We upgraded from v7.8.5 to v8.1 (HF4) two weeks ago. We started with one, then it progressed to 3 and now expanded to 4 today, Win2K servers that have spiked to and hold at 100% CPU, no memory graphed and all interfaces and volume state changes to "unknown". We also have discovered some (haven't worked the number yet) Cisco routers showing no CPU & memory (No Data for Selected Period) and all interfaces "unknown". When we try to rediscover the servers using SM (ListResources), we get "Interfaces-MIB not supported". When we try to rediscover the routers using SM (ListResources), we get "Routerx is currently down, unreachable, or xxxx is not a valid community string". We have double and triple-checked the SNMP configs are correct. We tried to Unmanage/Manage but no change. With that said, we took a sniffer capture of the problem. In our initial analysis looking at a good polling period before the change, we see the good SNMP Gets/Replys to the Host-Resources-MIB then, for some reason, the device starts responding to Gets with Reply SNMP General Error (5). Soon thereafter, Orion switches to Gets to the Enterprises-UCDavis-MIB (2021). So, our thought is, it's the SNMP Agent on the server so we stopped and restarted the SNMP service...no change. Due to the high visibility of the server we have not been able to reboot it to see if it clears the problem. We'll be opening a ticket soon and move on to looking at the router exchange.
Is anyone else experiencing this behavior?