Hi, looking for some assistance with our switch.
We are seeing an intermittent issue where our 8212 becomes intermittently unresponsive, we have noted that when this happens the switch CPU goes from the normal 2-3% to 65-70%
We are really struggling with where to start, any pointers?
Just make sure there are no physical loops and do ensure Spanning-Tree is enabled
If you are running with a very old code, do upgrade it to latest anything above Version K.15.17.0003
Same is fixed in this version if its a "False Reading"
CR_0000151164 The switch occasionally reports CPU utilization of 99%. This is a false reading and does not affect switch performance.
http://h20564.www2.hpe.com/hpsc/doc/public/display?docId=c04715502
Thanks Uma,
we have raised a change to restart the switch as we have noticed the log times are incorrect but the system click is fine.
When we view the CPU, we see:
CORE-SW1# show cpu 300
60/100 busy, for past 300 sec
Task usage for last 8 sec
% CPU | Description
-------+--------------------------
56.8 | Idle
0.2 | Hardware Mgmt
41.8 | System Services
0.2 | LLDP
0.8 | SNMP
0.2 | IP Host/Routing
Does anyone know if there is a way to drill down on the System Services? To see what this is doing?
Hello. In K.15.15.xxxx and greater there is enhanced process tracking available. It will provide output that would allow support to determine what internal task is consuming the CPU. If you are able to load new code and provide the output I can help figure that out. Enable "process-tracking" and then "show cpu process".
If upgrading the code is not feasible at the moment there is a way to get it via the support diagnostic context. But the safer way is to get it via the process tracking facility.