Near nightly httpd fail after upcp since version 11

[orty]

Every since cPanel 11 was installed a week or so ago, nearly every night about 5-10 minutes after the upcp runs, apache fails. I wake up in the morning with my upcp e-mail, with an e-mail with a time stamp five minutes afterwards that says "httpd failed @ Sun May 20 02:45:39 2007. A restart was attempted automagicly." The time on the e-mail is just about exactly when upcp gets done. Apache comes back up, fine and runs fine the rest of the day but what is causing this to die? I'm not noticing anything in the upcp that's restarting or killing the process. Is there someplace I can look?

[Infopro]

Open SSH and type this in:

/usr/sbin/apachectl configtest

Whats it tell you?

[orty]

Open SSH and type this in:

/usr/sbin/apachectl configtest

Whats it tell you?

Here's what I get:

root@servername [/usr/sbin]# /usr/sbin/apachectl configtest
-bash: /usr/sbin/apachectl: No such file or directory

I ran a "locate apachectl" just to see where it was, and got these:

/usr/local/apache/bin/apachectl
/usr/local/apache/htdocs/manual/programs/apachectl.html.en
/usr/local/apache/htdocs/manual/programs/apachectl.html.ja.jis
/usr/local/apache/htdocs/manual/programs/apachectl.html.html
/usr/local/apache/man/man8/apachectl.8
/usr/share/man/fr/man8/apachectl.8.gz

So I ran "/usr/local/apache/bin/apachectl configtest" and got the following:

Warning: DocumentRoot [/home/username/public_html/dev] does not exist
[Sun May 20 12:21:09 2007] [warn] NameVirtualHost xx.xx.xx.xx:80 has no VirtualHosts
[Sun May 20 12:21:09 2007] [warn] NameVirtualHost xx.xx.xx.xx:80 has no VirtualHosts
Syntax OK

I see that when I delete the dev.username.com subdomain from the username.com cPanel, it didn't delete it properly from the VirtualHost, so I'll manually delete that.

Thoughts otherwise?

[will_raymo2000]

I seem to be getting this failure as well.. not every night but intermittently every couple of days. I try to go onto the site and apache is not running at all. Does yours restart itself or do you have to do it manually?

Thanks,
Will

[orty]

I seem to be getting this failure as well.. not every night but intermittently every couple of days. I try to go onto the site and apache is not running at all. Does yours restart itself or do you have to do it manually?

Thanks,
Will

The the time I get online about 5 hours after the upcp, it appears to have been running fine as none of my various other monitoring services (siteuptime.com, monitor.us, etc...) pick up the site as being dead, so it must be just a temporary blip.

[Infopro]

My mistake sorry. Yes sure, clean it up so you get no errors and see if that helps.

[orty]

Sorry, I'm just getting back to replying on this. Cleared out the httpd.conf errors it was reporting, and also ran the httpdcleanup script, and my httpd.conf file is nice and clean.

While it's less frequent, I'm still getting an httpd failure every 4 nights or so after upcp runs.

Ideas?

[gfisch]

Hi Orty and others.

Did this get resolved? This just started happening to me four days ago. Intermitant apache failures and chkservd can't restart it with malloc errors.
Yet starting by hand is fine.

Thanks
Gord

[orty]

Hi Orty and others.

Did this get resolved? This just started happening to me four days ago. Intermitant apache failures and chkservd can't restart it with malloc errors.
Yet starting by hand is fine.

Thanks
Gord

Don't know if it'll help you at all, but I basically worked with cPanel and helped discover a bug in chksrvd that has since been resolved in the CURRENT branch. We also upgraded to Perl 5.8.8, and then, lastly, moved my cpbackup time so it was run after my upcp (it was running before it before) and spread it out a couple hours apart. It wasn't until after that last step that things were fixed. I haven't had the nightly fail like I was having before. So my guess it was some sort of resource problem that seems really odd considering the server has few clients and 2gigs of RAM and an AMD Athlon 3800+.