2016-09-08 02:32 PM
- WFA 3.1P2 - on Win2012R2
- OCUM 6.4 - vApp on ESXi
I recently upgraded to OCUM 6.4P1 and have noticed that if/when I go into OCUM->Health->Storage->Clusters and pick a custer to Action->Rediscover I'll see this error (see below) when I watch the progress. After completing rediscover of the cluster it seems to contact WFA to have WFA perform a data-source acquire-now (I think that is what it's doing).
I can understand the error because it says it is trying to trigger the WFA Data-Source called 'UM 6.4P1 : ocum.tnt.com' which does not exist. I know it doesn't exist because one day I say it and said 'What is this?' and deleted it ... since I already had a working ocum data-source under a different name.
So is there a way out of this? This would be very useful if doing an ocum rediscover also automatically triggered a WFA data-source 'acquire now' ... but I guess I've broken the feature by deleting the WFA data-source IT THINKS is the one to use. I can't find anyway to tell OCUM to use one named XYZ (or my name) instead.
Advise? (besides reading the manuals and starting all over)
Solved! See The Solution
2016-09-09 05:48 AM
That's a flaw in OCUM and has to do with the OCUM upgrade.
The name of the data source when OCUM triggers the aquisition seems to be hardcoded in OCUM. So OCUM always advises WFA to aquire the data source with the name of the following format:
"UM <um-version> : <um-ip-address or hostname>"
I bet your data source has a different name in WFA because
a) you added it manually with a different name
b) the automated peering of OCUM and WFA was done with a previous OCUM version
a) rename the existing data source to what OCUM expects, if you don't know
b) remove and re-create the WFA integration in OCUM. This will create a new data source with the correct name, then delete the old one.
I really wonder why this is and why it hasn't been changed as basically WFA integration will break (partly) as soon as OCUM is upgraded. I observed this with several versions now.