Identified VM's losing their information

[RobSmith]

I have many VM's, running on esx 2.5.2 that I am able to identify. They look ok for a while, then seem to lose their indentification info. They revert back to being "unknown", etc.

The protocol settings have the correct username/password, string and snmp settings on the vm's are correct. Any ideas what I'm doing wrong?

[Rob Buxton]

Have you changed from the default SNMP community names?
I've seen a "feature" in HPSIM where it reverts back to private and public as the community names in the global settings.

[RobSmith]

I dont think thats the case here. The only ones being dropped are the VM's. All my physical machines are still id'ed just fine. I'll double check...
Hmmm...my read string was still there, but there were anout 5 additional read strings that all looked like a slight variation of my normal string. Except I didnt put them there.
I removed them and will monitor for changes.

[RobSmith]

My workaround on this is as follows. I have set up a scheduled task to restart the Pegasus WMI Mapper services on the CMS once a day. This seems to have cleared up the problem for the most part. May need to do it more frequently, if they are still doing it.

Basically, as long as the WMI Mapper is working when device identification takes place, everything is happy. I might try to create a task in SIM to do this prior to any Auto Identification tasks.