[ngw] Linux Web access issue
MGlenn at cco.state.oh.us
Mon Apr 5 21:09:12 UTC 2010
The apache settings for logrotate (/etc/logrotate.d/apache) are very liberal. You might want to try tightening them up a bit.
>>> "Scott Kunau" <skunau at igtg.net> 04/05/2010 00:11 >>>
I have the same problem with a client. The problem is with the apache web server access_(something or other).log that grows and grows and doesn't get logrotated properly. Is your /dev/xvda3 represent the /var directory/partition? That is where the file in question lives.
I made the /var only 5GB in size when I built the server 18 months ago. It can't be re-sized so we wrote a script that can be run when needed to clear the log file and get us back to below 20% used space.
If you'd like to have the script, I can send it or post it Monday.
Scott Kunau, Sr. Consultant/Partner
Innovative Global Technology Group
skunau at igtg.net
Sent with GroupWise 8
>>> "Jeffrey Sessler" <Jeff at ScrippsCollege.edu> 4/1/2010 9:51 PM >>>
Well, when I run into this it's a case of the /tmp directory filling with webaccess temp files (tmp.filemane). Once /tmp runs out of space, users can't reply to messages.
>>> "John Dickey" <jddickey at mpsomaha.org> 4/1/2010 6:01 PM >>>
Having a problem lately on our Linux server that supports web access. The issue is that users web access at times can not reply to emails. Then after looking at the server df -h notice that /dev/xvda3 partition is at 100%. I go in and clean up some of the logs to bring it down in the 70% range and users can reply again. Today, the same thing but the usuage on /dev/xvda3 bounces around from 77% up to 97%.
My question is what is causing this to happen and how can I get it to stop?
thanks in advance for every ones help!!
ngw mailing list
ngw at ngwlist.com
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the ngw