Apache restarting regularly

[surfiez]

I have a box here which the apache restarts regularly. It show the error such as :-

[Sun Apr 9 04:04:43 2006] [warn] child process 3759 did not exit, sending another SIGHUP
[Sun Apr 9 04:04:43 2006] [warn] child process 3762 did not exit, sending another SIGHUP
[Sun Apr 9 04:04:43 2006] [warn] child process 3777 did not exit, sending another SIGHUP
[Sun Apr 9 04:04:43 2006] [warn] child process 3780 did not exit, sending another SIGHUP
[Sun Apr 9 04:04:43 2006] [warn] child process 3784 did not exit, sending another SIGHUP
[Sun Apr 9 04:04:43 2006] [warn] child process 3789 did not exit, sending another SIGHUP
[Sun Apr 9 04:04:43 2006] [warn] child process 3811 did not exit, sending another SIGHUP
[Sun Apr 9 04:04:43 2006] [warn] child process 3827 did not exit, sending another SIGHUP
[Sun Apr 9 04:04:45 2006] [notice] SIGHUP received. Attempting to restart
[Sun Apr 9 04:04:49 2006] [notice] Apache configured -- resuming normal operations
[Sun Apr 9 04:04:49 2006] [notice] suEXEC mechanism enabled (wrapper: /usr/sbin/suexec)
[Sun Apr 9 04:04:49 2006] [notice] Accept mutex: sysvsem (Default: sysvsem)


Then it restarts itself but during that period, all websites are not accessable. Its been like this since 2 weeks ago and we have tried whatever we could think of to fix it but to no avail.

What would be the problem ?

Thanks
Justin

 

[nobaloney]

This interesting thread on EV1 forums (on their Ensim forum) says a lot and your paste appears to be happening at about 4 am as well.

I'd guess it could be that the posters on the EV1 forum are right; that you're overpowering your server with what you're doing in a cron.daily set of scripts running about 4 am.

You might want to see what cronjobs are running on the server starting at about 4 am, and let us know.

Jeff

 

[surfiez]

This is in my main crontab

# run-parts
01 * * * * root run-parts /etc/cron.hourly
02 4 * * * root run-parts /etc/cron.daily
22 4 * * 0 root run-parts /etc/cron.weekly
42 4 1 * * root run-parts /etc/cron.monthly
*/10 * * * * root LANG="en_US" /usr/local/mrtg-2/bin/mrtg /var/www/html/mrtg/core/mrtg.cfg

whereby

total 96
drwxr-xr-x 2 root root 4096 Apr 3 14:54 .
drwxr-xr-x 72 root root 12288 Apr 9 10:31 ..
lrwxrwxrwx 1 root root 28 Mar 28 11:36 00-logwatch -> ../log.d/scripts/logwatch.pl
-rwxr-xr-x 1 root root 135 Aug 19 2004 00webalizer
-rwxr-xr-x 1 root root 276 Sep 29 2004 0anacron
-rw-r--r-- 1 root root 797 Oct 19 2004 certwatch
-rwx------ 1 root root 1115 Apr 3 14:54 backup.db.cron
-rwxr-xr-x 1 root root 180 Oct 20 2004 logrotate
-rwxr-xr-x 1 root root 2133 Nov 23 2004 prelink
-rwxr-xr-x 1 root root 104 Nov 2 2004 rpm
-rwxr-xr-x 1 root root 100 Jan 11 23:24 tetex.cron
-rwxr-xr-x 1 root root 286 Aug 14 2004 tmpwatch
-rwxr-xr-x 1 root root 136 Jul 6 2005 yum.cron

i removed slocate.cron and 00-makewhatis cronjob to weekly. If there's any update i'll post it

 

[surfiez]

Problem not fixed..

[Sat Apr 15 00:11:07 2006] [warn] child process 21523 did not exit, sending another SIGHUP
[Sat Apr 15 00:11:07 2006] [warn] child process 27129 did not exit, sending another SIGHUP
[Sat Apr 15 00:11:09 2006] [warn] child process 16360 still did not exit, sending a SIGTERM
[Sat Apr 15 00:11:09 2006] [warn] child process 16670 still did not exit, sending a SIGTERM
[Sat Apr 15 00:11:09 2006] [warn] child process 17892 still did not exit, sending a SIGTERM
[Sat Apr 15 00:11:09 2006] [warn] child process 23070 still did not exit, sending a SIGTERM
[Sat Apr 15 00:11:09 2006] [warn] child process 17008 still did not exit, sending a SIGTERM
[Sat Apr 15 00:11:09 2006] [warn] child process 16830 still did not exit, sending a SIGTERM
[Sat Apr 15 00:11:09 2006] [warn] child process 16985 still did not exit, sending a SIGTERM
[Sat Apr 15 00:11:09 2006] [warn] child process 17010 still did not exit, sending a SIGTERM
[Sat Apr 15 00:11:09 2006] [warn] child process 17011 still did not exit, sending a SIGTERM
[Sat Apr 15 00:11:09 2006] [warn] child process 17439 still did not exit, sending a SIGTERM
[Sat Apr 15 00:11:09 2006] [warn] child process 19582 still did not exit, sending a SIGTERM
[Sat Apr 15 00:11:09 2006] [warn] child process 18181 still did not exit, sending a SIGTERM
[Sat Apr 15 00:11:09 2006] [warn] child process 17596 still did not exit, sending a SIGTERM
[Sat Apr 15 00:11:09 2006] [warn] child process 16831 still did not exit, sending a SIGTERM
[Sat Apr 15 00:11:09 2006] [warn] child process 17298 still did not exit, sending a SIGTERM
[Sat Apr 15 00:11:09 2006] [warn] child process 17135 still did not exit, sending a SIGTERM
[Sat Apr 15 00:11:09 2006] [warn] child process 17136 still did not exit, sending a SIGTERM
[Sat Apr 15 00:11:09 2006] [warn] child process 15365 still did not exit, sending a SIGTERM
[Sat Apr 15 00:11:09 2006] [warn] child process 20387 still did not exit, sending a SIGTERM
[Sat Apr 15 00:11:09 2006] [warn] child process 20807 still did not exit, sending a SIGTERM
[Sat Apr 15 00:11:09 2006] [warn] child process 21416 still did not exit, sending a SIGTERM
*** glibc detected *** double free or corruption (fasttop): 0x09fcec80 ***
*** glibc detected *** double free or corruption (fasttop): 0x09fcec80 ***
[Sat Apr 15 00:11:13 2006] [error] child process 16831 still did not exit, sending a SIGKILL
[Sat Apr 15 00:11:13 2006] [error] child process 17135 still did not exit, sending a SIGKILL
[Sat Apr 15 00:11:13 2006] [notice] SIGHUP received. Attempting to restart
[Sat Apr 15 00:11:21 2006] [notice] Apache configured -- resuming normal operations
[Sat Apr 15 00:11:21 2006] [notice] suEXEC mechanism enabled (wrapper: /usr/sbin/suexec)
[Sat Apr 15 00:11:21 2006] [notice] Accept mutex: sysvsem (Default: sysvsem)

Please advice

 

[surfiez]

Hello ? No Solution available ?

 

[outpernet]

we are having the same cause: glibc detected *** double free or corruption (fasttop):

but the consecuences are high load. I think glibc 2.3.4 has a bug

 

[existenz]

For apache without looking at the server this looks like a timeout on a script. You may need to set a quicker timeout in Apache, PHP and possibly in MySQL. I can't tell you without check out the server to see exactly what that process is doing.

 

[daroch]

Hi.
We have the same problem. Apach is restarted every day at 00:10.

Is there any solution to disabled it?

 

[nobaloney]

This might have something to do with log rotation. Every daemon that writes to it's own logs using it's own routines must stop in order for the logs to be rotated.

Jeff

 

[PauGasol]

Same problem : http://www.directadmin.com/forum/showthread.php?s=&threadid=14524 ?

Do you use Debian 3.1 ?