Apache restarts every night

Portal Home > Knowledgebase > Articles Database > Apache restarts every night

Posted by BELLonline, 08-01-2007, 06:40 AM
Hi, I'm having a problem with Apache restarting every night - in the error logs I get the following message: [Tue Jul 31 21:39:02 2007] [notice] caught SIGTERM, shutting down [Tue Jul 31 21:39:03 2007] [notice] suEXEC mechanism enabled (wrapper: /usr/sbin/suexec) [Tue Jul 31 21:39:03 2007] [notice] Digest: generating secret for digest authentication ... [Tue Jul 31 21:39:03 2007] [notice] Digest: done [Tue Jul 31 21:39:03 2007] [notice] LDAP: Built with OpenLDAP LDAP SDK [Tue Jul 31 21:39:03 2007] [notice] LDAP: SSL support unavailable [Tue Jul 31 21:39:03 2007] [notice] Apache/2.0.52 (CentOS) configured -- resuming normal operations I know that this wouldn't normally be a problem but on this server I run Cast Control which uses PHP to start shoutcast servers - when Apache restarts it turns off all of the Shoutcast servers. Is there any way to stop Apache from restarting all the time? Alternatively, is there a way to run processes under a different user other than apache so that they don't go down with it? Any help is appreciated, thanks.
Posted by whmcsguru, 08-01-2007, 07:28 AM
Apache is shutting down because it's being told to: Many times the cause of this is simply logs being rotated, though occasionally the control panel you're using will restart apache nightly . I know this is the case for DA, though the fix I can't recall right off hand.
Posted by BELLonline, 08-01-2007, 07:45 AM
I'm not actually using a control panel at all on this server. This was happening before so after trying different things, I uninstalled the server and got a fresh OS installtion, but still the prolem is there. I've tried changing /etc/cron.daily/logrotate to the cron.weekly folder to see if that will work. We'll find out tomorrow morning I suppose
Posted by BELLonline, 08-01-2007, 08:28 AM
I think I've cracked it - I compared the config file from another server thats not having the same issue and found the following: User nobody Group apache On the servers with the problem it looked like this: User apache Group apache I changed it and restarted and now the shoutcast processes are all running under 'nobody' as with my other servers so I think that this will stop them from going down when apache restarts fingers crossed....

Add to Favourites Print this Article