Hello All I installed a brand new server Centos 6.4 64bit on a dedicated machine, I followed The Perfect Server - CentOS 6.4 x86_64 (nginx, Dovecot, ISPConfig 3) . I have a few wordpress sites and a few joomla sites that I moved from an older server running Apache to the new one I just installed. Everything seems to be running fine, sites load etc, but everything soon stops, I am getting 504 Gateway Time-out errors. I have to restart php-fpm service to get it to start working again, but in a few minutes after restarting I am getting the same error. The only things I did that were not part of the guide was: For wordpress sites, in ISPCONFIG/Options in the nginx Directives in order for URL rewriting to work, I added location / { index index.php index.html index.htm; try_files $uri $uri/ /index.php?$args; } and for Joomla sites in the same place I added: # Support Clean (aka Search Engine Friendly) URLs</tt> location / { try_files $uri $uri/ /index.php?q=$request_uri; } I am not sure what is the cause of this problem. Searching on google, I tried several things I found, but nothing seems to help resolve the problem. Can somebody help me get this resolved? Thanks!
Did you check the web site's error log and also the PHP-FPM log? Also check your PHP-FPM emergency settings (see chapter 3.3 on http://www.howtoforge.com/configuri...x-nginx-mysql-php-fpm-for-maximum-performance).
These are last entries of the error log for one of the wordpress sites 2013/10/02 19:30:53 [error] 2794#0: *2 connect() failed (111: Connection refused) while connecting to upstream, client: 192.168.1.14, server: oelsnet.net, request: "GET / HTTP/1.1", upstream: "fastcgi://127.0.0.1:9010", host: "oelsnet.net" 2013/10/02 19:30:57 [error] 2794#0: *2 connect() failed (111: Connection refused) while connecting to upstream, client: 192.168.1.14, server: oelsnet.net, request: "GET / HTTP/1.1", upstream: "fastcgi://127.0.0.1:9010", host: "oelsnet.net" 2013/10/02 19:31:08 [error] 2794#0: *2 connect() failed (111: Connection refused) while connecting to upstream, client: 192.168.1.14, server: oelsnet.net, request: "GET / HTTP/1.1", upstream: "fastcgi://127.0.0.1:9010", host: "oelsnet.net" 2013/10/02 19:31:09 [error] 2794#0: *2 connect() failed (111: Connection refused) while connecting to upstream, client: 192.168.1.14, server: oelsnet.net, request: "GET / HTTP/1.1", upstream: "fastcgi://127.0.0.1:9010", host: "oelsnet.net" 2013/10/02 19:31:09 [error] 2794#0: *2 connect() failed (111: Connection refused) while connecting to upstream, client: 192.168.1.14, server: oelsnet.net, request: "GET / HTTP/1.1", upstream: "fastcgi://127.0.0.1:9010", host: "oelsnet.net" 2013/10/02 21:50:44 [error] 2794#0: *140 recv() failed (104: Connection reset by peer) while reading response header from upstream, client: 192.168.1.122, server: oelsnet.net, request: "GET /wp-content/plugins/dropdown-menu-widget/css/shailan-dropdown.min.css HTTP/1.1", upstream: "fastcgi://127.0.0.1:9010", host: "oelsnet.net", referrer: "http://oelsnet.net/" 2013/10/02 21:55:43 [error] 2794#0: *139 upstream timed out (110: Connection timed out) while reading response header from upstream, client: 192.168.1.122, server: oelsnet.net, request: "GET /wp-content/plugins/ubm-premium/scripts.php?ver=3.6 HTTP/1.1", upstream: "fastcgi://127.0.0.1:9010", host: "oelsnet.net", referrer: "http://oelsnet.net/" These are last entries of php-fpm log [01-Oct-2013 22:30:00] NOTICE: fpm is running, pid 9322 [01-Oct-2013 22:30:00] NOTICE: ready to handle connections [01-Oct-2013 22:48:46] NOTICE: Reloading in progress ... [01-Oct-2013 22:48:46] NOTICE: reloading: execvp("php-fpm", {"php-fpm", "--daemonize"}) [01-Oct-2013 22:48:46] NOTICE: using inherited socket fd=8, "127.0.0.1:9000" [01-Oct-2013 22:48:46] NOTICE: using inherited socket fd=8, "127.0.0.1:9000" [01-Oct-2013 22:48:46] NOTICE: fpm is running, pid 13769 [01-Oct-2013 22:48:46] NOTICE: ready to handle connections [01-Oct-2013 22:50:03] WARNING: [pool ispconfig] child 13779 exited on signal 11 (SIGSEGV - core dumped) after 76.770434 seconds from start [01-Oct-2013 22:50:03] NOTICE: [pool ispconfig] child 14158 started [01-Oct-2013 22:54:03] NOTICE: Reloading in progress ... [01-Oct-2013 22:54:03] NOTICE: reloading: execvp("php-fpm", {"php-fpm", "--daemonize"}) [01-Oct-2013 22:54:03] NOTICE: using inherited socket fd=8, "127.0.0.1:9000" [01-Oct-2013 22:54:03] NOTICE: using inherited socket fd=8, "127.0.0.1:9000" [01-Oct-2013 22:54:03] NOTICE: using inherited socket fd=10, "/var/lib/php5-fpm/apps.sock" [01-Oct-2013 22:54:03] NOTICE: using inherited socket fd=10, "/var/lib/php5-fpm/apps.sock" [01-Oct-2013 22:54:03] NOTICE: using inherited socket fd=12, "/var/lib/php5-fpm/ispconfig.sock" [01-Oct-2013 22:54:03] NOTICE: using inherited socket fd=12, "/var/lib/php5-fpm/ispconfig.sock" [01-Oct-2013 22:54:03] NOTICE: fpm is running, pid 14430 [01-Oct-2013 22:54:03] NOTICE: ready to handle connections I went to the link you suggested about the emergency PHP-FPM settings, I made the adjustments in /etc/php-fpm.conf but this didn't help, I am still having the same problem.
I noticed that when the websites stop responding this error is present in /var/log/php-fpm/error.log [03-Oct-2013 21:26:18] WARNING: [pool web1] child 27029 exited on signal 11 (SIGSEGV - core dumped) after 555.135693 seconds from start I turned debug option and the log is as follows [03-Oct-2013 21:44:04.131474] DEBUG: pid 28203, fpm_scoreboard_proc_alloc(), line 296: [pool web1] the proc->free_slot was not free. Let's search [03-Oct-2013 21:44:04.132551] DEBUG: pid 28203, fpm_children_make(), line 421: [pool web1] child 28747 started [03-Oct-2013 21:44:04.132577] DEBUG: pid 28203, fpm_pctl_perform_idle_server_maintenance(), line 428: [pool web1] 1 child(ren) have been created dynamically [03-Oct-2013 21:44:04.132587] DEBUG: pid 28203, fpm_pctl_perform_idle_server_maintenance(), line 379: [pool web2] currently 0 active children, 2 spare children, 2 running children. Spawning rate 1 [03-Oct-2013 21:44:04.132596] DEBUG: pid 28203, fpm_pctl_perform_idle_server_maintenance(), line 379: [pool www] currently 0 active children, 5 spare children, 5 running children. Spawning rate 1 [03-Oct-2013 21:44:04.132608] DEBUG: pid 28203, fpm_pctl_perform_idle_server_maintenance(), line 379: [pool web4] currently 0 active children, 2 spare children, 2 running children. Spawning rate 1 [03-Oct-2013 21:44:04.132616] DEBUG: pid 28203, fpm_pctl_perform_idle_server_maintenance(), line 379: [pool ispconfig] currently 0 active children, 2 spare children, 2 running children. Spawning rate 1 [03-Oct-2013 21:44:04.132624] DEBUG: pid 28203, fpm_pctl_perform_idle_server_maintenance(), line 379: [pool web3] currently 0 active children, 2 spare children, 2 running children. Spawning rate 1 [03-Oct-2013 21:44:05.132706] DEBUG: pid 28203, fpm_pctl_perform_idle_server_maintenance(), line 379: [pool apps] currently 0 active children, 2 spare children, 2 running children. Spawning rate 1 [03-Oct-2013 21:44:05.132741] DEBUG: pid 28203, fpm_pctl_perform_idle_server_maintenance(), line 379: [pool web5] currently 0 active children, 2 spare children, 2 running children. Spawning rate 1 [03-Oct-2013 21:44:05.132755] DEBUG: pid 28203, fpm_pctl_perform_idle_server_maintenance(), line 379: [pool web1] currently 5 active children, 1 spare children, 6 running children. Spawning rate 2 [03-Oct-2013 21:44:05.132782] DEBUG: pid 28203, fpm_pctl_perform_idle_server_maintenance(), line 379: [pool web2] currently 0 active children, 2 spare children, 2 running children. Spawning At this point im stuck....