Hi everyone
I've inherited an installation of vCenter Server Appliance, managing 2 ESXi hosts. We only have 8 VMs running as most have now been migrated to our new platform in colocation. However the VMs we are running will need to stay on-site for a while.
We had a problem last weekend where the /storage/db partition ran out of space which I only noticed because our VM backups started failing (Unitrends).
We're currently running vCSA 5.0.0.3324 Build 472350 which uses the embedded DB2 database.
I tried updating the vCSA using the update tab of the web interface. That lists version 5.0.0.6194 Build 920217 as being available. That update eventually timed out after 5+ hours of running so I'll rolled back to a snapshot.
Then I also tried running the full upgrade to a fresh 5.1 appliance using this guide...http://kb.vmware.com/selfservice/microsites/search.do?language=en_US&cmd=displayKC&externalId=2033990
However this also failed over the weekend. On further investigation it failed because the appliance ran out of inodes on the / parition, caused by thousands of /new/export/dbxxxxxx.lob files
So first of all, I'd like to prune the database down, if possible, so the upgrade to Postgres doesn't take as long.
Is there a way to manually prune the DB2 database? It's currently at 19GB. I have reduced the tasks/events database retention through vSphere Client to 120 days but the database size remains constant at 19GB.
Secondly, is there a way to run the vCSA updates through the command line so I can actually see what's going on?
I'd like to run an manual update using the CLI to convert DB2 to Postgres so I can keep an eye on what's happening.
Once we are running a trimmed down Postgres database I will try the upgrade to a fresh 5.1 appliance again.
Any comments/suggestions/ideas?
Cheers, B