Databases Performance settings

[mguthrie]

I ran maintenance script, but it again have not deleted old information.

I'm looking at trying to recreate this, how are you verifying that the old data is not being deleted?

[alabzhinov]

I configured DB performance settings to retain 180 days for state history, but performance graphs still show the info starting the end of November 2010. It is the time i installed NagiosXI. After the cleanup performance info should start on April 2011.
Another criteria there should be more free hard drive space after cleanup, but it is not happening.

[mguthrie]

OH, ok. The performance data is not stored in the database at all. Performance data is stored in an rrd file that is a static size. Once the rrd is created the file size never changes. As long as the maintenance settings are set accordingly, the database will be pruned correctly. At this point you're probably not going to notice a change in total hard drive space because the dbmaint.php script has already trimmed the databases and you won't see much of a difference. The hard drive space is probably more affected by the quantity of hosts and services that you have than anything else.

[alabzhinov]

I had a default setting on database performance tab, then i lowered the settings to 90, 90 and 180 and i expected to free a lot of space because Nagios should have deleted the info from the DB for the period November 2010 - April 2011. But this change didn't affect the disk usage at all, it is constantly growing.
So now i'm wondering how do i free the disk space?

[alabzhinov]

How do i check the DB? I want to make sure that i don't have data older than specified in database performance settings.

[alabzhinov]

I found what my issue is. It is /var/spool/mail/root file , it was 14 Gb in size. I deleted it, but it is growing really fast - it's 55 Mb in 24 hours, so i'm going to find the way to drop mail to root.

Thanks for your help.

[mguthrie]

Thanks for the update, that's good to know for future reference. You may also want to look at the notifications settings for the nagiosadmin user to see if that user's email address is set to root@localhost. If that user is a contact for everything and is receiving email alerts for all of the state changes that would definitely create a large flood of emails.