,
    start = 0x2b50cc843010
"?u?~??{???^???????u?~??{???^???????u?~??{???^???????u?~??{???^???????u?~??{???^???????u?~??{???^???????u?~??{???^???????u?~??{???^???????u?~??{???^???????u?~??{???^???????u?~??{???^???????u?~??{???^??"...,
end = 0x2b50cc883010 "", tag = 0x5c8480, file = 0x0, shadow = 0x0,
temporary = 1, memory = 0, mmap = 0, recycled = 1, in_file = 0, flush =
0, sync = 0,
    last_buf = 0, last_in_chain = 0, last_shadow = 0, temp_file = 0, num
= 2}, length = 355230, out_bufs = 0x0, busy_bufs = 0x0, free_bufs =
0x0,
  input_filter_init = 0x47af80 <ngx_http_fastcgi_input_filter_init>,
input_filter = 0, input_filter_ctx = 0x0, create_key = 0x47c100
<ngx_http_fastcgi_create_key>,
  create_request = 0x47c3a0 <ngx_http_fastcgi_create_request>,
reinit_request = 0x47af50 <ngx_http_fastcgi_reinit_request>,
process_header = 0x47cf20 <ngx_http_fastcgi_process_header>,
  abort_request = 0x47b560 <ngx_http_fastcgi_abort_request>,
finalize_request = 0x47b530 <ngx_http_fastcgi_finalize_request>,
rewrite_redirect = 0, rewrite_cookie = 0, timeout = 0,
  state = 0x7478b8, method = {len = 0, data = 0x0}, schema = {len = 10,
data = 0x4a8c69 "fastcgi://"}, uri = {len = 0, data = 0x0}, cleanup =
0x7478f0, store = 0, cacheable = 0, accel = 0,
  ssl = 0, cache_status = 0, buffering = 1, keepalive = 0, request_sent
= 1, header_sent = 1}

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225420,225457#msg-225457


From 191919 at gmail.com  Wed Apr 18 06:07:00 2012
From: 191919 at gmail.com (191919)
Date: Wed, 18 Apr 2012 14:07:00 +0800
Subject: MP4 pseudostreaming - seek delay
In-Reply-To: <d30c68eab7e112b3fd2ca0511801ccc2.NginxMailingListEnglish@forum.nginx.org>
References: <CACDr0vFAzWMy6=cZTakZkpY6FN_mejb--yYn2LhnRAyj=jWRRQ@mail.gmail.com>
	<d30c68eab7e112b3fd2ca0511801ccc2.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <1ABA1FA6E6BE4F14B6F82D97B29B6C5D@gmail.com>

You can split the big movie file into small pieces, thus you have smaller index for each small movie file.

In the player, when current part is about to end, load next part, when current part is over, play the next part. So the movie is played seamlessly, you save network traffics, and your users feel better. You need to modify your movie player, but it is not a very hard work.

Be caution: you must split your movie file at offsets of key-frame.  

--  
191919


Le mercredi, avril 18, 2012 ? 9:23 AM, ragivan a ?crit :

> That was back in 2008, they didn't have 2 hour long videos. But what I
> was talking about what regarding the technology, lighttpd and nginx both
> do pseudo streaming. The moov atom is about 5 MB so it takes about 10
> seconds to download that before playing the video. Is there some
> technique to start playing the video after partial download of the moov
> atom? Or is it possible to cache the moov atom into memory for faster
> retrieval?
>  
> Posted at Nginx Forum: http://forum.nginx.org/read.php?2,223375,225453#msg-225453
>  
> _______________________________________________
> nginx mailing list
> nginx at nginx.org (mailto:nginx at nginx.org)
> http://mailman.nginx.org/mailman/listinfo/nginx




From nginx-forum at nginx.us  Wed Apr 18 06:29:22 2012
From: nginx-forum at nginx.us (ragivan)
Date: Wed, 18 Apr 2012 02:29:22 -0400 (EDT)
Subject: MP4 pseudostreaming - seek delay
In-Reply-To: <1ABA1FA6E6BE4F14B6F82D97B29B6C5D@gmail.com>
References: <1ABA1FA6E6BE4F14B6F82D97B29B6C5D@gmail.com>
Message-ID: <470a16e554fd989a5f1301211d1e6d91.NginxMailingListEnglish@forum.nginx.org>

Do you know a tool for doing this? Something that will work on a unix
system. Also, what if the user wants to jump to the end of the movie? I
suppose that won't be possible as the user must click to the end of each
segment until reaching the last segment.

191919 Wrote:
-------------------------------------------------------
> You can split the big movie file into small
> pieces, thus you have smaller index for each small
> movie file.
> 
> In the player, when current part is about to end,
> load next part, when current part is over, play
> the next part. So the movie is played seamlessly,
> you save network traffics, and your users feel
> better. You need to modify your movie player, but
> it is not a very hard work.
> 
> Be caution: you must split your movie file at
> offsets of key-frame.  
> 
> --  
> 191919
> 
> 
> Le mercredi, avril 18, 2012 ? 9:23 AM, ragivan a
> ?crit :
> 
> > That was back in 2008, they didn't have 2 hour
> long videos. But what I
> > was talking about what regarding the technology,
> lighttpd and nginx both
> > do pseudo streaming. The moov atom is about 5 MB
> so it takes about 10
> > seconds to download that before playing the
> video. Is there some
> > technique to start playing the video after
> partial download of the moov
> > atom? Or is it possible to cache the moov atom
> into memory for faster
> > retrieval?
> >  
> > Posted at Nginx Forum:
> http://forum.nginx.org/read.php?2,223375,225453#ms
> g-225453
> >  
> > _______________________________________________
> > nginx mailing list
> > nginx at nginx.org (mailto:nginx at nginx.org)
> > http://mailman.nginx.org/mailman/listinfo/nginx
> 
> 
> 
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,223375,225459#msg-225459


From igor at sysoev.ru  Wed Apr 18 06:38:12 2012
From: igor at sysoev.ru (Igor Sysoev)
Date: Wed, 18 Apr 2012 10:38:12 +0400
Subject: MP4 pseudostreaming - seek delay
In-Reply-To: <d30c68eab7e112b3fd2ca0511801ccc2.NginxMailingListEnglish@forum.nginx.org>
References: <CACDr0vFAzWMy6=cZTakZkpY6FN_mejb--yYn2LhnRAyj=jWRRQ@mail.gmail.com>
	<d30c68eab7e112b3fd2ca0511801ccc2.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20120418063812.GA25571@nginx.com>

On Tue, Apr 17, 2012 at 09:23:13PM -0400, ragivan wrote:
> That was back in 2008, they didn't have 2 hour long videos. But what I
> was talking about what regarding the technology, lighttpd and nginx both
> do pseudo streaming. The moov atom is about 5 MB so it takes about 10
> seconds to download that before playing the video. Is there some
> technique to start playing the video after partial download of the moov
> atom? Or is it possible to cache the moov atom into memory for faster
> retrieval?

I've just tried to watch 3 YouTube movies in Firefox 11 with Flash installed
and discovered using LiveHTTPHeaders that these movies are in FLV format.
Using this format player is smart enough to request just a file offset
but not movie time and it does not need movie metadata such as MOOV atom.
It need just a small static 13-bytes FLV header:
http://nginx.org/en/docs/http/ngx_http_flv_module.html


-- 
Igor Sysoev


From nginx-forum at nginx.us  Wed Apr 18 07:40:22 2012
From: nginx-forum at nginx.us (dagr)
Date: Wed, 18 Apr 2012 03:40:22 -0400 (EDT)
Subject: MP4 pseudostreaming - seek delay
In-Reply-To: <20120418063812.GA25571@nginx.com>
References: <20120418063812.GA25571@nginx.com>
Message-ID: <fc1db71f43a2ba943ab96d065ca2d189.NginxMailingListEnglish@forum.nginx.org>

If you try to download youtube movie (any movie which i tried so far)
with Download Master - then if you download 720p or higher,  you ll get 
MP4 file , not FLV. Download Master  emulates flash player to retrieve
video , so it looks like flash player in resolutions 720p 1080p plays
mp4 (not flv) too (may be i wrong). 

AFAIK youtube stores video under 720p as flv and >=720p as mp4 and
another batch of mp4 files  for ipads  which are encoded with main
profile, not high.

By the way, i was not able to find on youtube 720p  MP4 files longer
than 20 mins ( moov data is not huge), so their smooth playing doesnt
look very strange.


Anyway  - ipads  CAN seek smoothly mp4 files without all this shit with
moov data. 
And Adobe  flash player - doesnt - very strange.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,223375,225461#msg-225461


From 191919 at gmail.com  Wed Apr 18 08:27:42 2012
From: 191919 at gmail.com (191919)
Date: Wed, 18 Apr 2012 16:27:42 +0800
Subject: MP4 pseudostreaming - seek delay
In-Reply-To: <470a16e554fd989a5f1301211d1e6d91.NginxMailingListEnglish@forum.nginx.org>
References: <1ABA1FA6E6BE4F14B6F82D97B29B6C5D@gmail.com>
	<470a16e554fd989a5f1301211d1e6d91.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <467B949137834116A8EE2B666E00D3FD@gmail.com>

I wrote such a tool which generates mp4 file-series, it is simple if you are familiar with libgpac.  

You have to keep an array which maps time to file indexes, when a user wants to seek to a specific position, just convert the timestamp (e.g. 01:15:14) to file index (movie_2314_segment_519.mp4) then request it from the server.

You can also use FLVs which contain H264 and AAC if Flash Player is your only client.  

--  
191919


Le mercredi, avril 18, 2012 ? 2:29 PM, ragivan a ?crit :

> Do you know a tool for doing this? Something that will work on a unix
> system. Also, what if the user wants to jump to the end of the movie? I
> suppose that won't be possible as the user must click to the end of each
> segment until reaching the last segment.
>  
> 191919 Wrote:
> -------------------------------------------------------
> > You can split the big movie file into small
> > pieces, thus you have smaller index for each small
> > movie file.
> >  
> > In the player, when current part is about to end,
> > load next part, when current part is over, play
> > the next part. So the movie is played seamlessly,
> > you save network traffics, and your users feel
> > better. You need to modify your movie player, but
> > it is not a very hard work.
> >  
> > Be caution: you must split your movie file at
> > offsets of key-frame.  
> >  
> > --  
> > 191919
> >  
> >  
> > Le mercredi, avril 18, 2012 ? 9:23 AM, ragivan a
> > ?crit :
> >  
> > > That was back in 2008, they didn't have 2 hour
> > long videos. But what I
> > > was talking about what regarding the technology,
> >  
> >  
> > lighttpd and nginx both
> > > do pseudo streaming. The moov atom is about 5 MB
> >  
> >  
> > so it takes about 10
> > > seconds to download that before playing the
> >  
> >  
> > video. Is there some
> > > technique to start playing the video after
> >  
> >  
> > partial download of the moov
> > > atom? Or is it possible to cache the moov atom
> >  
> >  
> > into memory for faster
> > > retrieval?
> > >  
> > > Posted at Nginx Forum:
> >  
> > http://forum.nginx.org/read.php?2,223375,225453#ms
> > g-225453
> > >  
> > > _______________________________________________
> > > nginx mailing list
> > > nginx at nginx.org (mailto:nginx at nginx.org)
> > > http://mailman.nginx.org/mailman/listinfo/nginx
> >  
> >  
> >  
> >  
> >  
> > _______________________________________________
> > nginx mailing list
> > nginx at nginx.org (mailto:nginx at nginx.org)
> > http://mailman.nginx.org/mailman/listinfo/nginx
>  
>  
>  
> Posted at Nginx Forum: http://forum.nginx.org/read.php?2,223375,225459#msg-225459
>  
> _______________________________________________
> nginx mailing list
> nginx at nginx.org (mailto:nginx at nginx.org)
> http://mailman.nginx.org/mailman/listinfo/nginx




From luky-37 at hotmail.com  Wed Apr 18 09:00:53 2012
From: luky-37 at hotmail.com (Lukas Tribus)
Date: Wed, 18 Apr 2012 11:00:53 +0200
Subject: MP4 pseudostreaming - seek delay
In-Reply-To: <fc1db71f43a2ba943ab96d065ca2d189.NginxMailingListEnglish@forum.nginx.org>
References: <20120418063812.GA25571@nginx.com>,
	<fc1db71f43a2ba943ab96d065ca2d189.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <DUB107-W14DB8E222CE590BE6E14D0ED3C0@phx.gbl>


> Anyway - ipads CAN seek smoothly mp4 files without all this shit with> moov data.
Ipads don't use flash with pseudostreaming, but html5/native apps,which probably use HTTP Range requests instead.
If you embed MP4 content with html5 and play it in chrome or safari(they have html5 with h264 support), you will probably see similarperformance.
 		 	   		  

From nginx-forum at nginx.us  Wed Apr 18 10:09:51 2012
From: nginx-forum at nginx.us (VRider)
Date: Wed, 18 Apr 2012 06:09:51 -0400 (EDT)
Subject: nginx serving old files from cache
Message-ID: <59ab839ec58397d3a8ea72be3ad90408.NginxMailingListEnglish@forum.nginx.org>

I have nginx set as reverse proxy for Apache and for serving static
files.
Everything works great, except in the last few weeks I noticed that
sometimes nginx returns page from cache which is 3-4 hours old. I
checked the timestamp of the cache file, and it is old.

This is part of the config file related to caching in nginx.conf:
    proxy_cache off;
    proxy_buffering on;
    proxy_buffer_size 8k;
    proxy_buffers 100 8k;
    proxy_cache_path  /var/www/cache levels=1:2 keys_zone=my-cache:8m
max_size=1000m;
    proxy_temp_path /var/www/cache/tmp;
    proxy_cache_key "$host$request_uri";
    proxy_ignore_headers "Expires" "Cache-Control" "X-Accel-Expires";

and this is for particulat vhost:
            proxy_cache my-cache;
            proxy_cache_valid  200 404     1m; # <----- should be valid
for just one minute!
            proxy_ignore_headers Expires;
            proxy_ignore_headers Cache-Control;
            proxy_ignore_headers X-Accel-Expires;
            proxy_ignore_headers Set-Cookie;
            proxy_hide_header Set-Cookie;

Any ideas?

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225467,225467#msg-225467


From mdounin at mdounin.ru  Wed Apr 18 11:03:28 2012
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Wed, 18 Apr 2012 15:03:28 +0400
Subject: worker process exited on signal 11
In-Reply-To: <02299bd23d1683f18815cac8d2f70f37.NginxMailingListEnglish@forum.nginx.org>
References: <2bd9c9c86f68a802bb9544d26ad045bb.NginxMailingListEnglish@forum.nginx.org>
	<02299bd23d1683f18815cac8d2f70f37.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20120418110327.GG13466@mdounin.ru>

Hello!

On Wed, Apr 18, 2012 at 01:06:25AM -0400, mscheel wrote:

> Hello Maxim,
> 
> it appears only on two productive servers. Compiling without these
> modules is not possible at the moment but i've tried several times with
> each module deactivated (and without any of these three).
> 
> Full config is huge. I'm using a php-fpm backend with keepalive on and
> one upstream server with nginx and php-fpm.

[...]

> 	upstream be_server_nginx {
> 		server IP:80;
> 		keepalive 300;
> 	}
> 	}
> 	
> 	...
> 	
> 	location @be_nginx {
> 	proxy_pass http://be_server_nginx;
> 	proxy_set_header Connection "";	proxy_http_version 1.1;
> 	include /etc/nginx/proxy.conf;

This part of the config shows proxy_pass, while gdb prints 
provided say the request used fastcgi_pass.  Could you please:

1. Check processing flow of the 
"/nc/pdf-generation.html?pageID=776404" uri and provide relevant 
parts of the config.  According to gdb it's transformed to 
"/index.php?pageID=776404".

2. Check if the problem is reproduceable with the uri in question.

3. Check if the problem persists with the upstream keepalive 
turned off.

It would be also cool to obtain debug log if possible, it should 
be ok to grab one from a particular location with fastcgi_pass.

And see below from more gdb commands I would like to see output 
from.

> output of gdb:
> 
> (gdb) fr 0
> #0  0x000000000042cbe8 in ngx_readv_chain (c=0x2b508a1d43d0,
> chain=0x602348) at src/os/unix/ngx_readv_chain.c:195
> 195     in src/os/unix/ngx_readv_chain.c
> 
> (gdb) p chain
> $1 = (ngx_chain_t *) 0x602348
> 
> (gdb) p *chain
> $2 = {buf = 0x602460, next = 0x752628}

fr 0
p *chain->buf
p *chain->next
p *chain->next->buf
p *chain->next->next
p *chain->next->next->buf

(repeat last two prints with one more "->next" added till it fails)

> (gdb) fr 1
> #1  0x000000000042b133 in ngx_event_pipe (p=0x747170, do_write=<value
> optimized out>) at src/event/ngx_event_pipe.c:271
> 271     src/event/ngx_event_pipe.c: No such file or directory.
>         in src/event/ngx_event_pipe.c
> 
> (gdb) p *p
> $3 = {upstream = 0x2b508a1d43d0, downstream = 0x2b508a1d4550,
> free_raw_bufs = 0x0, in = 0x6024d0, last_in = 0x7523a0, out = 0x0, free
> = 0x602450, busy = 0x0,

fr 1
p *p->in
p *p->in->buf
p *p->in->next
p *p->in->next->buf

(same here, repeat with "->next" added till it fails)

p *p->free
p *p->free->buf
p *p->free->next
p *p->free->next->buf

(same here)

[...]

Maxim Dounin


From nginx-forum at nginx.us  Wed Apr 18 11:13:02 2012
From: nginx-forum at nginx.us (mscheel)
Date: Wed, 18 Apr 2012 07:13:02 -0400 (EDT)
Subject: worker process exited on signal 11
In-Reply-To: <20120418110327.GG13466@mdounin.ru>
References: <20120418110327.GG13466@mdounin.ru>
Message-ID: <c86ed64be6d186d9009f579f18c8374a.NginxMailingListEnglish@forum.nginx.org>

Hello Maxim, thanks alot for your help. I'll try to reproduce it.

(gdb) fr 0
#0  0x000000000042cbe8 in ngx_readv_chain (c=0x2b508a1d43d0,
chain=0x602348) at src/os/unix/ngx_readv_chain.c:195
195     src/os/unix/ngx_readv_chain.c: No such file or directory.
        in src/os/unix/ngx_readv_chain.c
		
(gdb) p *chain->buf
$6 = {pos = 0x0, last = 0x0, file_pos = 0, file_last = 0, start = 0x0,
end = 0x0, tag = 0x5c8480, file = 0x0, shadow = 0x0, temporary = 1,
memory = 0, mmap = 0, recycled = 1, in_file = 0,
  flush = 0, sync = 0, last_buf = 0, last_in_chain = 0, last_shadow = 0,
temp_file = 0, num = 0}

(gdb) p *chain->next
$7 = {buf = 0x7523d8, next = 0x752638}

(gdb) p *chain->next->buf
$8 = {pos = 0x0, last = 0x0, file_pos = 0, file_last = 0, start = 0x0,
end = 0x0, tag = 0x5c8480, file = 0x0, shadow = 0x0, temporary = 1,
memory = 0, mmap = 0, recycled = 1, in_file = 0,
  flush = 0, sync = 0, last_buf = 0, last_in_chain = 0, last_shadow = 0,
temp_file = 0, num = 0}
  
(gdb) p *chain->next->next
$9 = {buf = 0x746ff8, next = 0x0}

(gdb) p *chain->next->next->buf
$10 = {
  pos = 0x2b50cc843010
"?u?~??{???^???????u?~??{???^???????u?~??{???^???????u?~??{???^???????u?~??{???^???????u?~??{???^???????u?~??{???^???????u?~??{???^???????u?~??{???^???????u?~??{???^???????u?~??{???^???????u?~??{???^??"...,
  last = 0x2b50cc843010
"?u?~??{???^???????u?~??{???^???????u?~??{???^???????u?~??{???^???????u?~??{???^???????u?~??{???^???????u?~??{???^???????u?~??{???^???????u?~??{???^???????u?~??{???^???????u?~??{???^???????u?~??{???^??"...,
file_pos = 0, file_last = 0,
  start = 0x2b50cc843010
"?u?~??{???^???????u?~??{???^???????u?~??{???^???????u?~??{???^???????u?~??{???^???????u?~??{???^???????u?~??{???^???????u?~??{???^???????u?~??{???^???????u?~??{???^???????u?~??{???^???????u?~??{???^??"...,
end = 0x2b50cc883010 "", tag = 0x5c8480, file = 0x0, shadow = 0x0,
temporary = 1, memory = 0, mmap = 0, recycled = 1, in_file = 0, flush =
0, sync = 0,
  last_buf = 0, last_in_chain = 0, last_shadow = 0, temp_file = 0, num =
2}

(gdb) p *chain->next->next->next
Cannot access memory at address 0x0
(gdb) p *chain->next->next->next->buf
Cannot access memory at address 0x0


____________________________________________

(gdb) fr 1
#1  0x000000000042b133 in ngx_event_pipe (p=0x747170, do_write=<value
optimized out>) at src/event/ngx_event_pipe.c:271
271     src/event/ngx_event_pipe.c: No such file or directory.
        in src/event/ngx_event_pipe.c
(gdb) p *p->in
$11 = {buf = 0x7524d8, next = 0x7523a8}
(gdb) p *p->in->buf
$12 = {pos = 0x2b50cc908010 "?:??>G?\223?", last = 0x2b50cc918008
"\001\006", file_pos = 0, file_last = 0, start = 0x2b50cc908010
"?:??>G?\223?", end = 0x2b50cc948010 "", tag = 0x5c8480,
  file = 0x0, shadow = 0x752528, temporary = 1, memory = 0, mmap = 0,
recycled = 1, in_file = 0, flush = 0, sync = 0, last_buf = 0,
last_in_chain = 0, last_shadow = 0, temp_file = 0, num = 4}
(gdb) p *p->in->next
$13 = {buf = 0x752528, next = 0x752398}
(gdb) p *p->in->next->buf
$14 = {pos = 0x2b50cc918010 "t?|=\036o?G_?\227\027???U)?", last =
0x2b50cc91be90 "", file_pos = 0, file_last = 0, start = 0x2b50cc908010
"?:??>G?\223?", end = 0x2b50cc948010 "", tag = 0x5c8480,
  file = 0x0, shadow = 0x752478, temporary = 1, memory = 0, mmap = 0,
recycled = 1, in_file = 0, flush = 0, sync = 0, last_buf = 0,
last_in_chain = 0, last_shadow = 1, temp_file = 0, num = 4}
(gdb) p *p->in->next->next
$15 = {buf = 0x7525d8, next = 0x0}
(gdb) p *p->in->next->next->buf
$16 = {
  pos = 0x2b50cc949010
"g??d?z\211?\017\027??\026\225h^3?\217?}\221&NG\230N\033??.\207?\222?=R\215??\037\224|??N\224?\206?\205)J\003%dY\2348\235?5[UnJ+??6&?p?z.K\203?ZG?YJv?\025?bZM?\227,\016?\236'YrKFoqhN\223\022`L\213z{?i,\226?\b\t)\033\217?\225V?\005*??W?^>?U??$??",
last = 0x2b50cc9518f0 "", file_pos = 0, file_last = 0,
  start = 0x2b50cc949010
"g??d?z\211?\017\027??\026\225h^3?\217?}\221&NG\230N\033??.\207?\222?=R\215??\037\224|??N\224?\206?\205)J\003%dY\2348\235?5[UnJ+??6&?p?z.K\203?ZG?YJv?\025?bZM?\227,\016?\236'YrKFoqhN\223\022`L\213z{?i,\226?\b\t)\033\217?\225V?\005*??W?^>?U??$??",
end = 0x2b50cc989010 "", tag = 0x5c8480, file = 0x0, shadow = 0x752578,
temporary = 1, memory = 0, mmap = 0,
  recycled = 1, in_file = 0, flush = 0, sync = 0, last_buf = 0,
last_in_chain = 0, last_shadow = 1, temp_file = 0, num = 5}
(gdb) p *p->in->next->next->next
Cannot access memory at address 0x0

_______________________________________________


(gdb) p *p->free
$17 = {buf = 0x752428, next = 0x752290}
(gdb) p *p->free->buf
$18 = {
  pos = 0x2b50cc843010
"?u?~??{???^???????u?~??{???^???????u?~??{???^???????u?~??{???^???????u?~??{???^???????u?~??{???^???????u?~??{???^???????u?~??{???^???????u?~??{???^???????u?~??{???^???????u?~??{???^???????u?~??{???^??"...,
  last = 0x2b50cc843010
"?u?~??{???^???????u?~??{???^???????u?~??{???^???????u?~??{???^???????u?~??{???^???????u?~??{???^???????u?~??{???^???????u?~??{???^???????u?~??{???^???????u?~??{???^???????u?~??{???^???????u?~??{???^??"...,
file_pos = 0, file_last = 0,
  start = 0x2b50cc843010
"?u?~??{???^???????u?~??{???^???????u?~??{???^???????u?~??{???^???????u?~??{???^???????u?~??{???^???????u?~??{???^???????u?~??{???^???????u?~??{???^???????u?~??{???^???????u?~??{???^???????u?~??{???^??"...,
end = 0x2b50cc883010 "", tag = 0x5c8480, file = 0x0, shadow = 0x0,
temporary = 1, memory = 0, mmap = 0, recycled = 1, in_file = 0, flush =
0, sync = 0,
  last_buf = 0, last_in_chain = 0, last_shadow = 0, temp_file = 0, num =
2}
(gdb) p *p->free->next
$19 = {buf = 0x7523d8, next = 0x6024b0}
(gdb) p *p->free->next->buf
$20 = {pos = 0x0, last = 0x0, file_pos = 0, file_last = 0, start = 0x0,
end = 0x0, tag = 0x5c8480, file = 0x0, shadow = 0x0, temporary = 1,
memory = 0, mmap = 0, recycled = 1, in_file = 0,
  flush = 0, sync = 0, last_buf = 0, last_in_chain = 0, last_shadow = 0,
temp_file = 0, num = 0}
(gdb)
$21 = {pos = 0x0, last = 0x0, file_pos = 0, file_last = 0, start = 0x0,
end = 0x0, tag = 0x5c8480, file = 0x0, shadow = 0x0, temporary = 1,
memory = 0, mmap = 0, recycled = 1, in_file = 0,
  flush = 0, sync = 0, last_buf = 0, last_in_chain = 0, last_shadow = 0,
temp_file = 0, num = 0}
(gdb) p *p->free->next->next
$22 = {buf = 0x602460, next = 0x6024c0}
(gdb) p *p->free->next->next->buf
$23 = {pos = 0x0, last = 0x0, file_pos = 0, file_last = 0, start = 0x0,
end = 0x0, tag = 0x5c8480, file = 0x0, shadow = 0x0, temporary = 1,
memory = 0, mmap = 0, recycled = 1, in_file = 0,
  flush = 0, sync = 0, last_buf = 0, last_in_chain = 0, last_shadow = 0,
temp_file = 0, num = 0}
(gdb) p *p->free->next->next->next
$24 = {buf = 0x7510f0, next = 0x0}
(gdb) p *p->free->next->next->next->buf
$25 = {pos = 0x0, last = 0x0, file_pos = 0, file_last = 0, start = 0x0,
end = 0x0, tag = 0x5c8480, file = 0x0, shadow = 0x0, temporary = 1,
memory = 0, mmap = 0, recycled = 1, in_file = 0,
  flush = 0, sync = 0, last_buf = 0, last_in_chain = 0, last_shadow = 0,
temp_file = 0, num = 0}
(gdb) p *p->free->next->next->next->next
Cannot access memory at address 0x0
(gdb) p *p->free->next->next->next->next->buf
Cannot access memory at address 0x0

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225420,225474#msg-225474


From nginx-forum at nginx.us  Wed Apr 18 11:28:46 2012
From: nginx-forum at nginx.us (mscheel)
Date: Wed, 18 Apr 2012 07:28:46 -0400 (EDT)
Subject: worker process exited on signal 11
In-Reply-To: <c86ed64be6d186d9009f579f18c8374a.NginxMailingListEnglish@forum.nginx.org>
References: <20120418110327.GG13466@mdounin.ru>
	<c86ed64be6d186d9009f579f18c8374a.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <59a2b34633388aa6230a49eb0380adc3.NginxMailingListEnglish@forum.nginx.org>

Interesting, i'm unable to reproduce it with the uri. The uri generates
a pdf file (when it does not exist on the fs). I've deleted it on the fs
and it's created fine by our php application. I've also made some tests
with keepalive off, no change.
Maybe there is something missing in SLES 10 which is required by 1.1? A
manual configure and make (without install) with my nginx -V runs
without errors.

My fastcgi_params:
	set $php_root /docroot;
    root $php_root;
	try_files $uri $uri/ @404;
    fastcgi_pass php-fpm_backend;
	fastcgi_keep_conn on;
    fastcgi_index  index.php;
    fastcgi_param  SCRIPT_FILENAME  $php_root$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  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  REMOTE_ADDR	$HTTP_X_FORWARDED_FOR;
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  HOST_LSG           $http_host;
fastcgi_param  HTTP_REFERER       $http_referer;
fastcgi_param HTTPS $https_ms;

fastcgi_pass_header HTTP_HOST;
fastcgi_pass_header SERVER_NAME;
fastcgi_pass_header HOST;
fastcgi_pass_header SERVER_NAME;
fastcgi_pass_header X-REAL-IP;
fastcgi_pass_header X-FORWARDED-PROTO;
fastcgi_pass_header X-FORWARDED-FOR;

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225420,225475#msg-225475


From nginx-forum at nginx.us  Wed Apr 18 13:26:06 2012
From: nginx-forum at nginx.us (dagr)
Date: Wed, 18 Apr 2012 09:26:06 -0400 (EDT)
Subject: MP4 pseudostreaming - seek delay
In-Reply-To: <DUB107-W14DB8E222CE590BE6E14D0ED3C0@phx.gbl>
References: <DUB107-W14DB8E222CE590BE6E14D0ED3C0@phx.gbl>
Message-ID: <b5000a4eebd3d4428273eca7fc077ba7.NginxMailingListEnglish@forum.nginx.org>

> Ipads don't use flash with pseudostreaming, but html5/native apps

i know, i actually  dont understand why flash player cant do the same

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,223375,225476#msg-225476


From brian at akins.org  Wed Apr 18 14:02:08 2012
From: brian at akins.org (Brian Akins)
Date: Wed, 18 Apr 2012 10:02:08 -0400
Subject: nginx serving old files from cache
In-Reply-To: <59ab839ec58397d3a8ea72be3ad90408.NginxMailingListEnglish@forum.nginx.org>
References: <59ab839ec58397d3a8ea72be3ad90408.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <70228B38-8E8E-4D98-9D0B-7D6A42809963@akins.org>

Do you have proxy_cache_use_stale set?  Has any nginx process segfaulted?



From mdounin at mdounin.ru  Wed Apr 18 14:22:13 2012
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Wed, 18 Apr 2012 18:22:13 +0400
Subject: worker process exited on signal 11
In-Reply-To: <c86ed64be6d186d9009f579f18c8374a.NginxMailingListEnglish@forum.nginx.org>
References: <20120418110327.GG13466@mdounin.ru>
	<c86ed64be6d186d9009f579f18c8374a.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20120418142213.GI13466@mdounin.ru>

Hello!

On Wed, Apr 18, 2012 at 07:13:02AM -0400, mscheel wrote:

> Hello Maxim, thanks alot for your help. I'll try to reproduce it.
> 
> (gdb) fr 0
> #0  0x000000000042cbe8 in ngx_readv_chain (c=0x2b508a1d43d0,
> chain=0x602348) at src/os/unix/ngx_readv_chain.c:195
> 195     src/os/unix/ngx_readv_chain.c: No such file or directory.
>         in src/os/unix/ngx_readv_chain.c
> 		
> (gdb) p *chain->buf
> $6 = {pos = 0x0, last = 0x0, file_pos = 0, file_last = 0, start = 0x0,
> end = 0x0, tag = 0x5c8480, file = 0x0, shadow = 0x0, temporary = 1,
> memory = 0, mmap = 0, recycled = 1, in_file = 0,
>   flush = 0, sync = 0, last_buf = 0, last_in_chain = 0, last_shadow = 0,
> temp_file = 0, num = 0}
> 
> (gdb) p *chain->next
> $7 = {buf = 0x7523d8, next = 0x752638}
> 
> (gdb) p *chain->next->buf
> $8 = {pos = 0x0, last = 0x0, file_pos = 0, file_last = 0, start = 0x0,
> end = 0x0, tag = 0x5c8480, file = 0x0, shadow = 0x0, temporary = 1,
> memory = 0, mmap = 0, recycled = 1, in_file = 0,
>   flush = 0, sync = 0, last_buf = 0, last_in_chain = 0, last_shadow = 0,
> temp_file = 0, num = 0}
>   
> (gdb) p *chain->next->next
> $9 = {buf = 0x746ff8, next = 0x0}

[...]

Ok, thanks.  The buffer chain looks completely screwed up, and I 
suspect the culprit is substitutions4nginx module you use.  Quick 
look over it's sources suggests it at least changes the 
last_shadow flag of the buffers it gets, and this may cause 
undefined behaviour.

Additionally, the response in question has Content-Type set to 
"text/html" (though it looks like the actual response is pdf), which 
is handled by substitutions4nginx by default.

Could you please try to reproduce the problem without the module 
in question?

Maxim Dounin


From nginx-forum at nginx.us  Wed Apr 18 14:50:43 2012
From: nginx-forum at nginx.us (VRider)
Date: Wed, 18 Apr 2012 10:50:43 -0400 (EDT)
Subject: nginx serving old files from cache
In-Reply-To: <70228B38-8E8E-4D98-9D0B-7D6A42809963@akins.org>
References: <70228B38-8E8E-4D98-9D0B-7D6A42809963@akins.org>
Message-ID: <0cae86110bc8dbefacc2eb981091d4ef.NginxMailingListEnglish@forum.nginx.org>

No segfaults and proxy_cache_use_stale wasn't specified, so by default
it was off.

D

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225467,225482#msg-225482


From nginx-forum at nginx.us  Wed Apr 18 14:53:55 2012
From: nginx-forum at nginx.us (rob)
Date: Wed, 18 Apr 2012 10:53:55 -0400 (EDT)
Subject: How to optimize these config / rewrite rules
Message-ID: <966202bd319a1ca94d70368f6b70dbae.NginxMailingListEnglish@forum.nginx.org>

Hi there,

I'm running nginx and xenforo but need to rewrite some old legacy URL's
(left over from vbulletin + vbseo)..

Here is just a very small selection of my config which I want to
optimise as fully as possible:-
[code]
#Forum 26 main
location = /category-name-73/forum-name-26/  {
    rewrite ^ /forums/forum-name.26/ permanent;
}
 
#Forum 26
location /category-name-73/forum-name-26/ {
    #forum paging
    rewrite ^/category-name-73/forum-name-26/index([\d]+).html
/forums/forum-name.26/page-$1 permanent;
 
    #thread paging (modified version of kier's redirect script)
    rewrite [^/]+/([\d]+)-.+-([\d]+).html /showthread.php?t=$1&page=$2
last;
 
    #thread
    rewrite [^/]+/([\d]+)-.+.html /threads/redirectme.$1/ permanent;
}[/code]

If only I can optimise the above then I can transfer the same principle
to the rest of my (huge) config file.

I've been told that its possible to optimise the above so that a visit
to /category-name-73/forum-name-26/ or any path below it will run just
one regex instead of potentially 4 as per the above code...

So, optimising the above... does anyone have any clue how to do this?

Thank you for any help offered,

Rob

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225483,225483#msg-225483


From nginx-forum at nginx.us  Wed Apr 18 14:54:54 2012
From: nginx-forum at nginx.us (rob)
Date: Wed, 18 Apr 2012 10:54:54 -0400 (EDT)
Subject: How to optimize these config / rewrite rules
In-Reply-To: <966202bd319a1ca94d70368f6b70dbae.NginxMailingListEnglish@forum.nginx.org>
References: <966202bd319a1ca94d70368f6b70dbae.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <6cbee914ac9b02e540c1574c3fac6e48.NginxMailingListEnglish@forum.nginx.org>

bah... please ignore the [code][/code] tags above - assumed this forum
software used bbcode.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225483,225484#msg-225484


From nginx-forum at nginx.us  Wed Apr 18 15:12:11 2012
From: nginx-forum at nginx.us (mscheel)
Date: Wed, 18 Apr 2012 11:12:11 -0400 (EDT)
Subject: worker process exited on signal 11
In-Reply-To: <20120418142213.GI13466@mdounin.ru>
References: <20120418142213.GI13466@mdounin.ru>
Message-ID: <efbedf8e2159626158091385047992e3.NginxMailingListEnglish@forum.nginx.org>

Thanks. Do you think it's necessary to create a new rpm without this
module or will it be sufficient to turn it off in the configs?
I'll disable the config now.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225420,225485#msg-225485


From contact at jpluscplusm.com  Wed Apr 18 15:37:59 2012
From: contact at jpluscplusm.com (Jonathan Matthews)
Date: Wed, 18 Apr 2012 16:37:59 +0100
Subject: How to optimize these config / rewrite rules
In-Reply-To: <966202bd319a1ca94d70368f6b70dbae.NginxMailingListEnglish@forum.nginx.org>
References: <966202bd319a1ca94d70368f6b70dbae.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <CAKsTx7AR8TxgrKouoX95frQkgC7mH19FyQ67dD3AqhOv3pntDw@mail.gmail.com>

On 18 April 2012 15:53, rob <nginx-forum at nginx.us> wrote:
[snip]
> So, optimising the above... does anyone have any clue how to do this?

Optimising for what?
Runtime speed? Runtime memory use? Administrative efficiency?
Administrative clarity? Future flexibility?

J
-- 
Jonathan Matthews
Oxford, London, UK
http://www.jpluscplusm.com/contact.html


From nginx-forum at nginx.us  Wed Apr 18 15:48:53 2012
From: nginx-forum at nginx.us (rob)
Date: Wed, 18 Apr 2012 11:48:53 -0400 (EDT)
Subject: How to optimize these config / rewrite rules
In-Reply-To: <CAKsTx7AR8TxgrKouoX95frQkgC7mH19FyQ67dD3AqhOv3pntDw@mail.gmail.com>
References: <CAKsTx7AR8TxgrKouoX95frQkgC7mH19FyQ67dD3AqhOv3pntDw@mail.gmail.com>
Message-ID: <48d8dd62b58cce70ee93fcbb5015b268.NginxMailingListEnglish@forum.nginx.org>

As I said Jonathan,  I've been told that the above rules can be
optimised so that hitting any url in the path I mentioned would run just
one rex rule instead of a minimum of 3 currently.  I have been told that
by an nginx expert but what I have not been told is how to achieve it. I
suspect it will be by nesting location {} constructs (is thsat possible)
or by having more selective location {} constructs.

I want to start with as optimised a set of rewrite rules as possible.

Does this help any?

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225483,225487#msg-225487


From contact at jpluscplusm.com  Wed Apr 18 16:51:30 2012
From: contact at jpluscplusm.com (Jonathan Matthews)
Date: Wed, 18 Apr 2012 17:51:30 +0100
Subject: How to optimize these config / rewrite rules
In-Reply-To: <48d8dd62b58cce70ee93fcbb5015b268.NginxMailingListEnglish@forum.nginx.org>
References: <CAKsTx7AR8TxgrKouoX95frQkgC7mH19FyQ67dD3AqhOv3pntDw@mail.gmail.com>
	<48d8dd62b58cce70ee93fcbb5015b268.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <CAKsTx7BKtsnoymdRChVhvKAeM_PZK1QjfeSFmh2=BcYqNzqKyQ@mail.gmail.com>

On 18 April 2012 16:48, rob <nginx-forum at nginx.us> wrote:
> As I said Jonathan, ?I've been told that the above rules can be
> optimised so that hitting any url in the path I mentioned would run just
> one rex rule instead of a minimum of 3 currently. ?I have been told that
> by an nginx expert but what I have not been told is how to achieve it. I
> suspect it will be by nesting location {} constructs (is thsat possible)
> or by having more selective location {} constructs.
>
> I want to start with as optimised a set of rewrite rules as possible.
>
> Does this help any?

I still don't think you've defined the optimisation you're looking
for. Just to say "optimised" isn't specific enough IMHO. You optimise
/for/ a characteristic, almost always at the expense of other
characteristics.

It looks to me like you're trying to optimise for as few lines as
possible in your config file, at the expense of config file *clarity*
and runtime performance.
Also, just because you can technically combine multiple regular
expressions into a single one doesn't mean that the result will be any
faster than the multiple, simpler, regex route.

I can't personally be bothered to work out the complex regex that
would be required here for rewrites as it's *really* not something I'd
do with my nginx configs.
I want to understand them later and not worry if every minor change
might have some huge unintended impact - which is where I think you'll
end up if you carry on down this route.
However there are a few points I can give you, looking at what you've got.

1) Locations can be nested. This is valid:

location /foo {
  location /foo/bar {
    ...
  }
}

AFAICT, the intended use of this construct is to have non-regex outer
locations, preventing the evaluation of inner regex locations until
they're necessary e.g.

location /foo/ {
  location ~ ^/foo/(bar|baz) {
    ...
  }
  location ~ ^/foo/image(s)?/ {
    ...
  }
}

2) If you're using brackets solely for alternation (i.e. one thing OR
the other, e.g. "(bar|baz)", above) then consider using (?:bar|baz)
instead, as this stops the regex engine from capturing the match for
later use, which might save you valuable microseconds ;-)

3) Regex *locations* are, as per
http://wiki.nginx.org/HttpCoreModule#location, evaluated in order,
stopping after the first match. This probably isn't the case with
rewrites, which are cumulative I believe.

4) Here's how I'd structure what you're doing if my aim was to make it
forum-name agnostic. YMMV, 'cos it actually depends on the format
names of the forums. You may need to tweak it.
I wouldn't personally do this unless I *really* needed to make it
name-agnostic. It sacrifices readability for brevity.

----------------------
location ~ ^/category-name-73/(?<forum>[^/]+)/$  {
   rewrite ^ /forums/$forum/ permanent;
}
location ~ ^/category-name-73/(<?forum>[^/]+)/index(?<number>[\d]+).html {
   rewrite  /forums/$forum/page-$number permanent;
}
# Now /you/ redo these as locations; to do so requires a bit more
detail about the URI structures than you've posted :-)
#rewrite [^/]+/([\d]+)-.+-([\d]+).html /showthread.php?t=$1&page=$2 last;
#rewrite [^/]+/([\d]+)-.+.html /threads/redirectme.$1/ permanent;
----------------------

HTH,
Jonathan
-- 
Jonathan Matthews
Oxford, London, UK
http://www.jpluscplusm.com/contact.html


From nginx-forum at nginx.us  Wed Apr 18 17:14:11 2012
From: nginx-forum at nginx.us (rob)
Date: Wed, 18 Apr 2012 13:14:11 -0400 (EDT)
Subject: How to optimize these config / rewrite rules
In-Reply-To: <CAKsTx7BKtsnoymdRChVhvKAeM_PZK1QjfeSFmh2=BcYqNzqKyQ@mail.gmail.com>
References: <CAKsTx7BKtsnoymdRChVhvKAeM_PZK1QjfeSFmh2=BcYqNzqKyQ@mail.gmail.com>
Message-ID: <06b6f3c7cb4ed0c2eb8a9e003d01fc32.NginxMailingListEnglish@forum.nginx.org>

Hi Jonathan,

That was exactly the advice I needed.  I will spend some time rewriting
what I think to be a better way based on your advice and post back here
with more clearly defined rules using location a bit more.

Thanks :)

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225483,225492#msg-225492


From mdounin at mdounin.ru  Wed Apr 18 17:27:53 2012
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Wed, 18 Apr 2012 21:27:53 +0400
Subject: worker process exited on signal 11
In-Reply-To: <efbedf8e2159626158091385047992e3.NginxMailingListEnglish@forum.nginx.org>
References: <20120418142213.GI13466@mdounin.ru>
	<efbedf8e2159626158091385047992e3.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20120418172753.GK13466@mdounin.ru>

Hello!

On Wed, Apr 18, 2012 at 11:12:11AM -0400, mscheel wrote:

> Thanks. Do you think it's necessary to create a new rpm without this
> module or will it be sufficient to turn it off in the configs?
> I'll disable the config now.

It should be sufficient to disable it in config (that is, 
remove/comment out all "subs_filter" directives).

Maxim Dounin


From ashishs.dev at gmail.com  Thu Apr 19 01:29:23 2012
From: ashishs.dev at gmail.com (Ashish S)
Date: Wed, 18 Apr 2012 18:29:23 -0700
Subject: nginx upstream for broadcasting update messages ?
Message-ID: <CA+BpfUrjbnOQv5dLO62ktK_UPK1=XWKQiQVASRRyD2HUGdbfjg@mail.gmail.com>

Hi,

I have 2 memcache boxes behind nginx.  For reads, the upstream module
lets me round-robin between the backend memcache boxes. I want the
updates to go to all 2 memcache boxes. My question is, what is the
best way to broadcast the updates to the backend?  Is there a
reference implementation of some module which does similar stuff
(broadcast)?

Thanks
Ashish


From nginx-forum at nginx.us  Thu Apr 19 05:28:22 2012
From: nginx-forum at nginx.us (mscheel)
Date: Thu, 19 Apr 2012 01:28:22 -0400 (EDT)
Subject: worker process exited on signal 11
In-Reply-To: <20120418172753.GK13466@mdounin.ru>
References: <20120418172753.GK13466@mdounin.ru>
Message-ID: <a3bf6c15a5acee2be77b589155eade44.NginxMailingListEnglish@forum.nginx.org>

Hello Maxim, until now no signal 11 occured without substitutions4nginx.
It's interesting that such errors do not occur in 1.0. I'll contact the
developer of this module, maybe he can help or fix it. We're parsing the
content with regexp to rewrite links to another domain, mainly for
external CDN domains.

Thank you very much for your help.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225420,225498#msg-225498


From sammyraul1 at gmail.com  Thu Apr 19 08:39:22 2012
From: sammyraul1 at gmail.com (Sammy Raul)
Date: Thu, 19 Apr 2012 17:39:22 +0900
Subject: Registering Nginx callback in external library
Message-ID: <CACdz456WY-M6zF3tXz0ajCf_GXbeBLVNTC4pN3MX9wh_hc4LLw@mail.gmail.com>

Hi All,

Here's what I wanted to do:

1) I wanted to write a Nginx location handler for example "/mylocation"
which when invoked should call another program to get the response.

2) The other external program provides a client library where we can
register callbacks.

3) The external program will call the callback when the data is available.



I wanted to know if I am using an external library and one of the external
API is asking for registering callback, how i can achieve this in Nginx.

I cannot give any function from my handler, as that is called internally by
the nginx Core.



Please guide me how I can register a callback inside Nginx which will be
called by the external program.



Thanks,

Raul.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20120419/c5e9cb96/attachment.html>

From nginx-forum at nginx.us  Thu Apr 19 09:06:46 2012
From: nginx-forum at nginx.us (springbok)
Date: Thu, 19 Apr 2012 05:06:46 -0400 (EDT)
Subject: nginx segfault
Message-ID: <3c174a88bff7f85c8d642a4c8641c161.NginxMailingListEnglish@forum.nginx.org>

I tried to upgrade my passenger+nginx install to passenger 3.0.11 on our
staging server to match our production server,  I've also upgraded to
ruby 1.9.3. Previously I had successfully run passenger 3.0.3 with ruby
1.9.2.

After installing the new passenger gem 9.0.11, I install as usual using
passenger-install-nginx-module. Everything seems to build without any
problems but when I try and do:

/opt/nginx/sbin/nginx -V

I get "Segmentation fault", when I look in the log file I have:

Apr 19 19:00:28 jbdev1 kernel: nginx[32336]: segfault at 705740 ip
0000000000705740 sp 00007fff4673f478 error 15 in nginx[6f5000+11000]

I've tried building with the latest version of passenger 9.0.12 but I
get the same problem. I've installed passenger+nginx many times on many
server without this problem.

I'm running this particular build on suse  linux

Linux jbdev1 2.6.27.19-5-default #1 SMP 2009-02-28 04:40:21 +0100 x86_64
x86_64 x86_64 GNU/Linux
SUSE Linux Enterprise Server 11 (x86_64)
gcc version 4.3.2 [gcc-4_3-branch revision 141291] (SUSE Linux)

Any help appreciated.

Etienne.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225501,225501#msg-225501


From kjetil at pswin.com  Thu Apr 19 09:13:19 2012
From: kjetil at pswin.com (=?iso-8859-1?Q?Kjetil_M=E6land?=)
Date: Thu, 19 Apr 2012 11:13:19 +0200
Subject: Different Client certificate CA's in each server{} element?
Message-ID: <6C7F78F11731F14291BB7356B7A8AD101535DE@pro-srv4.provide.no>

 

Can we set up Nginx to handle Client verification from different CA's in each Server element?

Or do we need one NginX instance pr. CA?

 

Thanks

 

Kjetil 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20120419/4af2d7ed/attachment.html>

From edho at myconan.net  Thu Apr 19 09:15:16 2012
From: edho at myconan.net (Edho Arief)
Date: Thu, 19 Apr 2012 16:15:16 +0700
Subject: Different Client certificate CA's in each server{} element?
In-Reply-To: <6C7F78F11731F14291BB7356B7A8AD101535DE@pro-srv4.provide.no>
References: <6C7F78F11731F14291BB7356B7A8AD101535DE@pro-srv4.provide.no>
Message-ID: <CAAtReCnQWJyYX94xART533Zi0WxB-OoFVjaBNSQgr6NwBp+KAA@mail.gmail.com>

2012/4/19 Kjetil M?land <kjetil at pswin.com>:
>
>
> Can we set up Nginx to handle Client verification from different CA?s in
> each Server element?
>

Yes.


From nginx-forum at nginx.us  Thu Apr 19 10:23:30 2012
From: nginx-forum at nginx.us (wangbin579)
Date: Thu, 19 Apr 2012 06:23:30 -0400 (EDT)
Subject: Why the benchmark results are different?
In-Reply-To: <c0fcdfa70cee82a9d8d23a502359d23b.NginxMailingListEnglish@forum.nginx.org>
References: <c0fcdfa70cee82a9d8d23a502359d23b.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <fa3fe967a5a71d26dc6da2bb087a30ba.NginxMailingListEnglish@forum.nginx.org>

If you have online data,you can use tcpcopy to test your system.

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

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225415,225505#msg-225505


From nginx-forum at nginx.us  Thu Apr 19 11:51:38 2012
From: nginx-forum at nginx.us (keeef)
Date: Thu, 19 Apr 2012 07:51:38 -0400 (EDT)
Subject: Proxy_Pass & proxy_set_header problem
Message-ID: <78487d49512f812f81b51a2c3a204a39.NginxMailingListEnglish@forum.nginx.org>

I am trying to get a CMS (Liferay), a desktop application (Liferay Sync)
and WebDav to all work behind a Nginx reverse proxy and can't quite get
all 3 working although they all use the same domain. The following code
works BUT the only problem is that webpages that are returned contain
links to http://10.x.x.205:8080 If I put in a "proxy_set_header Host
$host" then the webpage links are correct and WebDav works but "Liferay
Sync" stops working.

    server {
       listen       80;
        server_name  liferay.mydomain.com;
        location / {
#               proxy_set_header Host $host;
                proxy_pass http://10.x.x.205:8080;
        }
    }

I think this is a HTTP header issue as I have tried passing all the
headers that I can find with "proxy_pass_header" and but I can't see why
the LiferaySync app isn't working.

Can anyone help ?

Cheers
Keith

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225506,225506#msg-225506


From gpakosz at yahoo.fr  Thu Apr 19 12:23:40 2012
From: gpakosz at yahoo.fr (=?ISO-8859-1?Q?Gr=E9gory_Pakosz?=)
Date: Thu, 19 Apr 2012 14:23:40 +0200
Subject: php-fpm backend, sigsegv and limit_conn
Message-ID: <CAC_01E3XjbPXAAfZdjueersS1qbyrp8VK7uzcFA-2+T=3=VbvA@mail.gmail.com>

Hello,

I'm facing the situation where worker process crashes in the php-fpm
backend with SIGSEGV: [alert] 5712#0: worker process 5713 exited on signal
11

In such a case, what happens to the connection? is it dangling?
Crashes in the backend are random, and I have the impression that when too
many crashes happen in a row, connection stay open and eventually I'm
hitting limit_conn.

What do you think?

Regards,
Gregory
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20120419/67585468/attachment.html>

From nginx-forum at nginx.us  Thu Apr 19 12:31:48 2012
From: nginx-forum at nginx.us (utahemre)
Date: Thu, 19 Apr 2012 08:31:48 -0400 (EDT)
Subject: Caching images with memcached
Message-ID: <893dbafae654ff3729e104451be26658.NginxMailingListEnglish@forum.nginx.org>

Hi everyone,

I want to cache png images with memcached and serve with nginx.

I configured nginx and i could write images to memcached. When i try to
read this images from cache with memcached client i can.

But when i try to read this images from nginx,  broken images comes.

For example : I put byte array to memcached with 115832 bytes but when i
try to read get 114940 bytes.

Anyone has idea?

Thanks

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225508,225508#msg-225508


From mdounin at mdounin.ru  Thu Apr 19 12:36:41 2012
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Thu, 19 Apr 2012 16:36:41 +0400
Subject: Default_server catch all block not working
In-Reply-To: <5c1276b46f9cd1e19931715fee0069ad.NginxMailingListEnglish@forum.nginx.org>
References: <CAKsTx7A+HKp65M0meMn5v1XQ2emLRGYj8Htj0+akL9=ZY_ZmMA@mail.gmail.com>
	<5c1276b46f9cd1e19931715fee0069ad.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20120419123641.GO13466@mdounin.ru>

Hello!

On Mon, Apr 16, 2012 at 04:43:46AM -0400, lockev3.0 wrote:

> Well, probably I'm, still novice at Nginx but, that default server block
> is meant to fit when no other block server (server_name)  fits. So my
> current problem is that 444 behaviour is arising all thru my nginx
> server, not just that default_server block/vhost 
> .....not quit sure If I explained my self better now ? or even
> understood you right ....

I belive Jonathan Matthews is right and the problem is as follows: 
your load balancer sees invalid response from your nginx server 
(as "return 444" just drops the connection) and starts to think 
it's not working (and starts to return errors to clients without 
even asking nginx).

There are two possible solutions:

1) Don't use "return 444" but return some correct response.  In 
any case "return 444" doesn't help at all as long as your nginx is 
behind load balancer (as your balancer will return response 
anyway).

2) Tune your balancer to ignore such connection closes (not sure 
this is possible).

Maxim Dounin


From am at simoon.de  Thu Apr 19 12:44:46 2012
From: am at simoon.de (Alexander Meis)
Date: Thu, 19 Apr 2012 14:44:46 +0200
Subject: Caching images with memcached
In-Reply-To: <893dbafae654ff3729e104451be26658.NginxMailingListEnglish@forum.nginx.org>
References: <893dbafae654ff3729e104451be26658.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <4F9008BE.2040108@simoon.de>

Hi,

i want to do the same setup, could you give me your configuration so 
that i can see how it works?

thanks
alex


From nginx-forum at nginx.us  Thu Apr 19 12:51:26 2012
From: nginx-forum at nginx.us (utahemre)
Date: Thu, 19 Apr 2012 08:51:26 -0400 (EDT)
Subject: Caching images with memcached
In-Reply-To: <893dbafae654ff3729e104451be26658.NginxMailingListEnglish@forum.nginx.org>
References: <893dbafae654ff3729e104451be26658.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <8e032c486e9817f59541ff54c7b18e3d.NginxMailingListEnglish@forum.nginx.org>

I did the configuration that is explained in this site;

http://f.souza.cc/2010/12/killer-java-applications-server-with-nginx-and-memcached/

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225508,225512#msg-225512


From mdounin at mdounin.ru  Thu Apr 19 13:35:18 2012
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Thu, 19 Apr 2012 17:35:18 +0400
Subject: php-fpm backend, sigsegv and limit_conn
In-Reply-To: <CAC_01E3XjbPXAAfZdjueersS1qbyrp8VK7uzcFA-2+T=3=VbvA@mail.gmail.com>
References: <CAC_01E3XjbPXAAfZdjueersS1qbyrp8VK7uzcFA-2+T=3=VbvA@mail.gmail.com>
Message-ID: <20120419133518.GU13466@mdounin.ru>

Hello!

On Thu, Apr 19, 2012 at 02:23:40PM +0200, Gr?gory Pakosz wrote:

> I'm facing the situation where worker process crashes in the php-fpm
> backend with SIGSEGV: [alert] 5712#0: worker process 5713 exited on signal
> 11

You mean nginx worker process crash, right?

> In such a case, what happens to the connection? is it dangling?
> Crashes in the backend are random, and I have the impression that when too
> many crashes happen in a row, connection stay open and eventually I'm
> hitting limit_conn.

The limit_conn numbers are kept in a shared memory and will not be 
decremented on nginx worker process crash, leading to a limit 
hits eventually.  To clear the numbers you may do an online 
upgrade, see http://nginx.org/en/docs/control.html#upgrade.

Or, better, you may want to debug crashes, see here for basic 
instructions:

http://wiki.nginx.org/Debugging

Note well that crashes might lead to much more severe 
consequences, e.g. if the crash happens during shared memory 
update it may be left in an inconsistent state, leading to 
unpredictable behaviour (usually more crashes).

Maxim Dounin


From mdounin at mdounin.ru  Thu Apr 19 13:40:39 2012
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Thu, 19 Apr 2012 17:40:39 +0400
Subject: Caching images with memcached
In-Reply-To: <893dbafae654ff3729e104451be26658.NginxMailingListEnglish@forum.nginx.org>
References: <893dbafae654ff3729e104451be26658.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20120419134039.GV13466@mdounin.ru>

Hello!

On Thu, Apr 19, 2012 at 08:31:48AM -0400, utahemre wrote:

> Hi everyone,
> 
> I want to cache png images with memcached and serve with nginx.
> 
> I configured nginx and i could write images to memcached. When i try to
> read this images from cache with memcached client i can.
> 
> But when i try to read this images from nginx,  broken images comes.
> 
> For example : I put byte array to memcached with 115832 bytes but when i
> try to read get 114940 bytes.
> 
> Anyone has idea?

Usually it happens due to compressed content being put into 
memcached by a memcached client.

Maxim Dounin


From gpakosz at yahoo.fr  Thu Apr 19 13:45:11 2012
From: gpakosz at yahoo.fr (=?ISO-8859-1?Q?Gr=E9gory_Pakosz?=)
Date: Thu, 19 Apr 2012 15:45:11 +0200
Subject: php-fpm backend, sigsegv and limit_conn
In-Reply-To: <20120419133518.GU13466@mdounin.ru>
References: <CAC_01E3XjbPXAAfZdjueersS1qbyrp8VK7uzcFA-2+T=3=VbvA@mail.gmail.com>
	<20120419133518.GU13466@mdounin.ru>
Message-ID: <CAC_01E3FbtckpUMnbDf=NcRvdiJ56=a8C5x=KUaM0piTmNuu=Q@mail.gmail.com>

On Thu, Apr 19, 2012 at 3:35 PM, Maxim Dounin <mdounin at mdounin.ru> wrote:

>
> You mean nginx worker process crash, right?
>
>
errors get logged into /var/log/nginx/error.log:
2012/04/19 15:34:41 [alert] 5712#0: worker process 9234 exited on signal 11
2012/04/19 15:34:48 [alert] 5712#0: worker process 9252 exited on signal 11
2012/04/19 15:34:57 [alert] 5712#0: worker process 9253 exited on signal 11
2012/04/19 15:35:06 [alert] 5712#0: worker process 9272 exited on signal 11
2012/04/19 15:36:11 [alert] 5712#0: worker process 9277 exited on signal 11

debian squeeze
nginx-full  1.1.19-1~bpo60+1  from squeeze-backports
php5-fpm 5.3.10-1~dotdeb.1 from dotdeb

The context: I'm experimenting with Piwik's new log analytics import
script. It's a Python script that parses server logs and does POST requests
to a Piwik instance.
I wouldn't say it's hammering much as it's able to post only around 10 log
lines per second.

It's those POST requests that randomly cause the SIGSEGV errors that got
logged into /var/log/nginx/error.log. After many worker process crashes, I
eventually hit a connection limit I set with limit_conn (of 10 simultaneous
connections per IP). I could disable limit_conn as a workaround until I
find out the root cause of those crashes.

About the crashes themselves, I really don't know what's going on so far.
PHP-FPM logs remain empty (or I wasn't able to enable them correctly).
Piwik's wiki states it's impossible Piwik itself is the root cause for
SIGSEGV; instead they say bugs in PHP itself, MySQL or accelerators like
APC are to blame. (note I disabled APC without much luck).


> The limit_conn numbers are kept in a shared memory and will not be
> decremented on nginx worker process crash, leading to a limit
> hits eventually.  To clear the numbers you may do an online
> upgrade, see http://nginx.org/en/docs/control.html#upgrade.
>
> Or, better, you may want to debug crashes, see here for basic
> instructions:
>
> http://wiki.nginx.org/Debugging
>
> Note well that crashes might lead to much more severe
> consequences, e.g. if the crash happens during shared memory
> update it may be left in an inconsistent state, leading to
> unpredictable behaviour (usually more crashes).
>
>
I'll try to give it a go. thank you for the reply

Gregory
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20120419/9f40449e/attachment.html>

From nginx-forum at nginx.us  Thu Apr 19 14:46:46 2012
From: nginx-forum at nginx.us (dignus)
Date: Thu, 19 Apr 2012 10:46:46 -0400 (EDT)
Subject: RHEL 6, nginx and parse fails
Message-ID: <9e3f559fb44adbae0fa8b2e838bb7a4b.NginxMailingListEnglish@forum.nginx.org>

Hi all,

Today I updated 2 of my RHEL 6 boxes to the latest version of all
installed packages including Nginx from the nginx repository.

Since then both reload & configtest won't work, the error I'm seeing is
this:

nginx object version 0.8.54 does not match bootstrap parameter 1.0.14 at
/usr/lib64/perl5/XSLoader.pm line 94.
Compilation failed in require.
BEGIN failed--compilation aborted.
2012/04/19 16:23:01 [alert] 11485#0: perl_parse() failed: 2

Any idea what's happening here?

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225519,225519#msg-225519


From nginx-forum at nginx.us  Thu Apr 19 14:56:21 2012
From: nginx-forum at nginx.us (utahemre)
Date: Thu, 19 Apr 2012 10:56:21 -0400 (EDT)
Subject: Caching images with memcached
In-Reply-To: <8e032c486e9817f59541ff54c7b18e3d.NginxMailingListEnglish@forum.nginx.org>
References: <893dbafae654ff3729e104451be26658.NginxMailingListEnglish@forum.nginx.org>
	<8e032c486e9817f59541ff54c7b18e3d.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <3774f3a7520338ec3cff2bedd966f269.NginxMailingListEnglish@forum.nginx.org>

I focused same topic that you are said about.

I am using spymemcached client to write the cache and nginx read wrong
data.

After your reply i tried to write with whalin memcached client and tried
to read both spymemcached and nginx. Neither nginx nor spymemcached can
read data from cache.

Does nginx have any configuration that i configure  which memcached
client nginx use? Or anyone know which memcached client can run with
nginx properly?

Emre

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225508,225520#msg-225520


From mdounin at mdounin.ru  Thu Apr 19 16:02:20 2012
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Thu, 19 Apr 2012 20:02:20 +0400
Subject: Caching images with memcached
In-Reply-To: <3774f3a7520338ec3cff2bedd966f269.NginxMailingListEnglish@forum.nginx.org>
References: <893dbafae654ff3729e104451be26658.NginxMailingListEnglish@forum.nginx.org>
	<8e032c486e9817f59541ff54c7b18e3d.NginxMailingListEnglish@forum.nginx.org>
	<3774f3a7520338ec3cff2bedd966f269.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20120419160220.GY13466@mdounin.ru>

Hello!

On Thu, Apr 19, 2012 at 10:56:21AM -0400, utahemre wrote:

> I focused same topic that you are said about.
> 
> I am using spymemcached client to write the cache and nginx read wrong
> data.
> 
> After your reply i tried to write with whalin memcached client and tried
> to read both spymemcached and nginx. Neither nginx nor spymemcached can
> read data from cache.
> 
> Does nginx have any configuration that i configure  which memcached
> client nginx use? Or anyone know which memcached client can run with
> nginx properly?

More or less any memcached client can work properly with nginx, 
you just need to figure out how to write raw data into memcached 
from your client, i.e. without compression, serialization and so 
on.  In case of Perl clients it's more or less default behaviour 
for all modules I know (Cache::Memcached, Cache::Memcached::Fast) 
as long as you write scalars to memcached.  No idea about Java.

Maxim Dounin


From zzz at zzz.org.ua  Thu Apr 19 16:11:55 2012
From: zzz at zzz.org.ua (Alexandr Gomoliako)
Date: Thu, 19 Apr 2012 19:11:55 +0300
Subject: RHEL 6, nginx and parse fails
In-Reply-To: <9e3f559fb44adbae0fa8b2e838bb7a4b.NginxMailingListEnglish@forum.nginx.org>
References: <9e3f559fb44adbae0fa8b2e838bb7a4b.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <CAKb9G9gKPGWi6=BKBmDN4E_kOXzYojGfGyMXP=2vmZ-LK=CQJA@mail.gmail.com>

On Thu, Apr 19, 2012 at 5:46 PM, dignus <nginx-forum at nginx.us> wrote:
> Hi all,
>
> Today I updated 2 of my RHEL 6 boxes to the latest version of all
> installed packages including Nginx from the nginx repository.
>
> Since then both reload & configtest won't work, the error I'm seeing is
> this:
>
> nginx object version 0.8.54 does not match bootstrap parameter 1.0.14 at
> /usr/lib64/perl5/XSLoader.pm line 94.
> Compilation failed in require.
> BEGIN failed--compilation aborted.
> 2012/04/19 16:23:01 [alert] 11485#0: perl_parse() failed: 2
>
> Any idea what's happening here?

Seems like you are trying to load older nginx.so with newer nginx.pm.


From nginx-forum at nginx.us  Thu Apr 19 17:18:14 2012
From: nginx-forum at nginx.us (sirvon)
Date: Thu, 19 Apr 2012 13:18:14 -0400 (EDT)
Subject: can't access sub directories
Message-ID: <84dea16a0832a02f92cd60e3b8e63ff4.NginxMailingListEnglish@forum.nginx.org>

with my current config, i can access the root directory
but i cant access any directory below it...
here's my vhost config:

http://pastebin.com/50pyYDtS

here's my nginx config:

http://pastebin.com/n1HXxVmu

thanks for any assistance

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225528,225528#msg-225528


From tdgh2323 at hotmail.com  Thu Apr 19 19:10:45 2012
From: tdgh2323 at hotmail.com (Micheal Wolfskill)
Date: Thu, 19 Apr 2012 19:10:45 +0000
Subject: Cache hit  statistics?
Message-ID: <BLU144-W2761BE8585FE06269EFD91D03D0@phx.gbl>


Does nginx have a cache hit ratio statistic... Ive been searching .. or perhaps a way to differentiate in the logs wchi was served from cache and which was not. Thanks

-Mike
 		 	   		  
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20120419/fbba1fd5/attachment.html>

From tdgh2323 at hotmail.com  Thu Apr 19 19:18:38 2012
From: tdgh2323 at hotmail.com (Micheal Wolfskill)
Date: Thu, 19 Apr 2012 19:18:38 +0000
Subject: How to umask nginx to make it write files in 770 (cache/logs)
Message-ID: <BLU144-W329FB098BB67F33706DD07D03D0@phx.gbl>


Hello,

Iam thinking of setting a umask in nginx... so that it writes files (logs ans cache) in 770 . Anybody know how to do this, or done it?

I need this because my environment needs access to all files produced by nginx and some are 700 .

Regards,
Mike
 		 	   		  
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20120419/54681de2/attachment.html>

From francis at daoine.org  Thu Apr 19 19:31:19 2012
From: francis at daoine.org (Francis Daly)
Date: Thu, 19 Apr 2012 20:31:19 +0100
Subject: can't access sub directories
In-Reply-To: <84dea16a0832a02f92cd60e3b8e63ff4.NginxMailingListEnglish@forum.nginx.org>
References: <84dea16a0832a02f92cd60e3b8e63ff4.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20120419193119.GC26729@craic.sysops.org>

On Thu, Apr 19, 2012 at 01:18:14PM -0400, sirvon wrote:

Hi there,

> with my current config, i can access the root directory
> but i cant access any directory below it...

With that configuration, I'd expect that any non-php request to a url
that starts http://sourcemarine.net/ would be served directly from the
filesystem below /usr/share/nginx/html/sourcemarine.net

I'm not able to work out how you want php requests to be handled.

After "fastcgi_pass", the configuration becomes unclear. By which I mean
"probably broken".

So, when you say "cant access any directory below it": can you nominate
one specific url, and say what you actually get when you fetch it,
and say what you expect to get when you fetch it?

I suspect that the output of something like

curl -i http://sourcemarine.net/one/two

will be useful.

If the problem is with the php configuration, then can you describe how
you would like the request to be handled?

As in, if I access http://sourcemarine.net/one/two.php/three, which php
file do you want "the system" to read and process?

	f
-- 
Francis Daly        francis at daoine.org


From ashishs.dev at gmail.com  Thu Apr 19 20:17:09 2012
From: ashishs.dev at gmail.com (Ashish S)
Date: Thu, 19 Apr 2012 13:17:09 -0700
Subject: nginx upstream for broadcasting update messages ?
In-Reply-To: <CA+BpfUrjbnOQv5dLO62ktK_UPK1=XWKQiQVASRRyD2HUGdbfjg@mail.gmail.com>
References: <CA+BpfUrjbnOQv5dLO62ktK_UPK1=XWKQiQVASRRyD2HUGdbfjg@mail.gmail.com>
Message-ID: <CA+BpfUqvxW5aDbs2D5To4r8PsGemKnCh9VSr4jupMv-2ehyVGQ@mail.gmail.com>

Hi,

Anyone with experience with a similar use-case?  Any recommendations here?

Thanks,
Ashish

On Wed, Apr 18, 2012 at 6:29 PM, Ashish S <ashishs.dev at gmail.com> wrote:
> Hi,
>
> I have 2 memcache boxes behind nginx. ?For reads, the upstream module
> lets me round-robin between the backend memcache boxes. I want the
> updates to go to all 2 memcache boxes. My question is, what is the
> best way to broadcast the updates to the backend? ?Is there a
> reference implementation of some module which does similar stuff
> (broadcast)?
>
> Thanks
> Ashish


From siefke_listen at web.de  Thu Apr 19 23:53:50 2012
From: siefke_listen at web.de (Silvio Siefke)
Date: Fri, 20 Apr 2012 01:53:50 +0200
Subject: Rewrite
Message-ID: <20120420015350.c2347e7e.siefke_listen@web.de>

Hello,

i want use rewrite to use article-"id".php to the post.php?id=1 for example. 
I has written the follow in the server section:

location /article-(\d+)\.php$ {
try_files $uri $uri/ /post.php?id=$1;
}

But it want not run. When i give http://example.com/blog/article-1.php, 
nginx not change to http://example.com/blog/post.php?id=1

192.168.0.20 - - [20/Apr/2012:01:52:16 +0200] "GET /blog/article-1.php HTTP/1.1" 
404 27 "http://silviosiefke_de/blog/pdo-read.php" "Mozilla/5.0 (Windows NT 5.1) 
AppleWebKit/535.19 (KHTML, like Gecko) Chrome/18.0.1025.162 Safari/535.19" "-"

Has someone idea whereis the mistake?


Thanks for help.

Regards
Silvio


From edho at myconan.net  Thu Apr 19 23:56:51 2012
From: edho at myconan.net (Edho Arief)
Date: Fri, 20 Apr 2012 06:56:51 +0700
Subject: Rewrite
In-Reply-To: <20120420015350.c2347e7e.siefke_listen@web.de>
References: <20120420015350.c2347e7e.siefke_listen@web.de>
Message-ID: <CAAtReC=4_D3bSkMdgeCb6wtesZD9k-UUOpCv=bFNNsHneEEgkQ@mail.gmail.com>

2012/4/20 Silvio Siefke <siefke_listen at web.de>:
> Hello,
>
> i want use rewrite to use article-"id".php to the post.php?id=1 for example.
> I has written the follow in the server section:
>
> location /article-(\d+)\.php$ {
> try_files $uri $uri/ /post.php?id=$1;
> }
>
> But it want not run. When i give http://example.com/blog/article-1.php,
> nginx not change to http://example.com/blog/post.php?id=1
>

1. you forgot to put ~ for regex match
2. the one you want is rewrite/return, not try_files

something like this may work:

location ~ ^/article-(\d+)\.php$ {
  rewrite ^ /post.php?id=$1 last;
}


From nginx-forum at nginx.us  Fri Apr 20 00:01:21 2012
From: nginx-forum at nginx.us (squimmy)
Date: Thu, 19 Apr 2012 20:01:21 -0400 (EDT)
Subject: Really bizarre problem with Sabredav
Message-ID: <312a7883911c6b7a9c81930afa760707.NginxMailingListEnglish@forum.nginx.org>

Hi everyone,

I've been desperately trying to get SabreDav working with Nginx for the
last 3 days almost to no avail. Well, with some success actually. By
following the instructions here:
http://blog.carlkoepke.com/index.php/2011/12/09/nginx-can-do-web-dav-with-a-little-help-from-a-friend/

I can set it up so I can login. I can create folders too. I can't upload
files though...well, I can't upload files with file extensions actually,
and it's this I find really real really bizarre. If I try and upload
blah.blah Transmit returns 405 Not Allowed and in the error logs in
Nginx I get something like "access forbidden by rule" but if I then
upload blah without the extension it works?!?! So i'm mostly getting
lots of 405 Not Allowed's :(

I'm pretty sure this is an Nginx issue not a Sabredav issue.

Help would be really, really, really appreciated.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225539,225539#msg-225539


From sammyraul1 at gmail.com  Fri Apr 20 00:11:21 2012
From: sammyraul1 at gmail.com (sammy_raul)
Date: Thu, 19 Apr 2012 17:11:21 -0700 (PDT)
Subject: Registering Nginx callback in external library
In-Reply-To: <CACdz456WY-M6zF3tXz0ajCf_GXbeBLVNTC4pN3MX9wh_hc4LLw@mail.gmail.com>
References: <CACdz456WY-M6zF3tXz0ajCf_GXbeBLVNTC4pN3MX9wh_hc4LLw@mail.gmail.com>
Message-ID: <1334880681719-7482648.post@n2.nabble.com>

Anything on this guys.

Raul

--
View this message in context: http://nginx.2469901.n2.nabble.com/Registering-Nginx-callback-in-external-library-tp7479794p7482648.html
Sent from the nginx mailing list archive at Nabble.com.


From nginx-forum at nginx.us  Fri Apr 20 00:20:33 2012
From: nginx-forum at nginx.us (springbok)
Date: Thu, 19 Apr 2012 20:20:33 -0400 (EDT)
Subject: nginx segfault
In-Reply-To: <3c174a88bff7f85c8d642a4c8641c161.NginxMailingListEnglish@forum.nginx.org>
References: <3c174a88bff7f85c8d642a4c8641c161.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <5a2d5f819003c964154382e2ffea7199.NginxMailingListEnglish@forum.nginx.org>

Ran gdb with a backtrace and got the following, is there an incompatible
library or something?

Program received signal SIGSEGV, Segmentation fault.
0x00000000006d2660 in strcmp@@GLIBC_2.2.5 ()
(gdb) bt full
#0  0x00000000006d2660 in strcmp@@GLIBC_2.2.5 ()
No symbol table info available.
#1  0x00007ffff7dea4a7 in _dl_relocate_object () from
/lib64/ld-linux-x86-64.so.2
No symbol table info available.
#2  0x00007fff00000001 in ?? ()
No symbol table info available.
#3  0x0000000000000000 in ?? ()
No symbol table info available.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225501,225541#msg-225541


From siefke_listen at web.de  Fri Apr 20 00:43:17 2012
From: siefke_listen at web.de (Silvio Siefke)
Date: Fri, 20 Apr 2012 02:43:17 +0200
Subject: Rewrite
In-Reply-To: <CAAtReC=4_D3bSkMdgeCb6wtesZD9k-UUOpCv=bFNNsHneEEgkQ@mail.gmail.com>
References: <20120420015350.c2347e7e.siefke_listen@web.de>
	<CAAtReC=4_D3bSkMdgeCb6wtesZD9k-UUOpCv=bFNNsHneEEgkQ@mail.gmail.com>
Message-ID: <20120420024317.5fa88fb8.siefke_listen@web.de>

On Fri, 20 Apr 2012 06:56:51 +0700
Edho Arief <edho at myconan.net> wrote:

> 1. you forgot to put ~ for regex match
> 2. the one you want is rewrite/return, not try_files

Thanks for help. So i have from this list, with the html files it run. 
But nginx want not use html inside php so i must change the rule.
 
> something like this may work:
> 
> location ~ ^/article-(\d+)\.php$ {
>   rewrite ^ /post.php?id=$1 last;
> }

Is the same, i become "File not found".

192.168.0.20 - - [20/Apr/2012:02:40:32 +0200] "GET /blog/article-1.php 
HTTP/1.1" 404 27 "http://silviosiefke_de/blog/index.php" "Mozilla/5.0 
(Windows NT 5.1) AppleWebKit/535.19 (KHTML, like Gecko) 
Chrome/18.0.1025.162 Safari/535.19" "-"

Regards
Silvio

gentoo-desk db # nginx -V
nginx version: nginx/1.0.15
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.pid --lock-path=/var/lock/nginx.lock 
--with-cc-opt=-I/usr/include --with-ld-opt=-L/usr/lib 
--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/fastcgi 
--http-scgi-temp-path=/var/tmp/nginx/scgi 
--http-uwsgi-temp-path=/var/tmp/nginx/uwsgi 
--with-http_addition_module --with-http_dav_module --with-http_realip_module 
--with-http_stub_status_module --with-http_sub_module --with-http_xslt_module 
--with-http_realip_module --with-pcre
--add-module=/var/tmp/portage/www-servers/nginx-1.0.15/work/agentzh-headers-more-nginx-module-137855d 
--add-module=/var/tmp/portage/www-servers/nginx-1.0.15/work/nginx_http_push_module-0.692 
--add-module=/var/tmp/portage/www-servers/nginx-1.0.15/work/nginx_upload_module-2.2.0 
--with-http_ssl_module --without-mail_imap_module --without-mail_pop3_module 
--without-mail_smtp_module --user=nginx --group=nginx


From edho at myconan.net  Fri Apr 20 00:44:57 2012
From: edho at myconan.net (Edho Arief)
Date: Fri, 20 Apr 2012 07:44:57 +0700
Subject: Rewrite
In-Reply-To: <20120420024317.5fa88fb8.siefke_listen@web.de>
References: <20120420015350.c2347e7e.siefke_listen@web.de>
	<CAAtReC=4_D3bSkMdgeCb6wtesZD9k-UUOpCv=bFNNsHneEEgkQ@mail.gmail.com>
	<20120420024317.5fa88fb8.siefke_listen@web.de>
Message-ID: <CAAtReCmMtcDASst8qdp-n4vc6JvRNUwKet7RWrWrsNskPHMfng@mail.gmail.com>

2012/4/20 Silvio Siefke <siefke_listen at web.de>:
> On Fri, 20 Apr 2012 06:56:51 +0700
> Edho Arief <edho at myconan.net> wrote:
>
>> 1. you forgot to put ~ for regex match
>> 2. the one you want is rewrite/return, not try_files
>
> Thanks for help. So i have from this list, with the html files it run.
> But nginx want not use html inside php so i must change the rule.
>
>> something like this may work:
>>
>> location ~ ^/article-(\d+)\.php$ {
>> ? rewrite ^ /post.php?id=$1 last;
>> }
>
> Is the same, i become "File not found".
>

Put the block before all other locations.


From siefke_listen at web.de  Fri Apr 20 01:23:15 2012
From: siefke_listen at web.de (Silvio Siefke)
Date: Fri, 20 Apr 2012 03:23:15 +0200
Subject: Rewrite
In-Reply-To: <CAAtReC=4_D3bSkMdgeCb6wtesZD9k-UUOpCv=bFNNsHneEEgkQ@mail.gmail.com>
References: <20120420015350.c2347e7e.siefke_listen@web.de>
	<CAAtReC=4_D3bSkMdgeCb6wtesZD9k-UUOpCv=bFNNsHneEEgkQ@mail.gmail.com>
Message-ID: <20120420032315.4b69f02d.siefke_listen@web.de>

Hello,

On Fri, 20 Apr 2012 06:56:51 +0700
Edho Arief <edho at myconan.net> wrote:

> 1. you forgot to put ~ for regex match
> 2. the one you want is rewrite/return, not try_files
> 
> something like this may work:
> 
> location ~ ^/article-(\d+)\.php$ {
>   rewrite ^ /post.php?id=$1 last;
> }


Ok i have it do, but the result is same, File not found.

The config complete: http://nopaste.info/af7861f553.html


Regards
Silvio


From agentzh at gmail.com  Fri Apr 20 07:21:34 2012
From: agentzh at gmail.com (agentzh)
Date: Fri, 20 Apr 2012 15:21:34 +0800
Subject: Registering Nginx callback in external library
In-Reply-To: <CACdz456WY-M6zF3tXz0ajCf_GXbeBLVNTC4pN3MX9wh_hc4LLw@mail.gmail.com>
References: <CACdz456WY-M6zF3tXz0ajCf_GXbeBLVNTC4pN3MX9wh_hc4LLw@mail.gmail.com>
Message-ID: <CAB4Tn6M857GC50sHT4+T1qE-9MWaXPgYAjMKMrPCLJcnQw6X+A@mail.gmail.com>

On Thu, Apr 19, 2012 at 4:39 PM, Sammy Raul <sammyraul1 at gmail.com> wrote:
> Here's what I wanted to do:
>
> 1) I wanted to write a Nginx location handler for example "/mylocation"
> which when invoked should call another program to get the response.
>
> 2) The other external program provides a client library where we can
> register callbacks.
>
> 3) The external program will call the callback when the data is available.
>
>
>
> I wanted to know if I am using an external library and one of the external
> API is asking for registering callback, how i can achieve this in Nginx.
>

Take a look at the ngx_drizzle module which integrates the external
client library libdrizzle into the Nginx core and generates outputs
for an nginx location:

    http://wiki.nginx.org/HttpDrizzleModule

Also take a look at the ngx_postgres module which integrates external
client library libpq into the Nginx core:

    https://github.com/FRiCKLE/ngx_postgres

Hope this helps,
-agentzh


From jdmls at yahoo.com  Fri Apr 20 11:55:42 2012
From: jdmls at yahoo.com (John Doe)
Date: Fri, 20 Apr 2012 04:55:42 -0700 (PDT)
Subject: How to umask nginx to make it write files in 770 (cache/logs)
In-Reply-To: <BLU144-W329FB098BB67F33706DD07D03D0@phx.gbl>
References: <BLU144-W329FB098BB67F33706DD07D03D0@phx.gbl>
Message-ID: <1334922942.14473.YahooMailNeo@web114703.mail.gq1.yahoo.com>

From: Micheal Wolfskill <tdgh2323 at hotmail.com>

>Iam thinking of setting a umask in nginx... so that it writes files (logs ans cache) in 770 . Anybody know how to do this, or done it?
>I need this because my environment needs access to all files produced by nginx and some are 700 .


Never tried but maybe set the umask at the beginning of the init.d script (but that would apply to all files creation)...
Or a quick/dirty way to do it would be to chmod 770 the log files every x hour or once a day from a cron job...


JD


From sparshgupta at gmail.com  Fri Apr 20 11:57:40 2012
From: sparshgupta at gmail.com (Sparsh Gupta)
Date: Fri, 20 Apr 2012 17:27:40 +0530
Subject: Permission Denied Error
Message-ID: <CAJi-N0FG008b0rwLzyt_F=k39xi1C2JC4SUC_z0qjvv4eAPz8w@mail.gmail.com>

I am seeing some error like below in my nginx error log:

2012/04/20 11:51:52 [crit] 21437#0: *2548 open()
"/var/cache/nginx/fastcgi_temp/5/00/0000000005" failed (13: Permission
denied) while reading upstream, client: 139.112.144.210, server: example.com,
request: "GET /index.php HTTP/1.0", upstream: "fastcgi://127.0.0.1:9000",
host: "example.com", referrer: "http://example.com/"

I have: user                    www-data; in my nginx.conf and
the owners of the  /var/cache/nginx/fastcgi_temp/5/00 are 'nginx:nginx'

How can I fix this?

Thanks
Sparsh Gupta
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20120420/90839cbd/attachment.html>

From mdounin at mdounin.ru  Fri Apr 20 13:06:23 2012
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Fri, 20 Apr 2012 17:06:23 +0400
Subject: Permission Denied Error
In-Reply-To: <CAJi-N0FG008b0rwLzyt_F=k39xi1C2JC4SUC_z0qjvv4eAPz8w@mail.gmail.com>
References: <CAJi-N0FG008b0rwLzyt_F=k39xi1C2JC4SUC_z0qjvv4eAPz8w@mail.gmail.com>
Message-ID: <20120420130623.GH13466@mdounin.ru>

Hello!

On Fri, Apr 20, 2012 at 05:27:40PM +0530, Sparsh Gupta wrote:

> I am seeing some error like below in my nginx error log:
> 
> 2012/04/20 11:51:52 [crit] 21437#0: *2548 open()
> "/var/cache/nginx/fastcgi_temp/5/00/0000000005" failed (13: Permission
> denied) while reading upstream, client: 139.112.144.210, server: example.com,
> request: "GET /index.php HTTP/1.0", upstream: "fastcgi://127.0.0.1:9000",
> host: "example.com", referrer: "http://example.com/"
> 
> I have: user                    www-data; in my nginx.conf and
> the owners of the  /var/cache/nginx/fastcgi_temp/5/00 are 'nginx:nginx'
> 
> How can I fix this?

With chown(1), see "man chown" for details.

Maxim Dounin


From mdounin at mdounin.ru  Fri Apr 20 13:14:53 2012
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Fri, 20 Apr 2012 17:14:53 +0400
Subject: Cache hit  statistics?
In-Reply-To: <BLU144-W2761BE8585FE06269EFD91D03D0@phx.gbl>
References: <BLU144-W2761BE8585FE06269EFD91D03D0@phx.gbl>
Message-ID: <20120420131453.GI13466@mdounin.ru>

Hello!

On Thu, Apr 19, 2012 at 07:10:45PM +0000, Micheal Wolfskill wrote:

> Does nginx have a cache hit ratio statistic... Ive been 
> searching .. or perhaps a way to differentiate in the logs wchi 
> was served from cache and which was not. Thanks

The $upstream_cache_status variable is what you are looking for.  
It's not yet described in docs, but here is a wiki entry:

http://wiki.nginx.org/HttpUpstreamModule#.24upstream_cache_status

Maxim Dounin


From sparshgupta at gmail.com  Fri Apr 20 18:57:40 2012
From: sparshgupta at gmail.com (Sparsh Gupta)
Date: Sat, 21 Apr 2012 00:27:40 +0530
Subject: Permission Denied Error
In-Reply-To: <20120420130623.GH13466@mdounin.ru>
References: <CAJi-N0FG008b0rwLzyt_F=k39xi1C2JC4SUC_z0qjvv4eAPz8w@mail.gmail.com>
	<20120420130623.GH13466@mdounin.ru>
Message-ID: <CAJi-N0H+Vh1a6BGcMxfKMz3-=d02PuNsxGrKjQ3wrAN05wNJqg@mail.gmail.com>

My concern was that when I have user www-data; mentioned in my config file,
why is nginx creating files for fastcgi_temp with user nginx and group nginx

I understand thats the default user:group while compiling was nginx:nginx
but shouldn't they be overridden by the option provided in nginx runtime
configuration file?

Sparsh Gupta


On 20 April 2012 18:36, Maxim Dounin <mdounin at mdounin.ru> wrote:

> Hello!
>
> On Fri, Apr 20, 2012 at 05:27:40PM +0530, Sparsh Gupta wrote:
>
> > I am seeing some error like below in my nginx error log:
> >
> > 2012/04/20 11:51:52 [crit] 21437#0: *2548 open()
> > "/var/cache/nginx/fastcgi_temp/5/00/0000000005" failed (13: Permission
> > denied) while reading upstream, client: 139.112.144.210, server:
> example.com,
> > request: "GET /index.php HTTP/1.0", upstream: "fastcgi://127.0.0.1:9000
> ",
> > host: "example.com", referrer: "http://example.com/"
> >
> > I have: user                    www-data; in my nginx.conf and
> > the owners of the  /var/cache/nginx/fastcgi_temp/5/00 are 'nginx:nginx'
> >
> > How can I fix this?
>
> With chown(1), see "man chown" for details.
>
> 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/20120421/42b1567b/attachment.html>

From nginx-forum at nginx.us  Fri Apr 20 19:38:59 2012
From: nginx-forum at nginx.us (catball123)
Date: Fri, 20 Apr 2012 15:38:59 -0400 (EDT)
Subject: nginx Unicode character bug in all win version
Message-ID: <0396d00e61b458ca0acd3ff485cf5abb.NginxMailingListEnglish@forum.nginx.org>

test in 0.7.69-1.1.9 / OS:WINXP SP2 CHINA

mkdir chinese Charset DIR like ????? 

you can see error

C:\?????\nginx.exe

2012/04/21 03:32:00 [emerg] 2060#3468: CreateFile()
"C:\?????\nginx-1.1.19/conf/nginx.conf" failed (1113: No
mapping for the Unicode character exists in the target multi-byte code
page)

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225564,225564#msg-225564


From cliff at develix.com  Fri Apr 20 20:05:40 2012
From: cliff at develix.com (Cliff Wells)
Date: Fri, 20 Apr 2012 13:05:40 -0700
Subject: Really bizarre problem with Sabredav
In-Reply-To: <312a7883911c6b7a9c81930afa760707.NginxMailingListEnglish@forum.nginx.org>
References: <312a7883911c6b7a9c81930afa760707.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <1334952340.28719.7.camel@portable-evil>

On Thu, 2012-04-19 at 20:01 -0400, squimmy wrote:
> Hi everyone,
> 
> I've been desperately trying to get SabreDav working with Nginx for the
> last 3 days almost to no avail. Well, with some success actually. By
> following the instructions here:
> http://blog.carlkoepke.com/index.php/2011/12/09/nginx-can-do-web-dav-with-a-little-help-from-a-friend/
> 
> I can set it up so I can login. I can create folders too. I can't upload
> files though...well, I can't upload files with file extensions actually,
> and it's this I find really real really bizarre. If I try and upload
> blah.blah Transmit returns 405 Not Allowed and in the error logs in
> Nginx I get something like "access forbidden by rule" but if I then
> upload blah without the extension it works?!?! So i'm mostly getting
> lots of 405 Not Allowed's :(

Without seeing any config or knowing which extensions you refer to, I'm
going to guess that you have a location that matches those extensions
(any chance they are .php?), and that location is handling the PUT
(hence, 405 method not allowed) rather than your DAV location.

Regards,
Cliff



> I'm pretty sure this is an Nginx issue not a Sabredav issue.
> 
> Help would be really, really, really appreciated.
> 
> Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225539,225539#msg-225539
> 
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx



From cliff at develix.com  Fri Apr 20 20:14:10 2012
From: cliff at develix.com (Cliff Wells)
Date: Fri, 20 Apr 2012 13:14:10 -0700
Subject: Really bizarre problem with Sabredav
In-Reply-To: <312a7883911c6b7a9c81930afa760707.NginxMailingListEnglish@forum.nginx.org>
References: <312a7883911c6b7a9c81930afa760707.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <1334952850.28719.8.camel@portable-evil>

You might try this instead:

https://github.com/arut/nginx-dav-ext-module

Cliff


On Thu, 2012-04-19 at 20:01 -0400, squimmy wrote:
> Hi everyone,
> 
> I've been desperately trying to get SabreDav working with Nginx for the
> last 3 days almost to no avail. Well, with some success actually. By
> following the instructions here:
> http://blog.carlkoepke.com/index.php/2011/12/09/nginx-can-do-web-dav-with-a-little-help-from-a-friend/
> 
> I can set it up so I can login. I can create folders too. I can't upload
> files though...well, I can't upload files with file extensions actually,
> and it's this I find really real really bizarre. If I try and upload
> blah.blah Transmit returns 405 Not Allowed and in the error logs in
> Nginx I get something like "access forbidden by rule" but if I then
> upload blah without the extension it works?!?! So i'm mostly getting
> lots of 405 Not Allowed's :(
> 
> I'm pretty sure this is an Nginx issue not a Sabredav issue.
> 
> Help would be really, really, really appreciated.
> 
> Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225539,225539#msg-225539
> 
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx



From nginx-forum at nginx.us  Fri Apr 20 20:16:46 2012
From: nginx-forum at nginx.us (haad)
Date: Fri, 20 Apr 2012 16:16:46 -0400 (EDT)
Subject: srcache + mogilefs module
Message-ID: <2077f3d358a87c93dcf94d8db97d653c.NginxMailingListEnglish@forum.nginx.org>

Hi,

I'm currently in a process of setting mogilefs + memcached + nginx
environment. My idea is to have nginx as a frontend server which will be
used to access files stored on a mogile filesystem. This is working
perfectly. To add some caching I would like to add support for memcached
to nginx itself. That way it will be able to cache some images for
itself and do not ask mogilefs for them.

My problem is that it seems that even if I put files to memcache
manually and it srcache will find it (see log below), my file is still
served from mogilefs and not from memcached.

    upstream mogilefs_trackers {
        server 192.168.206.3:6001;
        #server 192.168.206.2:6001;
    }

    upstream memcached_servers {
	server 192.168.206.3:11211;
    }

    server {
        listen       80;
        server_name  ws01x.test.reality.sk;

	root /var/www/html;

	#
	# Check if we can find anything on a memcache server if not fetch it
from a mogilefs.
	#
	location = /memc {
 	   internal;
	   memc_connect_timeout 100ms;
	   memc_send_timeout 100ms;
	   memc_read_timeout 100ms;

	   set $memc_key $query_string;
	   set $memc_exptime 300;

	   memc_pass memcached_servers;
	}

	location /share/ {
	    set $key $uri;

	    srcache_fetch GET /memc $key;
	    srcache_store PUT /memc $key;
    	    srcache_store_statuses 200 301 302;
	    srcache_response_cache_control off;

            mogilefs_tracker mogilefs_trackers;
            mogilefs_domain test;
            #mogilefs_class testclass;
	    mogilefs_methods GET;
            mogilefs_noverify on;

            mogilefs_pass {
                proxy_pass $mogilefs_path;
                proxy_hide_header Content-Type;
                proxy_buffering off;
            }
            
	    if ($request_uri ~* \.(ico|css|js)$) {
                expires 72h;
                break;
            }
	}
}



------ LOG

2012/04/20 21:28:21 [debug] 7385#0: *501 http request line: "GET
/share/file.html HTTP/1.0"
2012/04/20 21:28:21 [debug] 7385#0: *501 http uri: "/share/file.html"
2012/04/20 21:28:21 [debug] 7385#0: *501 http args: ""
2012/04/20 21:28:21 [debug] 7385#0: *501 http exten: "html"
2012/04/20 21:28:21 [debug] 7385#0: *501 http process request header
line
2012/04/20 21:28:21 [debug] 7385#0: *501 http header: "User-Agent:
Wget/1.11.4 Red Hat modified"
2012/04/20 21:28:21 [debug] 7385#0: *501 http header: "Accept: */*"
2012/04/20 21:28:21 [debug] 7385#0: *501 http header: "Host:
192.168.200.64"
2012/04/20 21:28:21 [debug] 7385#0: *501 http header: "Connection:
Keep-Alive"
2012/04/20 21:28:21 [debug] 7385#0: *501 http header done
2012/04/20 21:28:21 [debug] 7385#0: *501 event timer del: 3:
1334950161661
2012/04/20 21:28:21 [debug] 7385#0: *501 rewrite phase: 0
2012/04/20 21:28:21 [debug] 7385#0: *501 test location: "/memc"
2012/04/20 21:28:21 [debug] 7385#0: *501 test location: "/prehliadac"
2012/04/20 21:28:21 [debug] 7385#0: *501 test location: "/share/"
2012/04/20 21:28:21 [debug] 7385#0: *501 using configuration "/share/"
2012/04/20 21:28:21 [debug] 7385#0: *501 http cl:-1 max:1048576
2012/04/20 21:28:21 [debug] 7385#0: *501 rewrite phase: 2
2012/04/20 21:28:21 [debug] 7385#0: *501 http script complex value
2012/04/20 21:28:21 [debug] 7385#0: *501 http script var:
"/share/file.html"
2012/04/20 21:28:21 [debug] 7385#0: *501 http script set $key
2012/04/20 21:28:21 [debug] 7385#0: *501 http script var
2012/04/20 21:28:21 [debug] 7385#0: *501 http script var:
"/share/file.html"
2012/04/20 21:28:21 [debug] 7385#0: *501 http script regex:
"\.(ico|css|js)$"
2012/04/20 21:28:21 [notice] 7385#0: *501 "\.(ico|css|js)$" does not
match "/share/file.html", client: 192.168.200.64, server:
ws01x.test.reality.sk, request: "GET /share/file.html HTTP/1.0", host:
"192.168.200.64"
2012/04/20 21:28:21 [debug] 7385#0: *501 http script if
2012/04/20 21:28:21 [debug] 7385#0: *501 http script if: false
2012/04/20 21:28:21 [debug] 7385#0: *501 post rewrite phase: 3
2012/04/20 21:28:21 [debug] 7385#0: *501 generic phase: 4
2012/04/20 21:28:21 [debug] 7385#0: *501 generic phase: 5
2012/04/20 21:28:21 [debug] 7385#0: *501 access phase: 6
2012/04/20 21:28:21 [debug] 7385#0: *501 access phase: 7
2012/04/20 21:28:21 [debug] 7385#0: *501 access phase: 8
2012/04/20 21:28:21 [debug] 7385#0: *501 access phase: 9
2012/04/20 21:28:21 [debug] 7385#0: *501 http script var:
"/share/file.html"
2012/04/20 21:28:21 [debug] 7385#0: *501 posix_memalign:
0000000000EA31C0:4096 @16
2012/04/20 21:28:21 [debug] 7385#0: *501 http subrequest
"/memc?/share/file.html"
2012/04/20 21:28:21 [debug] 7385#0: *501 http posted request:
"/memc?/share/file.html"
2012/04/20 21:28:21 [debug] 7385#0: *501 rewrite phase: 0
2012/04/20 21:28:21 [debug] 7385#0: *501 test location: "/memc"
2012/04/20 21:28:21 [debug] 7385#0: *501 using configuration "=/memc"
2012/04/20 21:28:21 [debug] 7385#0: *501 http cl:-1 max:1048576
2012/04/20 21:28:21 [debug] 7385#0: *501 rewrite phase: 2
2012/04/20 21:28:21 [debug] 7385#0: *501 http script complex value
2012/04/20 21:28:21 [debug] 7385#0: *501 http script var:
"/share/file.html"
2012/04/20 21:28:21 [debug] 7385#0: *501 http script set $memc_key
2012/04/20 21:28:21 [debug] 7385#0: *501 http script value: "300"
2012/04/20 21:28:21 [debug] 7385#0: *501 http script set $memc_exptime
2012/04/20 21:28:21 [debug] 7385#0: *501 post rewrite phase: 3
2012/04/20 21:28:21 [debug] 7385#0: *501 generic phase: 4
2012/04/20 21:28:21 [debug] 7385#0: *501 generic phase: 5
2012/04/20 21:28:21 [debug] 7385#0: *501 http init upstream, client
timer: 0
2012/04/20 21:28:21 [debug] 7385#0: *501 epoll add event: fd:3 op:3
ev:80000005
2012/04/20 21:28:21 [debug] 7385#0: *501 http memcached request:
"/share/file.html"
2012/04/20 21:28:21 [debug] 7385#0: *501 http cleanup add:
0000000000EA4110
2012/04/20 21:28:21 [debug] 7385#0: *501 get rr peer, try: 1
2012/04/20 21:28:21 [debug] 7385#0: *501 socket 4
2012/04/20 21:28:21 [debug] 7385#0: *501 epoll add connection: fd:4
ev:80000005
2012/04/20 21:28:21 [debug] 7385#0: *501 connect to 192.168.206.3:11211,
fd:4 #502
2012/04/20 21:28:21 [debug] 7385#0: *501 http upstream connect: -2
2012/04/20 21:28:21 [debug] 7385#0: *501 event timer add: 4:
100:1334950101761
2012/04/20 21:28:21 [debug] 7385#0: *501 http finalize request: -4,
"/memc?/share/file.html" a:1, c:3
2012/04/20 21:28:21 [debug] 7385#0: *501 http request count:3 blk:0
2012/04/20 21:28:21 [debug] 7385#0: *501 http run request:
"/memc?/share/file.html"
2012/04/20 21:28:21 [debug] 7385#0: *501 http upstream check client,
write event:1, "/memc"
2012/04/20 21:28:21 [debug] 7385#0: *501 http upstream recv(): -1 (11:
Resource temporarily unavailable)
2012/04/20 21:28:21 [debug] 7385#0: *501 http upstream request:
"/memc?/share/file.html"
2012/04/20 21:28:21 [debug] 7385#0: *501 http upstream send request
handler
2012/04/20 21:28:21 [debug] 7385#0: *501 http upstream send request
2012/04/20 21:28:21 [debug] 7385#0: *501 chain writer buf fl:0 s:22
2012/04/20 21:28:21 [debug] 7385#0: *501 chain writer in:
0000000000EA4148
2012/04/20 21:28:21 [debug] 7385#0: *501 writev: 22
2012/04/20 21:28:21 [debug] 7385#0: *501 chain writer out:
0000000000000000
2012/04/20 21:28:21 [debug] 7385#0: *501 event timer del: 4:
1334950101761
2012/04/20 21:28:21 [debug] 7385#0: *501 event timer add: 4:
100:1334950101763
2012/04/20 21:28:21 [debug] 7385#0: *501 http upstream request:
"/memc?/share/file.html"
2012/04/20 21:28:21 [debug] 7385#0: *501 http upstream process header
2012/04/20 21:28:21 [debug] 7385#0: *501 malloc: 0000000000E989F0:4096
2012/04/20 21:28:21 [debug] 7385#0: *501 posix_memalign:
0000000000E99A00:4096 @16
2012/04/20 21:28:21 [debug] 7385#0: *501 recv: fd:4 46 of 4096
2012/04/20 21:28:21 [debug] 7385#0: *501 memcached: "VALUE
/share/file.html 0 10" <---- WE got it 
2012/04/20 21:28:21 [debug] 7385#0: *501 srcache_fetch: subrequest
returned status 200
2012/04/20 21:28:21 [debug] 7385#0: *501 srcache_fetch decides to send
the response in cache
2012/04/20 21:28:21 [debug] 7385#0: *501 tcp_nodelay
2012/04/20 21:28:21 [debug] 7385#0: *501 memcached filter bytes:17
size:17 length:17 rest:7
2012/04/20 21:28:21 [debug] 7385#0: *501 http upstream process non
buffered downstream
2012/04/20 21:28:21 [debug] 7385#0: *501 http output filter
"/memc?/share/file.html"
2012/04/20 21:28:21 [debug] 7385#0: *501 http copy filter:
"/memc?/share/file.html"
2012/04/20 21:28:21 [error] 7385#0: *501 srcache_fetch: cache sent
invalid status line while sending to client, client: 192.168.200.64,
server: ws01x.test.reality.sk, request: "GET /share/file.html HTTP/1.0",
subrequest: "/memc", upstream: "memcached://192.168.206.3:11211", host:
"192.168.200.64"
2012/04/20 21:28:21 [debug] 7385#0: *501 http copy filter: 0
"/memc?/share/file.html"
2012/04/20 21:28:21 [debug] 7385#0: *501 finalize http upstream request:
0
2012/04/20 21:28:21 [debug] 7385#0: *501 finalize http memcached
request
2012/04/20 21:28:21 [debug] 7385#0: *501 free rr peer 1 0
2012/04/20 21:28:21 [debug] 7385#0: *501 close http upstream connection:
4
2012/04/20 21:28:21 [debug] 7385#0: *501 event timer del: 4:
1334950101763
2012/04/20 21:28:21 [debug] 7385#0: *501 reusable connection: 0
2012/04/20 21:28:21 [debug] 7385#0: *501 http output filter
"/memc?/share/file.html"
2012/04/20 21:28:21 [debug] 7385#0: *501 http copy filter:
"/memc?/share/file.html"
2012/04/20 21:28:21 [debug] 7385#0: *501 http copy filter: 0
"/memc?/share/file.html"
2012/04/20 21:28:21 [debug] 7385#0: *501 http finalize request: 0,
"/memc?/share/file.html" a:1, c:2
2012/04/20 21:28:21 [debug] 7385#0: *501 http wake parent request:
"/share/file.html?"
2012/04/20 21:28:21 [debug] 7385#0: *501 http posted request:
"/share/file.html?"
2012/04/20 21:28:21 [debug] 7385#0: *501 access phase: 9
2012/04/20 21:28:21 [debug] 7385#0: *501 post access phase: 10
2012/04/20 21:28:21 [debug] 7385#0: *501 http set discard body
2012/04/20 21:28:21 [debug] 7385#0: *501 http init upstream, client
timer: 0
2012/04/20 21:28:21 [debug] 7385#0: *501 mogilefs request: "get_paths
key=file.html&domain=test&noverify=1"

>From this log I can see why is file can't serverd from memcached, server
status 200. Are srcache and mogilefs module somehow incompatible ?

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225568,225568#msg-225568


From adrian at navarro.at  Sat Apr 21 00:18:27 2012
From: adrian at navarro.at (=?UTF-8?Q?Adri=C3=A1n_Navarro?=)
Date: Sat, 21 Apr 2012 02:18:27 +0200
Subject: MP4 module and X-Accel-Redirect
Message-ID: <CAG7jnvup8TK7HBVEQ5gp1FVFUTffpM-CS=PfOBX+fP7ttDS4MA@mail.gmail.com>

Hello

I currently have a setup involving nginx-extras (1.1.19) and PHP,
making heavy use of the X-Accel-Redirect heades to serve files with
previous authentication. This also enables range requests and so on,
which work flawlessly.

Sadly, I can't seem to make this approach work with the MP4 streaming
module. It works as expected with a standalone mp4 file and the
?start= parameter works just fine. But passing a request like this:

/request?token=xxx&start=60

Does not work and yields the entire file instead.

My config approach is the following:

	location ~ \.php$ {
		internal;
		
		fastcgi_send_timeout 300;
		fastcgi_read_timeout 300;
		
		fastcgi_index index.php;
		
		include fastcgi_params;
		
		fastcgi_pass 127.0.0.1:9000;
	}
	
	location ~ \.mp4$ {
		mp4;
	}
	
	location /files/ {
		rewrite_by_lua '
.....misc script to track connections.....
			';
		internal;
		mp4;
		alias /home/service/files/;
	}

	location /priority {
		mp4;
		rewrite ^ /request.php last;
	}


I cannot try any more or different combinations for the mp4 directive,
I've tried to put it everywhere and just in a few places, and still,
can't manage to get anything. Some time Google shows some message
similar to this but dead in the wind without answer.

Thank you for your time.


From gpakosz at yahoo.fr  Sat Apr 21 00:47:23 2012
From: gpakosz at yahoo.fr (=?ISO-8859-1?Q?Gr=E9gory_Pakosz?=)
Date: Sat, 21 Apr 2012 02:47:23 +0200
Subject: error_page, try_files, internal redirect ends up in SIGSEGV
Message-ID: <CAC_01E1DGTEDquFcobFmMiSQLZP5-H5oj54niN2c8_bE4g2noQ@mail.gmail.com>

Hello,

I'm facing a SIGSEGV when trying to setup a custom error page using
the following (shortened for the sake of debugging) configuration.

/etc/nginx/sites-available/example.com:
server {
? listen 80;
? server_name example.com;

? access_log /var/log/nginx/example.com_access.log;
? error_log /var/log/nginx/example.com_error.log debug;

? root /var/www;

? error_page 404 @404;

? location @404 {
? ? try_files /404.html =404;
? }

? location / {
? ? try_files $uri $uri/ /index.php;
? }

? location ~ \.php$ {
? ? try_files $uri =404;

#? ? include php.conf;
#? ? fastcgi_pass unix:/var/run/php5-fpm/www.sock;
? }
}

The intent is to test whether $uri exists as a file, then a directory,
then if it doesn't exist defer to index.php if present. When index.php
doesn't exist, use the custom 404 page.

/var/www content:
  /var/www/index.html
  /var/www/404.html

There is no /var/www/index.php and I'm expecting my custom 404.html
page to be served.

Here is a debug log when accessing http://example.com/foo.txt

2012/04/21 02:38:41 [debug] 18030#0: *71 event timer del: 35: 3528919376
2012/04/21 02:38:41 [debug] 18030#0: *71 generic phase: 0
2012/04/21 02:38:41 [debug] 18030#0: *71 rewrite phase: 1
2012/04/21 02:38:41 [debug] 18030#0: *71 test location: "/"
2012/04/21 02:38:41 [debug] 18030#0: *71 test location: ~ "\.php$"
2012/04/21 02:38:41 [debug] 18030#0: *71 using configuration "/"
2012/04/21 02:38:41 [debug] 18030#0: *71 http cl:-1 max:524288
2012/04/21 02:38:41 [debug] 18030#0: *71 rewrite phase: 3
2012/04/21 02:38:41 [debug] 18030#0: *71 post rewrite phase: 4
2012/04/21 02:38:41 [debug] 18030#0: *71 generic phase: 5
2012/04/21 02:38:41 [debug] 18030#0: *71 generic phase: 6
2012/04/21 02:38:41 [debug] 18030#0: *71 generic phase: 7
2012/04/21 02:38:41 [debug] 18030#0: *71 access phase: 8
2012/04/21 02:38:41 [debug] 18030#0: *71 access phase: 9
2012/04/21 02:38:41 [debug] 18030#0: *71 access phase: 10
2012/04/21 02:38:41 [debug] 18030#0: *71 post access phase: 11
2012/04/21 02:38:41 [debug] 18030#0: *71 try files phase: 12
2012/04/21 02:38:41 [debug] 18030#0: *71 http script var: "/foo.txt"
2012/04/21 02:38:41 [debug] 18030#0: *71 trying to use file:
"/foo.txt" "/var/www/foo.txt"
2012/04/21 02:38:41 [debug] 18030#0: *71 add cleanup: 09D15FA8
2012/04/21 02:38:41 [debug] 18030#0: *71 malloc: 09C2E528:88
2012/04/21 02:38:41 [debug] 18030#0: *71 malloc: 09C2E5B0:55
2012/04/21 02:38:41 [debug] 18030#0: *71 cached open file:
/var/www/foo.txt, fd:-1, c:0, e:2, u:1
2012/04/21 02:38:41 [debug] 18030#0: *71 http script var: "/foo.txt"
2012/04/21 02:38:41 [debug] 18030#0: *71 trying to use dir: "/foo.txt"
"/var/www/foo.txt"
2012/04/21 02:38:41 [debug] 18030#0: *71 add cleanup: 09D15FC4
2012/04/21 02:38:41 [debug] 18030#0: *71 cached open file:
/var/www/foo.txt, fd:-1, c:0, e:2, u:2
2012/04/21 02:38:41 [debug] 18030#0: *71 trying to use file:
"/index.php" "/var/www/index.php"
2012/04/21 02:38:41 [debug] 18030#0: *71 internal redirect: "/index.php?"
2012/04/21 02:38:41 [debug] 18030#0: *71 rewrite phase: 1
2012/04/21 02:38:41 [debug] 18030#0: *71 test location: "/"
2012/04/21 02:38:41 [debug] 18030#0: *71 test location: ~ "\.php$"
2012/04/21 02:38:41 [debug] 18030#0: *71 using configuration "\.php$"
2012/04/21 02:38:41 [debug] 18030#0: *71 http cl:-1 max:524288
2012/04/21 02:38:41 [debug] 18030#0: *71 rewrite phase: 3
2012/04/21 02:38:41 [debug] 18030#0: *71 post rewrite phase: 4
2012/04/21 02:38:41 [debug] 18030#0: *71 generic phase: 5
2012/04/21 02:38:41 [debug] 18030#0: *71 generic phase: 6
2012/04/21 02:38:41 [debug] 18030#0: *71 generic phase: 7
2012/04/21 02:38:41 [debug] 18030#0: *71 access phase: 8
2012/04/21 02:38:41 [debug] 18030#0: *71 access phase: 9
2012/04/21 02:38:41 [debug] 18030#0: *71 access phase: 10
2012/04/21 02:38:41 [debug] 18030#0: *71 post access phase: 11
2012/04/21 02:38:41 [debug] 18030#0: *71 try files phase: 12
2012/04/21 02:38:41 [debug] 18030#0: *71 http script var: "/index.php"
2012/04/21 02:38:41 [debug] 18030#0: *71 trying to use file:
"/index.php" "/var/www/index.php"
2012/04/21 02:38:41 [debug] 18030#0: *71 add cleanup: 09D1602C
2012/04/21 02:38:41 [debug] 18030#0: *71 malloc: 09C2E5F0:88
2012/04/21 02:38:41 [debug] 18030#0: *71 malloc: 09C2E650:57
2012/04/21 02:38:41 [debug] 18030#0: *71 cached open file:
/var/www/index.php, fd:-1, c:0, e:2, u:1
2012/04/21 02:38:41 [debug] 18030#0: *71 trying to use file: "=404"
"/var/www=404"
2012/04/21 02:38:41 [debug] 18030#0: *71 http finalize request: 404,
"/index.php?" a:1, c:2
2012/04/21 02:38:41 [debug] 18030#0: *71 http special response: 404,
"/index.php?"
2012/04/21 02:38:41 [debug] 18030#0: *71 test location: "@404"
2012/04/21 02:38:41 [debug] 18030#0: *71 using location: @404 "/index.php?"
2012/04/21 02:38:41 [debug] 18030#0: *71 rewrite phase: 3
2012/04/21 02:38:41 [debug] 18030#0: *71 post rewrite phase: 4
2012/04/21 02:38:41 [debug] 18030#0: *71 generic phase: 5
2012/04/21 02:38:41 [debug] 18030#0: *71 generic phase: 6
2012/04/21 02:38:41 [debug] 18030#0: *71 generic phase: 7
2012/04/21 02:38:41 [debug] 18030#0: *71 access phase: 8
2012/04/21 02:38:41 [debug] 18030#0: *71 access phase: 9
2012/04/21 02:38:41 [debug] 18030#0: *71 access phase: 10
2012/04/21 02:38:41 [debug] 18030#0: *71 post access phase: 11
2012/04/21 02:38:41 [debug] 18030#0: *71 try files phase: 12

At which point nginx crashes, /var/log/nginx/error.log:
2012/04/21 02:38:41 [alert] 17786#0: worker process 18030 exited on signal 11

# nginx -V
nginx version: nginx/1.1.19
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=/tmp/buildd/nginx-1.1.19/debian/modules/nginx-auth-pam
--add-module=/tmp/buildd/nginx-1.1.19/debian/modules/nginx-echo
--add-module=/tmp/buildd/nginx-1.1.19/debian/modules/nginx-upstream-fair
--add-module=/tmp/buildd/nginx-1.1.19/debian/modules/nginx-dav-ext-module

I'm using nginx 1.1.19-1~bpo60+1 package from Debian squeeze backports.

Any idea please?
Is there anything else I can report?

Gregory


From tdgh2323 at hotmail.com  Sat Apr 21 02:09:30 2012
From: tdgh2323 at hotmail.com (Micheal Wolfskill)
Date: Sat, 21 Apr 2012 02:09:30 +0000
Subject: How to umask nginx to make it write files in 770 (cache/logs)
In-Reply-To: <1334922942.14473.YahooMailNeo@web114703.mail.gq1.yahoo.com>
References: <BLU144-W329FB098BB67F33706DD07D03D0@phx.gbl>,
	<1334922942.14473.YahooMailNeo@web114703.mail.gq1.yahoo.com>
Message-ID: <BLU144-W31F78791A176A0A924E574D0230@phx.gbl>


I tried this in the init script:

su - nginx -c "umask 002";
umask 002;


None of them work... new cache files/dir still get 700 permissions...

What could i be doing wrong=?



> Date: Fri, 20 Apr 2012 04:55:42 -0700
> From: jdmls at yahoo.com
> Subject: Re: How to umask nginx to make it write files in 770 (cache/logs)
> To: nginx at nginx.org
> 
> From: Micheal Wolfskill <tdgh2323 at hotmail.com>
> 
> >Iam thinking of setting a umask in nginx... so that it writes files (logs ans cache) in 770 . Anybody know how to do this, or done it?
> >I need this because my environment needs access to all files produced by nginx and some are 700 .
> 
> 
> Never tried but maybe set the umask at the beginning of the init.d script (but that would apply to all files creation)...
> Or a quick/dirty way to do it would be to chmod 770 the log files every x hour or once a day from a cron job...
> 
> 
> JD
> 
> _______________________________________________
> 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/20120421/45c218b8/attachment.html>

From nginx-forum at nginx.us  Sat Apr 21 02:22:25 2012
From: nginx-forum at nginx.us (nikita.tovstoles)
Date: Fri, 20 Apr 2012 22:22:25 -0400 (EDT)
Subject: HttpHealthcheckModule server not marked down
In-Reply-To: <CAB2wvz3k8J5FFBNUX=t1UsAaXcc-mavegJBTQgdVwttf34XUeA@mail.gmail.com>
References: <CAB2wvz3k8J5FFBNUX=t1UsAaXcc-mavegJBTQgdVwttf34XUeA@mail.gmail.com>
Message-ID: <98bff3fa433b3aa20090e187f022c094.NginxMailingListEnglish@forum.nginx.org>

Hello, 

I was seeing the same issue: when upstream node is down, healthcheck
correctly flags it as down yet subsequent requests are still routed to
that node (even though others are healthy). 

Tried this patch with 1.0.13 and .15 - no relief:
https://github.com/liseen/healthcheck_nginx_upstreams/blob/master/healthcheck.patch

Is the above supposed to work when ip_hash directive is used?

Thank you,

-nikita

My config:

        upstream admin-cluster {
                ip_hash; # clientIP-based session affinity
healthcheck_enabled;
healthcheck_send "GET /running HTTP/1.0" "Host: www.mydomain.com"
"User-Agent: FooBar/1.0 nginx" "Connection: close";
healthcheck_delay 2000;
healthcheck_failcount 2;
healthcheck_timeout 2000;
server admin1.staging.mydomain.com:8080 max_fails=1 fail_timeout=5; #max
sec to connect to upstream host
        }

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215656,225577#msg-225577


From tdgh2323 at hotmail.com  Sat Apr 21 02:38:53 2012
From: tdgh2323 at hotmail.com (Micheal Wolfskill)
Date: Sat, 21 Apr 2012 02:38:53 +0000
Subject: Change upstream server on the fly? (by command,
	instead of config editting)
Message-ID: <BLU144-W104F716F5C89A1FA8AFD0DD0230@phx.gbl>


Hello all,

Is there anyway I can change the upstream server of server{} on the fly... meaning without having to edit the config and reloading it?

Mike
 		 	   		  
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20120421/5749c570/attachment.html>

From ru at nginx.com  Sat Apr 21 03:28:30 2012
From: ru at nginx.com (Ruslan Ermilov)
Date: Sat, 21 Apr 2012 07:28:30 +0400
Subject: error_page, try_files, internal redirect ends up in SIGSEGV
In-Reply-To: <CAC_01E1DGTEDquFcobFmMiSQLZP5-H5oj54niN2c8_bE4g2noQ@mail.gmail.com>
References: <CAC_01E1DGTEDquFcobFmMiSQLZP5-H5oj54niN2c8_bE4g2noQ@mail.gmail.com>
Message-ID: <20120421032830.GA92227@lo0.su>

On Sat, Apr 21, 2012 at 02:47:23AM +0200, Gr?gory Pakosz wrote:
> I'm facing a SIGSEGV when trying to setup a custom error page using
> the following (shortened for the sake of debugging) configuration.

It's been fixed already:
http://trac.nginx.org/nginx/changeset/4601/nginx


From agentzh at gmail.com  Sat Apr 21 04:45:17 2012
From: agentzh at gmail.com (agentzh)
Date: Sat, 21 Apr 2012 12:45:17 +0800
Subject: srcache + mogilefs module
In-Reply-To: <2077f3d358a87c93dcf94d8db97d653c.NginxMailingListEnglish@forum.nginx.org>
References: <2077f3d358a87c93dcf94d8db97d653c.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <CAB4Tn6OeoKAAq-8GLvN973gn+f+NMaoD_Pd2kwh3rnvG6XnH5Q@mail.gmail.com>

On Sat, Apr 21, 2012 at 4:16 AM, haad <nginx-forum at nginx.us> wrote:
> My problem is that it seems that even if I put files to memcache
> manually and it srcache will find it (see log below), my file is still
> served from mogilefs and not from memcached.
>

What does the data you put into memcached look like? The ngx_srcache
module requires the value to be the *whole* response, that is,
including the status line and all those response headers.

Basically, you should rely on ngx_srcache's srcache_store to store the
responses automatically for you, rather than inserting values into
memcached yourself.

> 2012/04/20 21:28:21 [debug] 7385#0: *501 memcached: "VALUE
> /share/file.html 0 10" <---- WE got it

Yes, ngx_srcache got the value from memcached here.

> 2012/04/20 21:28:21 [error] 7385#0: *501 srcache_fetch: cache sent
> invalid status line while sending to client, client: 192.168.200.64,
> server: ws01x.test.reality.sk, request: "GET /share/file.html HTTP/1.0",
> subrequest: "/memc", upstream: "memcached://192.168.206.3:11211", host:
> "192.168.200.64"

But here, ngx_srcache then found the value is invalid in that it does
not have a valid HTTP status line at the beginning, so it immediately
abandoned the cached value, and treated it as a cache miss.

>
> From this log I can see why is file can't serverd from memcached, server
> status 200.

Maybe I should make the debug log more verbose a bit here ;)

> Are srcache and mogilefs module somehow incompatible ?
>

ngx_srcache is designed to be backend independent ;) If it's not, then
there must be a bug somewhere ;)

In your case here, it seems to be the bad format of the data that you
manually inserted into memcached.

Best regards,
-agentzh


From nginx-forum at nginx.us  Sat Apr 21 06:11:27 2012
From: nginx-forum at nginx.us (xxx)
Date: Sat, 21 Apr 2012 02:11:27 -0400 (EDT)
Subject: Rewrite in Ngnix
Message-ID: <fe6fc5606f7827d45aed72ae1d52b5cc.NginxMailingListEnglish@forum.nginx.org>

Hi,
How convert to httacces from apache to nginx

My file htaccess is:

#Options FollowSymLinks

RewriteEngine on

RewriteRule ^vardata/(.*).php$ [R=404,L]
RewriteRule ^([a-zA-Z0-9_-]+).html$ index.php?op=$1
RewriteRule ^([a-zA-Z0-9_-]+),([a-zA-Z0-9_-]+).html$
index.php?op=$1&do=$2
RewriteRule ^([a-zA-Z0-9_-]+),([a-zA-Z0-9_-]+),([a-zA-Z0-9_-]+).html$
index.php?op=$1&do=$2&id=$3
RewriteRule
^([a-zA-Z0-9_-]+),([a-zA-Z0-9_-]+),([a-zA-Z0-9_-]+),([a-zA-Z0-9_-]+).html$
index.php?op=$1&do=$2&id=$3&qs=$4
RewriteRule
^([a-zA-Z0-9_-]+),([a-zA-Z0-9_-]+),([a-zA-Z0-9_-]+),([a-zA-Z0-9_-]+),([a-zA-Z0-9_-]+).html$
index.php?op=$1&do=$2&id=$3&qs=$4&ne=$5
RewriteRule
^([a-zA-Z0-9_-]+),([a-zA-Z0-9_-]+),([a-zA-Z0-9_-]+),([a-zA-Z0-9_-]+),([a-zA-Z0-9_-]+),([a-zA-Z0-9_-]+).html$
index.php?op=$1&do=$2&id=$3&qs=$4&ne=$5&nr=$6
RewriteRule
^([a-zA-Z0-9_-]+),([a-zA-Z0-9_-]+),([a-zA-Z0-9_-]+),([a-zA-Z0-9_-]+),([a-zA-Z0-9_-]+),([a-zA-Z0-9_-]+),([a-zA-Z0-9_-]+).html$
index.php?op=$1&do=$2&id=$3&qs=$4&ne=$5&nr=$6&ha=$7
RewriteRule
^([a-zA-Z0-9_-]+),([a-zA-Z0-9_-]+),([a-zA-Z0-9_-]+),([a-zA-Z0-9_-]+),([a-zA-Z0-9_-]+),([a-zA-Z0-9_-]+),([a-zA-Z0-9_-]+),([a-zA-Z0-9_-]+).html$
index.php?op=$1&do=$2&id=$3&qs=$4&ne=$5&nr=$6&ha=$7&hs=$8 [L] 


HELP Me!

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225583,225583#msg-225583


From gpakosz at yahoo.fr  Sat Apr 21 07:55:09 2012
From: gpakosz at yahoo.fr (=?ISO-8859-1?Q?Gr=E9gory_Pakosz?=)
Date: Sat, 21 Apr 2012 09:55:09 +0200
Subject: error_page, try_files, internal redirect ends up in SIGSEGV
In-Reply-To: <20120421032830.GA92227@lo0.su>
References: <CAC_01E1DGTEDquFcobFmMiSQLZP5-H5oj54niN2c8_bE4g2noQ@mail.gmail.com>
	<20120421032830.GA92227@lo0.su>
Message-ID: <CAC_01E1=REzEoTgOszsV=vu4_6X8e2fZaeAjqBi9kssNRBgdYw@mail.gmail.com>

On Sat, Apr 21, 2012 at 5:28 AM, Ruslan Ermilov <ru at nginx.com> wrote:
> On Sat, Apr 21, 2012 at 02:47:23AM +0200, Gr?gory Pakosz wrote:
>> I'm facing a SIGSEGV when trying to setup a custom error page using
>> the following (shortened for the sake of debugging) configuration.
>
> It's been fixed already:
> http://trac.nginx.org/nginx/changeset/4601/nginx
>
Hi there!

Thank you for the answer. At least I managed to conduct my first nginx
debug session :)

Looking forward nginx 1.2.0 then.

Regards,
Gregory


From nginx-forum at nginx.us  Sat Apr 21 11:49:28 2012
From: nginx-forum at nginx.us (squimmy)
Date: Sat, 21 Apr 2012 07:49:28 -0400 (EDT)
Subject: Really bizarre problem with Sabredav
In-Reply-To: <312a7883911c6b7a9c81930afa760707.NginxMailingListEnglish@forum.nginx.org>
References: <312a7883911c6b7a9c81930afa760707.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <63e97d7a0e5dc0cbf0f3488738ebe6f3.NginxMailingListEnglish@forum.nginx.org>

Hi,

Yeh, I did think about using nginx modules, but that would require
recompiling nginx unless i'm mistaken? I'm also hoping to integrate this
with Owncloud, and that uses SabreDav, so getting that working is my
preferred option.

These are my Nginx conf files:
Nginx.conf: http://pastebin.com/fNPkw4fq

Virtual.conf: http://pastebin.com/N4PZXrBm

Are there any other conf files that are involved? I'm very new to
Nginx!

Thank you for your help!

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225539,225589#msg-225589


From contact at jpluscplusm.com  Sat Apr 21 13:01:12 2012
From: contact at jpluscplusm.com (Jonathan Matthews)
Date: Sat, 21 Apr 2012 14:01:12 +0100
Subject: Rewrite in Ngnix
In-Reply-To: <fe6fc5606f7827d45aed72ae1d52b5cc.NginxMailingListEnglish@forum.nginx.org>
References: <fe6fc5606f7827d45aed72ae1d52b5cc.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <CAKsTx7C6JmAK-q1ocyskZcQOABb+LSk+8mqJ8p6cCD9qaF3tDQ@mail.gmail.com>

On 21 April 2012 07:11, xxx <nginx-forum at nginx.us> wrote:
> Hi,
> How convert to httacces from apache to nginx
> HELP Me!

Step 1: Read and understand the documentation at http://nginx.org/r/rewrite
Step 2: Understand what your existing rewrite rules are doing
Step 3: Rewrite your rewrite rules

Let us know how you get on and also -- if you get stuck -- what you've
tried, what's worked, and what hasn't worked.

Cheers,
Jonathan
-- 
Jonathan Matthews
Oxford, London, UK
http://www.jpluscplusm.com/contact.html


From cliff at develix.com  Sat Apr 21 20:22:29 2012
From: cliff at develix.com (Cliff Wells)
Date: Sat, 21 Apr 2012 13:22:29 -0700
Subject: Really bizarre problem with Sabredav
In-Reply-To: <63e97d7a0e5dc0cbf0f3488738ebe6f3.NginxMailingListEnglish@forum.nginx.org>
References: <312a7883911c6b7a9c81930afa760707.NginxMailingListEnglish@forum.nginx.org>
	<63e97d7a0e5dc0cbf0f3488738ebe6f3.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <1335039749.2543.12.camel@portable-evil>

On Sat, 2012-04-21 at 07:49 -0400, squimmy wrote:
> Hi,
> 
> Yeh, I did think about using nginx modules, but that would require
> recompiling nginx unless i'm mistaken? I'm also hoping to integrate this
> with Owncloud, and that uses SabreDav, so getting that working is my
> preferred option.
> 
> These are my Nginx conf files:
> Nginx.conf: http://pastebin.com/fNPkw4fq
> 
> Virtual.conf: http://pastebin.com/N4PZXrBm

Here's what I suspect is happening:

You have location /dav, where I expect is where you want SabreDav to
handle requests.  After that you do:

include /usr/local/nginx/conf/php.conf;
include /include /usr/local/nginx/conf/staticfiles.conf;

I suspect these contain regex locations that are matching your
problematic extensions.  The reason for this is that these regex
locations will take precedence over your /dav location, so when you do a
DAV request such as "PUT /dav/image.jpg", you are really doing the PUT
to a regex location such as "~ \.(jpg|png|gif)$" in staticfiles.conf or
"~ \.php$" in php.conf.

The way Nginx processes requests is outlined here:

http://nginx.org/en/docs/http/request_processing.html

The end result of all of this is you will need to *not* include your
php.conf and staticfiles.conf as top-level locations.  Instead you will
need to include them inside other non-regex locations.

Alternatively, you could setup a subdomain such as dav.domain.com and do
all your DAV requests via that (this is probably the most foolproof).


Regards,
Cliff



From nginx-forum at nginx.us  Sat Apr 21 20:51:04 2012
From: nginx-forum at nginx.us (nikita.tovstoles)
Date: Sat, 21 Apr 2012 16:51:04 -0400 (EDT)
Subject: HttpHealthcheckModule server not marked down
In-Reply-To: <CAP0FwPR78VwaXg4qbftEZbsPFeen-xwrnGQU3wFgdd8rPu9XOg@mail.gmail.com>
References: <CAP0FwPR78VwaXg4qbftEZbsPFeen-xwrnGQU3wFgdd8rPu9XOg@mail.gmail.com>
Message-ID: <ea76c244ae51b2ad244079f9ce4a1693.NginxMailingListEnglish@forum.nginx.org>

I confirmed that this is an issue only when ip_hash is used - without it
healthcheck fail causes upstream to stop routing requests to failing
node. Could use a bit of help understanding:
- is this by design (i.e ip_hash support never implemented) or a bug?

I am happy to create an enhancement/fix but, as a c-newbie, would
appreciate a bit of guidance on where specifically to look in nginx or
https://github.com/liseen/healthcheck_nginx_upstreams/blob/master/ngx_http_healthcheck_module.c

thanks!

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,215656,225595#msg-225595


From nginx-forum at nginx.us  Sat Apr 21 21:49:23 2012
From: nginx-forum at nginx.us (haad)
Date: Sat, 21 Apr 2012 17:49:23 -0400 (EDT)
Subject: srcache + mogilefs module
In-Reply-To: <CAB4Tn6OeoKAAq-8GLvN973gn+f+NMaoD_Pd2kwh3rnvG6XnH5Q@mail.gmail.com>
References: <CAB4Tn6OeoKAAq-8GLvN973gn+f+NMaoD_Pd2kwh3rnvG6XnH5Q@mail.gmail.com>
Message-ID: <ff95461527422f4f9fc4db6d92d52ef3.NginxMailingListEnglish@forum.nginx.org>

agentzh Wrote:
-------------------------------------------------------
> On Sat, Apr 21, 2012 at 4:16 AM, haad
> <nginx-forum at nginx.us> wrote:
> > My problem is that it seems that even if I put
> files to memcache
> > manually and it srcache will find it (see log
> below), my file is still
> > served from mogilefs and not from memcached.
> >
> 
> What does the data you put into memcached look
> like? The ngx_srcache
> module requires the value to be the *whole*
> response, that is,
> including the status line and all those response
> headers.

It was simple string, so yes this was the problem.

> 
> Basically, you should rely on ngx_srcache's
> srcache_store to store the
> responses automatically for you, rather than
> inserting values into
> memcached yourself.

But from attached log or this new [1] dump it seems tat srcache_store is
never called.
I was not aware of this, it should be added to documentation :).

[1] https://gist.github.com/2439805


> 
> > 2012/04/20 21:28:21 [debug] 7385#0: *501
> memcached: "VALUE
> > /share/file.html 0 10" <---- WE got it
> 
> Yes, ngx_srcache got the value from memcached
> here.
> 
> > 2012/04/20 21:28:21 [error] 7385#0: *501
> srcache_fetch: cache sent
> > invalid status line while sending to client,
> client: 192.168.200.64,
> > server: ws01x.test.reality.sk, request: "GET
> /share/file.html HTTP/1.0",
> > subrequest: "/memc", upstream:
> "memcached://192.168.206.3:11211", host:
> > "192.168.200.64"
> 
> But here, ngx_srcache then found the value is
> invalid in that it does
> not have a valid HTTP status line at the
> beginning, so it immediately
> abandoned the cached value, and treated it as a
> cache miss.
> 
> >
> > From this log I can see why is file can't
> serverd from memcached, server
> > status 200.
> 
> Maybe I should make the debug log more verbose a
> bit here ;)
> 
> > Are srcache and mogilefs module somehow
> incompatible ?
> >
> 
> ngx_srcache is designed to be backend independent
> ;) If it's not, then
> there must be a bug somewhere ;)
> 
> In your case here, it seems to be the bad format
> of the data that you
> manually inserted into memcached.

I will try to add proper entry and try again, right now I would like to
know why
srcache_store do not work as expected. Here is config

	location = /memc {
 	   internal;
	   memc_connect_timeout 100ms;
	   memc_send_timeout 100ms;
	   memc_read_timeout 100ms;

	   set $memc_key $query_string;
	   set $memc_exptime 300;

	   memc_pass memcached_servers;
	}

	location /share/ {
	    set $key $uri;

	    srcache_fetch GET /memc $key;
	    srcache_store PUT /memc $key;
    	    srcache_store_statuses 200 301 302;
	    srcache_response_cache_control off;

      [snip]

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225568,225597#msg-225597


From cliff at develix.com  Sat Apr 21 22:28:27 2012
From: cliff at develix.com (Cliff Wells)
Date: Sat, 21 Apr 2012 15:28:27 -0700
Subject: Permission Denied Error
In-Reply-To: <CAJi-N0H+Vh1a6BGcMxfKMz3-=d02PuNsxGrKjQ3wrAN05wNJqg@mail.gmail.com>
References: <CAJi-N0FG008b0rwLzyt_F=k39xi1C2JC4SUC_z0qjvv4eAPz8w@mail.gmail.com>
	<20120420130623.GH13466@mdounin.ru>
	<CAJi-N0H+Vh1a6BGcMxfKMz3-=d02PuNsxGrKjQ3wrAN05wNJqg@mail.gmail.com>
Message-ID: <1335047307.2121.1.camel@portable-evil>

On Sat, 2012-04-21 at 00:27 +0530, Sparsh Gupta wrote:
> My concern was that when I have user www-data; mentioned in my config
> file, why is nginx creating files for fastcgi_temp with user nginx and
> group nginx

Probably they were created before you changed the nginx config file
settings.

Cliff



From francis at daoine.org  Sat Apr 21 22:30:18 2012
From: francis at daoine.org (Francis Daly)
Date: Sat, 21 Apr 2012 23:30:18 +0100
Subject: Really bizarre problem with Sabredav
In-Reply-To: <1335039749.2543.12.camel@portable-evil>
References: <312a7883911c6b7a9c81930afa760707.NginxMailingListEnglish@forum.nginx.org>
	<63e97d7a0e5dc0cbf0f3488738ebe6f3.NginxMailingListEnglish@forum.nginx.org>
	<1335039749.2543.12.camel@portable-evil>
Message-ID: <20120421223018.GE26729@craic.sysops.org>

On Sat, Apr 21, 2012 at 01:22:29PM -0700, Cliff Wells wrote:
> On Sat, 2012-04-21 at 07:49 -0400, squimmy wrote:

Hi there,

> Here's what I suspect is happening:
> 
> You have location /dav, where I expect is where you want SabreDav to
> handle requests.  

> The way Nginx processes requests is outlined here:
> 
> http://nginx.org/en/docs/http/request_processing.html

Possibly changing the line

  location /dav/

to

  location ^~ /dav/

will cause things to Just Work.

But it does depend on what exactly SabreDAV does, and what else is in
your included config files.

If it does work fully, it might be worth sending that information to the
SabreDAV people in case they want to include a "here is how to configure
your web server to use our application" note in their documentation
(assuming that that doesn't already exist).

Good luck with it,

	f
-- 
Francis Daly        francis at daoine.org


From mdounin at mdounin.ru  Sat Apr 21 22:32:54 2012
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Sun, 22 Apr 2012 02:32:54 +0400
Subject: Permission Denied Error
In-Reply-To: <CAJi-N0H+Vh1a6BGcMxfKMz3-=d02PuNsxGrKjQ3wrAN05wNJqg@mail.gmail.com>
References: <CAJi-N0FG008b0rwLzyt_F=k39xi1C2JC4SUC_z0qjvv4eAPz8w@mail.gmail.com>
	<20120420130623.GH13466@mdounin.ru>
	<CAJi-N0H+Vh1a6BGcMxfKMz3-=d02PuNsxGrKjQ3wrAN05wNJqg@mail.gmail.com>
Message-ID: <20120421223254.GL13466@mdounin.ru>

Hello!

On Sat, Apr 21, 2012 at 12:27:40AM +0530, Sparsh Gupta wrote:

> My concern was that when I have user www-data; mentioned in my config file,
> why is nginx creating files for fastcgi_temp with user nginx and group nginx
> 
> I understand thats the default user:group while compiling was nginx:nginx
> but shouldn't they be overridden by the option provided in nginx runtime
> configuration file?

They are overriden.  But if you change user/group set in config you 
have to fix permissions on already created subdirs (or just remove 
them, nginx will re-create them with correct permissions).

Maxim Dounin

> 
> Sparsh Gupta
> 
> 
> On 20 April 2012 18:36, Maxim Dounin <mdounin at mdounin.ru> wrote:
> 
> > Hello!
> >
> > On Fri, Apr 20, 2012 at 05:27:40PM +0530, Sparsh Gupta wrote:
> >
> > > I am seeing some error like below in my nginx error log:
> > >
> > > 2012/04/20 11:51:52 [crit] 21437#0: *2548 open()
> > > "/var/cache/nginx/fastcgi_temp/5/00/0000000005" failed (13: Permission
> > > denied) while reading upstream, client: 139.112.144.210, server:
> > example.com,
> > > request: "GET /index.php HTTP/1.0", upstream: "fastcgi://127.0.0.1:9000
> > ",
> > > host: "example.com", referrer: "http://example.com/"
> > >
> > > I have: user                    www-data; in my nginx.conf and
> > > the owners of the  /var/cache/nginx/fastcgi_temp/5/00 are 'nginx:nginx'
> > >
> > > How can I fix this?
> >
> > With chown(1), see "man chown" for details.
> >
> > 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 adrian at navarro.at  Sun Apr 22 02:03:26 2012
From: adrian at navarro.at (=?UTF-8?Q?Adri=C3=A1n_Navarro?=)
Date: Sun, 22 Apr 2012 04:03:26 +0200
Subject: MP4 module and X-Accel-Redirect
In-Reply-To: <CAG7jnvup8TK7HBVEQ5gp1FVFUTffpM-CS=PfOBX+fP7ttDS4MA@mail.gmail.com>
References: <CAG7jnvup8TK7HBVEQ5gp1FVFUTffpM-CS=PfOBX+fP7ttDS4MA@mail.gmail.com>
Message-ID: <CAG7jnvtcX2g3S_dsu2zWAPnbjiw+e=CBbWGpqUDWmph9zXcrZg@mail.gmail.com>

I have finally resorted to a mixed approach:

- I generate symbolic links (ln -s) from the concealed 'internal'
location to a folder in the public root, only for the files I need
- That folder has a location {} directive, containing the mp4;
parameter to enable streaming support
- This location has an access_by_lua directive, which passes, using
ngx.location.capture, a copy of the request to a local API (written in
PHP) and validates the access
- Also rate limiting

That way I can mimic the usual approach where PHP does all these
checks, in a much more complicated way? But still, it works.

On Sat, Apr 21, 2012 at 2:18 AM, Adri?n Navarro <adrian at navarro.at> wrote:
> Hello
>
> I currently have a setup involving nginx-extras (1.1.19) and PHP,
> making heavy use of the X-Accel-Redirect heades to serve files with
> previous authentication. This also enables range requests and so on,
> which work flawlessly.
>
> Sadly, I can't seem to make this approach work with the MP4 streaming
> module. It works as expected with a standalone mp4 file and the
> ?start= parameter works just fine. But passing a request like this:
>
> /request?token=xxx&start=60
>
> Does not work and yields the entire file instead.
>
> My config approach is the following:
>
> ? ? ? ?location ~ \.php$ {
> ? ? ? ? ? ? ? ?internal;
>
> ? ? ? ? ? ? ? ?fastcgi_send_timeout 300;
> ? ? ? ? ? ? ? ?fastcgi_read_timeout 300;
>
> ? ? ? ? ? ? ? ?fastcgi_index index.php;
>
> ? ? ? ? ? ? ? ?include fastcgi_params;
>
> ? ? ? ? ? ? ? ?fastcgi_pass 127.0.0.1:9000;
> ? ? ? ?}
>
> ? ? ? ?location ~ \.mp4$ {
> ? ? ? ? ? ? ? ?mp4;
> ? ? ? ?}
>
> ? ? ? ?location /files/ {
> ? ? ? ? ? ? ? ?rewrite_by_lua '
> .....misc script to track connections.....
> ? ? ? ? ? ? ? ? ? ? ? ?';
> ? ? ? ? ? ? ? ?internal;
> ? ? ? ? ? ? ? ?mp4;
> ? ? ? ? ? ? ? ?alias /home/service/files/;
> ? ? ? ?}
>
> ? ? ? ?location /priority {
> ? ? ? ? ? ? ? ?mp4;
> ? ? ? ? ? ? ? ?rewrite ^ /request.php last;
> ? ? ? ?}
>
>
> I cannot try any more or different combinations for the mp4 directive,
> I've tried to put it everywhere and just in a few places, and still,
> can't manage to get anything. Some time Google shows some message
> similar to this but dead in the wind without answer.
>
> Thank you for your time.



-- 
Adri?n Navarro / (+34) 608 831 094


From reallfqq-nginx at yahoo.fr  Sun Apr 22 02:11:14 2012
From: reallfqq-nginx at yahoo.fr (B.R.)
Date: Sat, 21 Apr 2012 22:11:14 -0400
Subject: Nginx rewrite to PHP
Message-ID: <CALqce=2NroC19YKUrh9UXWdvTVM=7c6ggsMCS+Kd_HbEPpHjjg@mail.gmail.com>

Hi,

I'm using Nginx linked to PHP using fastcgi proxy (to the FPM module of PHP
core btw).

I am trying to use some rewrite rules, which apparently work since I get
redirected (HTTP answer 200), but the content is not passed through the PHP
handler...
I am returned the file for downloading which content is the raw PHP code!

I am using a configuration like the following:

location / {
 rewrite ^/dunno/([^\.^/]+)$ /index.php?q=$1 break;
 return 404;
}

location ~ \.php$ {
 // PHP handling there
}

What am I doing wrong?
---
*B. R.*
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20120421/0b5d5da7/attachment-0001.html>

From edho at myconan.net  Sun Apr 22 02:17:34 2012
From: edho at myconan.net (Edho Arief)
Date: Sun, 22 Apr 2012 09:17:34 +0700
Subject: Nginx rewrite to PHP
In-Reply-To: <CALqce=2NroC19YKUrh9UXWdvTVM=7c6ggsMCS+Kd_HbEPpHjjg@mail.gmail.com>
References: <CALqce=2NroC19YKUrh9UXWdvTVM=7c6ggsMCS+Kd_HbEPpHjjg@mail.gmail.com>
Message-ID: <CAAtReCkkN_f4WShORmAtkRJ2VbmfVp7H2LD7Jbt2ABMbB1JinQ@mail.gmail.com>

2012/4/22 B.R. <reallfqq-nginx at yahoo.fr>:
> Hi,
>
> I'm using Nginx linked to PHP using fastcgi proxy (to the FPM module of PHP
> core btw).
>
> I am trying to use some rewrite rules, which apparently work since I get
> redirected (HTTP answer 200), but the content is not passed through the PHP
> handler...
> I am returned the file for downloading which content is the raw PHP code!
>
> I am using a configuration like the following:
>
> location / {
> ?rewrite ^/dunno/([^\.^/]+)$ /index.php?q=$1 break;
> ?return 404;
> }
>

break means the file is handled by current location block (location /
{ }). The one
you want is last.

reference: http://nginx.org/r/rewrite


From reallfqq-nginx at yahoo.fr  Sun Apr 22 02:23:00 2012
From: reallfqq-nginx at yahoo.fr (B.R.)
Date: Sat, 21 Apr 2012 22:23:00 -0400
Subject: Nginx rewrite to PHP
In-Reply-To: <CAAtReCkkN_f4WShORmAtkRJ2VbmfVp7H2LD7Jbt2ABMbB1JinQ@mail.gmail.com>
References: <CALqce=2NroC19YKUrh9UXWdvTVM=7c6ggsMCS+Kd_HbEPpHjjg@mail.gmail.com>
	<CAAtReCkkN_f4WShORmAtkRJ2VbmfVp7H2LD7Jbt2ABMbB1JinQ@mail.gmail.com>
Message-ID: <CALqce=0pgaJGt+0oe06QWc8Y+k33t=xN4gCi4i-1XNBEOG4HcQ@mail.gmail.com>

Thanks Edho,

I also use the official reference first-hand ;o)
>From the example given on the Wiki page of the Rewrite module you
quoted<http://wiki.nginx.org/HttpRewriteModule>,
they replace 'last' by 'break' when rewrite rules put inside a location
block...

Isn't it a problem to put 'last' flags inside the location block? What is
the 10-cycles-then-500-error referring to?
---
*B. R.*


On Sat, Apr 21, 2012 at 22:17, Edho Arief <edho at myconan.net> wrote:

> 2012/4/22 B.R. <reallfqq-nginx at yahoo.fr>:
> > Hi,
> >
> > I'm using Nginx linked to PHP using fastcgi proxy (to the FPM module of
> PHP
> > core btw).
> >
> > I am trying to use some rewrite rules, which apparently work since I get
> > redirected (HTTP answer 200), but the content is not passed through the
> PHP
> > handler...
> > I am returned the file for downloading which content is the raw PHP code!
> >
> > I am using a configuration like the following:
> >
> > location / {
> >  rewrite ^/dunno/([^\.^/]+)$ /index.php?q=$1 break;
> >  return 404;
> > }
> >
>
> break means the file is handled by current location block (location /
> { }). The one
> you want is last.
>
> reference: http://nginx.org/r/rewrite
>
> _______________________________________________
> 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/20120421/2b9816e7/attachment.html>

From edho at myconan.net  Sun Apr 22 02:32:32 2012
From: edho at myconan.net (Edho Arief)
Date: Sun, 22 Apr 2012 09:32:32 +0700
Subject: Nginx rewrite to PHP
In-Reply-To: <CALqce=0pgaJGt+0oe06QWc8Y+k33t=xN4gCi4i-1XNBEOG4HcQ@mail.gmail.com>
References: <CALqce=2NroC19YKUrh9UXWdvTVM=7c6ggsMCS+Kd_HbEPpHjjg@mail.gmail.com>
	<CAAtReCkkN_f4WShORmAtkRJ2VbmfVp7H2LD7Jbt2ABMbB1JinQ@mail.gmail.com>
	<CALqce=0pgaJGt+0oe06QWc8Y+k33t=xN4gCi4i-1XNBEOG4HcQ@mail.gmail.com>
Message-ID: <CAAtReCnzgFeNBvwcTAscoV8zfPD-bq0AohOGr1=LttsSOLyvVQ@mail.gmail.com>

2012/4/22 B.R. <reallfqq-nginx at yahoo.fr>:
> Thanks Edho,
>
> I also use the official reference first-hand ;o)
> From the example given on the Wiki page of the Rewrite module you quoted,
> they replace 'last' by 'break' when rewrite rules put inside a location
> block...
>
> Isn't it a problem to put 'last' flags inside the location block? What is
> the 10-cycles-then-500-error referring to?

The 500 error is cause by the rewrite: it rewrites /download/ (clean
url) to /download/ (actual file). If it use last flag, the /download/
location block will be searched again (and in fact it may return 403
instead of 500 if the rewritten url doesn't match the rewrite rule
anymore - it'll return 500 only on certain cases).

The one you want is for it to be handled by different location block
(\.php$) - hence last flag.


From reallfqq-nginx at yahoo.fr  Sun Apr 22 02:38:31 2012
From: reallfqq-nginx at yahoo.fr (B.R.)
Date: Sat, 21 Apr 2012 22:38:31 -0400
Subject: Nginx rewrite to PHP
In-Reply-To: <CAAtReCnzgFeNBvwcTAscoV8zfPD-bq0AohOGr1=LttsSOLyvVQ@mail.gmail.com>
References: <CALqce=2NroC19YKUrh9UXWdvTVM=7c6ggsMCS+Kd_HbEPpHjjg@mail.gmail.com>
	<CAAtReCkkN_f4WShORmAtkRJ2VbmfVp7H2LD7Jbt2ABMbB1JinQ@mail.gmail.com>
	<CALqce=0pgaJGt+0oe06QWc8Y+k33t=xN4gCi4i-1XNBEOG4HcQ@mail.gmail.com>
	<CAAtReCnzgFeNBvwcTAscoV8zfPD-bq0AohOGr1=LttsSOLyvVQ@mail.gmail.com>
Message-ID: <CALqce=3fXtovoA4zY3NBcs7UywqkvfHHh4eow95V3k6p0c+YWw@mail.gmail.com>

Well, I didn't take the required time to understand these examples...
Those are definitely hopefully not to be seen in real config files! ;o)

Thank you for having taken time to explain all that to me.
I just made the change and my requests end up at the right places.

Thanks again!
---
*B. R.*


On Sat, Apr 21, 2012 at 22:32, Edho Arief <edho at myconan.net> wrote:

> 2012/4/22 B.R. <reallfqq-nginx at yahoo.fr>:
> > Thanks Edho,
> >
> > I also use the official reference first-hand ;o)
> > From the example given on the Wiki page of the Rewrite module you quoted,
> > they replace 'last' by 'break' when rewrite rules put inside a location
> > block...
> >
> > Isn't it a problem to put 'last' flags inside the location block? What is
> > the 10-cycles-then-500-error referring to?
>
> The 500 error is cause by the rewrite: it rewrites /download/ (clean
> url) to /download/ (actual file). If it use last flag, the /download/
> location block will be searched again (and in fact it may return 403
> instead of 500 if the rewritten url doesn't match the rewrite rule
> anymore - it'll return 500 only on certain cases).
>
> The one you want is for it to be handled by different location block
> (\.php$) - hence last flag.
>
> _______________________________________________
> 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/20120421/3d85911b/attachment.html>

From agentzh at gmail.com  Sun Apr 22 03:31:53 2012
From: agentzh at gmail.com (agentzh)
Date: Sun, 22 Apr 2012 11:31:53 +0800
Subject: srcache + mogilefs module
In-Reply-To: <ff95461527422f4f9fc4db6d92d52ef3.NginxMailingListEnglish@forum.nginx.org>
References: <CAB4Tn6OeoKAAq-8GLvN973gn+f+NMaoD_Pd2kwh3rnvG6XnH5Q@mail.gmail.com>
	<ff95461527422f4f9fc4db6d92d52ef3.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <CAB4Tn6MHYc5C_muQJBipMP1+1VX9wr_Oe8arZDmmadj0Gjvi5g@mail.gmail.com>

On Sun, Apr 22, 2012 at 5:49 AM, haad <nginx-forum at nginx.us> wrote:
>> Basically, you should rely on ngx_srcache's
>> srcache_store to store the
>> responses automatically for you, rather than
>> inserting values into
>> memcached yourself.
>
> But from attached log or this new [1] dump it seems tat srcache_store is
> never called.
> I was not aware of this, it should be added to documentation :).
>

According to your debug log, the ngx_mogilefs module left your
location /share/ and did an internal redirect to an
automatically-generated location named /mogstored_spare_34991976/  or
so (maybe created by the mogilefs_pass directive):

[debug] 7200#0: *1058 internal redirect: "/mogstored_spare_34991976/?"

And you didn't configure ngx_srcache at all in this anonymous location
and hence no srcache_store love here.

AFAIK, the ngx_eval module also does something similar here (i.e.,
creating nested locations automatically).

> [1] https://gist.github.com/2439805
>

Just search for "internal redirect" in your debug log above.

Also, you should be careful when using the "if" directive in location
blocks because it also created a nested anonymous location. See the
discussion here:

http://agentzh.blogspot.com/2011/03/how-nginx-location-if-works.html

Best regards,
-agentzh


From nginx-forum at nginx.us  Sun Apr 22 04:04:18 2012
From: nginx-forum at nginx.us (ronin)
Date: Sun, 22 Apr 2012 00:04:18 -0400 (EDT)
Subject: File status about
Message-ID: <2c3c5394dd5b509a1a4a2328ebff1cc5.NginxMailingListEnglish@forum.nginx.org>

Web site is the Python, going to some type of the file will be forwarded
to the static files directory, the static directory should not return to
404, but there is a problem, the trouble you help me see
Web site configuration files:


        location ~* \.(php|php*|git*|htm|html|txt)$   {
                     root  /web/www/static;
                }

curl -I http://www.a.com/a.php
curl -I http://www.a.com/i.html

The result that come out: curl: (52) Empty reply from server 
why not 404?

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225612,225612#msg-225612


From nginx-forum at nginx.us  Sun Apr 22 06:14:43 2012
From: nginx-forum at nginx.us (haad)
Date: Sun, 22 Apr 2012 02:14:43 -0400 (EDT)
Subject: srcache + mogilefs module
In-Reply-To: <CAB4Tn6MHYc5C_muQJBipMP1+1VX9wr_Oe8arZDmmadj0Gjvi5g@mail.gmail.com>
References: <CAB4Tn6MHYc5C_muQJBipMP1+1VX9wr_Oe8arZDmmadj0Gjvi5g@mail.gmail.com>
Message-ID: <1ec69cd580ce8aedec369f4f05934c4e.NginxMailingListEnglish@forum.nginx.org>

agentzh Wrote:
-------------------------------------------------------
> On Sun, Apr 22, 2012 at 5:49 AM, haad
> <nginx-forum at nginx.us> wrote:
> >> Basically, you should rely on ngx_srcache's
> >> srcache_store to store the
> >> responses automatically for you, rather than
> >> inserting values into
> >> memcached yourself.
> >
> > But from attached log or this new [1] dump it
> seems tat srcache_store is
> > never called.
> > I was not aware of this, it should be added to
> documentation :).
> >
> 
> According to your debug log, the ngx_mogilefs
> module left your
> location /share/ and did an internal redirect to
> an
> automatically-generated location named
> /mogstored_spare_34991976/  or
> so (maybe created by the mogilefs_pass directive):
> 
> [debug] 7200#0: *1058 internal redirect:
> "/mogstored_spare_34991976/?"
> 
> And you didn't configure ngx_srcache at all in
> this anonymous location
> and hence no srcache_store love here.

So if I configure location with /mogstored and sets srcache_store there
it should be working ?

location ~ /mogstored_ {
   srcache_store PUT /memc
}

> AFAIK, the ngx_eval module also does something
> similar here (i.e.,
> creating nested locations automatically).

I'm not sure what you mean by this :) are you suggesting that 
I should use it ?

> 
> > [1] https://gist.github.com/2439805
> >
> 
> Just search for "internal redirect" in your debug
> log above.
> 
> Also, you should be careful when using the "if"
> directive in location
> blocks because it also created a nested anonymous
> location. See the
> discussion here:
> 
> http://agentzh.blogspot.com/2011/03/how-nginx-loca
> tion-if-works.html
> 
> Best regards,
> -agentzh
> 
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225568,225614#msg-225614


From tektonic_s at yahoo.com  Sun Apr 22 08:27:38 2012
From: tektonic_s at yahoo.com (Tektonic Tektonic)
Date: Sun, 22 Apr 2012 16:27:38 +0800 (SGT)
Subject: Rewrite Rules
Message-ID: <1335083258.34924.YahooMailNeo@web193102.mail.sg3.yahoo.com>

Hello,

New to this list so pardon me if I am asking a very common question (I am sure it's a common one though). It is about Rewrite Rules. We have a Magento website which use to respond slowly while working with Apache, so we decided to switch to Nginx and can see the difference already. However, the problem is the Apache Rewrite Rules which are causing all the links to redirect to "404 Not Found".

I have already tried more than a few solutions from Google but none worked.

Attached is the Rewrite Rules section of my .htaccess. Please check and suggest me a solution.

Regards.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20120422/7608e242/attachment.html>
-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: rewrite rules.txt
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20120422/7608e242/attachment.txt>

From nginx-forum at nginx.us  Sun Apr 22 08:51:32 2012
From: nginx-forum at nginx.us (xxx)
Date: Sun, 22 Apr 2012 04:51:32 -0400 (EDT)
Subject: Rewrite in Ngnix
In-Reply-To: <fe6fc5606f7827d45aed72ae1d52b5cc.NginxMailingListEnglish@forum.nginx.org>
References: <fe6fc5606f7827d45aed72ae1d52b5cc.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <e62d678ad04e1600dd968ee364cfb3de.NginxMailingListEnglish@forum.nginx.org>

Please help me, i can not run my nginx because i haven't good rewrite
rules

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225583,225616#msg-225616


From contact at jpluscplusm.com  Sun Apr 22 09:39:34 2012
From: contact at jpluscplusm.com (Jonathan Matthews)
Date: Sun, 22 Apr 2012 10:39:34 +0100
Subject: Rewrite in Ngnix
In-Reply-To: <e62d678ad04e1600dd968ee364cfb3de.NginxMailingListEnglish@forum.nginx.org>
References: <fe6fc5606f7827d45aed72ae1d52b5cc.NginxMailingListEnglish@forum.nginx.org>
	<e62d678ad04e1600dd968ee364cfb3de.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <CAKsTx7AC34CTJdiKeCNGYW8byFvg1FRYHV67_y2_h=pyPsYJwA@mail.gmail.com>

No. You must try translating your Apache rewrite rules to start with.
Show us what you try, and where you get stuck - then someone may be
willing to help you understand what you've got wrong.

No-one here should just do the whole thing for you if you're too lazy
even to try for yourself.

Jonathan
-- 
Jonathan Matthews
Oxford, London, UK
http://www.jpluscplusm.com/contact.html


From zzz at zzz.org.ua  Sun Apr 22 15:49:16 2012
From: zzz at zzz.org.ua (Alexandr Gomoliako)
Date: Sun, 22 Apr 2012 18:49:16 +0300
Subject: websockets, nginx-perl
Message-ID: <CAKb9G9hnZvHM6CeG7+4crYgZJD-y7G15y58Sg5Pg2p-oUPCq1w@mail.gmail.com>

Just a couple of examples on websockets in nginx-perl:

    https://github.com/zzzcpan/nginx-perl/tree/master/eg/websockets/lib/Websockets

Draw.pm - multiplayer drawing thingy
Echo.pm - echo server

Demo (works only in chrome and firefox 11):
   zzz.org.ua:55555/echo/
   zzz.org.ua:55555/draw/


From lists at ruby-forum.com  Sun Apr 22 16:48:06 2012
From: lists at ruby-forum.com (Jakub)
Date: Sun, 22 Apr 2012 18:48:06 +0200
Subject: NGINX and wrong memcached output
Message-ID: <89b67bf90e128310fea0429f089719aa@ruby-forum.com>

Hi,

At first let me tell my application was developped under PHP with
Symfony2, but I guess my problem is about nginx and memcached and not
about PHP. As I can see here is good community about nginx and I will be
glad for any help.

I have problem with nginx and memcached responses such as text/html or
application/json. My idea is:
-First request on nginx server (reverse proxy memcached server)
-Try if I have memcached response (memcached run on same server as
nginx)
-If not go to application server
-Application server makes response
-Stores response in to memcached
-And for next same request nginx hit response from memcached and return
to client.

But the problem is, when the application server stores response to
memcached (data in memcached are stored good and corectly) and NGINX hit
the request in memcached, but return totaly wrong hashed content.

For example:
First request, ngingx didnt find response in memcached, pass to
application server, and here is response ... good and correct JSON.
------- Headers
HTTP/1.0 200 OK
Date: Sun, 22 Apr 2012 16:13:58 GMT
Server: Apache/2.2.16 (Debian)
X-Powered-By: PHP/5.3.3-7+squeeze8
cache-control: no-cache
x-debug-token: 4f942e4dee10b
Content-Length: 476
Connection: close
Content-Type: application/json
-------- Content
[{"id":10,"title":"Cinema","url":"cinema","articles":null},{"id":2,"title":"Computers","url":"computer","articles":null},{"id":8,"title":"Foto","url":"foto","articles":null},{"id":12,"title":"Literature","url":"literature","articles":null},{"id":4,"title":"Music","url":"music","articles":null},{"id":18,"title":"Society","url":"society","articles":null},{"id":16,"title":"Sport","url":"sport","articles":null},{"id":6,"title":"Travelling","url":"travelling","articles":null}]

But for the second request, when NGINX find response in memcached, NGINX
serves totaly wrong.
-------- Headers
Connection:keep-alive
Content-Length:164
Content-Type:application/json
Date:Sun, 22 Apr 2012 16:17:15 GMT
Server:nginx/0.7.67
-------- Content
x?u?1?0??rs+"?Up?I???)??$A??n?59?????5o ?zU??0??=u?Z?
??E?c?%?O5??????`??N???}??3??a?????y+)`?????M8?H.?vJ???g???e?y?n??}?"?4???Df??????_?
?

With text/html I have exactly the same problem. Do you have anybody any
idea plese? Here is my nginx configuration:

# nginx virtual host from sites-available
server {
    listen          80;
    server_name     dp-xskrha.local;
    access_log    /var/log/nginx/dp-xskrha/access.log;
    error_log       /var/log/nginx/dp-xskrha/error.log;

    # Static content, images, css, js
    location ~*
^.+\.(jpe?g|gif|png|ico|css|zip|tgz|gz|rar|bz2|doc|xls|exe|pdf|ppt|txt|tar|mid|midi|wav|bmp|rtf|js|swf|avi|mp3)$
{
        proxy_redirect          off;
        proxy_pass              http://app.dp-xskrha.local:80;
        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_cache             dp-xskrha-static;
        roxy_cache_valid        200  10m;
    }

    # Model api json
    location /model-api {
        default_type          application/json;
        set $memcached_key    $uri;
        memcached_pass        127.0.0.1:11211;
        error_page            404 =200 @fallback;
    }

    # Dynamic content, all rest uri
    location / {
        default_type            text/html;
        charset                 utf-8;
        set $memcached_key      $uri;
        memcached_pass          127.0.0.1:11211;
        error_page              404 =200 @fallback;
      }

    # Memcached fallback for proxy pass
    location @fallback {
        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_redirect             off;
        proxy_pass                 http://app.dp-xskrha.local:80;
      }
}


#nginx.conf
user www-data;
worker_processes  1;

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

events {
    worker_connections  1024;
    multi_accept on;
}

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

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

    sendfile        on;
    tcp_nopush     on;

    #keepalive_timeout  0;
    keepalive_timeout  65;
    tcp_nodelay        on;

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

    # proxy settings
    proxy_cache_path  /data/nginx/cache  levels=1:2
keys_zone=dp-xskrha-static:10m
                                         inactive=24h  max_size=1g;

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

Thank you very much, for Your response :)

Jacob

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

From serga at seznam.cz  Sun Apr 22 17:02:33 2012
From: serga at seznam.cz (=?windows-1252?Q?Jakub_=8Akrha?=)
Date: Sun, 22 Apr 2012 19:02:33 +0200
Subject: Nginx and memcached response
Message-ID: <4F9439A9.3090306@seznam.cz>

An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20120422/25d1cc17/attachment.html>

From bpaquet at octo.com  Sun Apr 22 17:17:29 2012
From: bpaquet at octo.com (Bertrand Paquet)
Date: Sun, 22 Apr 2012 19:17:29 +0200
Subject: Nginx and memcached response
In-Reply-To: <4F9439A9.3090306@seznam.cz>
References: <4F9439A9.3090306@seznam.cz>
Message-ID: <CAPs0E=02iCkH4bGM6UqryPNS3cqhvxGp4SWt7NDAe6FNFeb1Jg@mail.gmail.com>

Hi,

Does your application server fill memcached correctly ? May be a binary /
encoding problem when connecting to memcached. Or the content is stored
gzipped ?

Regards

Bertrand


On Sun, Apr 22, 2012 at 19:02, Jakub ?krha <serga at seznam.cz> wrote:

> **
> Hi,
>
> I have problem with nginx and memcached responses such as text/html or
> application/json and I will be glad for any help. Thanks.
>
> My idea is:
> -First request on nginx server (reverse proxy memcached server)
> -Try if I have memcached response (memcached run on same server as
> nginx)
> -If not pass request to application server
> -Application server makes response
> -Stores response in to memcached
> -And for next same request nginx hit response from memcached and return
> to client.
>
> But the problem is, when the application server stores response to
> memcached (data in memcached are stored good and corectly) and NGINX hit
> the request in memcached, but return totaly wrong hashed content.
>
> For example:
> First request, ngingx didnt find response in memcached, pass to
> application server, and here is response ... good and correct JSON.
> ------- Headers
> HTTP/1.0 200 OK
> Date: Sun, 22 Apr 2012 16:13:58 GMT
> Server: Apache/2.2.16 (Debian)
> X-Powered-By: PHP/5.3.3-7+squeeze8
> cache-control: no-cache
> x-debug-token: 4f942e4dee10b
> Content-Length: 476
> Connection: close
> Content-Type: application/json
> -------- Content
>
> [{"id":10,"title":"Cinema","url":"cinema","articles":null},{"id":2,"title":"Computers","url":"computer","articles":null},{"id":8,"title":"Foto","url":"foto","articles":null},{"id":12,"title":"Literature","url":"literature","articles":null},{"id":4,"title":"Music","url":"music","articles":null},{"id":18,"title":"Society","url":"society","articles":null},{"id":16,"title":"Sport","url":"sport","articles":null},{"id":6,"title":"Travelling","url":"travelling","articles":null}]
>
> But for the second request, when NGINX find response in memcached, NGINX
> serves response totaly wrong.
> -------- Headers
> Connection:keep-alive
> Content-Length:164
> Content-Type:application/json
> Date:Sun, 22 Apr 2012 16:17:15 GMT
> Server:nginx/0.7.67
> -------- Content
> x?u?1?0??rs+"?Up?I???)??$A??n?59?????5o ?zU 0? =u?Z?
> ??E?c?%?O5??????`??N???}??3??a?????y+)`?????M8?H.?vJ???g???e?y?n??}
> "?4???Df??????_?
> ?
>
> With text/html I have exactly the same problem. Do you have anybody any
> idea plese? Here is my nginx configuration:
>
> # nginx virtual host from sites-available linked to sites-enabled
> server {
>     listen          80;
>     server_name     dp-xskrha.local;
>     access_log      /var/log/nginx/dp-xskrha/access.log;
>     error_log       /var/log/nginx/dp-xskrha/error.log;
>
>     # Static content, images, css, js
>     location ~*
> ^.+\.(jpe?g|gif|png|ico|css|zip|tgz|gz|rar|bz2|doc|xls|exe|pdf|ppt|txt|tar|mid|midi|wav|bmp|rtf|js|swf|avi|mp3)$
>
> {
>         proxy_redirect          off;
>         proxy_pass              http://app.dp-xskrha.local:80;
>         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_cache             dp-xskrha-static;
>         roxy_cache_valid        200  10m;
>     }
>
>     # Model api json
>     location /model-api {
>         default_type             application/json;
>         set $memcached_key       $uri;
>         memcached_pass           127.0.0.1:11211;
>         error_page               404 =200 @fallback;
>     }
>
>     # Dynamic content, all rest uri
>     location / {
>         default_type            text/html;
>         charset                 utf-8;
>         set $memcached_key      $uri;
>         memcached_pass          127.0.0.1:11211;
>         error_page              404 =200 @fallback;
>       }
>
>     # Memcached fallback for proxy pass
>     location @fallback {
>         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_redirect             off;
>         proxy_pass                 http://app.dp-xskrha.local:80;
>       }
> }
>
>
> #nginx.conf
> user www-data;
> worker_processes  1;
>
> error_log  /var/log/nginx/error.log;
> pid        /var/run/nginx.pid;
>
> events {
>     worker_connections  1024;
>     multi_accept on;
> }
>
> http {
>     include       /etc/nginx/mime.types;
>
>     access_log    /var/log/nginx/access.log;
>
>     sendfile        on;
>     tcp_nopush     on;
>
>     #keepalive_timeout  0;
>     keepalive_timeout  65;
>     tcp_nodelay        on;
>
>     gzip  on;
>     gzip_disable "MSIE [1-6]\.(?!.*SV1)";
>
>     # proxy settings
>     proxy_cache_path  /data/nginx/cache  levels=1:2
> keys_zone=dp-xskrha-static:10m
>                                          inactive=24h  max_size=1g;
>
>     include /etc/nginx/conf.d/*.conf;
>     include /etc/nginx/sites-enabled/*;
> }
>
> Thank you very much, for Your response :)
>
> Jacob
>
> _______________________________________________
> 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/20120422/7414b3aa/attachment-0001.html>

From serga at seznam.cz  Sun Apr 22 17:39:41 2012
From: serga at seznam.cz (=?windows-1252?Q?Jakub_=8Akrha?=)
Date: Sun, 22 Apr 2012 19:39:41 +0200
Subject: Nginx and memcached response
In-Reply-To: <mailman.872.1335115059.79859.nginx@nginx.org>
References: <mailman.872.1335115059.79859.nginx@nginx.org>
Message-ID: <4F94425D.5050607@seznam.cz>

On 22.4.2012 19:17, nginx-request at nginx.org wrote:
> ------------------------------
>
> Message: 2
> Date: Sun, 22 Apr 2012 18:48:06 +0200
> From: Jakub<lists at ruby-forum.com>
> To: nginx at nginx.org
> Subject: NGINX and wrong memcached output
> Message-ID:<89b67bf90e128310fea0429f089719aa at ruby-forum.com>
> Content-Type: text/plain; charset=UTF-8
>
> Hi,
>
> At first let me tell my application was developped under PHP with
> Symfony2, but I guess my problem is about nginx and memcached and not
> about PHP. As I can see here is good community about nginx and I will be
> glad for any help.
>
> I have problem with nginx and memcached responses such as text/html or
> application/json. My idea is:
> -First request on nginx server (reverse proxy memcached server)
> -Try if I have memcached response (memcached run on same server as
> nginx)
> -If not go to application server
> -Application server makes response
> -Stores response in to memcached
> -And for next same request nginx hit response from memcached and return
> to client.
>
> But the problem is, when the application server stores response to
> memcached (data in memcached are stored good and corectly) and NGINX hit
> the request in memcached, but return totaly wrong hashed content.
>
> For example:
> First request, ngingx didnt find response in memcached, pass to
> application server, and here is response ... good and correct JSON.
> ------- Headers
> HTTP/1.0 200 OK
> Date: Sun, 22 Apr 2012 16:13:58 GMT
> Server: Apache/2.2.16 (Debian)
> X-Powered-By: PHP/5.3.3-7+squeeze8
> cache-control: no-cache
> x-debug-token: 4f942e4dee10b
> Content-Length: 476
> Connection: close
> Content-Type: application/json
> -------- Content
> [{"id":10,"title":"Cinema","url":"cinema","articles":null},{"id":2,"title":"Computers","url":"computer","articles":null},{"id":8,"title":"Foto","url":"foto","articles":null},{"id":12,"title":"Literature","url":"literature","articles":null},{"id":4,"title":"Music","url":"music","articles":null},{"id":18,"title":"Society","url":"society","articles":null},{"id":16,"title":"Sport","url":"sport","articles":null},{"id":6,"title":"Travelling","url":"travelling","articles":null}]
>
> But for the second request, when NGINX find response in memcached, NGINX
> serves totaly wrong.
> -------- Headers
> Connection:keep-alive
> Content-Length:164
> Content-Type:application/json
> Date:Sun, 22 Apr 2012 16:17:15 GMT
> Server:nginx/0.7.67
> -------- Content
> x?u?1?0??rs+"?Up?I???)??$A??n?59?????5o ?zU??0??=u?Z?
> ??E?c?%?O5??????`??N???}??3??a?????y+)`?????M8?H.?vJ???g???e?y?n??}?"?4???Df??????_?
> ?
>
> With text/html I have exactly the same problem. Do you have anybody any
> idea plese? Here is my nginx configuration:
>
> # nginx virtual host from sites-available
> server {
>      listen          80;
>      server_name     dp-xskrha.local;
>      access_log    /var/log/nginx/dp-xskrha/access.log;
>      error_log       /var/log/nginx/dp-xskrha/error.log;
>
>      # Static content, images, css, js
>      location ~*
> ^.+\.(jpe?g|gif|png|ico|css|zip|tgz|gz|rar|bz2|doc|xls|exe|pdf|ppt|txt|tar|mid|midi|wav|bmp|rtf|js|swf|avi|mp3)$
> {
>          proxy_redirect          off;
>          proxy_pass              http://app.dp-xskrha.local:80;
>          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_cache             dp-xskrha-static;
>          roxy_cache_valid        200  10m;
>      }
>
>      # Model api json
>      location /model-api {
>          default_type          application/json;
>          set $memcached_key    $uri;
>          memcached_pass        127.0.0.1:11211;
>          error_page            404 =200 @fallback;
>      }
>
>      # Dynamic content, all rest uri
>      location / {
>          default_type            text/html;
>          charset                 utf-8;
>          set $memcached_key      $uri;
>          memcached_pass          127.0.0.1:11211;
>          error_page              404 =200 @fallback;
>        }
>
>      # Memcached fallback for proxy pass
>      location @fallback {
>          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_redirect             off;
>          proxy_pass                 http://app.dp-xskrha.local:80;
>        }
> }
>
>
> #nginx.conf
> user www-data;
> worker_processes  1;
>
> error_log  /var/log/nginx/error.log;
> pid        /var/run/nginx.pid;
>
> events {
>      worker_connections  1024;
>      multi_accept on;
> }
>
> http {
>      include       /etc/nginx/mime.types;
>
>      access_log    /var/log/nginx/access.log;
>
>      sendfile        on;
>      tcp_nopush     on;
>
>      #keepalive_timeout  0;
>      keepalive_timeout  65;
>      tcp_nodelay        on;
>
>      gzip  on;
>      gzip_disable "MSIE [1-6]\.(?!.*SV1)";
>
>      # proxy settings
>      proxy_cache_path  /data/nginx/cache  levels=1:2
> keys_zone=dp-xskrha-static:10m
>                                           inactive=24h  max_size=1g;
>
>      include /etc/nginx/conf.d/*.conf;
>      include /etc/nginx/sites-enabled/*;
> }
>
> Thank you very much, for Your response :)
>
> Jacob
>
Date: Sun, 22 Apr 2012 19:17:29 +0200
From: Bertrand Paquet<bpaquet at octo.com>
To:nginx at nginx.org
Subject: Re: Nginx and memcached response
Message-ID:
	<CAPs0E=02iCkH4bGM6UqryPNS3cqhvxGp4SWt7NDAe6FNFeb1Jg at mail.gmail.com>
Content-Type: text/plain; charset="utf-8"

Hi,

Does your application server fill memcached correctly ? May be a binary /
encoding problem when connecting to memcached. Or the content is stored
gzipped ?

Regards

Bertrand


Hi,

yes the server fill memcached correctly. For example, if I store on 
simple JSON value, it stores correctly. Or when I store plain text, 
there is not any problem. Just if store to memcached some content with 
headers. Maybe there is problem with headers. I will try some other 
memcached client than the the application server is using. But I am not 
sure if the problem is in store process. I think there is a problem with 
nginx read proces.

Thank you very much

Jacob


From mdounin at mdounin.ru  Sun Apr 22 20:49:48 2012
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Mon, 23 Apr 2012 00:49:48 +0400
Subject: NGINX and wrong memcached output
In-Reply-To: <89b67bf90e128310fea0429f089719aa@ruby-forum.com>
References: <89b67bf90e128310fea0429f089719aa@ruby-forum.com>
Message-ID: <20120422204948.GO13466@mdounin.ru>

Hello!

On Sun, Apr 22, 2012 at 06:48:06PM +0200, Jakub wrote:

> Hi,
> 
> At first let me tell my application was developped under PHP with
> Symfony2, but I guess my problem is about nginx and memcached and not
> about PHP. As I can see here is good community about nginx and I will be
> glad for any help.
> 
> I have problem with nginx and memcached responses such as text/html or
> application/json. My idea is:
> -First request on nginx server (reverse proxy memcached server)
> -Try if I have memcached response (memcached run on same server as
> nginx)
> -If not go to application server
> -Application server makes response
> -Stores response in to memcached
> -And for next same request nginx hit response from memcached and return
> to client.
> 
> But the problem is, when the application server stores response to
> memcached (data in memcached are stored good and corectly) and NGINX hit
> the request in memcached, but return totaly wrong hashed content.
> 
> For example:
> First request, ngingx didnt find response in memcached, pass to
> application server, and here is response ... good and correct JSON.
> ------- Headers
> HTTP/1.0 200 OK
> Date: Sun, 22 Apr 2012 16:13:58 GMT
> Server: Apache/2.2.16 (Debian)
> X-Powered-By: PHP/5.3.3-7+squeeze8
> cache-control: no-cache
> x-debug-token: 4f942e4dee10b
> Content-Length: 476
> Connection: close
> Content-Type: application/json
> -------- Content
> [{"id":10,"title":"Cinema","url":"cinema","articles":null},{"id":2,"title":"Computers","url":"computer","articles":null},{"id":8,"title":"Foto","url":"foto","articles":null},{"id":12,"title":"Literature","url":"literature","articles":null},{"id":4,"title":"Music","url":"music","articles":null},{"id":18,"title":"Society","url":"society","articles":null},{"id":16,"title":"Sport","url":"sport","articles":null},{"id":6,"title":"Travelling","url":"travelling","articles":null}]
> 
> But for the second request, when NGINX find response in memcached, NGINX
> serves totaly wrong.
> -------- Headers
> Connection:keep-alive
> Content-Length:164
> Content-Type:application/json
> Date:Sun, 22 Apr 2012 16:17:15 GMT
> Server:nginx/0.7.67
> -------- Content
> x?u?1?0??rs+"?Up?I???)??$A??n?59?????5o ?zU??0??=u?Z?
> ??E?c?%?O5??????`??N???}??3??a?????y+)`?????M8?H.?vJ???g???e?y?n??}?"?4???Df??????_?
> ?
> 
> With text/html I have exactly the same problem. Do you have anybody any
> idea plese? Here is my nginx configuration:

Content stored in memcached is compressed, and nginx just returns 
what you have in memcached.  You have to read docs on your 
memcached client to instruct it to don't do any 
compression/serialization and store raw data instead.

Maxim Dounin


From tdgh2323 at hotmail.com  Sun Apr 22 20:59:51 2012
From: tdgh2323 at hotmail.com (Micheal Wolfskill)
Date: Sun, 22 Apr 2012 20:59:51 +0000
Subject: Way to list current requestsw being served?
Message-ID: <BLU144-W106996693255CAB38CB30ED0200@phx.gbl>


Hello. Aside from the status page of nginx... is there any way i cant print the current REQUESTS it is attending .. like apaches status module it lists the URIs being served. Thanks

--Mike
 		 	   		  
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20120422/b734e5a0/attachment.html>

From tdgh2323 at hotmail.com  Sun Apr 22 21:09:07 2012
From: tdgh2323 at hotmail.com (Sam Ashmann)
Date: Sun, 22 Apr 2012 21:09:07 +0000
Subject: How to block non local refferers to an object?
Message-ID: <BLU144-W274AA80D65E009636733CFD0200@phx.gbl>


Hello Community,

Could anybody please give me an example (or point me to such) on best way to block non local referrers to my image files (jpg, jpeg, gif)

Thanks.. 

Regards,
Sam
 		 	   		  
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20120422/9bfaeced/attachment.html>

From tdgh2323 at hotmail.com  Sun Apr 22 21:22:22 2012
From: tdgh2323 at hotmail.com (Sam Ashmann)
Date: Sun, 22 Apr 2012 21:22:22 +0000
Subject: stub_status and other data to be displayed.
Message-ID: <BLU144-W3754338C2404936713DA86D0200@phx.gbl>


Hello all,

I made a debug html that displays some debugging information (additional nginx variables)... but Id also prefer to print along with the output produced by stub_status . Is there anyway to do this?

Something like this doesnt appear to work:


location  /static-test { 
                    default_type text/html;
                    stub_status on;
                    return 200 "test"                        
                    }

it either prints my test, or if I comment it .. it prints the status of nginx.


Regards,
Sam

 		 	   		  
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20120422/09f58cea/attachment.html>

From appa at perusio.net  Sun Apr 22 21:26:12 2012
From: appa at perusio.net (=?UTF-8?B?QW50w7NuaW8=?= P. P. Almeida)
Date: Sun, 22 Apr 2012 23:26:12 +0200
Subject: How to block non local refferers to an object?
In-Reply-To: <BLU144-W274AA80D65E009636733CFD0200@phx.gbl>
References: <BLU144-W274AA80D65E009636733CFD0200@phx.gbl>
Message-ID: <87ehrfxxu3.wl%appa@perusio.net>

On 22 Abr 2012 23h09 CEST, tdgh2323 at hotmail.com wrote:


>
> Hello Community,
>
> Could anybody please give me an example (or point me to such) on
> best way to block non local referrers to my image files (jpg, jpeg,
> gif)
>
> Thanks.. 

location ~* \.(?:gif|jpe?g|png)$ {
    valid_referers none blocked *.example.* *.google.* my.site.com;

    if ($invalid_referer) {
       return 200 "No image hotlinking allowed!\n";
    }
}

Cf. http://nginx.org/en/docs/http/ngx_http_referer_module.html

--- appa


From tdgh2323 at hotmail.com  Sun Apr 22 21:49:12 2012
From: tdgh2323 at hotmail.com (Joseph Cabezas)
Date: Sun, 22 Apr 2012 21:49:12 +0000
Subject: How to log only if $upstream_cache_status is not equal to "HIT"
Message-ID: <BLU144-W44F34B844411D33EE4296D0200@phx.gbl>


Hello,

Iam trying to find out if there is a way I can have an access logs only for the objects that did not produce a HIT from the cache (recorded in the nginx $upstream_cache_status var)

Is this possible?

Thanks and regards,
Joseph
 		 	   		  
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20120422/b5fe1485/attachment.html>

From lists at ruby-forum.com  Sun Apr 22 22:02:11 2012
From: lists at ruby-forum.com (Jakub)
Date: Mon, 23 Apr 2012 00:02:11 +0200
Subject: NGINX and wrong memcached output
In-Reply-To: <89b67bf90e128310fea0429f089719aa@ruby-forum.com>
References: <89b67bf90e128310fea0429f089719aa@ruby-forum.com>
Message-ID: <e30149a46676dde646ab05278396a708@ruby-forum.com>

Hi Maxim,

yes, you are right. That was the problem. I have just solved it and my 
app works correctly now. Thank you very much, for you response. And 
NGINX is strong tool :)

Regards,

Jacob

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


From nginx-forum at nginx.us  Mon Apr 23 02:25:16 2012
From: nginx-forum at nginx.us (JCR)
Date: Sun, 22 Apr 2012 22:25:16 -0400 (EDT)
Subject: upload module fails in some cases
Message-ID: <0b5caab894de6a90d1d03139c7133095.NginxMailingListEnglish@forum.nginx.org>

Hello,

I and using nginx 1.0.15 with the third-party module
nginx_upload_module-2.2.0/

When I am using an html form all works perfectly and I can see the
uploaded files and the post values as expected. 

But, when I send the HTTP Post request from a Poco application, it does
not work. (pocoproject.org)

I am pasting below the code of that C++ Poco app. I don't think that you
need to be an expert at Poco to understand it. 


114   std::ostringstream content;
115   content << "------WebKitFormBoundaryq2BIJhJChD7qIY1W\r\n";
116   content << "Content-Disposition: form-data;
name=\"submit\"\r\n\r\n"
117              "Upload\r\n"
118              "------WebKitFormBoundaryq2BIJhJChD7qIY1W\r\n";
119   content << "Content-Disposition: form-data; name=\"file\";
filename=\"test.txt\"\r\n\r\n"
120              "Content-Type: text/plain\r\n"
121              "This is a test.\r\n"
122              "------WebKitFormBoundaryq2BIJhJChD7qIY1W--";
123 
124   std::ostringstream os;
125   os << "POST /upload HTTP/1.0\r\n"
126         "Host: ip:31000\r\n"
127         "User-Agent: AAA\r\n"
128         "Accept:
text/plain,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8\r\n"
129         "Accept-Language:
en-us,ko;q=0.9,ja;q=0.7,cs;q=0.6,th;q=0.4,zh;q=0.3,en;q=0.1\r\n"
130         "Accept-Encoding: gzip,deflate\r\n"
131         "Accept-Charset: ISO-8859-1,utf-8;q=0.7,*;q=0.7\r\n"
132         "Referer: ip\r\n"
133         "Content-Type: multipart/form-data;
boundary=------WebKitFormBoundaryq2BIJhJChD7qIY1W\r\n";
134 
135   os << "Content-Length: " << strlen(content.str().c_str()) <<
"\r\n\r\n";
136   os << content.str();
137 
138   Poco::Net::SocketAddress sa("ip", 31000);
139   Poco::Net::StreamSocket socket(sa);
140   Poco::Net::SocketStream str(socket);
141   str << os.str();
142   str.flush();
143   Poco::StreamCopier::copyStream(str, std::cout);
144   return;

I am getting this from the server
HTTP/1.1 200 OK
Server: nginx/1.0.15
Date: Mon, 23 Apr 2012 02:00:45 GMT
Content-Type: text/html
Connection: close
X-Powered-By: PHP/5.1.6
Content-Length: 98

<br>POST<br>Array
(
    [<ngx_upload_module_dummy>] => 
)
<br>FILES<br>Array
(
)
<br><br><br><br>



Drilling into 
nginx_upload_module-2.2.0/ngx_http_upload_module.c
It seems that the problem is inside the function 
 static ngx_int_t upload_process_buf(ngx_http_upload_ctx_t *upload_ctx,
u_char *start, u_char *end)
and more precisely, it appears that from the html from the scope below
is entered while from the poco app it is never entered

3509      if(upload_ctx->boundary_pos == upload_ctx->boundary.data +
upload_ctx->boundary.len)
3510      { 
3512        upload_ctx->state = upload_state_after_boundary;
3513        upload_ctx->boundary_start = upload_ctx->boundary.data;
3514        upload_ctx->boundary_pos = upload_ctx->boundary_start;
3515      }

maybe my header is wrong... I am running out of ideas to make this work

I am pasting below the config file:
  1 #user  nobody;
  2 worker_processes  1;
  3 #error_log  logs/error.log;
  4 error_log  logs/debug.log  debug;
  5 #error_log  logs/error.log  info;
  6 #pid        logs/nginx.pid;
  7 events {
  8     worker_connections  1024;
  9 }
 10 http {
 11 
 12     proxy_buffering on;
 13     proxy_buffer_size 8k;
 14     proxy_buffers 2048 8k;
 15 
 16     include       mime.types;
 17     default_type  application/octet-stream;
 18     log_format  main  '$request_body  $remote_addr - $remote_user
[$time_local] "$request" '
 19                       '$status $body_bytes_sent "$http_referer" '
 20                       '"$http_user_agent" "$http_x_forwarded_for"';
 21     access_log  logs/access.log  main;
 22     sendfile        on;
 23     server_names_hash_max_size 1024;
 24     server_names_hash_bucket_size 128;
 25     client_body_temp_path /usr/local/nginx/uploads;
 26     keepalive_timeout  65;
 27 
 28     chunkin on;
 29 
 30     server { 
 31         client_max_body_size 100m;
 32         listen      ip:31000;
 33 
 34         #location /upload.php {
 35         #location ~ \.php$ { 
 36      
 37         error_page 411 = @my_error;
 38         location @my_error {
 39           chunkin_resume;
 40         }
 41         location /upload{
 42 
 43           chunkin_resume;
 44 
 45           # Pass altered request body to this location
 46           upload_pass  @uploadfiles;
 47           upload_store /usr/local/nginx/uploads;
 48           #upload_store_access user:r;
 49           # Set specified fields in request body
 50           upload_set_form_field $upload_field_name.name
"$upload_file_name";
 51           upload_set_form_field $upload_field_name.content_type
"$upload_content_type";
 52           upload_set_form_field $upload_field_name.path
"$upload_tmp_path";
 53           # Inform backend about hash and size of a file
 54           #upload_aggregate_form_field "$upload_field_name.md5"
"$upload_file_md5";
 55           upload_aggregate_form_field "$upload_field_name.size"
"$upload_file_size";
 56           upload_pass_form_field "^username$|^password$";
 57           #upload_pass_form_field ".*";
 58           upload_cleanup 400 404 499 500-505;
 59           upload_pass_args on;
 60           upload_max_part_header_len 100m;
 61           upload_max_output_body_len 0;
 62         }
 63         location ~ \.php$ {
 64             proxy_pass   http://127.0.0.1:80;
 65         }
 66         location = /favicon.ico {
 67           access_log off;
 68           log_not_found off;
 69         }
 70         location /hello{
 71           hello;
 72         }
 73         location @uploadfiles{
 74           rewrite ^ /upload.php last;
 75           #proxy_pass   http://127.0.0.1:80;
 76         }
 77         location = /boom{
 78         }
 79         #location / {
 80         #    root   html;
 81         #    index  index.html index.htm;
 82         #}
 83         error_page   500 502 503 504  /50x.html;
 84         #location = /50x.html {
 85         #    root   html;
 86         #}
 87     }
 88 }

Note that if I disable the module
'--add-module=/root/nginx/agentzh-chunkin-nginx-module-ddc0dd5", the
issue remains

Thanks in advance for any idea

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225636,225636#msg-225636


From agentzh at gmail.com  Mon Apr 23 03:41:17 2012
From: agentzh at gmail.com (agentzh)
Date: Mon, 23 Apr 2012 11:41:17 +0800
Subject: srcache + mogilefs module
In-Reply-To: <1ec69cd580ce8aedec369f4f05934c4e.NginxMailingListEnglish@forum.nginx.org>
References: <CAB4Tn6MHYc5C_muQJBipMP1+1VX9wr_Oe8arZDmmadj0Gjvi5g@mail.gmail.com>
	<1ec69cd580ce8aedec369f4f05934c4e.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <CAB4Tn6O22e7Fsta92hy+6S0sOekBo7DdtpFetNU9ZC+QSiRcMw@mail.gmail.com>

On Sun, Apr 22, 2012 at 2:14 PM, haad <nginx-forum at nginx.us> wrote:
>
> So if I configure location with /mogstored and sets srcache_store there
> it should be working ?
>
> location ~ /mogstored_ {
> ? srcache_store PUT /memc
> }
>

No, the /mogstored_spare_34991976/ location is automatically generated
by ngx_mogilefs. I'm not sure if it helps by putting your
srcache_store configuration into the "anonymous" location block under
the mogilefs_pass directive, i.e.,

           mogilefs_pass {
               proxy_pass $mogilefs_path;
               proxy_hide_header Content-Type;
               proxy_buffering off;
               srcache_store PUT /memc;
           }

But you can try it out anyway.

>> AFAIK, the ngx_eval module also does something
>> similar here (i.e.,
>> creating nested locations automatically).
>
> I'm not sure what you mean by this :) are you suggesting that
> I should use it ?
>

No.

I was just talking about the similarity in their implementations.

Regards,
-agentzh


From sammyraul1 at gmail.com  Mon Apr 23 04:26:37 2012
From: sammyraul1 at gmail.com (sammy_raul)
Date: Sun, 22 Apr 2012 21:26:37 -0700 (PDT)
Subject: Registering Nginx callback in external library
In-Reply-To: <CAB4Tn6M857GC50sHT4+T1qE-9MWaXPgYAjMKMrPCLJcnQw6X+A@mail.gmail.com>
References: <CACdz456WY-M6zF3tXz0ajCf_GXbeBLVNTC4pN3MX9wh_hc4LLw@mail.gmail.com>
	<CAB4Tn6M857GC50sHT4+T1qE-9MWaXPgYAjMKMrPCLJcnQw6X+A@mail.gmail.com>
Message-ID: <1335155197546-7490644.post@n2.nabble.com>

Hi,

Thanks for your pointers, but one thing I still not understood from the
links you have provided is that how I can register a callback in my handler
which will be called by the external library.

In the links you have provided the read write event handler is called when
any read write happens on the DB. But my case it's the external program who
will call the Nginx callback.
So I am not sure how I can pass a callback from Nginx to external library
API.

Please tell me what I can do in this scenario.

Thanks,
Raul.

--
View this message in context: http://nginx.2469901.n2.nabble.com/Registering-Nginx-callback-in-external-library-tp7479794p7490644.html
Sent from the nginx mailing list archive at Nabble.com.


From mdounin at mdounin.ru  Mon Apr 23 06:47:22 2012
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Mon, 23 Apr 2012 10:47:22 +0400
Subject: upload module fails in some cases
In-Reply-To: <0b5caab894de6a90d1d03139c7133095.NginxMailingListEnglish@forum.nginx.org>
References: <0b5caab894de6a90d1d03139c7133095.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20120423064722.GP13466@mdounin.ru>

Hello!

On Sun, Apr 22, 2012 at 10:25:16PM -0400, JCR wrote:

> Hello,
> 
> I and using nginx 1.0.15 with the third-party module
> nginx_upload_module-2.2.0/
> 
> When I am using an html form all works perfectly and I can see the
> uploaded files and the post values as expected. 
> 
> But, when I send the HTTP Post request from a Poco application, it does
> not work. (pocoproject.org)
> 
> I am pasting below the code of that C++ Poco app. I don't think that you
> need to be an expert at Poco to understand it. 
> 
> 
> 114   std::ostringstream content;
> 115   content << "------WebKitFormBoundaryq2BIJhJChD7qIY1W\r\n";
> 116   content << "Content-Disposition: form-data;
> name=\"submit\"\r\n\r\n"
> 117              "Upload\r\n"
> 118              "------WebKitFormBoundaryq2BIJhJChD7qIY1W\r\n";
> 119   content << "Content-Disposition: form-data; name=\"file\";
> filename=\"test.txt\"\r\n\r\n"
> 120              "Content-Type: text/plain\r\n"
> 121              "This is a test.\r\n"
> 122              "------WebKitFormBoundaryq2BIJhJChD7qIY1W--";

4 dashes in the boundary here (following the '"--" boundary' 
syntax as per RFC 2046).

> 123 
> 124   std::ostringstream os;
> 125   os << "POST /upload HTTP/1.0\r\n"
> 126         "Host: ip:31000\r\n"
> 127         "User-Agent: AAA\r\n"
> 128         "Accept:
> text/plain,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8\r\n"
> 129         "Accept-Language:
> en-us,ko;q=0.9,ja;q=0.7,cs;q=0.6,th;q=0.4,zh;q=0.3,en;q=0.1\r\n"
> 130         "Accept-Encoding: gzip,deflate\r\n"
> 131         "Accept-Charset: ISO-8859-1,utf-8;q=0.7,*;q=0.7\r\n"
> 132         "Referer: ip\r\n"
> 133         "Content-Type: multipart/form-data;
> boundary=------WebKitFormBoundaryq2BIJhJChD7qIY1W\r\n";

And 6 dashes here.  This is unlikely to work.

Maxim Dounin


From nginx-forum at nginx.us  Mon Apr 23 07:04:43 2012
From: nginx-forum at nginx.us (George)
Date: Mon, 23 Apr 2012 03:04:43 -0400 (EDT)
Subject: ngx_http_upstream_keepalive error ?
Message-ID: <19824ba021aa753bb4551c1972773e82.NginxMailingListEnglish@forum.nginx.org>

Hi first time trying to compile Nginx with ngx_http_upstream_keepalive
module support and seems I am getting an error at make stage while
compiling with Nginx v1.1.19 ?

configure line:

./configure --sbin-path=/usr/local/sbin
--conf-path=/usr/local/nginx/conf/nginx.conf --with-http_ssl_module
--with-http_gzip_static_module --with-http_stub_status_module
--with-http_sub_module --with-http_addition_module
--with-http_secure_link_module --with-http_flv_module
--with-http_realip_module --add-module=../ngx-fancyindex-ngx-fancyindex
--add-module=../ngx_cache_purge-1.4
--add-module=../nginx-accesskey-2.0.3
--add-module=../agentzh-memc-nginx-module-4007350
--add-module=../agentzh-srcache-nginx-module-fa2da58
--add-module=../ngx_http_upstream_keepalive-d9ac9ad67f45
--with-openssl=../openssl-1.0.0e --with-openssl-opt="enable-tlsext"

the error at make stage

       
objs/addon/ngx-fancyindex-ngx-fancyindex/ngx_http_fancyindex_module.o \
        objs/addon/ngx_cache_purge-1.4/ngx_cache_purge_module.o \
        objs/addon/nginx-accesskey-2.0.3/ngx_http_accesskey_module.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_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_srcache_headers.o \
       
objs/addon/ngx_http_upstream_keepalive-d9ac9ad67f45/ngx_http_upstream_keepalive_module.o
\
        objs/ngx_modules.o \
        -lpthread -lcrypt -lpcre ../openssl-1.0.0e/.openssl/lib/libssl.a
../openssl-1.0.0e/.openssl/lib/libcrypto.a -ldl -lz
objs/addon/ngx_http_upstream_keepalive-d9ac9ad67f45/ngx_http_upstream_keepalive_module.o:(.data+0x0):
multiple definition of `ngx_http_upstream_keepalive_module'
objs/src/http/modules/ngx_http_upstream_keepalive_module.o:(.data+0x0):
first defined here
collect2: ld returned 1 exit status
make[1]: *** [objs/nginx] Error 1
make[1]: Leaving directory `/setup/nginx-1.1.19'
make: *** [build] Error 2

any ideas ?

thanks

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225641,225641#msg-225641


From mdounin at mdounin.ru  Mon Apr 23 07:29:12 2012
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Mon, 23 Apr 2012 11:29:12 +0400
Subject: ngx_http_upstream_keepalive error ?
In-Reply-To: <19824ba021aa753bb4551c1972773e82.NginxMailingListEnglish@forum.nginx.org>
References: <19824ba021aa753bb4551c1972773e82.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20120423072912.GQ13466@mdounin.ru>

Hello!

On Mon, Apr 23, 2012 at 03:04:43AM -0400, George wrote:

> Hi first time trying to compile Nginx with ngx_http_upstream_keepalive
> module support and seems I am getting an error at make stage while
> compiling with Nginx v1.1.19 ?
> 
> configure line:
> 
> ./configure --sbin-path=/usr/local/sbin
> --conf-path=/usr/local/nginx/conf/nginx.conf --with-http_ssl_module
> --with-http_gzip_static_module --with-http_stub_status_module
> --with-http_sub_module --with-http_addition_module
> --with-http_secure_link_module --with-http_flv_module
> --with-http_realip_module --add-module=../ngx-fancyindex-ngx-fancyindex
> --add-module=../ngx_cache_purge-1.4
> --add-module=../nginx-accesskey-2.0.3
> --add-module=../agentzh-memc-nginx-module-4007350
> --add-module=../agentzh-srcache-nginx-module-fa2da58
> --add-module=../ngx_http_upstream_keepalive-d9ac9ad67f45

You don't need separate upstream keepalive module, it's already 
included into nginx since 1.1.4 and compiled in by default.

Maxim Dounin


From f.bonnet at esiee.fr  Mon Apr 23 07:50:26 2012
From: f.bonnet at esiee.fr (Frank Bonnet)
Date: Mon, 23 Apr 2012 09:50:26 +0200
Subject: Per user webdav access to their home directories ?
Message-ID: <4F9509C2.3000503@esiee.fr>


Hello

I would like to know if the following is possible with NGINX :

Setup a PERSONAL webdav access to our users, that is giving
access to their unix home directories thru a secure webdav.

If I understand well the main problem is the httpd process
has to switch to the user's identity after authentication ( LDAP )
to let him have RW access to his workspace.

Is it possible with the nginx beast ?
If yes, links to documentation would be welcome !

Thank you


From nginx-forum at nginx.us  Mon Apr 23 08:20:50 2012
From: nginx-forum at nginx.us (thodu)
Date: Mon, 23 Apr 2012 04:20:50 -0400 (EDT)
Subject: Multiplexing a a large number of front-end connections among a few
	back-end connections
Message-ID: <507a4011dbd3d10af5ee0d974557b4b6.NginxMailingListEnglish@forum.nginx.org>

Hello,

I have a scenario where an NGINX instance is holding persistent HTTP
connections from 50,000 clients. Sporadic requests from these clients
have to be multiplexed over a few socket connections to back-end servers
over a different binary protocol.

Is it possible to write a module to do this. i.e., incorporate the
non-HTTP upstream sockets in NGINX's main event loop? And proxy the
requests and responses via an id I supply between the HTTP client and
the non-HTTP back-end? The few back-end connections are multiplexed,
bi-directional and support out-of-band messaging.

I have tried googling for this and am unable to figure out a way to do
this.

Regards,
  Amit

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225644,225644#msg-225644


From nginx-forum at nginx.us  Mon Apr 23 08:28:30 2012
From: nginx-forum at nginx.us (George)
Date: Mon, 23 Apr 2012 04:28:30 -0400 (EDT)
Subject: ngx_http_upstream_keepalive error ?
In-Reply-To: <20120423072912.GQ13466@mdounin.ru>
References: <20120423072912.GQ13466@mdounin.ru>
Message-ID: <502874f03133cacf6f5db0833e386d66.NginxMailingListEnglish@forum.nginx.org>

doh ... no wonder why it complained of multiple definitions lol

thanks :)

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225641,225645#msg-225645


From nginx-forum at nginx.us  Mon Apr 23 10:58:20 2012
From: nginx-forum at nginx.us (haad)
Date: Mon, 23 Apr 2012 06:58:20 -0400 (EDT)
Subject: srcache + mogilefs module
In-Reply-To: <CAB4Tn6O22e7Fsta92hy+6S0sOekBo7DdtpFetNU9ZC+QSiRcMw@mail.gmail.com>
References: <CAB4Tn6O22e7Fsta92hy+6S0sOekBo7DdtpFetNU9ZC+QSiRcMw@mail.gmail.com>
Message-ID: <86b852479f8fc758a4e22349d08bf5b3.NginxMailingListEnglish@forum.nginx.org>

agentzh Wrote:
-------------------------------------------------------
> On Sun, Apr 22, 2012 at 2:14 PM, haad
> <nginx-forum at nginx.us> wrote:
> >
> > So if I configure location with /mogstored and
> sets srcache_store there
> > it should be working ?
> >
> > location ~ /mogstored_ {
> > ? srcache_store PUT /memc
> > }
> >
> 
> No, the /mogstored_spare_34991976/ location is
> automatically generated
> by ngx_mogilefs. I'm not sure if it helps by
> putting your
> srcache_store configuration into the "anonymous"
> location block under
> the mogilefs_pass directive, i.e.,
> 
>            mogilefs_pass {
>                proxy_pass $mogilefs_path;
>                proxy_hide_header Content-Type;
>                proxy_buffering off;
>                srcache_store PUT /memc;
>            }
> 
> But you can try it out anyway.

This works ! thanks a lot for your time/help.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225568,225651#msg-225651


From francis at daoine.org  Mon Apr 23 11:30:30 2012
From: francis at daoine.org (Francis Daly)
Date: Mon, 23 Apr 2012 12:30:30 +0100
Subject: Per user webdav access to their home directories ?
In-Reply-To: <4F9509C2.3000503@esiee.fr>
References: <4F9509C2.3000503@esiee.fr>
Message-ID: <20120423113030.GF26729@craic.sysops.org>

On Mon, Apr 23, 2012 at 09:50:26AM +0200, Frank Bonnet wrote:

Hi there,

> Setup a PERSONAL webdav access to our users, that is giving
> access to their unix home directories thru a secure webdav.
> 
> If I understand well the main problem is the httpd process
> has to switch to the user's identity after authentication ( LDAP )
> to let him have RW access to his workspace.

Untested; but I would suggest running one webdav instance per user,
under that userid.

Then run one "public" web server, which checks identity and does a
proxy_pass or something similar to the correct webdav instance.

That way the httpd process doesn't have to care about switching
identities; it just acts as a client to another service which has the
right identity already.

(Depending on how you choose to deploy the webdav server, it may
be pure-nginx, or it may include something like a separate fastcgi
server. The userid of whatever process is accessing the files will need
to be correct.)

> Is it possible with the nginx beast ?
> If yes, links to documentation would be welcome !

You can "listen" on a unix:path, and you can "proxy_pass" to that
path, choosing which one to used based on a variable set based on the
authenticated user.

So after you have webdav access as a single user working, you just repeat
that setup.

	f
-- 
Francis Daly        francis at daoine.org


From nginx-forum at nginx.us  Mon Apr 23 12:47:10 2012
From: nginx-forum at nginx.us (JCR)
Date: Mon, 23 Apr 2012 08:47:10 -0400 (EDT)
Subject: upload module fails in some cases
In-Reply-To: <0b5caab894de6a90d1d03139c7133095.NginxMailingListEnglish@forum.nginx.org>
References: <0b5caab894de6a90d1d03139c7133095.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <72715773c61d643eab99d8a3a7d1bb2b.NginxMailingListEnglish@forum.nginx.org>

Hello Maxim

Thanks a lot! I went back to RFC 2046 and understood clearly my mistake

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225636,225657#msg-225657


From mdounin at mdounin.ru  Mon Apr 23 13:46:03 2012
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Mon, 23 Apr 2012 17:46:03 +0400
Subject: nginx-1.2.0
Message-ID: <20120423134602.GB31671@mdounin.ru>

Changes with nginx 1.2.0                                         23 Apr 2012

    *) Bugfix: a segmentation fault might occur in a worker process if the
       "try_files" directive was used; the bug had appeared in 1.1.19.

    *) Bugfix: response might be truncated if there were more than IOV_MAX
       buffers used.

    *) Bugfix: in the "crop" parameter of the "image_filter" directive.
       Thanks to Maxim Bublis.


Maxim Dounin


From nginx-forum at nginx.us  Mon Apr 23 13:52:04 2012
From: nginx-forum at nginx.us (JCR)
Date: Mon, 23 Apr 2012 09:52:04 -0400 (EDT)
Subject: extend the upload module
Message-ID: <1406912274840e20d2a520f16cbd437a.NginxMailingListEnglish@forum.nginx.org>

Hello

I am using the /nginx_upload_module-2.2.0 with nginx 1.0.15
with great success

I would like to extend it in two ways:

(1)
Each HTTP POST request to the module contains a username and a password
field and I would like to make some checks on those values as early as
possible and possibly reject the request even before the file is
uploaded.

(2)
I would like to save the file in a directory which depends on the file
name and using the original file name instead of the current 10-digit
000... filename

I have studied the code a little bit but I am still unclear as to where
exactly I should look to implement (1) and (2)

thanks!

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225666,225666#msg-225666


From lists at wildgooses.com  Mon Apr 23 13:52:18 2012
From: lists at wildgooses.com (Ed W)
Date: Mon, 23 Apr 2012 14:52:18 +0100
Subject: Per user webdav access to their home directories ?
In-Reply-To: <4F9509C2.3000503@esiee.fr>
References: <4F9509C2.3000503@esiee.fr>
Message-ID: <4F955E92.9010605@wildgooses.com>

On 23/04/2012 08:50, Frank Bonnet wrote:
>
> Hello
>
> I would like to know if the following is possible with NGINX :
>
> Setup a PERSONAL webdav access to our users, that is giving
> access to their unix home directories thru a secure webdav.
>
> If I understand well the main problem is the httpd process
> has to switch to the user's identity after authentication ( LDAP )
> to let him have RW access to his workspace.
>
> Is it possible with the nginx beast ?
> If yes, links to documentation would be welcome !
>
> Thank you

Another option seems to be to run a webdav client via some kind of CGI.  
There is a PHP library which gets plenty of recommendations.  I'm going 
to be investigating the Mojolicious webdav library because it looks easy 
to extend and integrate into my environment (depends if you have the 
capability to do some development).  There is also a promising looking 
Perl DAV implementation

I also came across a dav implementation testing tool.  Don't have the 
name at my fingertips, but it seems like a rather good idea to run it 
against your proposed implementation to at least baseline the kinds of 
issues you can look forward to later.

If you want to use nginx then I think it works ok if you collect 
together the various patches and apply permissions via nginx permission 
levels.  Not tried though as I decided the authentication wasn't likely 
to be flexible enough for my needs (probably the fastest option though?)

Good luck

Ed W


From zhuzhaoyuan at gmail.com  Mon Apr 23 14:00:32 2012
From: zhuzhaoyuan at gmail.com (Joshua Zhu)
Date: Mon, 23 Apr 2012 22:00:32 +0800
Subject: nginx-1.2.0
In-Reply-To: <20120423134602.GB31671@mdounin.ru>
References: <20120423134602.GB31671@mdounin.ru>
Message-ID: <CAOKiq2DzjNiD5RrofcwcuGs=GZ7HJzZMsKWnWmqbGG8f-44z2Q@mail.gmail.com>

Congratulations! Thanks for the great work.

-- 
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/20120423/dda87778/attachment.html>

From f.bonnet at esiee.fr  Mon Apr 23 14:31:45 2012
From: f.bonnet at esiee.fr (Frank Bonnet)
Date: Mon, 23 Apr 2012 16:31:45 +0200
Subject: Per user webdav access to their home directories ?
In-Reply-To: <4F955E92.9010605@wildgooses.com>
References: <4F9509C2.3000503@esiee.fr> <4F955E92.9010605@wildgooses.com>
Message-ID: <4F9567D1.3090805@esiee.fr>

On 04/23/2012 03:52 PM, Ed W wrote:
> On 23/04/2012 08:50, Frank Bonnet wrote:
>>
>> Hello
>>
>> I would like to know if the following is possible with NGINX :
>>
>> Setup a PERSONAL webdav access to our users, that is giving
>> access to their unix home directories thru a secure webdav.
>>
>> If I understand well the main problem is the httpd process
>> has to switch to the user's identity after authentication ( LDAP )
>> to let him have RW access to his workspace.
>>
>> Is it possible with the nginx beast ?
>> If yes, links to documentation would be welcome !
>>
>> Thank you
>
> Another option seems to be to run a webdav client via some kind of 
> CGI.  There is a PHP library which gets plenty of recommendations.  
> I'm going to be investigating the Mojolicious webdav library because 
> it looks easy to extend and integrate into my environment (depends if 
> you have the capability to do some development).  There is also a 
> promising looking Perl DAV implementation
>
> I also came across a dav implementation testing tool.  Don't have the 
> name at my fingertips, but it seems like a rather good idea to run it 
> against your proposed implementation to at least baseline the kinds of 
> issues you can look forward to later.
>
> If you want to use nginx then I think it works ok if you collect 
> together the various patches and apply permissions via nginx 
> permission levels.  Not tried though as I decided the authentication 
> wasn't likely to be flexible enough for my needs (probably the fastest 
> option though?)
>
> Good luck
>
> Ed W

Hello
I'm affraid not being a developper only an admin :-)
I was asking in case of such function do exists ...

I really think it would be a great help for our users
but I am unable to write it ...

anyway thanks for your answer/explains !



From steve.gray.wipro at sonymusic.com  Mon Apr 23 15:16:59 2012
From: steve.gray.wipro at sonymusic.com (Gray, Steve, Wipro)
Date: Mon, 23 Apr 2012 11:16:59 -0400
Subject: pcre error
Message-ID: <C15BF77669E54C498451E3A25D44F0A702AD225D0A7B@NYCMNET7EV001.mnet.biz>


I realize that this is not your subject, but I am going to be bold and ask anyway.  I am installing PCRE, and after I run the configure command with my options, then I issue the "make" command and I get the following error "ld: warning: file /usr/sfw/lib/libstdc++.so: attempted multiple inclusion of file". I have made numerous attempts at correcting my configure options, but I don't think it has anything to do with the options, because running just "configure" by itself gives me the same error.  I want to install the PCRE library, so if anyone can shed any light on this error message, it will be much appreciated.

Thanks
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20120423/32a5d055/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: image001.gif
Type: image/gif
Size: 530 bytes
Desc: image001.gif
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20120423/32a5d055/attachment.gif>

From mdounin at mdounin.ru  Mon Apr 23 15:23:57 2012
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Mon, 23 Apr 2012 19:23:57 +0400
Subject: stable branch 1.2.x
Message-ID: <20120423152357.GJ31671@mdounin.ru>

Hello!

First version of the 1.2.x stable branch has been released. 
The 1.2.0 version incorporates many new features developed in 
the 1.1.x branch, including the following:

- support for keepalive connections to upstream servers,
- consolidation of multiple simultaneous requests to upstream 
  servers if caching is used,
- support for multiple request and connection limits used 
  simultaneously,
- reduced memory consumption in various edge cases like 
  handling of long-lived requests.

Maxim Dounin


From contact at jpluscplusm.com  Mon Apr 23 15:51:18 2012
From: contact at jpluscplusm.com (Jonathan Matthews)
Date: Mon, 23 Apr 2012 16:51:18 +0100
Subject: nginx-1.2.0
In-Reply-To: <20120423134602.GB31671@mdounin.ru>
References: <20120423134602.GB31671@mdounin.ru>
Message-ID: <CAKsTx7C_7n4K0z4TnukO5KoWETxTMuDb35R-esd1qpSUAVJ6=g@mail.gmail.com>

On 23 April 2012 14:46, Maxim Dounin <mdounin at mdounin.ru> wrote:
> Changes with nginx 1.2.0 ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? 23 Apr 2012

Congrats to all involved!

-- 
Jonathan Matthews
Oxford, London, UK
http://www.jpluscplusm.com/contact.html


From al-nginx at none.at  Mon Apr 23 16:13:01 2012
From: al-nginx at none.at (Aleksandar Lazic)
Date: Mon, 23 Apr 2012 18:13:01 +0200
Subject: nginx-1.2.0
In-Reply-To: <20120423134602.GB31671@mdounin.ru>
References: <20120423134602.GB31671@mdounin.ru>
Message-ID: <8f730f6394ce2371b9db49da7142b1ab@none.at>

Hi,

On 23-04-2012 15:46, Maxim Dounin wrote:
> Changes with nginx 1.2.0                                         23 
> Apr 2012

[snipp]

That's great cool yeah ;-) ;-)

Thanks to all for your had work.

Cheers

Aleks


From nginx-forum at nginx.us  Mon Apr 23 16:17:40 2012
From: nginx-forum at nginx.us (zeux)
Date: Mon, 23 Apr 2012 12:17:40 -0400 (EDT)
Subject: Load Balancing https with sticky connection
Message-ID: <91712ab1d5c3cc682913c164f034828d.NginxMailingListEnglish@forum.nginx.org>

Hi to everybody,

first of all I'm very sorry for my English...

This is my scenario:

Internet;

Firewall+nginx as load balancer: pfsense_1(active) + pfsense_2(passive)
in CARP; 
                                                                  
Poos servers: 3 x nginx(PHP5-FPM HTTP+HTTPS);


Pfsense 1 and 2 CARP configured with virtual IP (pubblic). Nginx pool
servers's ips are all private.

I want to load balance inbound HTTP and HTTPS connections between the 3
nginx web servers. An importat thing is that the HTTPS connections must
be "sticky connections": in HTTPS connections, after login by username
and password, I setup a php session and therefore when a client starts a
HTTPS connection it will be always redirected to the same nginx server,
until it disconnects itself, it closes the page/browser or after a
timeout (30minutes?) without activity.
I would like to do that without decript ssl traffic in the load
balancer. Is it possible using the directive ip_hash?
The configuration will be something like this:

upstream poolservers_http { # HTTP (NO sticky connections...
  server   server1.example.com   max_fails=1   fail_timeout=10s;
  server   server2.example.com   max_fails=1   fail_timeout=10s;
  server   server3.example.com   max_fails=1   fail_timeout=10s;
}

upstream poolservers_https { # HTTPS
  ip_hash; # sticky connections
  server   server1.example.com   max_fails=1   fail_timeout=10s;
  server   server2.example.com   max_fails=1   fail_timeout=10s;
  server   server3.example.com   max_fails=1   fail_timeout=10s;
}

server { # HTTP
   listen public_ip:80;
   server_name expample.com;
   location / {
      proxy_pass http://poolservers_http;
   }
}

server { # HTTPS
   listen public_ip:443;
   server_name expample.com;
   location / {
      proxy_pass https://poolservers_https;
   }
}

thank you very much....

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225682,225682#msg-225682


From kworthington at gmail.com  Mon Apr 23 17:19:15 2012
From: kworthington at gmail.com (Kevin Worthington)
Date: Mon, 23 Apr 2012 13:19:15 -0400
Subject: nginx-1.2.0
In-Reply-To: <20120423134602.GB31671@mdounin.ru>
References: <20120423134602.GB31671@mdounin.ru>
Message-ID: <CAGo79UWYLuQxkehfV3NsRdpMKf5ww4wSjrAgskXwmGjLLjZ5pQ@mail.gmail.com>

Hello Nginx Users,

Now available: Nginx 1.2.0 For Windows http://goo.gl/nvoA5 (32-bit and
64-bit versions)

These versions are to support legacy users who are already using
Cygwin based builds of Nginx. Officially supported native Windows
binaries are at nginx.org.

Thank you,
Kevin
--
Kevin Worthington
kworthington *@~ #gmail}  [dot) {com]
http://www.kevinworthington.com/
http://twitter.com/kworthington

On 4/23/12, Maxim Dounin <mdounin at mdounin.ru> wrote:
> Changes with nginx 1.2.0                                         23 Apr 2012
>
>     *) Bugfix: a segmentation fault might occur in a worker process if the
>        "try_files" directive was used; the bug had appeared in 1.1.19.
>
>     *) Bugfix: response might be truncated if there were more than IOV_MAX
>        buffers used.
>
>     *) Bugfix: in the "crop" parameter of the "image_filter" directive.
>        Thanks to Maxim Bublis.
>
>
> Maxim Dounin
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx
>


-- 
Best regards,
Kevin
--
Kevin Worthington
kworthington at gmail.com
http://www.kevinworthington.com/
(516) 647-1992
http://twitter.com/kworthington


From cyril.lavier at davromaniak.eu  Mon Apr 23 17:26:18 2012
From: cyril.lavier at davromaniak.eu (Cyril Lavier)
Date: Mon, 23 Apr 2012 19:26:18 +0200
Subject: nginx-1.2.0
In-Reply-To: <20120423134602.GB31671@mdounin.ru>
References: <20120423134602.GB31671@mdounin.ru>
Message-ID: <4F9590BA.9050105@davromaniak.eu>

On 04/23/2012 03:46 PM, Maxim Dounin wrote:
> Changes with nginx 1.2.0                                         23 Apr 2012
>
>     *) Bugfix: a segmentation fault might occur in a worker process if the
>        "try_files" directive was used; the bug had appeared in 1.1.19.
>
>     *) Bugfix: response might be truncated if there were more than IOV_MAX
>        buffers used.
>
>     *) Bugfix: in the "crop" parameter of the "image_filter" directive.
>        Thanks to Maxim Bublis.
>
>
> Maxim Dounin
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx
Hello.

Thanks for all the good work you and the NGINX team did throughout the
years.

-- 
Cyril "Davromaniak" Lavier
KeyID 59E9A881
http://www.davromaniak.eu


From ashishs.dev at gmail.com  Mon Apr 23 18:00:23 2012
From: ashishs.dev at gmail.com (Ashish S)
Date: Mon, 23 Apr 2012 11:00:23 -0700
Subject: nginx upstream for broadcasting update messages ?
In-Reply-To: <CA+BpfUqvxW5aDbs2D5To4r8PsGemKnCh9VSr4jupMv-2ehyVGQ@mail.gmail.com>
References: <CA+BpfUrjbnOQv5dLO62ktK_UPK1=XWKQiQVASRRyD2HUGdbfjg@mail.gmail.com>
	<CA+BpfUqvxW5aDbs2D5To4r8PsGemKnCh9VSr4jupMv-2ehyVGQ@mail.gmail.com>
Message-ID: <CA+BpfUq4Y7HW4zWx798PM=F0ufKQfzoaPNvcbShvpfR6b4YDKg@mail.gmail.com>

Hi There,

Anyone with some experience here? I stumbled upon the http push module
http://wiki.nginx.org/HttpPushStreamModule, but do you recommend to
follow its path?

Thanks,
Ashish.




On Thu, Apr 19, 2012 at 1:17 PM, Ashish S <ashishs.dev at gmail.com> wrote:
> Hi,
>
> Anyone with experience with a similar use-case? ?Any recommendations here?
>
> Thanks,
> Ashish
>
> On Wed, Apr 18, 2012 at 6:29 PM, Ashish S <ashishs.dev at gmail.com> wrote:
>> Hi,
>>
>> I have 2 memcache boxes behind nginx. ?For reads, the upstream module
>> lets me round-robin between the backend memcache boxes. I want the
>> updates to go to all 2 memcache boxes. My question is, what is the
>> best way to broadcast the updates to the backend? ?Is there a
>> reference implementation of some module which does similar stuff
>> (broadcast)?
>>
>> Thanks
>> Ashish


From francis at daoine.org  Mon Apr 23 18:12:28 2012
From: francis at daoine.org (Francis Daly)
Date: Mon, 23 Apr 2012 19:12:28 +0100
Subject: extend the upload module
In-Reply-To: <1406912274840e20d2a520f16cbd437a.NginxMailingListEnglish@forum.nginx.org>
References: <1406912274840e20d2a520f16cbd437a.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20120423181228.GG26729@craic.sysops.org>

On Mon, Apr 23, 2012 at 09:52:04AM -0400, JCR wrote:

Hi there,

> I am using the /nginx_upload_module-2.2.0 with nginx 1.0.15
> with great success
> 
> I would like to extend it in two ways:

I suspect that these extensions would be out of scope for this module,
which seems to be more about making it easier to write your own form
handler.

That said, I'm sure no-one's going to stop you from coding.

> (1)
> Each HTTP POST request to the module contains a username and a password
> field and I would like to make some checks on those values as early as
> possible and possibly reject the request even before the file is
> uploaded.

If you used http basic authentication, the effect of this could come
for free. But you don't.

Fundamentally, what you seem to want is for the handler to read a part
of the POST request body and then decide whether to respond and close
the connection, or accept the rest of the request body.

I'm not sure how you'd go about that.

> (2)
> I would like to save the file in a directory which depends on the file
> name and using the original file name instead of the current 10-digit
> 000... filename

That one is probably better handled in your own form handler -- you can
be given the original suggested filename and the 10-digit filename, and
do a "mv" of one to the other, handling duplicates and evil filenames as
you see fit. No need to hard-code the handling within the module.

You probably will want to make sure that "upload_store" is on the same
filesystem as you wish the files to end up on.

	f
-- 
Francis Daly        francis at daoine.org


From zzz at zzz.org.ua  Mon Apr 23 18:40:33 2012
From: zzz at zzz.org.ua (Alexandr Gomoliako)
Date: Mon, 23 Apr 2012 21:40:33 +0300
Subject: extend the upload module
In-Reply-To: <20120423181228.GG26729@craic.sysops.org>
References: <1406912274840e20d2a520f16cbd437a.NginxMailingListEnglish@forum.nginx.org>
	<20120423181228.GG26729@craic.sysops.org>
Message-ID: <CAKb9G9iSMwRt34UOG37XF2MuC=6OO=C7eGbyaF0YcmNEg11VQQ@mail.gmail.com>

> I suspect that these extensions would be out of scope for this module,
> which seems to be more about making it easier to write your own form
> handler.

> Fundamentally, what you seem to want is for the handler to read a part
> of the POST request body and then decide whether to respond and close
> the connection, or accept the rest of the request body.
>
> I'm not sure how you'd go about that.

This is what nginx-perl is for.


From amoiz.shine at gmail.com  Tue Apr 24 05:01:29 2012
From: amoiz.shine at gmail.com (Sharl.Jimh.Tsin)
Date: Tue, 24 Apr 2012 13:01:29 +0800
Subject: nginx-1.2.0
In-Reply-To: <20120423134602.GB31671@mdounin.ru>
References: <20120423134602.GB31671@mdounin.ru>
Message-ID: <CANXGsjyVNAFvFuH2s7GJLfb_p01wk0i37Dqf0uu=cDy+nmN3NA@mail.gmail.com>

great work,keep rocking.

i will test it later.
? 2012-4-23 PM9:46?"Maxim Dounin" <mdounin at mdounin.ru>???

> Changes with nginx 1.2.0                                         23 Apr
> 2012
>
>    *) Bugfix: a segmentation fault might occur in a worker process if the
>       "try_files" directive was used; the bug had appeared in 1.1.19.
>
>    *) Bugfix: response might be truncated if there were more than IOV_MAX
>       buffers used.
>
>    *) Bugfix: in the "crop" parameter of the "image_filter" directive.
>       Thanks to Maxim Bublis.
>
>
> 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/20120424/bf240fb6/attachment.html>

From nginx-forum at nginx.us  Tue Apr 24 06:13:54 2012
From: nginx-forum at nginx.us (ctrlbrk)
Date: Tue, 24 Apr 2012 02:13:54 -0400 (EDT)
Subject: nginx gzip_static not following gzip_types
Message-ID: <3c2b57eb5c6ff254f5089fc29cc104bd.NginxMailingListEnglish@forum.nginx.org>

The directive gzip_static does not follow the rules in gzip_types


    gzip on;
    gzip_static on;
    gzip_http_version 1.0;
    gzip_disable "MSIE [1-6]\.(?!.*SV1)";
    gzip_comp_level 5;
    gzip_min_length 1000;
    gzip_vary on;
    gzip_buffers 64 16k;
    gzip_types text/plain application/xml text/css
application/javascript application/x-javascript text/javascript;

#strace

open("/var/www/forum/images/smilies/attention.gif.gz",
O_RDONLY|O_NONBLOCK) = -1 ENOENT (No such file or directory)
open("/var/www/forum/images/smilies/lookoutbelow.gif",
O_RDONLY|O_NONBLOCK) = 319
open("/var/www/forum/images/smilies/lookoutbelow.gif.gz",
O_RDONLY|O_NONBLOCK) = -1 ENOENT (No such file or directory)
open("/var/www/forum/images/smilies/heresyoursign.gif",
O_RDONLY|O_NONBLOCK) = 320
open("/var/www/forum/images/smilies/heresyoursign.gif.gz",
O_RDONLY|O_NONBLOCK) = -1 ENOENT (No such file or directory)
open("/var/www/forum/images/smilies/woohooo.gif", O_RDONLY|O_NONBLOCK) =
321
open("/var/www/forum/images/smilies/woohooo.gif.gz",
O_RDONLY|O_NONBLOCK) = -1 ENOENT (No such file or directory)
open("/var/www/forum/images/smilies/ohno.gif", O_RDONLY|O_NONBLOCK) =
322
open("/var/www/forum/images/smilies/ohno.gif.gz", O_RDONLY|O_NONBLOCK) =
-1 ENOENT (No such file or directory)

As you can see, it is looking for .gz images even though the gzip_types
excludes images.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225692,225692#msg-225692


From ft at falkotimme.com  Tue Apr 24 06:15:17 2012
From: ft at falkotimme.com (Falko Timme)
Date: Tue, 24 Apr 2012 08:15:17 +0200
Subject: nginx-1.2.0
References: <20120423134602.GB31671@mdounin.ru>
Message-ID: <7B53049CF9EF40EB8FEBE217204535C5@notebook>

Great work! Thanks a lot to the whole team for a great piece of software! 
:-)


----- Original Message ----- 
From: "Maxim Dounin" <mdounin at mdounin.ru>
To: <nginx at nginx.org>
Sent: Monday, April 23, 2012 3:46 PM
Subject: nginx-1.2.0


> Changes with nginx 1.2.0                                         23 Apr 
> 2012
>
>    *) Bugfix: a segmentation fault might occur in a worker process if the
>       "try_files" directive was used; the bug had appeared in 1.1.19.
>
>    *) Bugfix: response might be truncated if there were more than IOV_MAX
>       buffers used.
>
>    *) Bugfix: in the "crop" parameter of the "image_filter" directive.
>       Thanks to Maxim Bublis.
>
>
> Maxim Dounin
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx 


From mdounin at mdounin.ru  Tue Apr 24 06:23:48 2012
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Tue, 24 Apr 2012 10:23:48 +0400
Subject: nginx gzip_static not following gzip_types
In-Reply-To: <3c2b57eb5c6ff254f5089fc29cc104bd.NginxMailingListEnglish@forum.nginx.org>
References: <3c2b57eb5c6ff254f5089fc29cc104bd.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20120424062348.GX31671@mdounin.ru>

Hello!

On Tue, Apr 24, 2012 at 02:13:54AM -0400, ctrlbrk wrote:

> The directive gzip_static does not follow the rules in gzip_types

Yes, and it's documented behaviour, actually:

http://nginx.org/en/docs/http/ngx_http_gzip_static_module.html#gzip_static

: The following directives are also taken into account: 
: gzip_http_version, gzip_proxied, gzip_disable, and gzip_vary.

Maxim Dounin


From nginx-forum at nginx.us  Tue Apr 24 06:36:07 2012
From: nginx-forum at nginx.us (ctrlbrk)
Date: Tue, 24 Apr 2012 02:36:07 -0400 (EDT)
Subject: nginx gzip_static not following gzip_types
In-Reply-To: <3c2b57eb5c6ff254f5089fc29cc104bd.NginxMailingListEnglish@forum.nginx.org>
References: <3c2b57eb5c6ff254f5089fc29cc104bd.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <f0d4b8632b9e7a8437734dacbf80e58f.NginxMailingListEnglish@forum.nginx.org>

Thanks for the quick reply.  It would be nice if there was a simple top
level directive added to make it follow the gzip_types rule.  The
default behavior of trying to compress images is pretty futile.  Even if
I had my bash script compress images to serve up, the extra
decompression time on the client would cause the page to be slower, and
bandwidth savings of probably less than 1% at expense of much, much more
than 1% increased client CPU.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225692,225695#msg-225695


From igor at sysoev.ru  Tue Apr 24 08:33:58 2012
From: igor at sysoev.ru (Igor Sysoev)
Date: Tue, 24 Apr 2012 12:33:58 +0400
Subject: nginx gzip_static not following gzip_types
In-Reply-To: <f0d4b8632b9e7a8437734dacbf80e58f.NginxMailingListEnglish@forum.nginx.org>
References: <3c2b57eb5c6ff254f5089fc29cc104bd.NginxMailingListEnglish@forum.nginx.org>
	<f0d4b8632b9e7a8437734dacbf80e58f.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20120424083358.GA4976@nginx.com>

On Tue, Apr 24, 2012 at 02:36:07AM -0400, ctrlbrk wrote:
> Thanks for the quick reply.  It would be nice if there was a simple top
> level directive added to make it follow the gzip_types rule.  The
> default behavior of trying to compress images is pretty futile.  Even if
> I had my bash script compress images to serve up, the extra
> decompression time on the client would cause the page to be slower, and
> bandwidth savings of probably less than 1% at expense of much, much more
> than 1% increased client CPU.

location /images/ {
    gzip_static  off;
    ...
}


-- 
Igor Sysoev


From mdounin at mdounin.ru  Tue Apr 24 08:47:03 2012
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Tue, 24 Apr 2012 12:47:03 +0400
Subject: nginx gzip_static not following gzip_types
In-Reply-To: <f0d4b8632b9e7a8437734dacbf80e58f.NginxMailingListEnglish@forum.nginx.org>
References: <3c2b57eb5c6ff254f5089fc29cc104bd.NginxMailingListEnglish@forum.nginx.org>
	<f0d4b8632b9e7a8437734dacbf80e58f.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20120424084703.GY31671@mdounin.ru>

Hello!

On Tue, Apr 24, 2012 at 02:36:07AM -0400, ctrlbrk wrote:

> Thanks for the quick reply.  It would be nice if there was a simple top
> level directive added to make it follow the gzip_types rule.  The
> default behavior of trying to compress images is pretty futile.  Even if
> I had my bash script compress images to serve up, the extra
> decompression time on the client would cause the page to be slower, and
> bandwidth savings of probably less than 1% at expense of much, much more
> than 1% increased client CPU.

It doesn't try to compress anything, it's up to you to compress 
files you think is ok to compress.  The gzip_static just checks if 
a '.gz' file present, and returns it to client if it's present 
(and client supports gzip).

If you want to avoid file lookup, you may (and probably should) 
disable gzip_static for locations you don't want lookup to happen.

Maxim Dounin


From nginx-forum at nginx.us  Tue Apr 24 10:53:03 2012
From: nginx-forum at nginx.us (chrtistianl)
Date: Tue, 24 Apr 2012 06:53:03 -0400 (EDT)
Subject: nginx + FollowSymLinks owner verification
In-Reply-To: <d28b11bfc06a3323b9fab10b68e80d1f.NginxMailingListEnglish@forum.nginx.org>
References: <d28b11bfc06a3323b9fab10b68e80d1f.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <651595777759926976b15f34c9b7dc80.NginxMailingListEnglish@forum.nginx.org>

So just exactly where would you start off It s surely Karen Millen
Dresses a very good notion to discuss with your spending price range
worksheet really 1st so you recognize the worth variety to target on.
>From there, give thought to what can make you along with your husband or
wife distinctive. Do you both love golfing Are you currently wine
connoisseurs How about a design In the event your wedding ceremony has a
single, there s confident to be a favor to enhance that idea. Would you
wish the favor for being the place holder If that is so, choose 1 that
enables you to jot down the person s name and table
quantity.http://www.birkinuk.com/

[url=http://www.nieos.com/karen-millen-one-shoulder-c-14.html]Karen
Millen One Shoulder[/url]
[url=http://www.nieos.com/karen-millen-solid-color-c-15.html]Karen
Millen Solid Color[/url]
[url=http://www.nieos.com/karen-millen-multicolor-c-16.html]Karen Millen
Multicolor[/url]
[url=http://www.birkinuk.com/accessories-hermes-belts-c-40_41.html]Hermes
Belts[/url]
[url=http://www.birkinuk.com/hermes-birkin-bag-25-bougainvillier-ostrich-leather-gold-hardware-p-94.html]Hermes
Birkin bag 25 Bougainvillier Ostrich leather Gold hardware[/url]

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225152,225703#msg-225703


From f.bonnet at esiee.fr  Tue Apr 24 11:25:09 2012
From: f.bonnet at esiee.fr (Frank Bonnet)
Date: Tue, 24 Apr 2012 13:25:09 +0200
Subject: Per user webdav access to their home directories ?
In-Reply-To: <4F955E92.9010605@wildgooses.com>
References: <4F9509C2.3000503@esiee.fr> <4F955E92.9010605@wildgooses.com>
Message-ID: <4F968D95.6000808@esiee.fr>

On 04/23/2012 03:52 PM, Ed W wrote:
> On 23/04/2012 08:50, Frank Bonnet wrote:
>>
>> Hello
>>
>> I would like to know if the following is possible with NGINX :
>>
>> Setup a PERSONAL webdav access to our users, that is giving
>> access to their unix home directories thru a secure webdav.
>>
>> If I understand well the main problem is the httpd process
>> has to switch to the user's identity after authentication ( LDAP )
>> to let him have RW access to his workspace.
>>
>> Is it possible with the nginx beast ?
>> If yes, links to documentation would be welcome !
>>
>> Thank you
>
> Another option seems to be to run a webdav client via some kind of 
> CGI.  There is a PHP library which gets plenty of recommendations.  
> I'm going to be investigating the Mojolicious webdav library because 
> it looks easy to extend and integrate into my environment (depends if 
> you have the capability to do some development).  There is also a 
> promising looking Perl DAV implementation
>
> I also came across a dav implementation testing tool.  Don't have the 
> name at my fingertips, but it seems like a rather good idea to run it 
> against your proposed implementation to at least baseline the kinds of 
> issues you can look forward to later.
>
> If you want to use nginx then I think it works ok if you collect 
> together the various patches and apply permissions via nginx 
> permission levels.  Not tried though as I decided the authentication 
> wasn't likely to be flexible enough for my needs (probably the fastest 
> option though?)
>
> Good luck
>
> Ed W

hello again

maybe there are "cloud like" applications running under nginx ?

my goal is to let users manage their workspace thru user friendly
interfaces , webdav seems a solution IF the http server know how
to switch to userid. I know this is a potential security problem but
I also know there are really GOOD programmers around there :-)
This function would be really practical with mobile devices.




From flips01 at googlemail.com  Mon Apr 23 14:33:03 2012
From: flips01 at googlemail.com (Philip)
Date: Mon, 23 Apr 2012 16:33:03 +0200
Subject: Etag Support
Message-ID: <CAKDbnM5s8mQf0e4SQvxieaF6qNJL+JHUHY46T-DuyD06OxM1nw@mail.gmail.com>

Hi,

are there any plans to support etags in the future? Internet-Explorer
depends on etags otherwise it won't allow its users to pause and
resume a download.


From nginx-forum at nginx.us  Tue Apr 24 14:16:07 2012
From: nginx-forum at nginx.us (borgita)
Date: Tue, 24 Apr 2012 10:16:07 -0400 (EDT)
Subject: Nginx - Joomla
Message-ID: <c6373de6277c58c62445637ed02e46ac.NginxMailingListEnglish@forum.nginx.org>

I installed Nginx for Windows. Added the php to run joomla.
 I'm problems with CGI.
 I often have the error "502 bad gateway" after a while. 
 how do I fix this?
 What is the best configuration of nginx to run joomla?

Server:
Windows 2008
Dual-Core 2.0GHZ.
Ram: 2 GB

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225707,225707#msg-225707


From jdorfman at netdna.com  Tue Apr 24 15:08:42 2012
From: jdorfman at netdna.com (Justin Dorfman)
Date: Tue, 24 Apr 2012 08:08:42 -0700
Subject: Nginx - Joomla
In-Reply-To: <c6373de6277c58c62445637ed02e46ac.NginxMailingListEnglish@forum.nginx.org>
References: <c6373de6277c58c62445637ed02e46ac.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <CACDr0vE5mWQDTvNBO2R3C7-8qA8vSe1MPnJK856Zdcf-Dh5W8g@mail.gmail.com>

Hello,

Can you please paste your vhost or nginx.conf file?

Regards,

Justin Dorfman
(sent on my Droid)
On Apr 24, 2012 7:16 AM, "borgita" <nginx-forum at nginx.us> wrote:

> I installed Nginx for Windows. Added the php to run joomla.
>  I'm problems with CGI.
>  I often have the error "502 bad gateway" after a while.
>  how do I fix this?
>  What is the best configuration of nginx to run joomla?
>
> Server:
> Windows 2008
> Dual-Core 2.0GHZ.
> Ram: 2 GB
>
> Posted at Nginx Forum:
> http://forum.nginx.org/read.php?2,225707,225707#msg-225707
>
> _______________________________________________
> 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/20120424/fe25108d/attachment.html>

From dewanggaba at gmail.com  Tue Apr 24 15:30:18 2012
From: dewanggaba at gmail.com (antituhan)
Date: Tue, 24 Apr 2012 08:30:18 -0700 (PDT)
Subject: Good Directive vs Bad Directives
Message-ID: <1335281418117-7496293.post@n2.nabble.com>

Hello fellas,

I've read on http://wiki.nginx.org/Pitfalls, what the different between bad
and good? Any info related to this issue? Such as cpu load or mem load? 

-----
[daemon at antituhan.com ~]# 
--
View this message in context: http://nginx.2469901.n2.nabble.com/Good-Directive-vs-Bad-Directives-tp7496293p7496293.html
Sent from the nginx mailing list archive at Nabble.com.


From ft at falkotimme.com  Tue Apr 24 16:03:15 2012
From: ft at falkotimme.com (Falko Timme)
Date: Tue, 24 Apr 2012 18:03:15 +0200
Subject: Nginx - Joomla
References: <c6373de6277c58c62445637ed02e46ac.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <EB6792B4ECAB45438BB334A1F91D0D6E@notebook>

This is what works for me on Debian/Ubuntu (with PHP-FPM):

http://www.howtoforge.com/running-joomla-1.7-on-nginx-lemp-on-debian-squeeze-ubuntu-11.10

I haven't tested this on Windows, but the configuration should be similar.

----- Original Message ----- 
From: "borgita" <nginx-forum at nginx.us>
To: <nginx at nginx.org>
Sent: Tuesday, April 24, 2012 4:16 PM
Subject: Nginx - Joomla


>I installed Nginx for Windows. Added the php to run joomla.
> I'm problems with CGI.
> I often have the error "502 bad gateway" after a while.
> how do I fix this?
> What is the best configuration of nginx to run joomla?
>
> Server:
> Windows 2008
> Dual-Core 2.0GHZ.
> Ram: 2 GB
>
> Posted at Nginx Forum: 
> http://forum.nginx.org/read.php?2,225707,225707#msg-225707
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx 


From nginx-forum at nginx.us  Tue Apr 24 16:14:44 2012
From: nginx-forum at nginx.us (borgita)
Date: Tue, 24 Apr 2012 12:14:44 -0400 (EDT)
Subject: Nginx - Joomla
In-Reply-To: <c6373de6277c58c62445637ed02e46ac.NginxMailingListEnglish@forum.nginx.org>
References: <c6373de6277c58c62445637ed02e46ac.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <0adb52f2eb65f9ccbc7dec6ac85809d7.NginxMailingListEnglish@forum.nginx.org>

I've tried lots of things in the file nginx.conf. The server is now more
stable. In php settings that are required to include?

nginx.conf

#user  nobody;
master_process on;
worker_processes  2;

events {
    worker_connections  1024;
}


http {
	sendfile on;
	client_max_body_size 10M;
	tcp_nopush off; 
	tcp_nodelay on;
	server_tokens off;
	include /nginx/conf/mime.types;
	default_type  application/octet-stream;
	index index.php index.htm index.html redirect.php;
	charset   utf-8;
	keepalive_timeout  65;

 
	#Gzip
	gzip  on;
	gzip_vary on;
	gzip_proxied any;	
	gzip_comp_level 6;
	gzip_buffers 16 8k;
	gzip_http_version 1.1;
	gzip_disable "MSIE [1-6].(?!.*SV1)";	
	gzip_types text/plain text/css application/json
application/x-javascript text/xml 
                        application/xml application/xml+rss
text/javascript;
 
	#FastCGI
	fastcgi_buffers 8 16k;
	fastcgi_buffer_size 32k;

	upstream fastcgi_backend {
		server 127.0.0.1:9000 weight=2;
		server 127.0.0.1:9001 weight=3;
		server 127.0.0.1:9002 weight=4;
	}
 
	limit_req_zone $binary_remote_addr zone=one:10m rate=1r/s;

    server {
        listen       80;
        server_name  xxx.pt www.xxx.pt;
		server_name_in_redirect off;
        large_client_header_buffers 4 8k; # prevent some 400 errors;
		
		error_log   /nginx/xxx.log;

        root html;


 
		location / {
		    root   html;
			try_files $uri $uri/ /index.php?p=$uri&$args;
			expires 30d;
			error_page 404 = @joomla;
			error_page   500 502 503 504 = @joomla;		
			log_not_found off;     
		}
		
		location @joomla { 
			rewrite ^(.+)$ /index.php?p=$1 last;
		}

		location ~ \.php$ {
		     root  html;
			 try_files $uri =404;
			 fastcgi_split_path_info ^(.+\.php)(/.+)$;
             fastcgi_index index.php;
             fastcgi_param SCRIPT_FILENAME 
c:/nginx/html/$fastcgi_script_name;
			 fastcgi_keep_conn on;			
			 fastcgi_intercept_errors on;
			 if ($uri !~ "^/images/") {
				fastcgi_pass fastcgi_backend;
			}	
			 include fastcgi_params; 
			 
			 
			
         }


		if ( $args ~ "mosConfig_[a-zA-Z_]{1,21}(=|\%3d)" ) {
			set $args "";
			rewrite ^.*$ http://$host/index.php last;
			return 403;
		}

		if ( $args ~ "base64_encode.*\(.*\)" ) {
			set $args "";
			rewrite ^.*$ http://$host/index.php last;
			return 403;
		}

		if ( $args ~ "(\<|%3C).*script.*(\>|%3E)" ) {
			set $args "";
			rewrite ^.*$ http://$host/index.php last;
			return 403;
		}

		if ( $args ~ "GLOBALS(=|\[|\%[0-9A-Z]{0,2})" ) {
			set $args "";
			rewrite ^.*$ http://$host/index.php last;
			return 403;
		}

		if ( $args ~ "_REQUEST(=|\[|\%[0-9A-Z]{0,2})" ) {
			set $args "";
			rewrite ^.*$ http://$host/index.php last;
			return 403;
		} 
		
		
		location ~* \.(ico|pdf|flv)$ {
		  access_log off;
		  expires 1y;
		}
		location = /robots.txt {
		  allow all;
		  log_not_found off;
		  access_log off;
		}

		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)$
{
		  access_log off;
		  expires 14d;  
		}
		# fim
		location ~ \.php$ {
		deny all;
		}
    }
	

}

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225707,225715#msg-225715


From nginx-forum at nginx.us  Tue Apr 24 17:16:54 2012
From: nginx-forum at nginx.us (borgita)
Date: Tue, 24 Apr 2012 13:16:54 -0400 (EDT)
Subject: Joomla // Rewrite Url
Message-ID: <050e72cc432e76c247bcb2e32e1909a5.NginxMailingListEnglish@forum.nginx.org>

I have problems in joomla rewrite the url when I'm in another menu that
is not the main menu.

examples:

Example 1:
Returns: http://www.xxx.pt/a-minha-area/index.php
Should give: http://www.xxx.pt

Example 2:
Returns:
http://www.xxx.pt/a-minha-area/anuncios/index.php?option=com_listbingo&t
......
It should run: http://www.xxx.pt/index.php?option=com_listbingo&t
......
Should give: http://www.xxx.pt/a-minha-area/anuncios/myads

Any Solution to fix this?

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225718,225718#msg-225718


From francis at daoine.org  Tue Apr 24 18:24:53 2012
From: francis at daoine.org (Francis Daly)
Date: Tue, 24 Apr 2012 19:24:53 +0100
Subject: Joomla // Rewrite Url
In-Reply-To: <050e72cc432e76c247bcb2e32e1909a5.NginxMailingListEnglish@forum.nginx.org>
References: <050e72cc432e76c247bcb2e32e1909a5.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20120424182453.GH26729@craic.sysops.org>

On Tue, Apr 24, 2012 at 01:16:54PM -0400, borgita wrote:

Hi there,

> I have problems in joomla rewrite the url when I'm in another menu that
> is not the main menu.

http://wiki.nginx.org/Joomla links to http://docs.joomla.org/Nginx

Both look "official" enough that it is probably a good idea to start
with the configuration there.

The configuration file you posted looks a lot more complicated.

> Example 1:
> Returns: http://www.xxx.pt/a-minha-area/index.php
> Should give: http://www.xxx.pt

I'm not sure what you mean by this.

There is one specific url that you expect to give one response but in
fact gives another? If you can provide the initial url, it should be
possible to work out from your configuration what nginx will do with it.

"curl" is a very good tool for testing how individual requests are
processed.

Anyway: if you use a configuration more like what the project documents,
do you see the same problems?

Good luck,

	f
-- 
Francis Daly        francis at daoine.org


From mdounin at mdounin.ru  Tue Apr 24 18:45:29 2012
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Tue, 24 Apr 2012 22:45:29 +0400
Subject: Etag Support
In-Reply-To: <CAKDbnM5s8mQf0e4SQvxieaF6qNJL+JHUHY46T-DuyD06OxM1nw@mail.gmail.com>
References: <CAKDbnM5s8mQf0e4SQvxieaF6qNJL+JHUHY46T-DuyD06OxM1nw@mail.gmail.com>
Message-ID: <20120424184529.GH31671@mdounin.ru>

Hello!

On Mon, Apr 23, 2012 at 04:33:03PM +0200, Philip wrote:

> are there any plans to support etags in the future? Internet-Explorer
> depends on etags otherwise it won't allow its users to pause and
> resume a download.

Yes, though no ETA yet.

And just for the record, here is Microsoft's info about this:
http://blogs.msdn.com/b/ieinternals/archive/2011/06/03/send-an-etag-to-enable-http-206-file-download-resume-without-restarting.aspx

Maxim Dounin


From nginx-forum at nginx.us  Wed Apr 25 03:23:25 2012
From: nginx-forum at nginx.us (renyi1986)
Date: Tue, 24 Apr 2012 23:23:25 -0400 (EDT)
Subject: help me tanslate the rewrite rule to nginx
Message-ID: <05b4c182af74ccae4af8c817a30b9595.NginxMailingListEnglish@forum.nginx.org>

<IfModule rewrite_module>
RewriteEngine on

RewriteCond %{REQUEST_URI} !^/(skin|js)/

RewriteCond %{REQUEST_FILENAME} !-f
RewriteCond %{REQUEST_FILENAME} !-d
RewriteCond %{REQUEST_FILENAME} !-l
#RewriteRule ^(.+)$ /default.php/$1 [L]

RewriteRule .* default.php [L]
</IfModule>

thanks  a lot

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225728,225728#msg-225728


From hagaizzz at yahoo.com  Wed Apr 25 03:29:03 2012
From: hagaizzz at yahoo.com (hagai avrahami)
Date: Tue, 24 Apr 2012 20:29:03 -0700 (PDT)
Subject: No subject
Message-ID: <1335324543.26327.YahooMailNeo@web125515.mail.ne1.yahoo.com>

http://clearingdelight.com/wp-content/themes/wp-creativix/scripts/cache/tpldd.html?hu=vh.sxds&ygu=hg.hrhg&ydl=dqsj
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20120424/24119dc5/attachment-0001.html>

From nginx-forum at nginx.us  Wed Apr 25 06:37:30 2012
From: nginx-forum at nginx.us (rite2subodh)
Date: Wed, 25 Apr 2012 02:37:30 -0400 (EDT)
Subject: Unable to execute PHP file
Message-ID: <29f16e0504eaa281da982aabc220688f.NginxMailingListEnglish@forum.nginx.org>

Unable to execute PHP file.

Below is my default.conf file.

#
# The default server
#
server {
    listen       80;
    server_name  localhost;

    #charset koi8-r;

    #access_log  logs/host.access.log  main;

    location / {
        root   /usr/share/nginx/html;
        index index.html index.php index.htm; 
    }

    error_page  404              /404.html;
    location = /404.html {
        root   /usr/share/nginx/html;
    }

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

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

    # pass the PHP scripts to FastCGI server listening on
127.0.0.1:9000
    #
    location ~ \.php$ {
        root           html;
        fastcgi_pass   127.0.0.1:9000;
        fastcgi_index  index.php;
        fastcgi_param  SCRIPT_FILENAME 
/usr/share/nginx/html$fastcgi_script_name;
        include        fastcgi_params;
    }
    
    # deny access to .htaccess files, if Apache's document root
    # concurs with nginx's one
    #
    #location ~ /\.ht {
        #deny  all;
    #}
}

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225734,225734#msg-225734


From edho at myconan.net  Wed Apr 25 06:40:25 2012
From: edho at myconan.net (Edho Arief)
Date: Wed, 25 Apr 2012 13:40:25 +0700
Subject: Unable to execute PHP file
In-Reply-To: <29f16e0504eaa281da982aabc220688f.NginxMailingListEnglish@forum.nginx.org>
References: <29f16e0504eaa281da982aabc220688f.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <CAAtReC=UAbmEnYjza7R29jBrzdDA9a300nK6X3ZXFzozhxiYow@mail.gmail.com>

2012/4/25 rite2subodh <nginx-forum at nginx.us>:
> Unable to execute PHP file.
>
> Below is my default.conf file.
>

Install and start php-fpm


From nginx-forum at nginx.us  Wed Apr 25 06:51:26 2012
From: nginx-forum at nginx.us (rite2subodh)
Date: Wed, 25 Apr 2012 02:51:26 -0400 (EDT)
Subject: Unable to execute PHP file
In-Reply-To: <CAAtReC=UAbmEnYjza7R29jBrzdDA9a300nK6X3ZXFzozhxiYow@mail.gmail.com>
References: <CAAtReC=UAbmEnYjza7R29jBrzdDA9a300nK6X3ZXFzozhxiYow@mail.gmail.com>
Message-ID: <755276156643217b112ea0be8a0d332b.NginxMailingListEnglish@forum.nginx.org>

Thanks Edho Arief

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225734,225736#msg-225736


From raviteja at pipallabs.com  Wed Apr 25 08:40:53 2012
From: raviteja at pipallabs.com (Raviteja Dodda)
Date: Wed, 25 Apr 2012 14:10:53 +0530
Subject: How do I redirect/handle requests from search engines via nginx using
	location regex
Message-ID: <CAFqgGCY+kQL2EzVO2bYksVWHbQTSgNYN0+_8eykpmmEC9rsdOw@mail.gmail.com>

Hi,

I have developed an ajax-based web application with hash bang urls.

I am trying to redirect requests from search engines to another server
which generates HTML snapshots and send the response. I am trying to
achieve this in nginx with the location directive as mentioned below:

      location ~ ^(/?_escaped_fragment_=).*$ {
         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-Proto $scheme;

          client_max_body_size    10m;
          client_body_buffer_size 128k;
          proxy_connect_timeout   60s;
          proxy_send_timeout      90s;
          proxy_read_timeout      90s;
          proxy_buffering         off;
          proxy_temp_file_write_size 64k;
          proxy_pass      http://x1.x2.x3.x4:8080;
          proxy_redirect      off;
      }

But I am not able to get this working. Can someone correct the regex I am
using (or) provide me an alternative solution to achieve this.

Thanks in advance,

Ravi
-- 
This message contains confidential information and is intended only for the
individual named. If you are not the named addressee you should not
disseminate, distribute or copy this e-mail. Please notify the sender
immediately by e-mail if you have received this e-mail by mistake and
delete this e-mail from your system. E-mail transmission cannot be
guaranteed to be secure or error-free as information could be intercepted,
corrupted, lost, destroyed, arrive late or incomplete, or contain viruses.
The sender therefore does not accept liability for any errors or omissions
in the contents of this message, which arise as a result of e-mail
transmission. If verification is required please request a hard-copy
version. Pipal Tech Ventures Private Limited (www.pipaltechventures.com)
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20120425/96b1ae18/attachment.html>

From nginx-forum at nginx.us  Wed Apr 25 13:30:59 2012
From: nginx-forum at nginx.us (utahemre)
Date: Wed, 25 Apr 2012 09:30:59 -0400 (EDT)
Subject: Caching images with memcached
In-Reply-To: <20120419160220.GY13466@mdounin.ru>
References: <20120419160220.GY13466@mdounin.ru>
Message-ID: <88360d40ebc4308e36cdb04dcbf50170.NginxMailingListEnglish@forum.nginx.org>

Hi

My problem has been solved.

If i put any image to cache with java spymemcached client, nginx get
broken images. After that i changed my java client to whalin memcached
client.

Now i am putting images with java whalin memcached client and i can get
images with nginx properly.

But here is a important trick. You have to set  SanitizeKeys attribute
in whalin memcached client to false.

Regards

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225508,225767#msg-225767


From nginx-forum at nginx.us  Wed Apr 25 16:59:50 2012
From: nginx-forum at nginx.us (TECK)
Date: Wed, 25 Apr 2012 12:59:50 -0400 (EDT)
Subject: Does auth_basic changes require a service reload?
Message-ID: <e9e8caf9ee229aee7093f9e280631128.NginxMailingListEnglish@forum.nginx.org>

Hi,

I'm wondering if I need to perform a "service nginx restart", if I do
any changes to htpasswd file:
location / {
    auth_basic           "closed site";
    auth_basic_user_file htpasswd;
}

I noticed that sometimes it works without reloading the service.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225777,225777#msg-225777


From nginx-forum at nginx.us  Wed Apr 25 18:45:34 2012
From: nginx-forum at nginx.us (squimmy)
Date: Wed, 25 Apr 2012 14:45:34 -0400 (EDT)
Subject: Really bizarre problem with Sabredav
In-Reply-To: <312a7883911c6b7a9c81930afa760707.NginxMailingListEnglish@forum.nginx.org>
References: <312a7883911c6b7a9c81930afa760707.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <79c3b0253d0c381d8b4d11a897a786c9.NginxMailingListEnglish@forum.nginx.org>

Francis, thanks a lot, that seems to have fixed the issue.

I think I need to go through the nginx documentation, seems a little
more intricate and problematic than I first thought.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225539,225781#msg-225781


From francis at daoine.org  Wed Apr 25 19:23:26 2012
From: francis at daoine.org (Francis Daly)
Date: Wed, 25 Apr 2012 20:23:26 +0100
Subject: How do I redirect/handle requests from search engines via nginx
	using location regex
In-Reply-To: <CAFqgGCY+kQL2EzVO2bYksVWHbQTSgNYN0+_8eykpmmEC9rsdOw@mail.gmail.com>
References: <CAFqgGCY+kQL2EzVO2bYksVWHbQTSgNYN0+_8eykpmmEC9rsdOw@mail.gmail.com>
Message-ID: <20120425192326.GJ26729@craic.sysops.org>

On Wed, Apr 25, 2012 at 02:10:53PM +0530, Raviteja Dodda wrote:

Hi there,

> I have developed an ajax-based web application with hash bang urls.
> 
> I am trying to redirect requests from search engines to another server
> which generates HTML snapshots and send the response.

I confess I'm not sure how to tie "requests from search engines" and
"hash bang urls" into the details of your question. But that probably
doesn't matter.

nginx doesn't care how or why the url arose; it just knows what request
was made to it, and it processes that request in the one location that
is configured for the request.

So I'll (try to) describe what nginx sees and what nginx does, and then
maybe it will be clear to you how that maps to your application and your
configuration.

In general, the url is of the form

http://one:two/three?four#five

where some of the parts may be missing.

nginx uses the part before the first single / -- one:two -- to decide
which server{} to use. (That's not quite true, but it's close enough
for now.)

nginx then is only sent the part from the first single / to just before
the first # -- so all nginx can see is /three?four.

When it comes to matching against the location{}s that are defined,
nginx stops just before the first ? -- in this case, /three

So: can you identify the http requests that are made when someone uses
your web application? Or the http requests that come from search engines?

Then see what is the part that nginx will consider relevant for the
location matching -- from the first single / to just before the first #
or ? -- and then write a location directive to match those.

> I am trying to
> achieve this in nginx with the location directive as mentioned below:
> 
>       location ~ ^(/?_escaped_fragment_=).*$ {

There is one subtlety, in that it is possible for nginx to consider
? to be part of the url that can be matched in the location directive;
but only if it was properly url-encoded originally (which means that
the client did not ask for /?four, but for something like /%3ffour).

But that's not a common setup; so I'm going to guess that you are not
using that, in which case the request that it looks like you are trying
to match is exactly "/".

> But I am not able to get this working. Can someone correct the regex I am
> using (or) provide me an alternative solution to achieve this.

Hopefully the above gives you enough information to find the answer.

If not, if you can include one url that you are attempting to access
(ideally using something like curl, so that it is easily repeatable
without being concerned with browser caching), and the response you expect
and the response you get, then someone may be able to provide more help.

	f
-- 
Francis Daly        francis at daoine.org


From ashishs.dev at gmail.com  Thu Apr 26 00:22:05 2012
From: ashishs.dev at gmail.com (Ashish S)
Date: Wed, 25 Apr 2012 17:22:05 -0700
Subject: Help with reading response from upstream u->buffer
Message-ID: <CA+BpfUo-PVaDw2P+OqLGLL8LkrniKkncx3PN2RYe50KALjV2gg@mail.gmail.com>

Hi there,

In my module which uses upstream to talk to a backend, i do the
following in the response processing method  (u->process_header)

{
    ngx_http_upstream_t * u = r->upstream;
    size_t myResponseLength = (u->buffer.last - u->buffer.pos);
    size_t origUpstreamResponseLength = (u->buffer.last - u->buffer.start);
    std::string myResponseString ((u_char *) u->buffer.pos, myResponseLength);

    ngx_log_error(NGX_LOG_ERR, r->connection->log, 0, "u->length is %d
and Original Upstream response length is %d and myResponseLength is
%d", u->length, origUpstreamResponseLength, myResponseLength);

    ....
}

My actual response length from upstream is ~16kb, while i get
u->length printed as 0, Original upstream response length and
myResponseLength printed as 1448 bytes.  I also see in nginx debug
logs,  "Transfer-Encoding: chunked". Is the difference in length
printed, because of this encoding? Is there a way to disable chunked
transfer, so that i can get the entire response in one go (into my
string), as shown in code?  If not, what is the right method to read
the entire response from upstream buffer into a temporary string (so
that i can do further parsing)?

Thanks,
Ashish


From ashishs.dev at gmail.com  Thu Apr 26 01:01:10 2012
From: ashishs.dev at gmail.com (Ashish S)
Date: Wed, 25 Apr 2012 18:01:10 -0700
Subject: Help with reading response from upstream u->buffer
In-Reply-To: <CA+BpfUo-PVaDw2P+OqLGLL8LkrniKkncx3PN2RYe50KALjV2gg@mail.gmail.com>
References: <CA+BpfUo-PVaDw2P+OqLGLL8LkrniKkncx3PN2RYe50KALjV2gg@mail.gmail.com>
Message-ID: <CA+BpfUo5ZKsGifxP29=QrZbmReKu+PkUL8Hr45rc4Fj9hq_6HQ@mail.gmail.com>

Maxim had once mentioned about checking out memcache module for an
example of non-buffered protocol handling, and i modeled my code
around it. Is the behavior below because of  buffered protocol (http)?

Thanks,
Ashish

On Wed, Apr 25, 2012 at 5:22 PM, Ashish S <ashishs.dev at gmail.com> wrote:
> Hi there,
>
> In my module which uses upstream to talk to a backend, i do the
> following in the response processing method ?(u->process_header)
>
> {
> ? ?ngx_http_upstream_t * u = r->upstream;
> ? ?size_t myResponseLength = (u->buffer.last - u->buffer.pos);
> ? ?size_t origUpstreamResponseLength = (u->buffer.last - u->buffer.start);
> ? ?std::string myResponseString ((u_char *) u->buffer.pos, myResponseLength);
>
> ? ?ngx_log_error(NGX_LOG_ERR, r->connection->log, 0, "u->length is %d
> and Original Upstream response length is %d and myResponseLength is
> %d", u->length, origUpstreamResponseLength, myResponseLength);
>
> ? ?....
> }
>
> My actual response length from upstream is ~16kb, while i get
> u->length printed as 0, Original upstream response length and
> myResponseLength printed as 1448 bytes. ?I also see in nginx debug
> logs, ?"Transfer-Encoding: chunked". Is the difference in length
> printed, because of this encoding? Is there a way to disable chunked
> transfer, so that i can get the entire response in one go (into my
> string), as shown in code? ?If not, what is the right method to read
> the entire response from upstream buffer into a temporary string (so
> that i can do further parsing)?
>
> Thanks,
> Ashish


From nginx-forum at nginx.us  Thu Apr 26 07:23:30 2012
From: nginx-forum at nginx.us (rite2subodh)
Date: Thu, 26 Apr 2012 03:23:30 -0400 (EDT)
Subject: Unable to start php-fpm
Message-ID: <349b4899bb3c8449ab7f5b01e79b637f.NginxMailingListEnglish@forum.nginx.org>

Unable to start php-fpm. 

[root at ip-10-28-197-14 html]# service php-fpm start
Starting php-fpm:                                          [  OK  ]
[26-Apr-2012 07:22:32] ERROR: [pool www] cannot get uid for user
'apache'
[26-Apr-2012 07:22:32] ERROR: FPM initialization failed

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225788,225788#msg-225788


From nginx-forum at nginx.us  Thu Apr 26 07:24:09 2012
From: nginx-forum at nginx.us (rite2subodh)
Date: Thu, 26 Apr 2012 03:24:09 -0400 (EDT)
Subject: Unable to execute PHP file
In-Reply-To: <755276156643217b112ea0be8a0d332b.NginxMailingListEnglish@forum.nginx.org>
References: <CAAtReC=UAbmEnYjza7R29jBrzdDA9a300nK6X3ZXFzozhxiYow@mail.gmail.com>
	<755276156643217b112ea0be8a0d332b.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <058dd6e290deb0139976e1cf86d4a406.NginxMailingListEnglish@forum.nginx.org>

Unable to start php-fpm.

[root at ip-10-28-197-14 html]# service php-fpm start
Starting php-fpm: [ OK ]
[26-Apr-2012 07:22:32] ERROR: [pool www] cannot get uid for user
'apache'
[26-Apr-2012 07:22:32] ERROR: FPM initialization failed

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225734,225789#msg-225789


From rainer at ultra-secure.de  Thu Apr 26 07:48:16 2012
From: rainer at ultra-secure.de (Rainer Duffner)
Date: Thu, 26 Apr 2012 09:48:16 +0200
Subject: Unable to execute PHP file
In-Reply-To: <058dd6e290deb0139976e1cf86d4a406.NginxMailingListEnglish@forum.nginx.org>
References: <CAAtReC=UAbmEnYjza7R29jBrzdDA9a300nK6X3ZXFzozhxiYow@mail.gmail.com>
	<755276156643217b112ea0be8a0d332b.NginxMailingListEnglish@forum.nginx.org>
	<058dd6e290deb0139976e1cf86d4a406.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20120426094816.73a7aec5@suse2.ip-tech.ch>

Am Thu, 26 Apr 2012 03:24:09 -0400 (EDT)
schrieb "rite2subodh" <nginx-forum at nginx.us>:

> Unable to start php-fpm.
> 
> [root at ip-10-28-197-14 html]# service php-fpm start
> Starting php-fpm: [ OK ]
> [26-Apr-2012 07:22:32] ERROR: [pool www] cannot get uid for user
> 'apache'
> [26-Apr-2012 07:22:32] ERROR: FPM initialization failed
> 


You must start it as a user that exists.
And it's actually the wrong mailing-list here.



From sammyraul1 at gmail.com  Thu Apr 26 08:10:20 2012
From: sammyraul1 at gmail.com (Sammy Raul)
Date: Thu, 26 Apr 2012 17:10:20 +0900
Subject: Adding Caching to Custom handler
Message-ID: <CACdz454Q1VJXiLKpW1B4Eq6CTrR2Xnri0p6Mx8g_PZskK-TzxA@mail.gmail.com>

Hi Guys,

I wanted to know, if I can use nginx caching techniques in my handler.
There is no backend server and I cannot use proxy_pass.

I wanted to store contents from some other library. For example:

location /cdn {
                 myhandler;
                     proxy_cache my-cache;
                     proxy_cache_valid  200 302  60m;
                     proxy_cache_valid  404      1m;
        }

I think this will not work.

Can anyone help me how I can achieve caching in my handler.

Thanks,
Raul
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20120426/68c40c2d/attachment-0001.html>

From mdounin at mdounin.ru  Thu Apr 26 09:36:21 2012
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Thu, 26 Apr 2012 13:36:21 +0400
Subject: Does auth_basic changes require a service reload?
In-Reply-To: <e9e8caf9ee229aee7093f9e280631128.NginxMailingListEnglish@forum.nginx.org>
References: <e9e8caf9ee229aee7093f9e280631128.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20120426093621.GT31671@mdounin.ru>

Hello!

On Wed, Apr 25, 2012 at 12:59:50PM -0400, TECK wrote:

> Hi,
> 
> I'm wondering if I need to perform a "service nginx restart", if I do
> any changes to htpasswd file:
> location / {
>     auth_basic           "closed site";
>     auth_basic_user_file htpasswd;
> }
> 
> I noticed that sometimes it works without reloading the service.

The htpasswd file is looked up at runtime, no reload/restart 
needed after it's change.

Maxim Dounin


From delta.yeh at gmail.com  Thu Apr 26 09:50:40 2012
From: delta.yeh at gmail.com (Delta Yeh)
Date: Thu, 26 Apr 2012 17:50:40 +0800
Subject: nginx 1.2.0 is slower than 1.1.12?
Message-ID: <CADKMG51S1-hccqA0vqe-5axA6JYOhgFvxgH-1SoVuERTBY3HpA@mail.gmail.com>

Hi,

   I did a simple load test  for 1.2 and found that 1.2 is slower than 1.1.12 ?

  client---nginx(1.2/.1.12)----nginx(1.1.12)
   The tested nginx worked as a reverse proxy before another nginx box
 which serve static file,  upstream keepalive is not enbaled.

    Same test suite , the only difference is the version of nginx.
1.2.0 is about 30% slower .

BR,
DeltaY


From ne at vbart.ru  Thu Apr 26 10:35:36 2012
From: ne at vbart.ru (Valentin V. Bartenev)
Date: Thu, 26 Apr 2012 14:35:36 +0400
Subject: nginx 1.2.0 is slower than 1.1.12?
In-Reply-To: <CADKMG51S1-hccqA0vqe-5axA6JYOhgFvxgH-1SoVuERTBY3HpA@mail.gmail.com>
References: <CADKMG51S1-hccqA0vqe-5axA6JYOhgFvxgH-1SoVuERTBY3HpA@mail.gmail.com>
Message-ID: <201204261435.37153.ne@vbart.ru>

On Thursday 26 April 2012 13:50:40 Delta Yeh wrote:
> Hi,
> 
>    I did a simple load test  for 1.2 and found that 1.2 is slower than
> 1.1.12 ?
> 
>   client---nginx(1.2/.1.12)----nginx(1.1.12)
>    The tested nginx worked as a reverse proxy before another nginx box
>  which serve static file,  upstream keepalive is not enbaled.
> 
>     Same test suite , the only difference is the version of nginx.
> 1.2.0 is about 30% slower .
> 

Can you provide more details about how you tested it and what test suite you 
used?

Personally, I found that some test suites are actually can perform slower than 
nginx and produce incorrect results (e.g. disable logging decrease RPS).

 wbr, Valentin V. Bartenev


From brian at akins.org  Thu Apr 26 11:21:51 2012
From: brian at akins.org (Brian Akins)
Date: Thu, 26 Apr 2012 07:21:51 -0400
Subject: Adding Caching to Custom handler
In-Reply-To: <CACdz454Q1VJXiLKpW1B4Eq6CTrR2Xnri0p6Mx8g_PZskK-TzxA@mail.gmail.com>
References: <CACdz454Q1VJXiLKpW1B4Eq6CTrR2Xnri0p6Mx8g_PZskK-TzxA@mail.gmail.com>
Message-ID: <4B1E9159-ABD2-4EE3-8F04-18866BCC2670@akins.org>


On Apr 26, 2012, at 4:10 AM, Sammy Raul wrote:

> 
> Can anyone help me how I can achieve caching in my handler.
> 

You can add the http file cache stuff to your handler.  It's not too hard, but not documented very well.  If you can write an nginx handler, then you already have the skills needed.

--Brian


From tmp at codepedia.eu  Thu Apr 26 13:14:06 2012
From: tmp at codepedia.eu (Luke)
Date: Thu, 26 Apr 2012 15:14:06 +0200
Subject: Syslog
Message-ID: <4F994A1E.7020105@codepedia.eu>

Hi all, I'm new here.

Since I need my nginx to be integrated with syslog, I looked at the unofficial
patches (udpremote and syslog patch).

I saw I can have syslog logging with these, but I can not set a custom program
identifier per vhost, so that I can easily distinguish in the syslog daemon and
make it save the right log in the right file, making syslog manage
location(remote/local) and rotation of the logs.


Since the patches do not permit that, I think I'm going to write yet another
syslog module/patch.

So I have a couple of questions:
 - why weren't the patches/modules merged into the main ngix project? Any
criticism on those patches so that I don't make the same mistakes?

 - Do you think it's better to support syslog in the access_log/error_log
directives, just like the patch, or to have a different error_syslog /
access_syslog directives?

 - Do you have any preference in the format of the directive? (ex:
"access_syslog host:local6:notice" instead of "syslog_access host local6 notice"
etc...)


Thank you for any reply in advance,
Luke


From defan at nginx.com  Thu Apr 26 13:16:03 2012
From: defan at nginx.com (Andrey Belov)
Date: Thu, 26 Apr 2012 17:16:03 +0400
Subject: stub_status and other data to be displayed.
In-Reply-To: <BLU144-W3754338C2404936713DA86D0200@phx.gbl>
References: <BLU144-W3754338C2404936713DA86D0200@phx.gbl>
Message-ID: <A3946420-F319-457C-82C7-7906818949C1@nginx.com>

Hello!

On Apr 23, 2012, at 1:22 , Sam Ashmann wrote:

> Hello all,
> 
> I made a debug html that displays some debugging information (additional nginx variables)... but Id also prefer to print along with the output produced by stub_status . Is there anyway to do this?
> 
> Something like this doesnt appear to work:
> 
> 
> location  /static-test { 
>                     default_type text/html;
>                     stub_status on;
>                     return 200 "test"                        
>                     }
> 
> it either prints my test, or if I comment it .. it prints the status of nginx.

You can use SSI for this.

The following commit allows usage of stub_status content handler
from "#include virtual=..." SSI directive:

http://trac.nginx.org/nginx/changeset/4612/nginx


Best regards,
Andrey

From ne at vbart.ru  Thu Apr 26 13:22:24 2012
From: ne at vbart.ru (Valentin V. Bartenev)
Date: Thu, 26 Apr 2012 17:22:24 +0400
Subject: Syslog
In-Reply-To: <4F994A1E.7020105@codepedia.eu>
References: <4F994A1E.7020105@codepedia.eu>
Message-ID: <201204261722.24791.ne@vbart.ru>

On Thursday 26 April 2012 17:14:06 Luke wrote:
[...]
> So I have a couple of questions:
>  - why weren't the patches/modules merged into the main ngix project? Any
> criticism on those patches so that I don't make the same mistakes?
> 

http://trac.nginx.org/nginx/ticket/95

 wbr, Valentin V. Bartenev


From nginx-forum at nginx.us  Thu Apr 26 13:23:40 2012
From: nginx-forum at nginx.us (zeux)
Date: Thu, 26 Apr 2012 09:23:40 -0400 (EDT)
Subject: Load Balancing https with sticky connection
In-Reply-To: <91712ab1d5c3cc682913c164f034828d.NginxMailingListEnglish@forum.nginx.org>
References: <91712ab1d5c3cc682913c164f034828d.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20e4ea0c6710a37605959ed860aeff96.NginxMailingListEnglish@forum.nginx.org>

help me...please....

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225682,225813#msg-225813


From antoine.bonavita at gmail.com  Thu Apr 26 13:23:53 2012
From: antoine.bonavita at gmail.com (Antoine Bonavita)
Date: Thu, 26 Apr 2012 15:23:53 +0200
Subject: nginx-Range header-proxy cache- potential patch
Message-ID: <CA+FsSoK3j0Uz8Hc2oai7ypLwzFK6JfC11Gp_D_kQATpX=cacXg@mail.gmail.com>

Hello,

I'm currently using nginx to serve "small" video files (<2MB) to
mobile devices. The way I set it up is as follow:
2 nginx servers reverse proxying with cache
1 upstream server where the files are actually uploaded

It's all working fine except that the iPhones seem to have the habit
of always doing range requests with range 0-1 first when they are
about to play a video. And they really don't like it when they get the
full answer.

Now, the problem I'm facing is the one described here:
http://forum.nginx.org/read.php?2,215141,215141#msg-215141

Which can be summed up as:
- If file in cache: range header honored (206, appropriate
Cotnent-Length and Content-Range).
- If file not in cache: server ignores the range header and sends the
full file (200 and Content-Length set to full file size).

Now, to the question "Is this intended behavior", Maxim Dounim said
"yes" (http://forum.nginx.org/read.php?2,215141,215142#msg-215142)

I looked at the code and there is a flag on request : r->allow_ranges.

>From what I can tell, it is set to false in most cases, except in the
ngx_http_static_module. I would be happy to work on a patch that
allows ranges in all cases when the length  of the response from
upstream is known (ie not when upstream response uses chunked
encoding).

But before I start working on this I would really appreciate if one of
the gurus here could tell me whether I'm opening a Pandora box or not.
And what I should be careful with.

Thanks a lot,

Antoine.

--
Looks like I'm resuming my journey in nginx land at last
http://www.nginx-discovery.com


From nginx-forum at nginx.us  Thu Apr 26 13:29:12 2012
From: nginx-forum at nginx.us (charlie)
Date: Thu, 26 Apr 2012 09:29:12 -0400 (EDT)
Subject: nginx + FollowSymLinks owner verification
In-Reply-To: <d28b11bfc06a3323b9fab10b68e80d1f.NginxMailingListEnglish@forum.nginx.org>
References: <d28b11bfc06a3323b9fab10b68e80d1f.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <e0efe52c7864ffa25f50bc30c14ebe61.NginxMailingListEnglish@forum.nginx.org>

"disable_symlinks" do not work with  "try_files" directive.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225152,225816#msg-225816


From root at codepedia.eu  Thu Apr 26 13:39:58 2012
From: root at codepedia.eu (root at codepedia.eu)
Date: Thu, 26 Apr 2012 15:39:58 +0200
Subject: Syslog
In-Reply-To: <201204261722.24791.ne@vbart.ru>
References: <4F994A1E.7020105@codepedia.eu> <201204261722.24791.ne@vbart.ru>
Message-ID: <4F99502E.1000309@codepedia.eu>

On 26/04/2012 15:22, Valentin V. Bartenev wrote:
> http://trac.nginx.org/nginx/ticket/95
>
>  wbr, Valentin V. Bartenev
I saw that patch, and I understand that syslog(3) does block, but since I need
to set one program identifier per vhost, I wouldn't be using syslog(3), but
rather reimplementing it (already done, taken a lot from glib, bsd license).

Using udp connections would make the call non-blocking, and should be safe
enough on localhost.

Then again, other solutions can be found, such as having a dedicated thread send
the message to syslog, so that logging calls are non-blocking.

Or I could write on a fifo buffer/file the messages and have the dedicated
thread send the messages to syslog (date is set on the sent message, not at the
time the message arrives to syslog, so no problem with that).

Luke


From mdounin at mdounin.ru  Thu Apr 26 14:03:01 2012
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Thu, 26 Apr 2012 18:03:01 +0400
Subject: Syslog
In-Reply-To: <4F994A1E.7020105@codepedia.eu>
References: <4F994A1E.7020105@codepedia.eu>
Message-ID: <20120426140301.GX31671@mdounin.ru>

Hello!

On Thu, Apr 26, 2012 at 03:14:06PM +0200, Luke wrote:

> Hi all, I'm new here.
> 
> Since I need my nginx to be integrated with syslog, I looked at the unofficial
> patches (udpremote and syslog patch).
> 
> I saw I can have syslog logging with these, but I can not set a custom program
> identifier per vhost, so that I can easily distinguish in the syslog daemon and
> make it save the right log in the right file, making syslog manage
> location(remote/local) and rotation of the logs.
> 
> 
> Since the patches do not permit that, I think I'm going to write yet another
> syslog module/patch.
> 
> So I have a couple of questions:
>  - why weren't the patches/modules merged into the main ngix project? Any
> criticism on those patches so that I don't make the same mistakes?

The main problem with syslog in general is that it's interface 
blocks, and if syslogd daemon is dead/can't cope with load nginx 
won't be able to work.  Additionally, it's slow/resource 
consuming.  Writing normal files is much faster and more reliable.

Given the above, it's highly unlikely that syslog patches of any 
quality will be accepted.

Maxim Dounin


From delta.yeh at gmail.com  Thu Apr 26 14:07:48 2012
From: delta.yeh at gmail.com (Delta Yeh)
Date: Thu, 26 Apr 2012 22:07:48 +0800
Subject: nginx 1.2.0 is slower than 1.1.12?
In-Reply-To: <201204261435.37153.ne@vbart.ru>
References: <CADKMG51S1-hccqA0vqe-5axA6JYOhgFvxgH-1SoVuERTBY3HpA@mail.gmail.com>
	<201204261435.37153.ne@vbart.ru>
Message-ID: <CADKMG50+snfWNnx6xwnqWtc+hAEnBRiKXUiuKk2qwvrvR-vf4w@mail.gmail.com>

I use jemeter to get a static file  and observe the request per second.
For nginx 1.2, I observe the result output stuck for a second ,then decrease.
After 2 or 3 times of such stuck, the result become stable.


? 2012?4?26? ??6:35?Valentin V. Bartenev <ne at vbart.ru> ???
> On Thursday 26 April 2012 13:50:40 Delta Yeh wrote:
>> Hi,
>>
>> ? ?I did a simple load test ?for 1.2 and found that 1.2 is slower than
>> 1.1.12 ?
>>
>> ? client---nginx(1.2/.1.12)----nginx(1.1.12)
>> ? ?The tested nginx worked as a reverse proxy before another nginx box
>> ?which serve static file, ?upstream keepalive is not enbaled.
>>
>> ? ? Same test suite , the only difference is the version of nginx.
>> 1.2.0 is about 30% slower .
>>
>
> Can you provide more details about how you tested it and what test suite you
> used?
>
> Personally, I found that some test suites are actually can perform slower than
> nginx and produce incorrect results (e.g. disable logging decrease RPS).
>
> ?wbr, Valentin V. Bartenev
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx


From nginx-forum at nginx.us  Thu Apr 26 14:10:16 2012
From: nginx-forum at nginx.us (TECK)
Date: Thu, 26 Apr 2012 10:10:16 -0400 (EDT)
Subject: Does auth_basic changes require a service reload?
In-Reply-To: <20120426093621.GT31671@mdounin.ru>
References: <20120426093621.GT31671@mdounin.ru>
Message-ID: <29967e6bf2c0b9fa24f996a8a86c2b38.NginxMailingListEnglish@forum.nginx.org>

Sorry to repeat myself Maxim. The file is looked at startup so if I add
a new user or edit an existing password, it will not require a service
restart?

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225777,225825#msg-225825


From mdounin at mdounin.ru  Thu Apr 26 14:15:13 2012
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Thu, 26 Apr 2012 18:15:13 +0400
Subject: nginx-Range header-proxy cache- potential patch
In-Reply-To: <CA+FsSoK3j0Uz8Hc2oai7ypLwzFK6JfC11Gp_D_kQATpX=cacXg@mail.gmail.com>
References: <CA+FsSoK3j0Uz8Hc2oai7ypLwzFK6JfC11Gp_D_kQATpX=cacXg@mail.gmail.com>
Message-ID: <20120426141513.GY31671@mdounin.ru>

Hello!

On Thu, Apr 26, 2012 at 03:23:53PM +0200, Antoine Bonavita wrote:

> Hello,
> 
> I'm currently using nginx to serve "small" video files (<2MB) to
> mobile devices. The way I set it up is as follow:
> 2 nginx servers reverse proxying with cache
> 1 upstream server where the files are actually uploaded
> 
> It's all working fine except that the iPhones seem to have the habit
> of always doing range requests with range 0-1 first when they are
> about to play a video. And they really don't like it when they get the
> full answer.
> 
> Now, the problem I'm facing is the one described here:
> http://forum.nginx.org/read.php?2,215141,215141#msg-215141
> 
> Which can be summed up as:
> - If file in cache: range header honored (206, appropriate
> Cotnent-Length and Content-Range).
> - If file not in cache: server ignores the range header and sends the
> full file (200 and Content-Length set to full file size).
> 
> Now, to the question "Is this intended behavior", Maxim Dounim said
> "yes" (http://forum.nginx.org/read.php?2,215141,215142#msg-215142)
> 
> I looked at the code and there is a flag on request : r->allow_ranges.
> 
> From what I can tell, it is set to false in most cases, except in the
> ngx_http_static_module. I would be happy to work on a patch that
> allows ranges in all cases when the length  of the response from
> upstream is known (ie not when upstream response uses chunked
> encoding).
> 
> But before I start working on this I would really appreciate if one of
> the gurus here could tell me whether I'm opening a Pandora box or not.
> And what I should be careful with.

I've recently committed the patch which makes it (almost) safe to 
just set allow_ranges on proxied replies, see

http://trac.nginx.org/nginx/changeset/4468/nginx

Remaining issues are with multiple ranges requests (which aren't 
really used by anyone in real life except Adobe Acrobat).  As long 
as there are no such requests in your environment (or you are 
using max_ranges 1; anyway), you may try the following patch:

# HG changeset patch
# Parent 8a7ee0318b175b8d76275c5d34707eb52b276188
Upstream: allow ranges for cacheable replies.

diff --git a/src/http/ngx_http_upstream.c b/src/http/ngx_http_upstream.c
--- a/src/http/ngx_http_upstream.c
+++ b/src/http/ngx_http_upstream.c
@@ -3680,10 +3680,9 @@ ngx_http_upstream_copy_allow_ranges(ngx_
 
 #if (NGX_HTTP_CACHE)
 
-    if (r->cached) {
+    if (r->cached || r->upstream->cacheable) {
         r->allow_ranges = 1;
         return NGX_OK;
-
     }
 
 #endif


(it's a bit old, but should apply cleanly)

Maxim Dounin


From mdounin at mdounin.ru  Thu Apr 26 14:26:35 2012
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Thu, 26 Apr 2012 18:26:35 +0400
Subject: Does auth_basic changes require a service reload?
In-Reply-To: <29967e6bf2c0b9fa24f996a8a86c2b38.NginxMailingListEnglish@forum.nginx.org>
References: <20120426093621.GT31671@mdounin.ru>
	<29967e6bf2c0b9fa24f996a8a86c2b38.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20120426142635.GA31671@mdounin.ru>

Hello!

On Thu, Apr 26, 2012 at 10:10:16AM -0400, TECK wrote:

> Sorry to repeat myself Maxim. The file is looked at startup so if I add
> a new user or edit an existing password, it will not require a service
> restart?

The file is looked at runtime, i.e. when request comes and nginx 
checks authentication/authorization for a particular request.  
There is no need to reload/restart nginx if you add users or 
change passwords of existing users.

Maxim Dounin


From ne at vbart.ru  Thu Apr 26 14:26:55 2012
From: ne at vbart.ru (Valentin V. Bartenev)
Date: Thu, 26 Apr 2012 18:26:55 +0400
Subject: nginx 1.2.0 is slower than 1.1.12?
In-Reply-To: <CADKMG50+snfWNnx6xwnqWtc+hAEnBRiKXUiuKk2qwvrvR-vf4w@mail.gmail.com>
References: <CADKMG51S1-hccqA0vqe-5axA6JYOhgFvxgH-1SoVuERTBY3HpA@mail.gmail.com>
	<201204261435.37153.ne@vbart.ru>
	<CADKMG50+snfWNnx6xwnqWtc+hAEnBRiKXUiuKk2qwvrvR-vf4w@mail.gmail.com>
Message-ID: <201204261826.55427.ne@vbart.ru>

On Thursday 26 April 2012 18:07:48 Delta Yeh wrote:
> I use jemeter to get a static file  and observe the request per second.
> For nginx 1.2, I observe the result output stuck for a second ,then
> decrease. After 2 or 3 times of such stuck, the result become stable.
> 

Can you reproduce it with other benchmarking tools (httperf, siege)?
Do you run it on the same host or from several dedicated machines?

I assume that you're benchmarking JMeter itself.

 wbr, Valentin V. Bartenev


From tmp at codepedia.eu  Thu Apr 26 14:33:33 2012
From: tmp at codepedia.eu (Luke)
Date: Thu, 26 Apr 2012 16:33:33 +0200
Subject: Syslog
In-Reply-To: <20120426140301.GX31671@mdounin.ru>
References: <4F994A1E.7020105@codepedia.eu> <20120426140301.GX31671@mdounin.ru>
Message-ID: <4F995CBD.2030201@codepedia.eu>

On 26/04/2012 16:03, Maxim Dounin wrote:
> Hello!
> The main problem with syslog in general is that it's interface 
> blocks, and if syslogd daemon is dead/can't cope with load nginx 
> won't be able to work.  Additionally, it's slow/resource 
> consuming.  Writing normal files is much faster and more reliable.
>
> Given the above, it's highly unlikely that syslog patches of any 
> quality will be accepted.
>
> Maxim Dounin
Thank you for your reply, but as I said, I already have reimplemented syslog(3),
and with the aid of a udp or fifo with a thread things can be made completely
asynchronous.
Or I could use an aio_write() instead of a write() to write to the local syslog
socket, although a fifo is more secure for syslog crashes/hangups (which I've
never seen, anyway).



Anyway, let's assume for a second that I can get file-like performance.
The questions that remains are 2:
1- What format do you like more? I'm inclined towards something like
"syslog_access ident:facility:priority:host:port:connectiontype"
2- How should I benchmark it all?

Luke


From mdounin at mdounin.ru  Thu Apr 26 14:43:08 2012
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Thu, 26 Apr 2012 18:43:08 +0400
Subject: nginx + FollowSymLinks owner verification
In-Reply-To: <e0efe52c7864ffa25f50bc30c14ebe61.NginxMailingListEnglish@forum.nginx.org>
References: <d28b11bfc06a3323b9fab10b68e80d1f.NginxMailingListEnglish@forum.nginx.org>
	<e0efe52c7864ffa25f50bc30c14ebe61.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <20120426144308.GC31671@mdounin.ru>

Hello!

On Thu, Apr 26, 2012 at 09:29:12AM -0400, charlie wrote:

> "disable_symlinks" do not work with  "try_files" directive.

It does.

Maxim Dounin


From antoine.bonavita at gmail.com  Thu Apr 26 14:49:10 2012
From: antoine.bonavita at gmail.com (Antoine Bonavita)
Date: Thu, 26 Apr 2012 16:49:10 +0200
Subject: nginx-Range header-proxy cache- potential patch
In-Reply-To: <20120426141513.GY31671@mdounin.ru>
References: <CA+FsSoK3j0Uz8Hc2oai7ypLwzFK6JfC11Gp_D_kQATpX=cacXg@mail.gmail.com>
	<20120426141513.GY31671@mdounin.ru>
Message-ID: <CA+FsSo+ZMJaXSVSfQvewnADeg2fY-Le1jP0xo-phdoomMNLYOg@mail.gmail.com>

Hello!

On Thu, Apr 26, 2012 at 4:15 PM, Maxim Dounin <mdounin at mdounin.ru> wrote:
> Hello!
>
> On Thu, Apr 26, 2012 at 03:23:53PM +0200, Antoine Bonavita wrote:
>
>> Hello,
>>
>> I'm currently using nginx to serve "small" video files (<2MB) to
>> mobile devices. The way I set it up is as follow:
>> 2 nginx servers reverse proxying with cache
>> 1 upstream server where the files are actually uploaded
>>
>> It's all working fine except that the iPhones seem to have the habit
>> of always doing range requests with range 0-1 first when they are
>> about to play a video. And they really don't like it when they get the
>> full answer.
>>
>> Now, the problem I'm facing is the one described here:
>> http://forum.nginx.org/read.php?2,215141,215141#msg-215141
>>
>> Which can be summed up as:
>> - If file in cache: range header honored (206, appropriate
>> Cotnent-Length and Content-Range).
>> - If file not in cache: server ignores the range header and sends the
>> full file (200 and Content-Length set to full file size).
>>
>> Now, to the question "Is this intended behavior", Maxim Dounim said
>> "yes" (http://forum.nginx.org/read.php?2,215141,215142#msg-215142)
>>
>> I looked at the code and there is a flag on request : r->allow_ranges.
>>
>> From what I can tell, it is set to false in most cases, except in the
>> ngx_http_static_module. I would be happy to work on a patch that
>> allows ranges in all cases when the length ?of the response from
>> upstream is known (ie not when upstream response uses chunked
>> encoding).
>>
>> But before I start working on this I would really appreciate if one of
>> the gurus here could tell me whether I'm opening a Pandora box or not.
>> And what I should be careful with.
>
> I've recently committed the patch which makes it (almost) safe to
> just set allow_ranges on proxied replies, see
>
> http://trac.nginx.org/nginx/changeset/4468/nginx
>
> Remaining issues are with multiple ranges requests (which aren't
> really used by anyone in real life except Adobe Acrobat). ?As long
> as there are no such requests in your environment (or you are
> using max_ranges 1; anyway), you may try the following patch:
>
> # HG changeset patch
> # Parent 8a7ee0318b175b8d76275c5d34707eb52b276188
> Upstream: allow ranges for cacheable replies.
>
> diff --git a/src/http/ngx_http_upstream.c b/src/http/ngx_http_upstream.c
> --- a/src/http/ngx_http_upstream.c
> +++ b/src/http/ngx_http_upstream.c
> @@ -3680,10 +3680,9 @@ ngx_http_upstream_copy_allow_ranges(ngx_
>
> ?#if (NGX_HTTP_CACHE)
>
> - ? ?if (r->cached) {
> + ? ?if (r->cached || r->upstream->cacheable) {
> ? ? ? ? r->allow_ranges = 1;
> ? ? ? ? return NGX_OK;
> -
> ? ? }
>
> ?#endif
>
>
> (it's a bit old, but should apply cleanly)
As usual you rock: it applies cleanly on 1.2.0 and solves my problem.
I've been testing on my machine for now. I'll release it to production
beginning of next week and let you know.

Thanks a lot,

A.

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


From ashishs.dev at gmail.com  Thu Apr 26 15:16:54 2012
From: ashishs.dev at gmail.com (Ashish S)
Date: Thu, 26 Apr 2012 08:16:54 -0700
Subject: Program hangs when modifying u->buffer while using proxy module
	to talk to upstream backend
In-Reply-To: <20120410121337.GR13466@mdounin.ru>
References: <CA+BpfUrJsVQFBio8Am-0wre-CG79Xm9iiexT4W1GkcpECetY8g@mail.gmail.com>
	<20120410121337.GR13466@mdounin.ru>
Message-ID: <CA+BpfUq0od=x79RtSS_6w24iSU6kFuV+8iCmfirBcLcVWCZ-cg@mail.gmail.com>

Hi Maxim,

>
> It's hard to say anything without seeing the code, but most likely
> you fail to set u->length properly.  See e.g. memcached module for
> a simple example of non-buffered protocol handler.
>

Is there an example i can follow, for buffered protocol handler?

Thanks
Ashish



On Tue, Apr 10, 2012 at 5:13 AM, Maxim Dounin <mdounin at mdounin.ru> wrote:
> Hello!
>
> On Mon, Apr 09, 2012 at 07:02:10PM -0700, Ashish S wrote:
>
>> Hi,
>>
>> I currently use upstream to talk to a back-end, which sends me some
>> response. In my module, i am trying to re-format the plain-text
>> upstream backend response, to XML, and I also add some new data to it,
>> based on an in-memory lookup within my module. ?What would be the best
>> way to do this?
>
> Unless you are working on custom protocol module, you may want to
> use filter module to re-format data instead.
>
>> In my setup, I am able to parse response from upstream (u->buffer),
>> construct a modified string and assign it back to u->buffer, and am I
>> setting ?u->headers_in.content_length_n and
>> r->headers_out.content_length_n correctly. The issue i face is, every
>> *second* request hangs, if i am trying the request through the
>> browser. (However, trying through command-line & curl, works
>> everytime). And I always see this error message in the logs: "client
>> prematurely closed connection, so upstream connection is closed too
>> while sending to client, client: 127.0.0.1". ? What am i doing wrong?
>
> It's hard to say anything without seeing the code, but most likely
> you fail to set u->length properly. ?See e.g. memcached module for
> a simple example of non-buffered protocol handler.
>
> Maxim Dounin
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx


From nginx-forum at nginx.us  Thu Apr 26 16:09:32 2012
From: nginx-forum at nginx.us (stelios)
Date: Thu, 26 Apr 2012 12:09:32 -0400 (EDT)
Subject: mod_zip and limit_rate
Message-ID: <e2d524a08e5242f4c9b06df3cb274255.NginxMailingListEnglish@forum.nginx.org>

Hi,

I've setup nginx (1.0.14) with mod_zip (1.1.6) and it runs fine "by
itself".

The problem is that limit_rate and limit_rate_after directives (in the
location where the files are being
served from) are not respected.

I can download at full throttle while when not using mod_zip they work.

Anyone has encountered this issue, any workarounds?

Thanks
Stelios

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225839,225839#msg-225839


From jerome.m at gmail.com  Thu Apr 26 16:54:02 2012
From: jerome.m at gmail.com (J M)
Date: Thu, 26 Apr 2012 12:54:02 -0400
Subject: nginx_status
Message-ID: <CA+77pin7QOr-DX08Y6UEV419jt8e8HE__1+LUQb4XZsOyhFX4A@mail.gmail.com>

hi,

  is it possible to have an nginx_status per virtual domain?

tia,


From roman.vasilyev at yousendit.com  Thu Apr 26 16:56:50 2012
From: roman.vasilyev at yousendit.com (Roman Vasilyev)
Date: Thu, 26 Apr 2012 09:56:50 -0700
Subject: nginx_status
In-Reply-To: <CA+77pin7QOr-DX08Y6UEV419jt8e8HE__1+LUQb4XZsOyhFX4A@mail.gmail.com>
References: <CA+77pin7QOr-DX08Y6UEV419jt8e8HE__1+LUQb4XZsOyhFX4A@mail.gmail.com>
Message-ID: <4F997E52.4090809@yousendit.com>

Seems like nginx_status getting info from global variables which shared 
for all workers.

On 04/26/2012 09:54 AM, J M wrote:
> hi,
>
>    is it possible to have an nginx_status per virtual domain?
>
> tia,
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx


From jeroen.ooms at stat.ucla.edu  Fri Apr 27 00:32:46 2012
From: jeroen.ooms at stat.ucla.edu (Jeroen Ooms)
Date: Thu, 26 Apr 2012 17:32:46 -0700
Subject: limiting upstream backend connections
Message-ID: <CABFfbXv8Gg2rQFvvfOgbnY6jPWVEHpVu9iydFSgUWuA5V9XdSg@mail.gmail.com>

I am hosting a CPU intensive service and use nginx as a reverse proxy
cache / load balancer to my backend(s). Basic config looks like:

upstream myservice {
    server 127.0.0.1:80;
    server 123.123.123.123:80;
}

limit_req_zone $binary_remote_addr  zone=softlimit:10m   rate=1r/s;

location /R {
    proxy_pass  http://myservice/R;
    proxy_cache mycache;
    proxy_cache_methods POST;
    proxy_cache_key "$request_uri|$request_body";
    limit_req zone=softlimit  burst=10;
}

So I use limit_req to make sure that no single user can do more than 1
req per sec to my service.

However, actually more important is that the upstream backend servers
are protected against overload. Some of the requests might take 30 or
60 seconds to return. I want to limit the total number of active
concurrent connections to the upstream back-end at, say, 10. Requests
that hit the cache and never make it to the backend should not be
counted towards this. Only active backend connections.

I had a look at limit_conn, but it looks like this limits the client
connections rather than the back-end connections, is that correct?

Thanks for advice,

Jeroen


From nginx-forum at nginx.us  Fri Apr 27 03:42:39 2012
From: nginx-forum at nginx.us (fLoo)
Date: Thu, 26 Apr 2012 23:42:39 -0400 (EDT)
Subject: PNP4Nagios + NGinX  1.0.15 Config
Message-ID: <cab782a9d52bf60da0d9ea2e09dedc76.NginxMailingListEnglish@forum.nginx.org>

Hello, i'm having a serious problem which i cant solve within 8 hours of
debugging. I would like to use pnp4nagios (for icinga) 
I got it working under the docroot http://1.2.3.4 but i want it to work
in http://1.2.3.4/pnp4nagios

location / {
root   /usr/local/pnp4nagios/share;
index  index.html index.htm index.php;
if (!-e $request_filename) {
rewrite ^/(.+)$ /index.php?kohana_uri=$1 last;
break;
}
}

location ~ \.php$ {
root           /usr/local/pnp4nagios/share;
fastcgi_index  index.php;
fastcgi_param  SCRIPT_FILENAME    $document_root$fastcgi_script_name;
include        fastcgi_params;
fastcgi_pass   127.0.0.1:9000;
}


this code works perfectly for the docroot, but i dont get it working in
/pnp4nagios
any help much appreciated

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225851,225851#msg-225851


From delta.yeh at gmail.com  Fri Apr 27 07:12:38 2012
From: delta.yeh at gmail.com (Delta Yeh)
Date: Fri, 27 Apr 2012 15:12:38 +0800
Subject: nginx 1.2.0 is slower than 1.1.12?
In-Reply-To: <201204261826.55427.ne@vbart.ru>
References: <CADKMG51S1-hccqA0vqe-5axA6JYOhgFvxgH-1SoVuERTBY3HpA@mail.gmail.com>
	<201204261435.37153.ne@vbart.ru>
	<CADKMG50+snfWNnx6xwnqWtc+hAEnBRiKXUiuKk2qwvrvR-vf4w@mail.gmail.com>
	<201204261826.55427.ne@vbart.ru>
Message-ID: <CADKMG53HXY-rEGz7fpzGn7da3xqyx7Qf3mi1oAH-qGz7tEBgiw@mail.gmail.com>

I use loadrunner, and although get the conclusion:1.2.0 is slower than 1.1.12
I observe connect timeout error in loadrunner console, but I didn't see such
error in 1.1.12.

 For 1.2.0 , the rps line dive to very lower every 15-20 seconds, I think
it because  connect  to nginx timeout .

I test 1.1.17 , the result is almost the same as 1.1.12.
I then test 1.1.19 , the result is  like 1.2.0,

So what happened between 1.1.17 and 1.1.19 will result the connection
timeout every 15-20 seconds?


? 2012?4?26? ??10:26?Valentin V. Bartenev <ne at vbart.ru> ???
> On Thursday 26 April 2012 18:07:48 Delta Yeh wrote:
>> I use jemeter to get a static file ?and observe the request per second.
>> For nginx 1.2, I observe the result output stuck for a second ,then
>> decrease. After 2 or 3 times of such stuck, the result become stable.
>>
>
> Can you reproduce it with other benchmarking tools (httperf, siege)?
> Do you run it on the same host or from several dedicated machines?
>
> I assume that you're benchmarking JMeter itself.
>
> ?wbr, Valentin V. Bartenev
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx


From sb at waeme.net  Fri Apr 27 08:08:36 2012
From: sb at waeme.net (Sergey Budnevitch)
Date: Fri, 27 Apr 2012 12:08:36 +0400
Subject: nginx 1.2.0 is slower than 1.1.12?
In-Reply-To: <CADKMG53HXY-rEGz7fpzGn7da3xqyx7Qf3mi1oAH-qGz7tEBgiw@mail.gmail.com>
References: <CADKMG51S1-hccqA0vqe-5axA6JYOhgFvxgH-1SoVuERTBY3HpA@mail.gmail.com>
	<201204261435.37153.ne@vbart.ru>
	<CADKMG50+snfWNnx6xwnqWtc+hAEnBRiKXUiuKk2qwvrvR-vf4w@mail.gmail.com>
	<201204261826.55427.ne@vbart.ru>
	<CADKMG53HXY-rEGz7fpzGn7da3xqyx7Qf3mi1oAH-qGz7tEBgiw@mail.gmail.com>
Message-ID: <FF919B63-A52E-4482-88F7-F99E1F41024E@waeme.net>


On 27.04.2012, at 11:12, Delta Yeh wrote:

> I use loadrunner, and although get the conclusion:1.2.0 is slower than 1.1.12
> I observe connect timeout error in loadrunner console, but I didn't see such
> error in 1.1.12.
> 
> For 1.2.0 , the rps line dive to very lower every 15-20 seconds, I think
> it because  connect  to nginx timeout .
> 
> I test 1.1.17 , the result is almost the same as 1.1.12.
> I then test 1.1.19 , the result is  like 1.2.0,
> 
> So what happened between 1.1.17 and 1.1.19 will result the connection
> timeout every 15-20 seconds?

Could you show your config? And what OS are you using?


From delta.yeh at gmail.com  Fri Apr 27 08:43:09 2012
From: delta.yeh at gmail.com (Delta Yeh)
Date: Fri, 27 Apr 2012 16:43:09 +0800
Subject: nginx 1.2.0 is slower than 1.1.12?
In-Reply-To: <FF919B63-A52E-4482-88F7-F99E1F41024E@waeme.net>
References: <CADKMG51S1-hccqA0vqe-5axA6JYOhgFvxgH-1SoVuERTBY3HpA@mail.gmail.com>
	<201204261435.37153.ne@vbart.ru>
	<CADKMG50+snfWNnx6xwnqWtc+hAEnBRiKXUiuKk2qwvrvR-vf4w@mail.gmail.com>
	<201204261826.55427.ne@vbart.ru>
	<CADKMG53HXY-rEGz7fpzGn7da3xqyx7Qf3mi1oAH-qGz7tEBgiw@mail.gmail.com>
	<FF919B63-A52E-4482-88F7-F99E1F41024E@waeme.net>
Message-ID: <CADKMG53M02h3edQQKjVjTWR1qMSChBxW7F4OiKNU1rT8eGU0kQ@mail.gmail.com>

The loadrunner output of 1.1.19 is attached, there are many connection
timeout error  reported by loadrunner.

 My latest test show that 1.1.18 is OK , so changes between 1.1.18 and 1.1.19
make things worse in my test lab.

 the os is debian squeeze 32bit, the config is :

user root root;
worker_rlimit_nofile 81920;
worker_processes 2;
pid /secone/var/run/nginx.pid;
error_log /dev/null error;
pcre_jit on;
events {
	use epoll;
	worker_connections 12000;
	multi_accept on;
	accept_mutex off;
}
http {
	include mime.types;
	default_type application/octet-stream;
	sendfile on;
	tcp_nopush on;
	reset_timedout_connection on;
	ignore_invalid_headers on;
	underscores_in_headers on;
	proxy_buffering on;
	proxy_ignore_client_abort on;
	proxy_intercept_errors on;
	client_header_buffer_size 16k;
	large_client_header_buffers 4 16k;
	msie_padding off;
	msie_refresh off;
	access_log off;
	
	upstream upstream_www {
		server 192.168.2.3:80 max_fails=0 ;
		server 192.168.2.4:80 max_fails=0 ;
		server 192.168.2.5:80 max_fails=0 ;
		server 192.168.2.6:80 max_fails=0 ;
	}
	upstream upstream_www2 {
		server 192.168.2.254:80 max_fails=0 ;
	}

	server {
		
		listen 192.168.1.1:80 ;
		server_name _;
		proxy_set_header Host $http_host;
		location / {
			proxy_pass http://upstream_www2;
			proxy_redirect default;
		}
	}

	server {
	
		client_header_timeout 60;
		client_body_timeout 120;
		client_max_body_size 100m;
		client_header_buffer_size 16k;
		large_client_header_buffers 4 16k;
		proxy_buffer_size 16k;
		proxy_buffers 4 16k;
		listen 192.168.1.1:80 ;
		server_name www.test.com;
		
		ssl off;
		keepalive_timeout 0;
		proxy_cache_key $scheme$host$request_uri;

		location / {
			proxy_set_header Accept-Encoding identity;
			proxy_set_header Host $http_host;
			proxy_next_upstream error timeout http_500 http_502 http_503 http_504;
			proxy_cache off;
			proxy_pass http://upstream_www;
		}
	}

}

? 2012?4?27? ??4:08?Sergey Budnevitch <sb at waeme.net> ???
>
> On 27.04.2012, at 11:12, Delta Yeh wrote:
>
>> I use loadrunner, and although get the conclusion:1.2.0 is slower than 1.1.12
>> I observe connect timeout error in loadrunner console, but I didn't see such
>> error in 1.1.12.
>>
>> For 1.2.0 , the rps line dive to very lower every 15-20 seconds, I think
>> it because ?connect ?to nginx timeout .
>>
>> I test 1.1.17 , the result is almost the same as 1.1.12.
>> I then test 1.1.19 , the result is ?like 1.2.0,
>>
>> So what happened between 1.1.17 and 1.1.19 will result the connection
>> timeout every 15-20 seconds?
>
> Could you show your config? And what OS are you using?
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx
-------------- next part --------------
A non-text attachment was scrubbed...
Name: nginx-1.1.19.jpg
Type: image/jpeg
Size: 30524 bytes
Desc: not available
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20120427/27914a82/attachment-0001.jpg>

From delta.yeh at gmail.com  Fri Apr 27 10:51:04 2012
From: delta.yeh at gmail.com (Delta Yeh)
Date: Fri, 27 Apr 2012 18:51:04 +0800
Subject: nginx 1.2.0 is slower than 1.1.12?
In-Reply-To: <CADKMG53M02h3edQQKjVjTWR1qMSChBxW7F4OiKNU1rT8eGU0kQ@mail.gmail.com>
References: <CADKMG51S1-hccqA0vqe-5axA6JYOhgFvxgH-1SoVuERTBY3HpA@mail.gmail.com>
	<201204261435.37153.ne@vbart.ru>
	<CADKMG50+snfWNnx6xwnqWtc+hAEnBRiKXUiuKk2qwvrvR-vf4w@mail.gmail.com>
	<201204261826.55427.ne@vbart.ru>
	<CADKMG53HXY-rEGz7fpzGn7da3xqyx7Qf3mi1oAH-qGz7tEBgiw@mail.gmail.com>
	<FF919B63-A52E-4482-88F7-F99E1F41024E@waeme.net>
	<CADKMG53M02h3edQQKjVjTWR1qMSChBxW7F4OiKNU1rT8eGU0kQ@mail.gmail.com>
Message-ID: <CADKMG53QxfwyD2UG3YY18JU+3G0bcfCnypgMU6v+skcQYkVEog@mail.gmail.com>

After revert svn r4577 of 1.2.0 , I get almost the same benchmark
result with 1.2.0 and 1.1.17.

So it seems r4577 descrease nginx 1.2.0 performance by 30-40%( a lot
of connect timeout reported by loadrunner) in my test lab.

nginx box  OS is debian squeeze 32bit, CPU is intel core2 .


? 2012?4?27? ??4:43?Delta Yeh <delta.yeh at gmail.com> ???
> The loadrunner output of 1.1.19 is attached, there are many connection
> timeout error ?reported by loadrunner.
>
> ?My latest test show that 1.1.18 is OK , so changes between 1.1.18 and 1.1.19
> make things worse in my test lab.
>
> ?the os is debian squeeze 32bit, the config is :
>
> user root root;
> worker_rlimit_nofile 81920;
> worker_processes 2;
> pid /secone/var/run/nginx.pid;
> error_log /dev/null error;
> pcre_jit on;
> events {
> ? ? ? ?use epoll;
> ? ? ? ?worker_connections 12000;
> ? ? ? ?multi_accept on;
> ? ? ? ?accept_mutex off;
> }
> http {
> ? ? ? ?include mime.types;
> ? ? ? ?default_type application/octet-stream;
> ? ? ? ?sendfile on;
> ? ? ? ?tcp_nopush on;
> ? ? ? ?reset_timedout_connection on;
> ? ? ? ?ignore_invalid_headers on;
> ? ? ? ?underscores_in_headers on;
> ? ? ? ?proxy_buffering on;
> ? ? ? ?proxy_ignore_client_abort on;
> ? ? ? ?proxy_intercept_errors on;
> ? ? ? ?client_header_buffer_size 16k;
> ? ? ? ?large_client_header_buffers 4 16k;
> ? ? ? ?msie_padding off;
> ? ? ? ?msie_refresh off;
> ? ? ? ?access_log off;
>
> ? ? ? ?upstream upstream_www {
> ? ? ? ? ? ? ? ?server 192.168.2.3:80 max_fails=0 ;
> ? ? ? ? ? ? ? ?server 192.168.2.4:80 max_fails=0 ;
> ? ? ? ? ? ? ? ?server 192.168.2.5:80 max_fails=0 ;
> ? ? ? ? ? ? ? ?server 192.168.2.6:80 max_fails=0 ;
> ? ? ? ?}
> ? ? ? ?upstream upstream_www2 {
> ? ? ? ? ? ? ? ?server 192.168.2.254:80 max_fails=0 ;
> ? ? ? ?}
>
> ? ? ? ?server {
>
> ? ? ? ? ? ? ? ?listen 192.168.1.1:80 ;
> ? ? ? ? ? ? ? ?server_name _;
> ? ? ? ? ? ? ? ?proxy_set_header Host $http_host;
> ? ? ? ? ? ? ? ?location / {
> ? ? ? ? ? ? ? ? ? ? ? ?proxy_pass http://upstream_www2;
> ? ? ? ? ? ? ? ? ? ? ? ?proxy_redirect default;
> ? ? ? ? ? ? ? ?}
> ? ? ? ?}
>
> ? ? ? ?server {
>
> ? ? ? ? ? ? ? ?client_header_timeout 60;
> ? ? ? ? ? ? ? ?client_body_timeout 120;
> ? ? ? ? ? ? ? ?client_max_body_size 100m;
> ? ? ? ? ? ? ? ?client_header_buffer_size 16k;
> ? ? ? ? ? ? ? ?large_client_header_buffers 4 16k;
> ? ? ? ? ? ? ? ?proxy_buffer_size 16k;
> ? ? ? ? ? ? ? ?proxy_buffers 4 16k;
> ? ? ? ? ? ? ? ?listen 192.168.1.1:80 ;
> ? ? ? ? ? ? ? ?server_name www.test.com;
>
> ? ? ? ? ? ? ? ?ssl off;
> ? ? ? ? ? ? ? ?keepalive_timeout 0;
> ? ? ? ? ? ? ? ?proxy_cache_key $scheme$host$request_uri;
>
> ? ? ? ? ? ? ? ?location / {
> ? ? ? ? ? ? ? ? ? ? ? ?proxy_set_header Accept-Encoding identity;
> ? ? ? ? ? ? ? ? ? ? ? ?proxy_set_header Host $http_host;
> ? ? ? ? ? ? ? ? ? ? ? ?proxy_next_upstream error timeout http_500 http_502 http_503 http_504;
> ? ? ? ? ? ? ? ? ? ? ? ?proxy_cache off;
> ? ? ? ? ? ? ? ? ? ? ? ?proxy_pass http://upstream_www;
> ? ? ? ? ? ? ? ?}
> ? ? ? ?}
>
> }
>
> ? 2012?4?27? ??4:08?Sergey Budnevitch <sb at waeme.net> ???
>>
>> On 27.04.2012, at 11:12, Delta Yeh wrote:
>>
>>> I use loadrunner, and although get the conclusion:1.2.0 is slower than 1.1.12
>>> I observe connect timeout error in loadrunner console, but I didn't see such
>>> error in 1.1.12.
>>>
>>> For 1.2.0 , the rps line dive to very lower every 15-20 seconds, I think
>>> it because ?connect ?to nginx timeout .
>>>
>>> I test 1.1.17 , the result is almost the same as 1.1.12.
>>> I then test 1.1.19 , the result is ?like 1.2.0,
>>>
>>> So what happened between 1.1.17 and 1.1.19 will result the connection
>>> timeout every 15-20 seconds?
>>
>> Could you show your config? And what OS are you using?
>>
>> _______________________________________________
>> nginx mailing list
>> nginx at nginx.org
>> http://mailman.nginx.org/mailman/listinfo/nginx

From mdounin at mdounin.ru  Fri Apr 27 10:57:05 2012
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Fri, 27 Apr 2012 14:57:05 +0400
Subject: nginx 1.2.0 is slower than 1.1.12?
In-Reply-To: <CADKMG53QxfwyD2UG3YY18JU+3G0bcfCnypgMU6v+skcQYkVEog@mail.gmail.com>
References: <CADKMG51S1-hccqA0vqe-5axA6JYOhgFvxgH-1SoVuERTBY3HpA@mail.gmail.com>
	<201204261435.37153.ne@vbart.ru>
	<CADKMG50+snfWNnx6xwnqWtc+hAEnBRiKXUiuKk2qwvrvR-vf4w@mail.gmail.com>
	<201204261826.55427.ne@vbart.ru>
	<CADKMG53HXY-rEGz7fpzGn7da3xqyx7Qf3mi1oAH-qGz7tEBgiw@mail.gmail.com>
	<FF919B63-A52E-4482-88F7-F99E1F41024E@waeme.net>
	<CADKMG53M02h3edQQKjVjTWR1qMSChBxW7F4OiKNU1rT8eGU0kQ@mail.gmail.com>
	<CADKMG53QxfwyD2UG3YY18JU+3G0bcfCnypgMU6v+skcQYkVEog@mail.gmail.com>
Message-ID: <20120427105705.GM31671@mdounin.ru>

Hello!

On Fri, Apr 27, 2012 at 06:51:04PM +0800, Delta Yeh wrote:

> After revert svn r4577 of 1.2.0 , I get almost the same benchmark
> result with 1.2.0 and 1.1.17.
> 
> So it seems r4577 descrease nginx 1.2.0 performance by 30-40%( a lot
> of connect timeout reported by loadrunner) in my test lab.
> 
> nginx box  OS is debian squeeze 32bit, CPU is intel core2 .

Which compiler do you use?  Which compiler flags used during 
compilation?

Maxim Dounin


From delta.yeh at gmail.com  Fri Apr 27 11:06:02 2012
From: delta.yeh at gmail.com (Delta Yeh)
Date: Fri, 27 Apr 2012 19:06:02 +0800
Subject: nginx 1.2.0 is slower than 1.1.12?
In-Reply-To: <20120427105705.GM31671@mdounin.ru>
References: <CADKMG51S1-hccqA0vqe-5axA6JYOhgFvxgH-1SoVuERTBY3HpA@mail.gmail.com>
	<201204261435.37153.ne@vbart.ru>
	<CADKMG50+snfWNnx6xwnqWtc+hAEnBRiKXUiuKk2qwvrvR-vf4w@mail.gmail.com>
	<201204261826.55427.ne@vbart.ru>
	<CADKMG53HXY-rEGz7fpzGn7da3xqyx7Qf3mi1oAH-qGz7tEBgiw@mail.gmail.com>
	<FF919B63-A52E-4482-88F7-F99E1F41024E@waeme.net>
	<CADKMG53M02h3edQQKjVjTWR1qMSChBxW7F4OiKNU1rT8eGU0kQ@mail.gmail.com>
	<CADKMG53QxfwyD2UG3YY18JU+3G0bcfCnypgMU6v+skcQYkVEog@mail.gmail.com>
	<20120427105705.GM31671@mdounin.ru>
Message-ID: <CADKMG50qghZthHHj+a-SR89_YQav8k+swxx63HH8czYzbr9WaQ@mail.gmail.com>

Using built-in specs.
Target: i486-linux-gnu
Configured with: ../src/configure -v --with-pkgversion='Debian
4.3.5-4' --with-bugurl=file:///usr/share/doc/gcc-4.3/README.Bugs
--enable-languages=c,c++,fortran,objc,obj-c++ --prefix=/usr
--enable-shared --enable-multiarch --enable-linker-build-id
--with-system-zlib --libexecdir=/usr/lib --without-included-gettext
--enable-threads=posix --enable-nls
--with-gxx-include-dir=/usr/include/c++/4.3 --program-suffix=-4.3
--enable-clocale=gnu --enable-libstdcxx-debug --enable-objc-gc
--enable-mpfr --enable-targets=all --with-tune=generic
--enable-checking=release --build=i486-linux-gnu --host=i486-linux-gnu
--target=i486-linux-gnu
Thread model: posix
gcc version 4.3.5 (Debian 4.3.5-4)




./nginx  -V
nginx version: nginx/2.1.2.0
built by gcc 4.3.5 (Debian 4.3.5-4)
TLS SNI support enabled
configure arguments: --prefix=/usr/local/nginx --pid-path=/var
--with-cpu-opt=pentium4 --with-ipv6 --with-pcre=/usr/local/pcre
--with-pcre-jit --without-http_autoindex_module
--without-http_ssi_module --without-http_referer_module
--without-http_userid_module --without-http_empty_gif_module
--without-http_limit_req_module --without-http_browser_module
--without-http_memcached_module --without-http_charset_module
--without-http_split_clients_module --with-http_stub_status_module
--with-http_ssl_module --with-http_realip_module
--with-http_sub_module --with-http_geoip_module -


? 2012?4?27? ??6:57?Maxim Dounin <mdounin at mdounin.ru> ???
> Hello!
>
> On Fri, Apr 27, 2012 at 06:51:04PM +0800, Delta Yeh wrote:
>
>> After revert svn r4577 of 1.2.0 , I get almost the same benchmark
>> result with 1.2.0 and 1.1.17.
>>
>> So it seems r4577 descrease nginx 1.2.0 performance by 30-40%( a lot
>> of connect timeout reported by loadrunner) in my test lab.
>>
>> nginx box ?OS is debian squeeze 32bit, CPU is intel core2 .
>
> Which compiler do you use? ?Which compiler flags used during
> compilation?
>
> Maxim Dounin
>
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx


From mdounin at mdounin.ru  Fri Apr 27 12:22:41 2012
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Fri, 27 Apr 2012 16:22:41 +0400
Subject: nginx 1.2.0 is slower than 1.1.12?
In-Reply-To: <CADKMG50qghZthHHj+a-SR89_YQav8k+swxx63HH8czYzbr9WaQ@mail.gmail.com>
References: <CADKMG51S1-hccqA0vqe-5axA6JYOhgFvxgH-1SoVuERTBY3HpA@mail.gmail.com>
	<201204261435.37153.ne@vbart.ru>
	<CADKMG50+snfWNnx6xwnqWtc+hAEnBRiKXUiuKk2qwvrvR-vf4w@mail.gmail.com>
	<201204261826.55427.ne@vbart.ru>
	<CADKMG53HXY-rEGz7fpzGn7da3xqyx7Qf3mi1oAH-qGz7tEBgiw@mail.gmail.com>
	<FF919B63-A52E-4482-88F7-F99E1F41024E@waeme.net>
	<CADKMG53M02h3edQQKjVjTWR1qMSChBxW7F4OiKNU1rT8eGU0kQ@mail.gmail.com>
	<CADKMG53QxfwyD2UG3YY18JU+3G0bcfCnypgMU6v+skcQYkVEog@mail.gmail.com>
	<20120427105705.GM31671@mdounin.ru>
	<CADKMG50qghZthHHj+a-SR89_YQav8k+swxx63HH8czYzbr9WaQ@mail.gmail.com>
Message-ID: <20120427122241.GQ31671@mdounin.ru>

Hello!

On Fri, Apr 27, 2012 at 07:06:02PM +0800, Delta Yeh wrote:

> Using built-in specs.
> Target: i486-linux-gnu
> Configured with: ../src/configure -v --with-pkgversion='Debian
> 4.3.5-4' --with-bugurl=file:///usr/share/doc/gcc-4.3/README.Bugs
> --enable-languages=c,c++,fortran,objc,obj-c++ --prefix=/usr
> --enable-shared --enable-multiarch --enable-linker-build-id
> --with-system-zlib --libexecdir=/usr/lib --without-included-gettext
> --enable-threads=posix --enable-nls
> --with-gxx-include-dir=/usr/include/c++/4.3 --program-suffix=-4.3
> --enable-clocale=gnu --enable-libstdcxx-debug --enable-objc-gc
> --enable-mpfr --enable-targets=all --with-tune=generic
> --enable-checking=release --build=i486-linux-gnu --host=i486-linux-gnu
> --target=i486-linux-gnu
> Thread model: posix
> gcc version 4.3.5 (Debian 4.3.5-4)
> 
> 
> 
> 
> ./nginx  -V
> nginx version: nginx/2.1.2.0

This doesn't looks like vanilla nginx.  Could you please test 
vanilla one, without any patches?

> built by gcc 4.3.5 (Debian 4.3.5-4)
> TLS SNI support enabled
> configure arguments: --prefix=/usr/local/nginx --pid-path=/var
> --with-cpu-opt=pentium4 --with-ipv6 --with-pcre=/usr/local/pcre
> --with-pcre-jit --without-http_autoindex_module
> --without-http_ssi_module --without-http_referer_module
> --without-http_userid_module --without-http_empty_gif_module
> --without-http_limit_req_module --without-http_browser_module
> --without-http_memcached_module --without-http_charset_module
> --without-http_split_clients_module --with-http_stub_status_module
> --with-http_ssl_module --with-http_realip_module
> --with-http_sub_module --with-http_geoip_module -

The output looks truncated (note trailing "-"), is it?

In any case, please provide sample of an actual gcc command as 
used during compilation of nginx object files, it may contain 
various flags not visible in "nginx -V" output.

Maxim Dounin

> 
> 
> ? 2012?4?27? ??6:57?Maxim Dounin <mdounin at mdounin.ru> ???
> > Hello!
> >
> > On Fri, Apr 27, 2012 at 06:51:04PM +0800, Delta Yeh wrote:
> >
> >> After revert svn r4577 of 1.2.0 , I get almost the same benchmark
> >> result with 1.2.0 and 1.1.17.
> >>
> >> So it seems r4577 descrease nginx 1.2.0 performance by 30-40%( a lot
> >> of connect timeout reported by loadrunner) in my test lab.
> >>
> >> nginx box ?OS is debian squeeze 32bit, CPU is intel core2 .
> >
> > Which compiler do you use? ?Which compiler flags used during
> > compilation?
> >
> > 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 delta.yeh at gmail.com  Fri Apr 27 13:08:45 2012
From: delta.yeh at gmail.com (Delta Yeh)
Date: Fri, 27 Apr 2012 21:08:45 +0800
Subject: nginx 1.2.0 is slower than 1.1.12?
In-Reply-To: <20120427122241.GQ31671@mdounin.ru>
References: <CADKMG51S1-hccqA0vqe-5axA6JYOhgFvxgH-1SoVuERTBY3HpA@mail.gmail.com>
	<201204261435.37153.ne@vbart.ru>
	<CADKMG50+snfWNnx6xwnqWtc+hAEnBRiKXUiuKk2qwvrvR-vf4w@mail.gmail.com>
	<201204261826.55427.ne@vbart.ru>
	<CADKMG53HXY-rEGz7fpzGn7da3xqyx7Qf3mi1oAH-qGz7tEBgiw@mail.gmail.com>
	<FF919B63-A52E-4482-88F7-F99E1F41024E@waeme.net>
	<CADKMG53M02h3edQQKjVjTWR1qMSChBxW7F4OiKNU1rT8eGU0kQ@mail.gmail.com>
	<CADKMG53QxfwyD2UG3YY18JU+3G0bcfCnypgMU6v+skcQYkVEog@mail.gmail.com>
	<20120427105705.GM31671@mdounin.ru>
	<CADKMG50qghZthHHj+a-SR89_YQav8k+swxx63HH8czYzbr9WaQ@mail.gmail.com>
	<20120427122241.GQ31671@mdounin.ru>
Message-ID: <CADKMG53LW55-xD1DCq5=VTZkj=U+WTHrY+XF=tH4c2o=hZFEEg@mail.gmail.com>

? 2012?4?27? ??8:22?Maxim Dounin <mdounin at mdounin.ru> ???
> Hello!
>
> On Fri, Apr 27, 2012 at 07:06:02PM +0800, Delta Yeh wrote:
>
>> Using built-in specs.
>> Target: i486-linux-gnu
>> Configured with: ../src/configure -v --with-pkgversion='Debian
>> 4.3.5-4' --with-bugurl=file:///usr/share/doc/gcc-4.3/README.Bugs
>> --enable-languages=c,c++,fortran,objc,obj-c++ --prefix=/usr
>> --enable-shared --enable-multiarch --enable-linker-build-id
>> --with-system-zlib --libexecdir=/usr/lib --without-included-gettext
>> --enable-threads=posix --enable-nls
>> --with-gxx-include-dir=/usr/include/c++/4.3 --program-suffix=-4.3
>> --enable-clocale=gnu --enable-libstdcxx-debug --enable-objc-gc
>> --enable-mpfr --enable-targets=all --with-tune=generic
>> --enable-checking=release --build=i486-linux-gnu --host=i486-linux-gnu
>> --target=i486-linux-gnu
>> Thread model: posix
>> gcc version 4.3.5 (Debian 4.3.5-4)
>>
>>
>>
>>
>> ./nginx ?-V
>> nginx version: nginx/2.1.2.0
>
> This doesn't looks like vanilla nginx. ?Could you please test
> vanilla one, without any patches?
>

It's an openresty like nginx bundle ,consist of modules for reverse proxy only.
No thirdparty patch applied except  the one "proxy_bind support variable".
Maxim, I remember it is from you, :)

The major 2 in version string means upstream keepalive feaure only .



>> built by gcc 4.3.5 (Debian 4.3.5-4)
>> TLS SNI support enabled
>> configure arguments: --prefix=/usr/local/nginx --pid-path=/var
>> --with-cpu-opt=pentium4 --with-ipv6 --with-pcre=/usr/local/pcre
>> --with-pcre-jit --without-http_autoindex_module
>> --without-http_ssi_module --without-http_referer_module
>> --without-http_userid_module --without-http_empty_gif_module
>> --without-http_limit_req_module --without-http_browser_module
>> --without-http_memcached_module --without-http_charset_module
>> --without-http_split_clients_module --with-http_stub_status_module
>> --with-http_ssl_module --with-http_realip_module
>> --with-http_sub_module --with-http_geoip_module -
>
> The output looks truncated (note trailing "-"), is it?
>
> In any case, please provide sample of an actual gcc command as
> used during compilation of nginx object files, it may contain
> various flags not visible in "nginx -V" output.
>

The shell script to build nginx is :

DEPLOY_TARGET="/home/ngxproxy"
INSTALL_DIR="/home/ngxproxy"
# tell nginx's build system where to find lua:
#export LUA_LIB=/usr/local/lib
#export LUA_INC=/usr/local/include
#export LUA_LIB=/usr/lib
#export LUA_INC=/usr/include/lua5.1

# or tell where to find LuaJIT when you want to use JIT instead
export LUAJIT_LIB=$DEPLOY_TARGET/lib
export LUAJIT_INC=$DEPLOY_TARGET/include/luajit-2.0


./configure --prefix=$INSTALL_DIR  \
--with-cpu-opt=pentium4 \
--with-ipv6  \
--with-pcre=$DEPLOY_TARGET/pcre  \
--with-pcre-jit \
--with-pcre-jit \
--without-http_autoindex_module \
--without-http_ssi_module \
--without-http_referer_module  \
--without-http_userid_module   \
--without-http_empty_gif_module  \
--without-http_limit_req_module  \
--without-http_browser_module    \
--without-http_memcached_module \
--without-http_charset_module \
--without-http_split_clients_module \
--with-http_stub_status_module \
--with-http_ssl_module \
--with-http_realip_module \
--with-http_sub_module \
--with-http_geoip_module \
--add-module=src/http/modules/ngx_devel_kit \
--add-module=src/http/modules/lua-nginx  \
--add-module=src/http/modules/headers_more \
--add-module=src/http/modules/substitute \
--add-module=src/http/modules/ngx_iconv


make
make install



> Maxim Dounin
>
>>
>>
>> ? 2012?4?27? ??6:57?Maxim Dounin <mdounin at mdounin.ru> ???
>> > Hello!
>> >
>> > On Fri, Apr 27, 2012 at 06:51:04PM +0800, Delta Yeh wrote:
>> >
>> >> After revert svn r4577 of 1.2.0 , I get almost the same benchmark
>> >> result with 1.2.0 and 1.1.17.
>> >>
>> >> So it seems r4577 descrease nginx 1.2.0 performance by 30-40%( a lot
>> >> of connect timeout reported by loadrunner) in my test lab.
>> >>
>> >> nginx box ?OS is debian squeeze 32bit, CPU is intel core2 .
>> >
>> > Which compiler do you use? ?Which compiler flags used during
>> > compilation?
>> >
>> > 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 weiyue at taobao.com  Fri Apr 27 13:18:15 2012
From: weiyue at taobao.com (=?gb2312?B?zsDUvQ==?=)
Date: Fri, 27 Apr 2012 21:18:15 +0800
Subject: =?UTF-8?Q?=E7=AD=94=E5=A4=8D=3A_limiting_upstream_backend_connections?=
In-Reply-To: <CABFfbXv8Gg2rQFvvfOgbnY6jPWVEHpVu9iydFSgUWuA5V9XdSg@mail.gmail.com>
References: <CABFfbXv8Gg2rQFvvfOgbnY6jPWVEHpVu9iydFSgUWuA5V9XdSg@mail.gmail.com>
Message-ID: <8B08B66A4C097F428BF7BCC5B245019B617855F640@CNHZ-EXCMS-04.ali.com>

I think this module will fit your demands:
https://github.com/cfsego/nginx-limit-upstream

Sunny Chen

-----????-----
???: nginx-bounces at nginx.org [mailto:nginx-bounces at nginx.org] ?? Jeroen Ooms
????: 2012?4?27? 8:33
???: nginx at nginx.org
??: limiting upstream backend connections

I am hosting a CPU intensive service and use nginx as a reverse proxy
cache / load balancer to my backend(s). Basic config looks like:

upstream myservice {
    server 127.0.0.1:80;
    server 123.123.123.123:80;
}

limit_req_zone $binary_remote_addr  zone=softlimit:10m   rate=1r/s;

location /R {
    proxy_pass  http://myservice/R;
    proxy_cache mycache;
    proxy_cache_methods POST;
    proxy_cache_key "$request_uri|$request_body";
    limit_req zone=softlimit  burst=10;
}

So I use limit_req to make sure that no single user can do more than 1
req per sec to my service.

However, actually more important is that the upstream backend servers
are protected against overload. Some of the requests might take 30 or
60 seconds to return. I want to limit the total number of active
concurrent connections to the upstream back-end at, say, 10. Requests
that hit the cache and never make it to the backend should not be
counted towards this. Only active backend connections.

I had a look at limit_conn, but it looks like this limits the client
connections rather than the back-end connections, is that correct?

Thanks for advice,

Jeroen

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

This email (including any attachments) is confidential and may be legally privileged. If you received this email in error, please delete it immediately and do not copy it or use it for any purpose or disclose its contents to any other person. Thank you.

???(??????)??????????????????????????????????????????????????????????????????????

From mdounin at mdounin.ru  Fri Apr 27 14:00:42 2012
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Fri, 27 Apr 2012 18:00:42 +0400
Subject: nginx 1.2.0 is slower than 1.1.12?
In-Reply-To: <CADKMG53LW55-xD1DCq5=VTZkj=U+WTHrY+XF=tH4c2o=hZFEEg@mail.gmail.com>
References: <CADKMG50+snfWNnx6xwnqWtc+hAEnBRiKXUiuKk2qwvrvR-vf4w@mail.gmail.com>
	<201204261826.55427.ne@vbart.ru>
	<CADKMG53HXY-rEGz7fpzGn7da3xqyx7Qf3mi1oAH-qGz7tEBgiw@mail.gmail.com>
	<FF919B63-A52E-4482-88F7-F99E1F41024E@waeme.net>
	<CADKMG53M02h3edQQKjVjTWR1qMSChBxW7F4OiKNU1rT8eGU0kQ@mail.gmail.com>
	<CADKMG53QxfwyD2UG3YY18JU+3G0bcfCnypgMU6v+skcQYkVEog@mail.gmail.com>
	<20120427105705.GM31671@mdounin.ru>
	<CADKMG50qghZthHHj+a-SR89_YQav8k+swxx63HH8czYzbr9WaQ@mail.gmail.com>
	<20120427122241.GQ31671@mdounin.ru>
	<CADKMG53LW55-xD1DCq5=VTZkj=U+WTHrY+XF=tH4c2o=hZFEEg@mail.gmail.com>
Message-ID: <20120427140041.GS31671@mdounin.ru>

Hello!

On Fri, Apr 27, 2012 at 09:08:45PM +0800, Delta Yeh wrote:

> ? 2012?4?27? ??8:22?Maxim Dounin <mdounin at mdounin.ru> ???
> > Hello!
> >
> > On Fri, Apr 27, 2012 at 07:06:02PM +0800, Delta Yeh wrote:
> >
> >> Using built-in specs.
> >> Target: i486-linux-gnu
> >> Configured with: ../src/configure -v --with-pkgversion='Debian
> >> 4.3.5-4' --with-bugurl=file:///usr/share/doc/gcc-4.3/README.Bugs
> >> --enable-languages=c,c++,fortran,objc,obj-c++ --prefix=/usr
> >> --enable-shared --enable-multiarch --enable-linker-build-id
> >> --with-system-zlib --libexecdir=/usr/lib --without-included-gettext
> >> --enable-threads=posix --enable-nls
> >> --with-gxx-include-dir=/usr/include/c++/4.3 --program-suffix=-4.3
> >> --enable-clocale=gnu --enable-libstdcxx-debug --enable-objc-gc
> >> --enable-mpfr --enable-targets=all --with-tune=generic
> >> --enable-checking=release --build=i486-linux-gnu --host=i486-linux-gnu
> >> --target=i486-linux-gnu
> >> Thread model: posix
> >> gcc version 4.3.5 (Debian 4.3.5-4)
> >>
> >>
> >>
> >>
> >> ./nginx ?-V
> >> nginx version: nginx/2.1.2.0
> >
> > This doesn't looks like vanilla nginx. ?Could you please test
> > vanilla one, without any patches?
> >
> 
> It's an openresty like nginx bundle ,consist of modules for reverse proxy only.
> No thirdparty patch applied except  the one "proxy_bind support variable".
> Maxim, I remember it is from you, :)
> 
> The major 2 in version string means upstream keepalive feaure only .

Again: it's really good idea to test vanilla nginx, without any 
patches and 3rd party modules, especially when debugging such 
subtle problems.  Even trivial patch might cause unspecified 
behaviour e.g. if it applied incorrectly due to context code 
changes.  And the similar thing applies to 3rd party modules.

> >> built by gcc 4.3.5 (Debian 4.3.5-4)
> >> TLS SNI support enabled
> >> configure arguments: --prefix=/usr/local/nginx --pid-path=/var
> >> --with-cpu-opt=pentium4 --with-ipv6 --with-pcre=/usr/local/pcre
> >> --with-pcre-jit --without-http_autoindex_module
> >> --without-http_ssi_module --without-http_referer_module
> >> --without-http_userid_module --without-http_empty_gif_module
> >> --without-http_limit_req_module --without-http_browser_module
> >> --without-http_memcached_module --without-http_charset_module
> >> --without-http_split_clients_module --with-http_stub_status_module
> >> --with-http_ssl_module --with-http_realip_module
> >> --with-http_sub_module --with-http_geoip_module -
> >
> > The output looks truncated (note trailing "-"), is it?
> >
> > In any case, please provide sample of an actual gcc command as
> > used during compilation of nginx object files, it may contain
> > various flags not visible in "nginx -V" output.
> >
> 
> The shell script to build nginx is :

I need a gcc command as shown during "make", like this:

gcc -c -pipe  -O -W -Wall -Wpointer-arith -Wno-unused-parameter -Wunused-function -Wunused-variable -Wunused-value -Werror -g  -I src/core -I src/event -I src/event/modules -I src/os/unix -I objs \
		-o objs/src/core/nginx.o \
		src/core/nginx.c

Maxim Dounin


From nginx-forum at nginx.us  Fri Apr 27 18:29:27 2012
From: nginx-forum at nginx.us (Nullivex)
Date: Fri, 27 Apr 2012 14:29:27 -0400 (EDT)
Subject: no mp4 trak atoms were found in
Message-ID: <44f9f08954470017315ce66556a3c10d.NginxMailingListEnglish@forum.nginx.org>

Hey Guys,

I am getting a ton of these errors on certain videos.

"no mp4 trak atoms were found in"

I have tried changing the mp4_max_buffer_size. However, this has not
helped me solve the problem.

Is there something else I can try to solve this?

Version
~# /usr/local/nginx/sbin/nginx -v
nginx version: nginx/1.0.15

OS
Debian 6 x86_64 (Fully updated)

Complete log entry
2012/04/27 11:27:39 [error] 19642#0: *1714719 no mp4 trak atoms were
found in "/cache/7b74b8312d200e9db307a4c26c39121f", client:
190.66.16.106, server: _, request: "GET
/cache/stream/mp4/7b74b8312d200e9db307a4c26c39121f?start=0&file_type=video/mp4&file_name=047.mp4&file_size=33046909
HTTP/1.1", host: "xxx", referrer: "http://xxx/jwplayer/player.swf"

Let me know if there is anything you need.

Thanks

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225883,225883#msg-225883


From nginx-forum at nginx.us  Fri Apr 27 18:51:52 2012
From: nginx-forum at nginx.us (Nullivex)
Date: Fri, 27 Apr 2012 14:51:52 -0400 (EDT)
Subject: no mp4 trak atoms were found in
In-Reply-To: <44f9f08954470017315ce66556a3c10d.NginxMailingListEnglish@forum.nginx.org>
References: <44f9f08954470017315ce66556a3c10d.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <bef0a9e4d253b479819e8dc503897336.NginxMailingListEnglish@forum.nginx.org>

Actually,

I did a little more testing it seems this is coming from video files
that missed bytes on the upload.

I was able to fix this by re-uploading the file.

Thanks

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225883,225884#msg-225884


From nginx-forum at nginx.us  Sat Apr 28 00:12:42 2012
From: nginx-forum at nginx.us (ctrlbrk)
Date: Fri, 27 Apr 2012 20:12:42 -0400 (EDT)
Subject: Does nginx plan to support spdy?
In-Reply-To: <C2B88226-C48B-460C-8CC0-523EF83E599D@nginx.com>
References: <C2B88226-C48B-460C-8CC0-523EF83E599D@nginx.com>
Message-ID: <d201bd7fdd2bb596d64d3cb9143d5600.NginxMailingListEnglish@forum.nginx.org>

Firefox 13 (in beta today, scheduled for public release near end of May)
will enable SPDY by default.

Any update regarding nginx support?

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,217299,225885#msg-225885


From admin at yqed.com  Sat Apr 28 01:47:03 2012
From: admin at yqed.com (Floren Munteanu)
Date: Fri, 27 Apr 2012 21:47:03 -0400
Subject: Does nginx plan to support spdy?
In-Reply-To: <d201bd7fdd2bb596d64d3cb9143d5600.NginxMailingListEnglish@forum.nginx.org>
References: <C2B88226-C48B-460C-8CC0-523EF83E599D@nginx.com>
	<d201bd7fdd2bb596d64d3cb9143d5600.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <jnfi6j$7t1$1@dough.gmane.org>

http://trac.nginx.org/nginx/ticket/30

On 4/27/2012 8:12 PM, ctrlbrk wrote:
> Firefox 13 (in beta today, scheduled for public release near end of May)
> will enable SPDY by default.
>
> Any update regarding nginx support?
>
> Posted at Nginx Forum: http://forum.nginx.org/read.php?2,217299,225885#msg-225885


From steve.gray.wipro at sonymusic.com  Sat Apr 28 01:59:15 2012
From: steve.gray.wipro at sonymusic.com (Gray, Steve, Wipro)
Date: Fri, 27 Apr 2012 21:59:15 -0400
Subject: nginx: [crit] pread() "/opt/nginx" failed (21: Is a directory)
Message-ID: <C15BF77669E54C498451E3A25D44F0A702AD2268D578@NYCMNET7EV001.mnet.biz>


Hi,
I am using nginx 1.0.15 running on Solaris 10.  When I issue my "/opt/nginx/sbin/nginx -s stop" command, I get this error "nginx: [crit] pread() "/opt/nginx" failed (21: Is a directory)". The funny thing is that I can issue "/opt/nginx/sbin/nginx" to start nginx and get no errors.  Nginx starts okay.  I can also issue "/opt/nginx/sbin/nginx -V or v" and get no errors.  The "nginx: [crit] pread() "/opt/nginx" failed (21: Is a directory)" error happens when I issue the command with the "-s stop|reload|quit|reopen" options.
My configure argument is --prefix=/opt/nginx. Has anyone seen this or know why this is happening?  Let me know if you need any more information from me.

Thanks
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20120427/5b1d04c9/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: image001.gif
Type: image/gif
Size: 530 bytes
Desc: image001.gif
URL: <http://mailman.nginx.org/pipermail/nginx/attachments/20120427/5b1d04c9/attachment.gif>

From nginx-forum at nginx.us  Sat Apr 28 02:05:09 2012
From: nginx-forum at nginx.us (dbanks)
Date: Fri, 27 Apr 2012 22:05:09 -0400 (EDT)
Subject: Inconsistent latency to sub_status
In-Reply-To: <bd7c9b57b6cc36a0f64b87bc5736aef7.NginxMailingListEnglish@forum.nginx.org>
References: <bd7c9b57b6cc36a0f64b87bc5736aef7.NginxMailingListEnglish@forum.nginx.org>
Message-ID: <db8d7ecaeb0ad982356b3c7175e24c3f.NginxMailingListEnglish@forum.nginx.org>

I found the issue -- this server handles a mix of SSL and regular
traffic.  When initially configuring SSL, I used a 4096-bit CSR and did
not manage cypher selection carefully.  As a result, nginx workers spent
a lot of time in OpenSSL, calls unable to service other requests.  The
ratio of SSL to regular traffic made the ping latency erratic.  

The fix was straightforward:  use a smaller CSR and select
server-preferred cyphers that were less computationally intensive.

Cheers,

Dean

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,221270,225889#msg-225889


From andrew at nginx.com  Sat Apr 28 06:50:44 2012
From: andrew at nginx.com (Andrew Alexeev)
Date: Sat, 28 Apr 2012 10:50:44 +0400
Subject: Does nginx plan to support spdy?
In-Reply-To: <jnfi6j$7t1$1@dough.gmane.org>
References: <C2B88226-C48B-460C-8CC0-523EF83E599D@nginx.com>
	<d201bd7fdd2bb596d64d3cb9143d5600.NginxMailingListEnglish@forum.nginx.org>
	<jnfi6j$7t1$1@dough.gmane.org>
Message-ID: <8CB46FEB-3D67-4E93-BACB-161F091D0D9B@nginx.com>

On Apr 28, 2012, at 5:47 AM, Floren Munteanu wrote:

> http://trac.nginx.org/nginx/ticket/30

Also this
http://trac.nginx.org/nginx/roadmap

The work's ongoing.

> On 4/27/2012 8:12 PM, ctrlbrk wrote:
>> Firefox 13 (in beta today, scheduled for public release near end of May)
>> will enable SPDY by default.
>> 
>> Any update regarding nginx support?
>> 
>> Posted at Nginx Forum: http://forum.nginx.org/read.php?2,217299,225885#msg-225885
> 
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx


From agentzh at gmail.com  Sun Apr 29 03:58:42 2012
From: agentzh at gmail.com (agentzh)
Date: Sun, 29 Apr 2012 11:58:42 +0800
Subject: [ANN] ngx_openresty devel version 1.0.15.1 released!
In-Reply-To: <CAB4Tn6OCm+BdX9tvAUnznvGWY3xPx6NOvTiQcJri7jGE=GO=oA@mail.gmail.com>
References: <CAB4Tn6OCm+BdX9tvAUnznvGWY3xPx6NOvTiQcJri7jGE=GO=oA@mail.gmail.com>
Message-ID: <CAB4Tn6MsD=QpV=YSK=pQL6V2htt7aK0ep3w1+bWd=0+DHOP6PA@mail.gmail.com>

Hi, folks!

After more than one month's hacking, I'm happy to announce the new
development release of ngx_openresty, 1.0.15.1:

? ?http://openresty.org/#Download

Complete change log (compared to 1.0.11.28) can be seen here:

? ?http://openresty.org/#ChangeLog1000015

This is the first ngx_openresty release based on the Nginx 1.0.15
core. (And yeah, we will eventually move onto the 1.2.x stable branch
soon!)

Special thanks go to all our contributors and users for helping make
this happen :)

We've started running extensive regression tests in our test cluster
based on Amazon EC2. And you can always view the latest test report
here:

    http://qa.openresty.org/

We're still adding more operating systems and more Nginx versions to the list!

OpenResty (aka. ngx_openresty) is a full-fledged web application
server by bundling the standard Nginx core, lots of 3rd-party Nginx
modules, as well as most of their external dependencies.

By taking advantage 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.

See OpenResty's web site for more details:

? ?http://openresty.org/

Have fun!
-agentzh


From siefke_listen at web.de  Sun Apr 29 17:31:34 2012
From: siefke_listen at web.de (Silvio Siefke)
Date: Sun, 29 Apr 2012 19:31:34 +0200
Subject: HTML as PHP
In-Reply-To: <20120418001749.8d31e902.siefke_listen@web.de>
References: <20120418001749.8d31e902.siefke_listen@web.de>
Message-ID: <20120429193134.06273ca0.siefke_listen@web.de>

Hello, 

when used Nginx with PHP-FPM, should set the follow in your fpm config. 

security.limit_extensions = .php .html


Regards
Silvio


From dmiller at amfes.com  Sun Apr 29 20:06:03 2012
From: dmiller at amfes.com (Daniel L. Miller)
Date: Sun, 29 Apr 2012 13:06:03 -0700
Subject: Location - or how to setup sites in subfolders
Message-ID: <assp.046696f332.4F9D9F2B.5050605@amfes.com>

I'm trying to give nginx a try as an alternative to my current server 
solutions.  Named-based virtual hosting SEEMS straightforward enough - 
but I'm having a devil of a time trying to get subfolder based sites to 
work.  In other words -

egroupware.mydomain.com
roundcubemail.mydomain.com
ldap-account-manager.mydomain.com

for some examples, work just fine.  But trying to get -

www.mydomain.com/egroupware
www.mydomain.com/roundcubemail
www.mydomain.com/ldap-account-manager

to work is an exercise in extreme frustration.  I'm sure it doesn't HAVE 
to be - but Google thus far has let me down in retrieving the 
information I need.  What I've done thus far, for egroupware for example:

I have a site file with:
server {
     server_name www.amfeslan.local egroupware.amfeslan.local;
     root /opt/egroupware;
     index index.php;

     client_max_body_size 8M;
     rewrite ^/egroupware/(.*)$ /$1 last;
     try_files $uri $uri/ /index.php$args;

     include global/php.conf;
}

my global/php.conf has:
location ~ \.php$ {
         # There's gotta be a better way to do this - try and find it.
         # Since nginx's "best" match is this one - need to do rewrites
         # here to get subfolders to work.
         set $php_root $document_root;
         if ($request_uri ~* /egroupware) {
             set $php_root /opt/egroupware;
         }

         # Zero-day exploit defense.
         try_files $uri =404;

         fastcgi_split_path_info ^(.+\.php)(/.+)$;

         include fastcgi_params;
         fastcgi_index index.php;
         fastcgi_param SCRIPT_FILENAME $php_root$fastcgi_script_name;
         fastcgi_pass php;
}

Note that I tried to add the "if" construct in php.conf based on a 
previous post - and I gotta believe there's a better way to do this.  
With or without the if construct, the above site works - but only for 
egroupware.  Revising the site file to:
server {
     server_name www.amfeslan.local egroupware.amfeslan.local;
     root /var/www;
     index index.html;

     client_max_body_size 8M;

     location / {
        try_files $uri $uri/ /index.php?$args;
     }

     location ^~ /egroupware {
        root /opt/egroupware;
        index index.php;
        rewrite ^/egroupware/(.*)$ /$1 last;
        try_files $uri $uri/ /index.php$args;
     }

     include global/php.conf;
}

Doesn't work right.  Can some kind soul help me not only fix this - but 
get a better understanding of how to build these generic constructs?  
Based on experience with other servers, I can't believe folder-based 
controls in nginx can be as difficult as I seem to be making it!
-- 
Daniel


From edho at myconan.net  Sun Apr 29 20:22:56 2012
From: edho at myconan.net (Edho Arief)
Date: Mon, 30 Apr 2012 03:22:56 +0700
Subject: Location - or how to setup sites in subfolders
In-Reply-To: <assp.046696f332.4F9D9F2B.5050605@amfes.com>
References: <assp.046696f332.4F9D9F2B.5050605@amfes.com>
Message-ID: <CAAtReCk+d+qcGs2=OuHi+ymmjygdtscbwMMPdSsBJ31sdfTJ1Q@mail.gmail.com>

2012/4/30 Daniel L. Miller <dmiller at amfes.com>:
> Note that I tried to add the "if" construct in php.conf based on a previous
> post - and I gotta believe there's a better way to do this. ?With or without
> the if construct, the above site works - but only for egroupware. ?Revising
> the site file to:
> server {
> ? ?server_name www.amfeslan.local egroupware.amfeslan.local;
> ? ?root /var/www;
> ? ?index index.html;
>
> ? ?client_max_body_size 8M;
>
> ? ?location / {
> ? ? ? try_files $uri $uri/ /index.php?$args;
> ? ?}
>
> ? ?location ^~ /egroupware {
> ? ? ? root /opt/egroupware;
> ? ? ? index index.php;
> ? ? ? rewrite ^/egroupware/(.*)$ /$1 last;
> ? ? ? try_files $uri $uri/ /index.php$args;
> ? ?}
>
> ? ?include global/php.conf;
> }
>
> Doesn't work right. ?Can some kind soul help me not only fix this - but get
> a better understanding of how to build these generic constructs? ?Based on
> experience with other servers, I can't believe folder-based controls in
> nginx can be as difficult as I seem to be making it!

Put "location ~ \.php$ { }" inside each "location ^~ /appname/ { }"
block instead of globally so it looks like this:

location ^~ /egroupware/ {
  # if index.php's path is /opt/egroupware/index.php,
  # use "root /opt" instead
  root /opt/egroupware;
  index index.php;
  # and remove this
  rewrite ^/egroupware/(.*)$ /$1 last;
  # and update this accordingly
  try_files $uri $uri/ /index.php$args;
  location ~ \.php($|/) {
    ...
  }
}

Also remember that some applications may need to know that it's
running in a subdirectory or it will generate incorrect link (eg.
/something.php instead of /app/something.php).


From dmiller at amfes.com  Sun Apr 29 21:17:29 2012
From: dmiller at amfes.com (Daniel L. Miller)
Date: Sun, 29 Apr 2012 14:17:29 -0700
Subject: Location - or how to setup sites in subfolders
In-Reply-To: <CAAtReCk+d+qcGs2=OuHi+ymmjygdtscbwMMPdSsBJ31sdfTJ1Q@mail.gmail.com>
References: <4F9D9F2B.5050605@amfes.com>
	<CAAtReCk+d+qcGs2=OuHi+ymmjygdtscbwMMPdSsBJ31sdfTJ1Q@mail.gmail.com>
Message-ID: <assp.14667e3450.4F9DAFE9.2090303@amfes.com>

On 4/29/2012 1:22 PM, Edho Arief wrote:
> 2012/4/30 Daniel L. Miller<dmiller at amfes.com>:
>> Note that I tried to add the "if" construct in php.conf based on a previous
>> post - and I gotta believe there's a better way to do this.  With or without
>> the if construct, the above site works - but only for egroupware.
[..]
>> Doesn't work right.  Can some kind soul help me not only fix this - but get
>> a better understanding of how to build these generic constructs?  Based on
>> experience with other servers, I can't believe folder-based controls in
>> nginx can be as difficult as I seem to be making it!
> Put "location ~ \.php$ { }" inside each "location ^~ /appname/ { }"
> block instead of globally so it looks like this:
>
> location ^~ /egroupware/ {
>    # if index.php's path is /opt/egroupware/index.php,
>    # use "root /opt" instead
>    root /opt/egroupware;
>    index index.php;
>    # and remove this
>    rewrite ^/egroupware/(.*)$ /$1 last;
>    # and update this accordingly
>    try_files $uri $uri/ /index.php$args;
>    location ~ \.php($|/) {
>      ...
>    }
> }
>

Thank you! - that helped quite a bit.  That way of thinking - of 
including the php handler inside each location, instead of a global php 
handler, just didn't seem as "elegant" to me - but I can see the power 
potential.

However - I still don't understand WHY it works - specifically, why I 
need to have "root /opt" instead of "root /opt/egroupware".  I'm 
inferring that the location path is automatically added to the declared 
root in processing.  But if that's the case, then to be able to declare, 
"location ^~ /lam/", and be able to reach a physical path of 
"/opt/ldap-account-manager-3.7", I need to add a symlink from the 
extended path to "/opt/lam".  And while I may in fact do so for other 
reasons - I WOULD like to be able to control exactly what nginx is 
doing, and better yet - understand the how/why.
-- 
Daniel


From nginx-forum at nginx.us  Mon Apr 30 08:20:31 2012
From: nginx-forum at nginx.us (springbok)
Date: Mon, 30 Apr 2012 04:20:31 -0400 (EDT)
Subject: nginx 1.2.0 - lots of 'client prematurely closed connection while
	reading client request line' errors
Message-ID: <fa86011485a1346ab7619b9c6b47bfe0.NginxMailingListEnglish@forum.nginx.org>

Hi,

I've upgraded to nginx 1.2.0, I'm using passenger 3.0.12 on a Suse SLE
11.1 server.

Since upgrading from nginx 1.0.10 I'm getting lots of errors like
these:

2012/04/30 18:13:37 [info] 10435#0: *379 client prematurely closed
connection while reading client request line, client: xxx.xxx.xxx.xxx,
server: localhost

There are 6-7 of these generated for almost every request.

Any ideas on what is causing these, is there something I need to change
or adjust?

Thanks,

Etienne.

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,225913,225913#msg-225913


From mdounin at mdounin.ru  Mon Apr 30 11:39:03 2012
From: mdounin at mdounin.ru (Maxim Dounin)
Date: Mon, 30 Apr 2012 15:39:03 +0400
Subject: nginx: [crit] pread() "/opt/nginx" failed (21: Is a directory)
In-Reply-To: <C15BF77669E54C498451E3A25D44F0A702AD2268D578@NYCMNET7EV001.mnet.biz>
References: <C15BF77669E54C498451E3A25D44F0A702AD2268D578@NYCMNET7EV001.mnet.biz>
Message-ID: <20120430113903.GE31671@mdounin.ru>

Hello!

On Fri, Apr 27, 2012 at 09:59:15PM -0400, Gray, Steve, Wipro wrote:

> 
> Hi,
> I am using nginx 1.0.15 running on Solaris 10.  When I issue my 
> "/opt/nginx/sbin/nginx -s stop" command, I get this error 
> "nginx: [crit] pread() "/opt/nginx" failed (21: Is a 
> directory)". The funny thing is that I can issue 
> "/opt/nginx/sbin/nginx" to start nginx and get no errors.  Nginx 
> starts okay.  I can also issue "/opt/nginx/sbin/nginx -V or v" 
> and get no errors.  The "nginx: [crit] pread() "/opt/nginx" 
> failed (21: Is a directory)" error happens when I issue the 
> command with the "-s stop|reload|quit|reopen" options.
> My configure argument is --prefix=/opt/nginx. Has anyone seen 
> this or know why this is happening?  Let me know if you need any 
> more information from me.

The message suggests you have pid (either in config via 
http://nginx.org/r/pid or with --pid-path configure argument) set 
to "/opt/nginx".  The fact that nginx starts is actually strange 
though.  Could you please provide "nginx -V" output and full 
config used?

Maxim Dounin


From antoine.bonavita at gmail.com  Mon Apr 30 16:30:12 2012
From: antoine.bonavita at gmail.com (Antoine Bonavita)
Date: Mon, 30 Apr 2012 18:30:12 +0200
Subject: nginx-Range header-proxy cache- potential patch
In-Reply-To: <CA+FsSo+ZMJaXSVSfQvewnADeg2fY-Le1jP0xo-phdoomMNLYOg@mail.gmail.com>
References: <CA+FsSoK3j0Uz8Hc2oai7ypLwzFK6JfC11Gp_D_kQATpX=cacXg@mail.gmail.com>
	<20120426141513.GY31671@mdounin.ru>
	<CA+FsSo+ZMJaXSVSfQvewnADeg2fY-Le1jP0xo-phdoomMNLYOg@mail.gmail.com>
Message-ID: <CA+FsSo+-1_zeXkfqEDV_=SR8yiN+0Dfy4rERCiohHCZ44SznJA@mail.gmail.com>

Hello,

Just a quick update as promised: I applied Maxim patch on 1.2.0 and
deployed on my production servers today.
Everything is smooth and the original problem fixed.

Thanks again,

Antoine.

On Thu, Apr 26, 2012 at 4:49 PM, Antoine Bonavita
<antoine.bonavita at gmail.com> wrote:
> Hello!
>
> On Thu, Apr 26, 2012 at 4:15 PM, Maxim Dounin <mdounin at mdounin.ru> wrote:
>> Hello!
>>
>> On Thu, Apr 26, 2012 at 03:23:53PM +0200, Antoine Bonavita wrote:
>>
>>> Hello,
>>>
>>> I'm currently using nginx to serve "small" video files (<2MB) to
>>> mobile devices. The way I set it up is as follow:
>>> 2 nginx servers reverse proxying with cache
>>> 1 upstream server where the files are actually uploaded
>>>
>>> It's all working fine except that the iPhones seem to have the habit
>>> of always doing range requests with range 0-1 first when they are
>>> about to play a video. And they really don't like it when they get the
>>> full answer.
>>>
>>> Now, the problem I'm facing is the one described here:
>>> http://forum.nginx.org/read.php?2,215141,215141#msg-215141
>>>
>>> Which can be summed up as:
>>> - If file in cache: range header honored (206, appropriate
>>> Cotnent-Length and Content-Range).
>>> - If file not in cache: server ignores the range header and sends the
>>> full file (200 and Content-Length set to full file size).
>>>
>>> Now, to the question "Is this intended behavior", Maxim Dounim said
>>> "yes" (http://forum.nginx.org/read.php?2,215141,215142#msg-215142)
>>>
>>> I looked at the code and there is a flag on request : r->allow_ranges.
>>>
>>> From what I can tell, it is set to false in most cases, except in the
>>> ngx_http_static_module. I would be happy to work on a patch that
>>> allows ranges in all cases when the length ?of the response from
>>> upstream is known (ie not when upstream response uses chunked
>>> encoding).
>>>
>>> But before I start working on this I would really appreciate if one of
>>> the gurus here could tell me whether I'm opening a Pandora box or not.
>>> And what I should be careful with.
>>
>> I've recently committed the patch which makes it (almost) safe to
>> just set allow_ranges on proxied replies, see
>>
>> http://trac.nginx.org/nginx/changeset/4468/nginx
>>
>> Remaining issues are with multiple ranges requests (which aren't
>> really used by anyone in real life except Adobe Acrobat). ?As long
>> as there are no such requests in your environment (or you are
>> using max_ranges 1; anyway), you may try the following patch:
>>
>> # HG changeset patch
>> # Parent 8a7ee0318b175b8d76275c5d34707eb52b276188
>> Upstream: allow ranges for cacheable replies.
>>
>> diff --git a/src/http/ngx_http_upstream.c b/src/http/ngx_http_upstream.c
>> --- a/src/http/ngx_http_upstream.c
>> +++ b/src/http/ngx_http_upstream.c
>> @@ -3680,10 +3680,9 @@ ngx_http_upstream_copy_allow_ranges(ngx_
>>
>> ?#if (NGX_HTTP_CACHE)
>>
>> - ? ?if (r->cached) {
>> + ? ?if (r->cached || r->upstream->cacheable) {
>> ? ? ? ? r->allow_ranges = 1;
>> ? ? ? ? return NGX_OK;
>> -
>> ? ? }
>>
>> ?#endif
>>
>>
>> (it's a bit old, but should apply cleanly)
> As usual you rock: it applies cleanly on 1.2.0 and solves my problem.
> I've been testing on my machine for now. I'll release it to production
> beginning of next week and let you know.
>
> Thanks a lot,
>
> A.
>
>>
>> Maxim Dounin
>>
>> _______________________________________________
>> nginx mailing list
>> nginx at nginx.org
>> http://mailman.nginx.org/mailman/listinfo/nginx


From dewanggaba at gmail.com  Mon Apr 30 21:24:15 2012
From: dewanggaba at gmail.com (antituhan)
Date: Mon, 30 Apr 2012 14:24:15 -0700 (PDT)
Subject: Is it possible applied external module from packages?
Message-ID: <1335821055357-7515095.post@n2.nabble.com>

Hello,

I've found WAF Module for NginX called Naxsi http://code.google.com/p/naxsi/
And I've read the manual to compile it, but is it possible to
applied/recompile naxsi module to NginX from packages?

Reference: how to install naxsi from tarball
http://code.google.com/p/naxsi/wiki/Howto#Installing_nginx_+_naxsi

-----
[daemon at antituhan.com ~]# 
--
View this message in context: http://nginx.2469901.n2.nabble.com/Is-it-possible-applied-external-module-from-packages-tp7515095.html
Sent from the nginx mailing list archive at Nabble.com.


From dewanggaba at gmail.com  Mon Apr 30 21:27:08 2012
From: dewanggaba at gmail.com (antituhan)
Date: Mon, 30 Apr 2012 14:27:08 -0700 (PDT)
Subject: [ASK] Good Directive vs Bad Directives
In-Reply-To: <1335281418117-7496293.post@n2.nabble.com>
References: <1335281418117-7496293.post@n2.nabble.com>
Message-ID: <1335821228308-7515099.post@n2.nabble.com>

*bump*

-----
[daemon at antituhan.com ~]# 
--
View this message in context: http://nginx.2469901.n2.nabble.com/ASK-Good-Directive-vs-Bad-Directives-tp7496293p7515099.html
Sent from the nginx mailing list archive at Nabble.com.