Worker dies with a segfault error

[greekduke]

Hello,

Hello,

I have an issue with segfaults especially when I send lot's of traffic to
the server. The hardware is an HP proliant G8 with 24 cores but I am using
only one worker because the problem becames worse with more than one. I have
also modified the ulimit for the nginx user up to 65535 but I am still
getting the segfaults. I have also tried various nginx versions from 1.3.9
up to the latest 1.5.9. In one of the tests I have also removed the two
modules I am using. Finally I am using nginx as a proxy only.

Feb 24 13:08:19 -00 kernel: nginx[23543]: segfault at 20 ip 0000000000405d3f
sp 00007fffc66ab600 error 4 in nginx[400000+92000]
Feb 24 13:09:12 a1-00 kernel: nginx[23547]: segfault at 20 ip
0000000000405d3f sp 00007fffc66ab600 error 4 in nginx[400000+92000]
Feb 24 13:09:43 a1-00 kernel: nginx[23584]: segfault at 20 ip
0000000000405d3f sp 00007fff0f2f3b80 error 4 in nginx[400000+92000]
Feb 24 13:10:20 a1-00 kernel: nginx[23592]: segfault at 20 ip
0000000000405d3f sp 00007fff0f2f3be0 error 4 in nginx[400000+92000]
Feb 24 13:12:06 a1-00 kernel: nginx[23671]: segfault at 20 ip
0000000000405d3f sp 00007fff0f2f3be0 error 4 in nginx[400000+92000]

nginx version: nginx/1.5.9
built by gcc 4.4.7 20120313 (Red Hat 4.4.7-4) (GCC) 
configure arguments: --add-module=/software/ngx_postgres-master-1.5
--add-module=/software/nginx-eval-module-master --user=nginx --with-debug

Red Hat Enterprise Linux Server release 6.5 (Santiago)
Linux ape-00 2.6.32-431.el6.x86_64 #1 SMP Sun Nov 10 22:19:54 EST 2013
x86_64 x86_64 x86_64 GNU/Linux

Thanks

BR/Kostas

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,247886,247886#msg-247886