This has been a long-standing issue that I'm finally getting around to troubleshooting. We have recurring issues monitoring our Windows servers where SNMP (native Windows SNMP service) responds for a while and then simply stops responding. The Windows services are running, but Orion NPM's queries to the servers apparently are not answered. Restarting the Windows SNMP service resolves the issue temporarily, but after a few days (time varies), it stops responding again.
At this time, I have roughly half of my 95 Windows nodes flashing with interfaces in an "unknown" state. And actually, everything dependent on SNMP (CPU, memory, disks, interfaces, etc) are "unknown", NPM just doesn't state that (for some reason, it thinks it is still gathering data).
We have seen this issue on Windows 2003, 2003 R2, 2008, and 2008 R2. Most of our servers are now 2008 R2. From a quick glance, I see 2008 and 2008 R2 in "unknown" states, while an equal number are polling fine.
We are running NPM 10.1.2 (SolarWinds Orion Core 2011.1.0, IPSLAMGR 3.5.1, NPM 10.1.2, IVIM 1.1.0). Anyone else out there seeing this? SW Staff: any ideas for troubleshooting?
Thanks,
Chris Gurley
www.bctechnet.com