TECHZEN Zenoss User Community ARCHIVE  

Cleaning/Shrinking the Event Database went wrong

Subject: Cleaning/Shrinking the Event Database went wrong
Author: [Not Specified]
Posted: 2014-02-12 09:21

Running Zenoss 3.2.1 on SLES

So my Zenoss drive was getting full and I saw that the database was taking up 75% of the space available on the virtual machine. So I searched around and found that I could shrink the event database by following the steps here: https://community.zenoss.org/message/60349

Well everything went as expected until I got to the steps for restoring the database, then when it was importing the database from the file that I used to backup the prior events database, the error came after it processed events for about 15minutes. Now when I attempt to access the dashboard I get error after error (as expected since the restore failed).

So since I have already deleted the files that I needed I'm thinking that my options are as follows:
1) Try and recreate an empty events database
2) Try and upgrade to v4 and see if in the process of the upgrade it might fix the database problem
3) Scrap everything since I appear to not have a good export of the event database and start from scratch with a new system.

Obviously I would have liked to keep the logs and events on everything that was up/down so that I can report to my bosses on our up-time. However they would rather see me spend my time wisely and resolve the issue as quickly as possible and move forward. If I go with option #3, is there a way to export all of the devices to easily import into the new system That is my hangup with option #3 is having to re-enter/re-setup everything again from scratch and adding devices takes like 15seconds a device...

Please advise what you would do if you were in my situation...
Tim



Subject: Restore from backup
Author: [Not Specified]
Posted: 2014-02-14 10:57

Your best bet would have been to make a backup before you got started, in case things like this happen. If you have no backup.....

Try #1, if that fails then do a fresh install and export the events database back into your existing install

The ultimate solution though would be #3, using the current 4.x branch!

Hydruid



< Previous
Using an agent instead of zenperfsnmp
  Next
Processes being monitored always show status "up" even if down?
>