Hi,
In order to determine what exactly might be causing the clog in the SNMP engine, we need to gather an SNMP stack output while the issue is occurring. To gather this stack you need to do issue the below command:
CR-NMS-2811#show proc cpu sort | inc CPU|PID|SNMP ENG
CPU utilization for five seconds: 2%/0%; one minute: 1%; five minutes: 1%
PID Runtime(ms) Invoked uSecs 5Sec 1Min 5Min TTY Process
303 137304 51825 2649 0.00% 0.01% 0.00% 0 SNMP ENGINE
Then, using the PID you get in return (in my case 303) get an SNMP stack this way:
gin-pv0-har1# show stack 303
Process 153: SNMP ENGINE
Stack segment 0x4A0A9CCC - 0x4A0ACBAC
FP: 0x4A0ACAB8, RA: 0x41473078
FP: 0x4A0ACAF8, RA: 0x4127C8D8
FP: 0x4A0ACB70, RA: 0x4142AA74
FP: 0x4A0ACB88, RA: 0x4142AA60
The above stack with only 4 RA's means the engine is idle. You want to get several of these over 2 or 3 second intervals until you see a longer list of RA's and the list is repeating itself. This means it is stuck there.
I can decode that stack and check exactly where the problem is. You might just want to log a putty with all the outputs you get. I also need the full "show version" of this device to decode.
Looking forward to your reply,
Claudio Gonzalez S
