Nginx Forum - Php-fpm Mailing List - English

ERROR: failed to ptrace(PT_IO) pid 81718: Bad address (14) (no replies)

RedFoxy — Sun, 19 May 2013 09:18:00 -0400

Hi all!
I don't know from where that error comes, but I've a lot of "ERROR: failed to ptrace(PT_IO) pid 81718: Bad address (14)" in my php-fpm-error.log

Any ideas?

--

---
You received this message because you are subscribed to the Google Groups "highload-php-en" group.
To unsubscribe from this group and stop receiving emails from it, send an email to highload-php-en+unsubscribe@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

PHP-FPM exited on signal 10 (SIGBUS)? (2 replies)

RedFoxy — Sun, 19 May 2013 09:16:00 -0400

Hello!
I'm tring to install again openx after the free service of openx.com clese… But I've some strange trouble, after the installation on a regular server with FreeBSD 9.1 + Nginx 1.4.1 + PHP 5.3.24, an online server that I use for some website like www.softairmania.it, www.redfoxy.it and other, i've a strange trouble, when I try to add a new zone I get a 502 bad gateway error…
Reading the Nginx error log I see:
2013/05/18 19:30:25 [error] 80541#0: *8500 upstream prematurely closed connection while reading response header from upstream, client: 87.10.110.222, server: samspot.softairmania.it, request: "GET /www/admin/zone-edit.php?affiliateid=1 HTTP/1.1", upstream: "fastcgi://unix:/home/serverweb/tmp/php-fpm.sock:", host: "samspot.softairmania.it", referrer: "http://samspot.softairmania.it/www/admin/affiliate-zones.php"

And in the php-fpm error log (in debug mode) I read:
[18-May-2013 19:30:25.116401] DEBUG: pid 80481, fpm_got_signal(), line 72: received SIGCHLD
[18-May-2013 19:30:25.116452] WARNING: pid 80481, fpm_children_bury(), line 252: [pool www] child 80483 exited on signal 10 (SIGBUS) after 918.184752 seconds from start
[18-May-2013 19:30:25.117784] NOTICE: pid 80481, fpm_children_make(), line 421: [pool www] child 80857 started
[18-May-2013 19:30:25.117824] DEBUG: pid 80481, fpm_event_loop(), line 411: event module triggered 1 events

my php-fpm.conf is: http://pastebin.com/ADnfNXxb
and my php.ini: http://pastebin.com/LvedL8hk
and my php extension: http://pastebin.com/pJdECN0B

thank's for all

--

---
You received this message because you are subscribed to the Google Groups "highload-php-en" group.
To unsubscribe from this group and stop receiving emails from it, send an email to highload-php-en+unsubscribe@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

FastCGI - implementation (no replies)

Anonymous User — Mon, 13 May 2013 07:20:02 -0400

PHP-FPM is an externally spawned FastCGI-aware process.

Over one fastCGI transport connection PHP-FPM can handle multiple
concurrent requests.
Web server does not have to spawn new processes with PHP FCGI, PHP-FPM will
spawn and control if necessary.

Going into details ....

Using FastCGI on *nix platform ones can handle concurrent requests that way:

socket() / bind() / listen()

// BEGIN thread 1
accept()
// END thread 1

// BEGIN thread 2
accept()
// END thread 2

I'm wondering is it possible, on windows platform, to handle multiple
concurrent fastcgi requests with only one process and multiple threads in
it (or with one management process and multiple spawned children) ?

Can FastCGI module for IIS handle fastCGI application that way ?
Let's focus on pipes and leave tcp kind of sockets for a moment...

Does anyone can point me in to the right direction.

Is it possible on Windows (without Cygwin, etc...) at all ?

I don't want to investigate if it's possible with PHP-FPM but if it's
possible at all.

Any help appreciated.

/Adam

--

---
You received this message because you are subscribed to the Google Groups "highload-php-en" group.
To unsubscribe from this group and stop receiving emails from it, send an email to highload-php-en+unsubscribe@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

PHP-FPM documentation (no replies)

Renzo Racioppi — Thu, 09 May 2013 13:48:02 -0400

Hello all,
dou
Does anybody have a documentation which describes PHP-FPM software
architecture?

I have some doubts regarding pools and master process interactions.

Regards.

--

---
You received this message because you are subscribed to the Google Groups "highload-php-en" group.
To unsubscribe from this group and stop receiving emails from it, send an email to highload-php-en+unsubscribe@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

PHP-FPM documentation (3 replies)

Renzo Racioppi — Tue, 14 May 2013 06:08:01 -0400

Hello all, does anybody have a documentation which describes PHP-FPM
software architecture?

I have some doubts concerning pools and master process interactions...

--

---
You received this message because you are subscribed to the Google Groups "highload-php-en" group.
To unsubscribe from this group and stop receiving emails from it, send an email to highload-php-en+unsubscribe@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

apache+php-fpm and security in shared enviroment compared to apache+suexec+mod_fastcgi (4 replies)

Anonymous User — Tue, 07 May 2013 10:11:50 -0400

I've been using apache with suexec and mod_fcgid. It starts php handlers as
selected system user and listen to request via PIPES (fastcgi protocol). It
works well but can't share opcode between many handlers.
I've been investigating php-fpm aproach. What I can see is that in only
runs as tcp fastcgi server (like 127.0.0.1:9999) (under selected system
user) and apache has to conenct to it using FastCGIExternalServer or
fastcgi proxy mod. What the hell will stop other local users from
connecting to 127.0.0.1:9999 and passing their own code to handler that is
running as other system user? That seems totaly not secure in shared
enviroment. Am I missing something or what?

--

---
You received this message because you are subscribed to the Google Groups "highload-php-en" group.
To unsubscribe from this group and stop receiving emails from it, send an email to highload-php-en+unsubscribe@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

php-fpm workers eating lots of memory even on zero traffic (1 reply)

Michal Novotný — Mon, 29 Apr 2013 23:22:56 -0400

Hello, my php-fpm workers are taking large amount of memory even if there
has been zero traffic on server for some time. What is this memory? Is it
leaked memory or is it some sort of php cache (I use just APC cache which
should be in a shared memory somewhere though)? Or is it normal behaviour
of php-fpm that it does not realease memory?

A defailed report with my config is here:
http://serverfault.com/questions/503185/php-fpm-workers-eating-lots-of-memory-even-on-zero-traffic

Any help greatly appreciated!
clime

--

---
You received this message because you are subscribed to the Google Groups "highload-php-en" group.
To unsubscribe from this group and stop receiving emails from it, send an email to highload-php-en+unsubscribe@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

PHP5-FPM not producing an errors on startup (no replies)

txsl — Mon, 15 Apr 2013 21:56:56 -0400

So I'm just setting up nginx and php5-fpm on my VPS, having moved over from Apache. I'm running Debian Squeeze, with the latest PHP from the dotdeb.org repo (that means PHP 5.4.13-1~dotdeb.1).

I am having trouble getting php5-fpm going. When run /etc/init.d/php5-fpm start, I don't get any confirmation, nor any error signals. Same for running /etc/init.d/php5-fpm stop. If I pass it the status arg, it returns php5-fpm is not running ... failed!

Now I can't see any errors in any log, so I'm at a losss as to how to even start to debug the error.

Has anyone experience similar problems, or knows what I should do next?

Many thanks!

PHP Sessions not "saving" / "working" (no replies)

binbash — Thu, 11 Apr 2013 20:58:27 -0400

Hi all,

I am running nginx and php-fpm on Centos 6.3.

I am noticing a strange behaviour where php sessions are not working. To troubleshoot it, I did the basics:

1. Made sure php-fpm was using the /tmp directory
2. chmod 777 /tmp
3. Set php-fpm to use the same user as nginx.

None of this helped, so then I created a simple PHP file to help diagnose: sessiontest.php

session_start();
$_SESSION["count"]++;
print_r($_SESSION);
echo session_id();
?>

When I run this file, in theory, the value should increase by 1 every single time. However, it does not.

Does anybody have any suggestions?

PHP5-FPM + Litespeed? (no replies)

Luck Piecuch — Mon, 01 Apr 2013 03:34:00 -0400

Hello there,
anyone have by a chance tried to setup Litespeed with its "External App"
"Fast CGI App" along with remote PHP-FPM ?
If so - would it pass -user -group to php-fpm like the old mod_fastcgi did?

--

---
You received this message because you are subscribed to the Google Groups "highload-php-en" group.
To unsubscribe from this group and stop receiving emails from it, send an email to highload-php-en+unsubscribe@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

Please tell me how can I configure php-fpm on dedicate server. (no replies)

Anonymous User — Wed, 20 Mar 2013 08:00:00 -0400

Please tell me how can I configure php-fpm on dedicate server.
I am looking for php-fpm and nginx settings.

I have 3 raspberry pi small servers with 512MB and 8GB usb sticks.
I wish to install nginx on one.
I wish to install php, php-fpm and php modules stuff on second.
Last one with mysql server only.

Best regards
Antonio Apostoliu.

--

---
You received this message because you are subscribed to the Google Groups "highload-php-en" group.
To unsubscribe from this group and stop receiving emails from it, send an email to highload-php-en+unsubscribe@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

Environment variables, php-fpm & Nginx (no replies)

Pooky — Wed, 20 Mar 2013 06:54:00 -0400

I have an ubuntu server with a handful of custom environment set in
/etc/environment as per the ubuntu community recommendation

When I use php from the command line I can use php's getenv() function to
access this variables.

Also, if I run phpinfo() from the command line I see all of my variables in
the ENVIRONMENT section.

HOWEVER:

When trying to access the same data inside processes being run by php5-fpm
this data is not available. All I can see in the ENVIRONMENT section of
phpinfo() is:

USER www-data
HOME /var/www

I know the command line uses this ini:

/etc/php5/cli/php.ini

And fpm uses:

/etc/php5/fpm/php.ini

I've not managed to find any differences between the two that would explain
why the ENV variables are not coming through in both.

Also if I run:

sudo su www-data

and then echo the environment variables I am expecting they are indeed
available to the www-data user.

What do I need to do to get my environment variables into the php processes
run by fpm?

--

---
You received this message because you are subscribed to the Google Groups "highload-php-en" group.
To unsubscribe from this group and stop receiving emails from it, send an email to highload-php-en+unsubscribe@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

Re: Digest for highload-php-en@googlegroups.com - 1 Message in 1 Topic (no replies)

Brad Chesney — Wed, 13 Mar 2013 06:10:01 -0400

On many of the systems I work on (mostly Debian based) you will find error
messages in /var/log/php5-fpm.log , /var/log/apache/error.log , or custom
locations which I usually find with a recursive grep (a feature sadly
missing on my OpenIndiana machine, look for a find - exec work around in
that case). There are also the syslog and messages you can check in the
/var/log/ directory. Without further details on your system, it is hard to
say.

-BradChesney79

On Mar 13, 2013 5:25 AM, wrote:
>
> Today's Topic Summary
>
> Group: http://groups.google.com/group/highload-php-en/topics
>
> Core dumps location? [1 Update]
> Core dumps location?
> Robert Mar 11 09:56PM -0700
>
> Hi all, just a quick noob question: I'm getting a bunch of SIGBUSs
> happening from a php app I'm working on, I'm not sure what's causing this
> so I want to look at core dumps. Am I right in thinking that "log_level =
> debug" in will generate core dumps that I can look at? If so where would
I
> find these?
>
>
>
> You received this message because you are subscribed to the Google Group
highload-php-en.
> You can post via email.
> To unsubscribe from this group, send an empty message.
> For more options, visit this group.
>
> --
>
> ---
> You received this message because you are subscribed to the Google Groups
"highload-php-en" group.
> To unsubscribe from this group and stop receiving emails from it, send an
email to highload-php-en+unsubscribe@googlegroups.com.
> For more options, visit https://groups.google.com/groups/opt_out.
>
>

--

---
You received this message because you are subscribed to the Google Groups "highload-php-en" group.
To unsubscribe from this group and stop receiving emails from it, send an email to highload-php-en+unsubscribe@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

Core dumps location? (3 replies)

Robert — Wed, 13 Mar 2013 07:56:01 -0400

Hi all, just a quick noob question: I'm getting a bunch of SIGBUSs
happening from a php app I'm working on, I'm not sure what's causing this
so I want to look at core dumps. Am I right in thinking that "log_level =
debug" in will generate core dumps that I can look at? If so where would I
find these?

--

---
You received this message because you are subscribed to the Google Groups "highload-php-en" group.
To unsubscribe from this group and stop receiving emails from it, send an email to highload-php-en+unsubscribe@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

Many POST requests, fastcgi cache delay? (no replies)

maur — Sat, 09 Mar 2013 03:44:10 -0500

Hi,

I have a site with many POST requests.Each visitor is generating one POST requests per second to get status of few things.
With status Im returning current datetime.

Im reading about different options for caching and made few changes:
fastcgi_cache microcache;
fastcgi_cache_methods POST;
fastcgi_cache_bypass $no_cache;
fastcgi_no_cache $no_cache;
#fastcgi_cache_key $scheme$host$request_method$request_uri;
fastcgi_cache_key $server_name|$request_uri;
fastcgi_cache_valid 200 302 1s;
fastcgi_cache_valid 301 1s;
fastcgi_cache_valid any 1s;
fastcgi_max_temp_file_size 1M;
#fastcgi_cache_use_stale updating;
#fastcgi_cache_use_stale updating error timeout invalid_header http_500;
fastcgi_cache_use_stale error;
conf I pasted is in location /status.php

However in firebug I see each second new request - but timer has 1 second delay.
I see for example:
09.03.2013 03:34:00
09.03.2013 03:34:00
09.03.2013 03:34:01
09.03.2013 03:34:01
09.03.2013 03:34:02
09.03.2013 03:34:02
and so on. While without cache I saw each second in response.

Do you have any hints? :)

why php-fpm use some udp port? (3 replies)

peixinchen — Thu, 07 Mar 2013 05:20:01 -0500

like this:

$ netstat -nlp

udp 0 0 10.168.16.205:54144 0.0.0.0:*
20456/php-fpm: pool
udp 5664 0 0.0.0.0:54086 0.0.0.0:*
9687/php-fpm: pool
udp 0 0 0.0.0.0:54088 0.0.0.0:*
9687/php-fpm: pool
udp 5664 0 0.0.0.0:54103 0.0.0.0:*
19114/php-fpm: pool
udp 0 0 10.168.16.205:54104 0.0.0.0:*
19114/php-fpm: pool
udp 0 0 0.0.0.0:54105 0.0.0.0:*
19114/php-fpm: pool
udp 0 0 10.168.16.205:54109 0.0.0.0:*
17451/php-fpm: pool
udp 0 0 10.168.16.205:54110 0.0.0.0:*
18259/php-fpm: pool
udp 0 0 0.0.0.0:54111 0.0.0.0:*
18259/php-fpm: pool
udp 0 0 10.168.16.205:54135 0.0.0.0:*
23890/php-fpm: pool
udp 0 0 10.168.16.205:54136 0.0.0.0:*
22221/php-fpm: pool
udp 5664 0 0.0.0.0:54141 0.0.0.0:*
26125/php-fpm: pool
udp 0 0 10.168.16.205:54142 0.0.0.0:*
26125/php-fpm: pool
udp 0 0 0.0.0.0:54143 0.0.0.0:*
26125/php-fpm: pool

does these for master/worker communication？

--

---
You received this message because you are subscribed to the Google Groups "highload-php-en" group.
To unsubscribe from this group and stop receiving emails from it, send an email to highload-php-en+unsubscribe@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

php-fpm is mixing configs of different vhosts. (4 replies)

bvidinli — Tue, 07 May 2013 14:49:27 -0400

The newest problem is: php-fpm is mixing configs of different vhosts.
let me explain,
I have a vhost section like:

server {
listen 80;
server_name sub1.something.com;
.....
fastcgi_param PHP_ADMIN_VALUE "open_basedir=somedir11";
}

and another section similar but different server_name:

server {ne
listen 80;
server_name sub2.something.com;
.....
fastcgi_param PHP_ADMIN_VALUE "open_basedir=somedir22";
}

the problem is this:
even the user goes to sub1.something.com, the open_basedir of second vhost
section (somedir22) is applied. It seems that, php or php-fpm or nginx or
whatever it is, remembering some previous state of PHP_ADMIN_VALUE.
what may be the problem ?
I also suspect this before, because I see logs in syslog, showing wrong
filenames.
for ex: php syslog(LOG_WARNING,"log from file1") generates a log in syslog,
syslog(LOG_WARNING,"log from file2")
I saw in past, logs are mixed somehow. wrong message is displayed with
wrong prefix (of syslog line).

Is there a setting that causes php-fpm (or anything) remembers some other
state ?
Any suggestion ?

--

---
You received this message because you are subscribed to the Google Groups "highload-php-en" group.
To unsubscribe from this group and stop receiving emails from it, send an email to highload-php-en+unsubscribe@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

Heavy load because of concurrent users on Nginx and php-fpm causes 502 bad gateway error. (no replies)

ehussain — Wed, 20 Feb 2013 07:27:18 -0500

Hi,

I have Nginx + PHP FPM installed on my server. We are putting load on server for long period for 30 concurrent users.

For initial user it works fine but after some time it starts throwing 502 bad gateway error.

I have placed some of the log of nginx php-fpm and slow log of php-fpm.

There are entries getting logged in slow log of php-fpm because of long running script and load on server. I think this is the reason for 502 bad gateway error. But I dont know how to solve that problem.

- What are the tweaks I need to make in php-fpm.conf so that this errors gets resolved?
- How to make nginx wait for long time for response from php-fpm?
- How to increase php-fpm max execution time?

Here are the logs attached.

========================= NGINX LOG=====================================

2013/01/29 15:03:38 [error] 2493#0: *1046562 recv() failed (104: Connection reset by peer) while reading response header from upstream, client: 49.248.0.2, server: ***********.com, request: "GET MY_SCRIPT_URI HTTP/1.1", upstream: "fastcgi://127.0.0.1:9000", host: "***********.com", referrer: "MY_SCRIPT_URL"

2013/01/29 15:03:39 [error] 2493#0: *1046561 recv() failed (104: Connection reset by peer) while reading response header from upstream, client: 49.248.0.2, server: **********.com, request: "GET MY_SCRIPT_URI HTTP/1.1", upstream: "fastcgi://127.0.0.1:9000", host: "***********.com", referrer: "MY_SCRIPT_URL"

There are many errors of this type and they are repeating in entire file.

============================== PHP FPM LOG =================================

[14-Feb-2013 12:54:13] ERROR: failed to ptrace(PEEKDATA) pid 10748: Input/output error (5)
[14-Feb-2013 12:54:18] ERROR: failed to ptrace(PEEKDATA) pid 10112: Input/output error (5)
[14-Feb-2013 12:54:18] ERROR: failed to ptrace(PEEKDATA) pid 12147: Input/output error (5)
[14-Feb-2013 12:54:19] ERROR: failed to ptrace(PEEKDATA) pid 30857: Input/output error (5)
[14-Feb-2013 12:54:19] ERROR: failed to ptrace(PEEKDATA) pid 9568: Input/output error (5)
[14-Feb-2013 12:54:19] ERROR: failed to ptrace(PEEKDATA) pid 32519: Input/output error (5)
[14-Feb-2013 12:54:23] ERROR: failed to ptrace(PEEKDATA) pid 31361: Input/output error (5)
[14-Feb-2013 12:54:23] ERROR: failed to ptrace(PEEKDATA) pid 9014: Input/output error (5)
[14-Feb-2013 12:54:23] ERROR: failed to ptrace(PEEKDATA) pid 10789: Input/output error (5)
[14-Feb-2013 12:54:23] ERROR: failed to ptrace(PEEKDATA) pid 840: Input/output error (5)
[14-Feb-2013 12:54:23] ERROR: failed to ptrace(PEEKDATA) pid 953: Input/output error (5)
[14-Feb-2013 12:54:26] ERROR: failed to ptrace(PEEKDATA) pid 10927: Input/output error (5)
[14-Feb-2013 12:54:26] ERROR: failed to ptrace(PEEKDATA) pid 9012: Input/output error (5)
[14-Feb-2013 12:54:26] ERROR: failed to ptrace(PEEKDATA) pid 10861: Input/output error (5)
[14-Feb-2013 12:54:39] ERROR: failed to ptrace(PEEKDATA) pid 31361: Input/output error (5)
[14-Feb-2013 12:54:58] ERROR: failed to ptrace(PEEKDATA) pid 10102: Input/output error (5)
[14-Feb-2013 12:54:58] ERROR: failed to ptrace(PEEKDATA) pid 10918: Input/output error (5)
[14-Feb-2013 12:54:58] ERROR: failed to ptrace(PEEKDATA) pid 32519: Input/output error (5)
[14-Feb-2013 12:54:59] ERROR: failed to ptrace(PEEKDATA) pid 9326: Input/output error (5)
[14-Feb-2013 12:54:59] ERROR: failed to ptrace(PEEKDATA) pid 10789: Input/output error (5)
[14-Feb-2013 12:55:08] ERROR: failed to ptrace(PEEKDATA) pid 10943: Input/output error (5)
[14-Feb-2013 12:55:08] ERROR: failed to ptrace(PEEKDATA) pid 31850: Input/output error (5)
[14-Feb-2013 12:55:08] ERROR: failed to ptrace(PEEKDATA) pid 31898: Input/output error (5)
[14-Feb-2013 12:55:08] ERROR: failed to ptrace(PEEKDATA) pid 32737: Input/output error (5)
[14-Feb-2013 12:55:13] ERROR: failed to ptrace(PEEKDATA) pid 10748: Input/output error (5)
[14-Feb-2013 12:55:13] ERROR: failed to ptrace(PEEKDATA) pid 10117: Input/output error (5)

============================== PHP FPM SLOW LOG =================================

[14-Feb-2013 12:55:13] [pool www] pid 10748
script_filename = MY_SCRIPT_PATH
[0x00007f446e8e06b0] curl_exec() MY_SCRIPT_PATH_1.php:317
[0x00007f446e8e0490] callService() MY_SCRIPT_PATH_2:1331
[0x00007f446e8e0148] convertToPurchaseOrders() MY_SCRIPT_PATH_3:15
[0x00007fff0102b4d0] convertToPurchaseOrders() unknown:0
[0x00007f446e8de0d8] call_user_func_array() MY_SCRIPT_PATH_4:359
[0x00007f446e8dd4d0] +++ dump failed

[14-Feb-2013 12:55:13] [pool www] pid 10117
script_filename = MY_SCRIPT_PATH
[0x00007f446e8e06b0] curl_exec() MY_SCRIPT_PATH_1.php:317
[0x00007f446e8e0490] callService() MY_SCRIPT_PATH_2:1331
[0x00007f446e8e0148] convert() MY_SCRIPT_PATH_3:15
[0x00007fff0102b4d0] convert() unknown:0
[0x00007f446e8de0d8] call_user_func_array() MY_SCRIPT_PATH_4:359
[0x00007f446e8dd4d0] +++ dump failed

Help is appriciated.

Thanks in advance.

Strange .ini file parsing (3 replies)

Anoop Alias — Fri, 26 Apr 2013 05:50:00 -0400

Hi,

I have a strange issue of php.fpm parsing a file ending in .ini whereas
this .ini file doesnt have anything to do with php

*26 FastCGI sent in stderr: "PHP message: WARNING: You have errors in you
INI file (/home/xxxxx/public_html/xxx/etc/xxxxx/magiczoomplus.settings.ini)
on line 1!"

I renamed the file and the error was gone .

So the question is - does php-fpm look for files ending in .ini ; if so how
to disable that and does this default behavior slow down things ?

Thanks
--
Anoop P Alias (PGP Key ID : 0x014F9953)
GNU system administrator
http://UniversalAdm.in

--

---
You received this message because you are subscribed to the Google Groups "highload-php-en" group.
To unsubscribe from this group and stop receiving emails from it, send an email to highload-php-en+unsubscribe@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

FastCGI: comm with server "/usr/lib/cgi-bin/php5-fcgi" aborted: idle timeout (30 sec), FastCGI: incomplete headers (0 bytes) received from server "/usr/lib/cgi-bin/php5-fcgi" (2 replies)

otherjohn — Fri, 01 Feb 2013 17:34:05 -0500

Hi all,
I cannot track down what is causing this error! I get the following on a
regular basis, plus I have extremely high CPU usage too (around 80%)!

[Fri Feb 01 11:51:43 2013] [error] [client xx.xx.xx.xx] FastCGI: comm with
server "/usr/lib/cgi-bin/php5-fcgi" aborted: idle timeout (30 sec),
referer: http://mydomain.com/
[Fri Feb 01 11:51:43 2013] [error] [client xx.xx.xx.xx] FastCGI: incomplete
headers (0 bytes) received from server "/usr/lib/cgi-bin/php5-fcgi",
referer: http://mydomain.com/

I have apache2 installed with php-fpm on an amazon m1.large ec2 ubuntu 64
12.04 server.
7.5 GiB memory
4 EC2 Compute Units (2 virtual cores with 2 EC2 Compute Units each)
64-bit platform
I/O Performance: High

I also have APC installed and set to use up to 1gig of ram

Here is some of my config settings:
/etc/apache2/httpd.conf

AddHandler php5-fcgi .php
Action php5-fcgi /php5-fcgi
Alias /php5-fcgi /usr/lib/cgi-bin/php5-fcgi
FastCgiExternalServer /usr/lib/cgi-bin/php5-fcgi -socket
/tmp/php5-fpm.sock -pass-header Authorization

/etc/apache2/apache.conf

MaxKeepAliveRequests 100
KeepAliveTimeout 5
# prefork MPM

StartServers 5
MinSpareServers 5
MaxSpareServers 10
MaxClients 150
MaxRequestsPerChild 0

StartServers 2
MinSpareThreads 25
MaxSpareThreads 75
ThreadLimit 64
ThreadsPerChild 25
MaxClients 150
MaxRequestsPerChild 0

/etc/php5/fpm/php-fpm.conf
pid = /var/run/php5-fpm.pid

/etc/php5/fpm/pool.d/www.conf

listen = /tmp/php5-fpm.sock

pm = dynamic
pm.max_children =150
pm.start_servers = 75
pm.min_spare_servers = 20
pm.max_spare_servers = 150

Can anyone help?
John

--

---
You received this message because you are subscribed to the Google Groups "highload-php-en" group.
To unsubscribe from this group and stop receiving emails from it, send an email to highload-php-en+unsubscribe@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

Minimal Debian Wheezy Setup Using Stock Packages (no replies)

Brad Chesney — Sat, 26 Jan 2013 14:02:01 -0500

You will need all the official repos for the packages specified. But a
working super basic Debian Wheezy system can be running fpm relatively easy.

It took a while to get to this point because there is a lot of information
out there that does not work.
Most of this stuff should apply to the associated Ubuntu systems that will
be based on Debian Wheezy.

And without delay:

apt-get install libapache2-mod-fastcgi apache2-mpm-worker php5-fpm

a2enmod actions

vi /etc/apache2/mods-available/fastcgi.conf
FastCgiIpcDir /var/lib/apache2/fastcgi
AddHandler php5-fcgi .php
Action php5-fcgi /fcgi-bin

vi /etc/apache2/sites-available/default

ServerAdmin webmaster@localhost
DocumentRoot /var/www

Options Indexes FollowSymLinks MultiViews
AllowOverride All
Order allow,deny
allow from all

Order allow,deny
Allow from all

FastCgiExternalServer /tmp/imaginary-file -socket /var/www/default.sock
-user www-data -group www-data
Alias /fcgi-bin /tmp/imaginary-file
LogLevel warn
ErrorLog /var/log/apache2/error.log
CustomLog /var/log/apache2/access.log combined

vi /etc/php5/fpm/pool.d/www.conf
[www]
user = www-data
group = www-data
listen = /var/www/default.sock
pm = ondemand
pm.max_children = 5
pm.process_idle_timeout = 10s;
pm.max_requests = 500

chown www-data:www-data /var/log/apache2/*.log
chown -R www-data:www-data /var/www
service php5-fpm restart
service apache2 restart

https://gist.github.com/4642771

--

session_start() slows my servers?? (4 replies)

crirus — Mon, 28 Jan 2013 03:30:00 -0500

Hello

I keep getting occasional slow downs on the server, with no apparant
relation to user count.
It was slow at 800 users per second and now with 1300 is instant.

I started logging slow scripts in php-fpm and I catch, this :

[26-Jan-2013 19:02:42] [pool www] pid 78697
script_filename = /var/www/html/v.php
[0x00007fe0dea8eb28] session_start() /var/www/html/include/config.php:62
[0x00007fe0dea8a5a8] +++ dump failed

This is something in the beginning of any script, my slow down is 30!!!
seconds

What is going on in there?

Thanks for any tips

---------------------------------------------------------------
Cristian Rusu
Web Developement & Electronic Publishing

======
Crilance.com
Crilance.blogspot.com

--

php-fpm processes max out even though CPU is not that high (6 replies)

Jim Hackett — Thu, 24 Jan 2013 04:52:00 -0500

My company is looking at a combination of PHP-FPM/APC/nginx to respond to a
very large number of requests, very quickly, globally. In general we see a
huge performance increase over lighttpd, however once we get to around 200
req/sec things get wacky. All of the idle processes are taken up and we
start responding with 502 errors. We've increased shared memory, toyed
around with nginx/php-fpm config to no avail. Until we high around 200
req/sec everything looks really great, CPU levels are fine, memory is fine,
but once we hit that mark we get a lot of socket errors in the nginx error
log and many many 502 error. Any help would be GREATLY appreciated. Below I
have included some of our config, please let me know what else you need:

php-fpm.conf

[global]
pid = /var/run/php-fpm/php-fpm.pid
error_log = log/php-fpm_error.log
log_level = debug
;emergency_restart_threshold = 0
;emergency_restart_interval = 0
;process_control_timeout = 0

process.max = 500

include=/etc/php-fpm.d/*.conf

[www]

user = nobody
group = nobody
listen = /tmp/pool1.socket
listen.allowed_clients = 127.0.0.1
slowlog = /var/log/php-fpm/www-slow.log
request_slowlog_timeout = 1
catch_workers_output = yes
php_admin_value[error_log] = /var/log/php-fpm/www-error.log
php_admin_flag[log_errors] = on
rlimit_files = 50000
request_terminate_timeout = 30s

pm = ondemand
pm.max_children = 300
pm.process_idle_timeout = 2s
pm.max_requests = 5000

pm.status_path = /monitor/status.fpm
listen.backlog=0

nginx.conf

user nobody nobody;
worker_processes 10;
pid /var/run/nginx.pid;

events {
worker_connections 1024;
}

http {
include mime.types;
default_type application/octet-stream;
index index.html index.php;
log_format main '$remote_addr - $remote_user [$time_local]
"$request" '
'$status $body_bytes_sent "$http_referer" '
'"$http_user_agent" "$http_x_forwarded_for"
$request_time';
error_log /var/log/nginx/error.log;
access_log off;
sendfile on;

keepalive_timeout 0s;
fastcgi_read_timeout 360s;

upstream phpfarm {
server unix:/tmp/pool1.socket weight=100 max_fails=5 fail_timeout=5;
}

server {
listen 80;
listen 443 default_server ssl;
root /srv/www/;
fastcgi_busy_buffers_size 256k;
fastcgi_buffers 4 256k;
fastcgi_buffer_size 128k;
fastcgi_temp_file_write_size 256k;
proxy_buffering off;
tcp_nopush on;
tcp_nodelay on;
auth_digest_user_file user.passwd;
auth_digest_expires 300s;
location /monitor {
auth_digest 'Authorized users only';
location /monitor/status {
extended_status on;
access_log off;
allow all;
}
location /monitor/status.fpm {
fastcgi_pass phpfarm;
fastcgi_param SCRIPT_FILENAME $document_root$fastcgi_script_name;
fastcgi_param PATH_INFO $fastcgi_script_name;
include fastcgi_params;
}
location /monitor/status/php-fpm {
alias /usr/share/fpm/;
allow all;
}
location ~ \.php$ {
try_files $uri =404;
fastcgi_pass phpfarm;
fastcgi_param SCRIPT_FILENAME $document_root$fastcgi_script_name;
fastcgi_param PATH_INFO $fastcgi_script_name;
include fastcgi_params;
}
}

location ~ \.php$ {
try_files $uri =404;
fastcgi_pass phpfarm;
fastcgi_param SCRIPT_FILENAME $document_root$fastcgi_script_name;
fastcgi_param PATH_INFO $fastcgi_script_name;
include fastcgi_params;
}

location = /favicon.ico {
return 204;
access_log off;
log_not_found off;
}
}
}

--

Announcing cpXstack - The complete nginX+ PHP-FPM stack for cpanel (no replies)

Anoop Alias — Sat, 19 Jan 2013 04:16:01 -0500

cpXstack

cpXstack is a GPL licensed cpanel plugin developed by SysAlly (
http://SysAlly.net ) that implements the full LEMP ( Linux- nginX - MySQL
- PHP-FPM ) stack .

More information about the project can be found at
http://cpxstack.sysally.net/

Documentation at
http://manage.piserve.com/projects/cpxstack/wiki/Documentation

The software is FREE (as in FREE beer ) for use .

SysAlly provides :

1. Installation Support - 10 USD
2. Technical Support - 10 USD/ Hr
3. Value Added Service - 25 checkpoint Server Optimization and security
hardening with PDF report + FREE cpXstack installation - worth 45 USD at
30 USD/server .

About SysAlly
SysAlly is a division of PiServe Technologies Private Limited that provides
RIM (Remote IT Infrastructure Management Services). At SysAlly, we are
proud of our ability to timely deliver Superior and cost-effective services
and products, thereby enhancing the business value to our esteemed clients.
Our products and services offer true value for money and involve low cost
of ownership.

--
Anoop P Alias (PGP Key ID : 0x014F9953)
GNU system administrator
http://UniversalAdm.in

Restrict PHP execution (2 replies)

David Blank-Edelman — Thu, 17 Jan 2013 04:00:00 -0500

Hi-
I suspect this is a simple (read: stupid) question, but is there a
PHP-FPM equivalent to the mod_php directive for restricting PHP execution,
i.e.:

...
php_admin_flag engine off

...

I realize I can block access to files with a .php suffix so it never reaches FPM, but is there a closer equivalent? Thanks!

-- dNb

open_basedir messes up all sites (3 replies)

mkn — Sun, 05 May 2013 15:55:51 -0400

Hi,

My fresh installation worked well for two weeks, and suddenly all sites started crashing to "open_basedir_restriction in effect".

What I had done was simply add a directive to couple of the domains (just an example below):

server {
listen 81;
server_name .thiscustomer.com;
# set variable first
set $user_lib /home/thiscustomer/libs/;
...
location ~ \.php$ {
# now set open_basedir for THIS site
fastcgi_param PHP_ADMIN_VALUE "open_basedir =$user_lib:/tmp:/usr/local/lib/php:";
.. normal stuff
}
}

This had the effect that most sites got the wrong open_basedir after a hit to the above domain, as if the open_basedir value had gone global! It seemed to switch between the two sites I had made the change to, probably depending on which of them had got a visitor last.

After removing the fastgi_param everything moved to normal again. After adding it again everything weng haywire again so the problem was very easy to reproduce. So the PHP_ADMIN_VALUE seems to be pretty dangerous and should be avoided.

regards
Markku

For the community (no replies)

Anoop Alias — Sat, 05 Jan 2013 01:58:00 -0500

Hello,

Here is a sneak preview of cpXstack . The plugin for cpanel powered servers
to use the full potential of nginX + PHP-FPM

===
http://youtu.be/UAmXOJIC93o
===

--
Anoop P Alias (PGP Key ID : 0x014F9953)
GNU system administrator
http://UniversalAdm.in

Suggest best php-fpm configuration (1 reply)

ssatturi — Sat, 29 Dec 2012 01:25:59 -0500

Hi,

I am srikanth working with a ecommerce company. We have a hardware loadbalancer which sends traffic to 2 web servers in round robin fashion. Two web servers which are dedicated to serve web requests. Database servers are seperate.
Here is the configuration of web boxes I have.

Intel E5620 processor 2.40Ghz, 32GB (3*8GB+2*4GB) RAM , 300GB *4 HDD

Also when i do nproc .. output is 8.

We have ubuntu 12.04 OS, nginx[nginx/1.1.19] web server, php-fpm [PHP 5.3.10-1ubuntu3.4 (fpm-fcgi) ].

Please guide me with the best configuration to handle 10000 requests per second.

Srikanth

For nginx server php setcookie() is not working in chrome and IE (no replies)

shashikant — Wed, 19 Dec 2012 23:59:35 -0500

Hi,

I am Shashikant, deployed the application on nginx server but I am facing a problem while setting cookies.
Cookie is successfully set on Firefox browser but failed to set in IE and chrome.

Can you please help me, if any setting is required to set for nginx server server.

Thanks & Regards,
Shashikant

unix socket creation upon SIGUSR2 (1 reply)

Ro Achterberg — Wed, 19 Dec 2012 18:12:01 -0500

php-fpm seems to be missing a feature that could be really useful in a
shared hosting environment like the one I'm currently setting up.

When dynamically creating pools for new domains, the corresponding unix
socket isn't being created after having issued a SIGUSR2. It seems that
currently it is impossible to gracefully create a communication socket
on-the-fly, which means that php-fpm would have to be stopped/restarted.
This isn't very handy, for obvious reasons.

Any insights into a graceful solution to this problem would be most
appreciated!