Lighttpd permission denied on php-fastcgi.socket

[sandrokeil]

I've tried this image with Docker 1.5.0. The Zend Server backend is not reachable, because of the following error (from lighttpd.error.log):

2015-02-23 21:15:33: (mod_fastcgi.c.2550) unexpected end-of-file (perhaps the fastcgi process died): 11 Permission denied
2015-02-23 21:15:33: (mod_fastcgi.c.3346) response not received, request sent: 830 on socket: unix:/usr/local/zend/gui/lighttpd/tmp/php-fastcgi.socket for /ZendServer/index.php?, closing connection

Is this a known problem?

[dror-g]

Hi Sandro, This issue occurred at times when a Zend Server container was stopped and started again. was this the case for you or did it fail on the first "run" attempt? On Feb 23, 2015 11:27 PM, "Sandro Keil" notifications@github.com wrote:

I've tried this image with Docker 1.5.0. The Zend Server backend is not reachable, because of the following error (from lighttpd.error.log):

2015-02-23 21:15:33: (mod_fastcgi.c.2550) unexpected end-of-file (perhaps the fastcgi process died): 11 Permission denied 2015-02-23 21:15:33: (mod_fastcgi.c.3346) response not received, request sent: 830 on socket: unix:/usr/local/zend/gui/lighttpd/tmp/php-fastcgi.socket for /ZendServer/index.php?, closing connection

Is this a known problem?

— Reply to this email directly or view it on GitHub https://github.com/zendtech/php-zendserver-docker/issues/1.

[sandrokeil]

Hello, it occurs on the first start. I use docker run -p 80:80 -p 10081:10081 zend/php-zendserver and get the following output:

Starting Zend Server 8.0.1 ..

Starting Zend Server Monitor node [OK]
[24.02.2015 19:10:31 SYSTEM] watchdog for monitor is running. 
[24.02.2015 19:10:31 SYSTEM] monitor is running. 
 * Starting web server apache2
AH00558: apache2: Could not reliably determine the server's fully qualified domain name, using 172.17.0.2. Set the 'ServerName' directive globally to suppress this message
 * 
spawn-fcgi: child spawned successfully: PID: 167
Starting Zend Server GUI [Lighttpd] [OK]
[24.02.2015 19:10:37 SYSTEM] watchdog for lighttpd is running. 
[24.02.2015 19:10:37 SYSTEM] lighttpd is running. 
Starting Session Clustering daemon [OK]
[24.02.2015 19:10:39 SYSTEM] watchdog for scd is running. 
[24.02.2015 19:10:39 SYSTEM] scd is running. 
Starting Deployment [OK]
[24.02.2015 19:10:40 SYSTEM] watchdog for zdd is running. 
[24.02.2015 19:10:40 SYSTEM] zdd is running. 
Starting JobQueue [OK]
[24.02.2015 19:10:41 SYSTEM] watchdog for jqd is running. 
[24.02.2015 19:10:41 SYSTEM] jqd is running. 
Starting ZendServerDaemon [OK]
[24.02.2015 19:10:42 SYSTEM] watchdog for zsd is running. 
[24.02.2015 19:10:42 SYSTEM] zsd is running. 

Zend Server started...
Bootstrapping single server
Error: API error: Unexpected response content type: text/html
restarting ZS
Error: API error: Unexpected response content type: text/html
Zend Server is ready for use
Your application is available at http://172.17.0.2
To access Zend Server, navigate to http://172.17.0.2:10081

Notice: Undefined offset: 1 in /usr/local/bin/sanity.php on line 7

Are these permissions ok?

root@b364ac35c37f:/usr/local/zend/gui# ls -la

-rw-r--r--  1 root root 2602 Dec 15  2011 .zfproject.xml
drwxr-sr-x  2 root root 4096 Jan 28 11:13 3rdparty
drwxrwsr-x  3 root zend 4096 Feb 24 19:33 config
drwxr-sr-x  4 root root 4096 Jan 28 11:13 data
-rw-r--r--  1 root root  879 Nov 20 08:31 init_application_config.php
-rw-r--r--  1 root root  281 Sep 24  2012 init_autoloader.php
drwxr-xr-x 12 zend zend 4096 Feb 24 19:32 lighttpd
drwxr-sr-x 19 root root 4096 Jan 28 11:13 module
drwxr-sr-x  9 root root 4096 Jan 28 11:13 public
drwxr-sr-x  3 root root 4096 Jan 28 11:13 tests
drwxr-xr-x  5 root root 4096 Jan 28 11:13 vendor

[dror-g]

Yes, the permissions seem ok. Unfortunately I am unable to reproduce this error, and our QA personnel did not encounter it during their tests.

Does this happen every time you run the php-zendserver image consistently? Could you share you system architecture please (32/64 OS, linux/mac/win, which distro, available RAM)?

If you dont mind please reply by e-mail to dror.g@zend.com . Thank you very much for this important report!

[sandrokeil]

I've sent you an email with detailed information. In short, it's an Ubuntu 14.10 (Linux 3.16.0-30-generic (x86_64)) with 8x Intel(R) Core(TM) i7 CPU Q 720 @ 1.60GHz and 8 GB RAM (~5 GB avialable).

Does this happen every time you run the php-zendserver image consistently?

Yes, this problem occurs every time.

This issue occurred at times when a Zend Server container was stopped and started again.

I've found one interesting point in the lighttpd.error.log file. There is an entry with "server stopped" from 2015-01-28 11:16:32. Could this be the problem?

root@9f93e217ae3d:/# cat /usr/local/zend/gui/lighttpd/logs/lighttpd.error.log 
2015-01-28 11:15:56: (log.c.164) server started 
2015-01-28 11:16:32: (server.c.1558) server stopped by UID = 0 PID = 5711 
2015-02-25 20:07:15: (log.c.164) server started 
2015-02-25 20:07:28: (mod_fastcgi.c.2550) unexpected end-of-file (perhaps the fastcgi process died): 13 Permission denied 
2015-02-25 20:07:28: (mod_fastcgi.c.3346) response not received, request sent: 1089 on socket: unix:/usr/local/zend/gui/lighttpd/tmp/php-fastcgi.socket for /ZendServer/index.php?, closing connection 

[sandrokeil]

Good news @dror-g , I've upgraded to kernel 3.19.0-031900-generic (x86_64) and there are no problems. It seems it's a kernel version issue. So it should be reproducible with kernel Linux 3.16.0-30-generic (x86_64).

[dror-g]

Great! very glad to hear that. thanks again for your detailed report and assistance. On Feb 26, 2015 7:36 PM, "Sandro Keil" notifications@github.com wrote:

Good news @dror-g https://github.com/dror-g , I've upgraded to kernel 3.19.0-031900-generic (x86_64) and there are no problems. It seems it's a kernel version issue. So it should be reproducible with kernel Linux 3.16.0-30-generic (x86_64).

— Reply to this email directly or view it on GitHub https://github.com/zendtech/php-zendserver-docker/issues/1#issuecomment-76225119 .