Hi community,
I have network of 4 x SG500X. One 48-port modell is the middle of the star-topology, the other 3 switches are connected via a 10GB/s connection (fiber) to the center switch. No stacking.
One of the 3 "satelite" switches is showing nearly permanent 100% cpu load - the webinterface and also ssh-cli is very slow. After about 30-40sec the cpu load drops down to 4-5% - just a few seconds later it goes up to 100% again. The cpu input rate is always at about 15-20pps:
<-- snip -->
switchac9699#show cpu utilization
CPU utilization service is on.
CPU utilization
---------------
five seconds: 100%; one minute: 75%; five minutes: 78%
switchac9699#show cpu input rate
Input Rate to CPU is 17 pps
<-- snap -->
This seems to happen regardless of connected clients. It also seems to work/switch fine - but we have complaints of the users that a few times within a week all connections are horribly slow. I don't know of this is related to cpu load (we found the cpu load problem while investigating the slowness problem). Is there a way to find out what is causing this cpu load?
The other switches don't show this behaviour...
Thanks a lot for any thoughts and ideas.
Best regards,
Martin
