Atomicorp

Hello,

Since recently I have one problem with Apache server on CentOS 5.5
The load average goes sometimes up to 200
that happens not every day but randomly
sometimes two or tree times daily sometimes two or three times weekly
after Apache restart everything goes normal.

I found following lines in error.log
....
[Mon Dec 27 15:29:16 2010] [warn] Init: SSL server IP/port conflict: default-94-126-17-98:443 (/etc/httpd/conf.d/zz010_psa_httpd.conf:123) vs. synect.ch:443 (/var/www/vhosts/synect.ch/conf/httpd.include:13)
[Mon Dec 27 15:29:16 2010] [warn] Init: SSL server IP/port conflict: default-94-126-17-97:443 (/etc/httpd/conf.d/zz010_psa_httpd.conf:104) vs. praderfaltdisplay.ch:443 (/var/www/vhosts/praderfaltdisplay.ch/conf/httpd.include:13)
[Mon Dec 27 15:29:16 2010] [warn] Init: SSL server IP/port conflict: default-94-126-17-96:443 (/etc/httpd/conf.d/zz010_psa_httpd.conf:85) vs. horde.webmail:443 (/etc/httpd/conf.d/zzz_horde_vhost.conf:41)
[Mon Dec 27 15:29:16 2010] [warn] Init: You should not use name-based virtual hosts in conjunction with SSL!!
[Mon Dec 27 15:29:17 2010] [notice] mod_bw : Memory Allocated 0 bytes (each conf takes 32 bytes)
[Mon Dec 27 15:29:18 2010] [notice] mod_bw : Version 0.8 - Initialized [0 Confs]
[Mon Dec 27 15:29:18 2010] [notice] mod_python: Creating 4 session mutexes based on 256 max processes and 0 max threads.
[Mon Dec 27 15:29:18 2010] [notice] Apache configured -- resuming normal operations
[Mon Dec 27 15:29:39 2010] [error] [client 67.195.115.94] File does not exist: /var/www/vhosts/paritasalariale.ch/httpdocs/robots.txt
[Mon Dec 27 15:29:40 2010] [error] [client 67.195.115.94] File does not exist: /var/www/vhosts/paritasalariale.ch/httpdocs/homeframeset-i.html
[Mon Dec 27 15:32:56 2010] [error] [client 85.2.160.111] File does not exist: /var/www/vhosts/yakaree.com/httpdocs/favicon.ico
[Mon Dec 27 15:39:17 2010] [error] [client 208.115.111.244] File does not exist: /var/www/vhosts/yakaree.com/httpdocs/robots.txt
[Mon Dec 27 15:41:12 2010] [warn] child process 10887 still did not exit, sending a SIGTERM
[Mon Dec 27 15:41:12 2010] [warn] child process 9367 still did not exit, sending a SIGTERM
[Mon Dec 27 15:41:12 2010] [warn] child process 11292 still did not exit, sending a SIGTERM
.....

[Mon Dec 27 15:41:18 2010] [error] child process 10887 still did not exit, sending a SIGKILL
[Mon Dec 27 15:41:18 2010] [error] child process 9367 still did not exit, sending a SIGKILL
[Mon Dec 27 15:41:18 2010] [error] child process 11292 still did not exit, sending a SIGKILL
....

and this lines repeat every time when that happens
did anybody of you experienced this problem before or
has any idea or solution because I'm not really sure what causes the problem

System information

Operating system CentOS Linux 5.5
Kernel and CPU Linux 2.6.32.21-3.art.x86_64 on x86_64
Processor information Intel(R) Core(TM)2 Duo CPU E4600 @ 2.40GHz, 2 cores
Apache version Apache 2.2.3-43.el5.centos.3

I'm having a similar type problem right now. Check your /var/log/messages file for a kernel bug. What seems to be happening on mine is this kernel hangs for a while, so processes pile up, and the load shoots up. I'm on a hamstringed kernel because of being in a Xen VM. Maybe not the same problem as you, but similar symptoms with httpd shooting the load up.

Yes we found today a lot of kernel messages during server overload for a half hour. We have ASL and use latest ASL Kernel. Scott, Mike any idea?