annuler
Affichage des résultats de 
Rechercher plutôt 
Vouliez-vous dire : 
cancel
486
Visites
0
Compliment
4
Réponses

High CPU usage after upgrading to 15.2(7)E9

Hello guys,

 

I upgraded my 8 WS-C2960X-48FPD-L to the 15.2(7)E9 and since then, all my switches are experiencing high CPU usage from time to time

abdellahhamdani_0-1719233485793.png

 

with sh processes cpu sorted, there is not much important information (as the peak is happening from time to time and cannot expect when it will happen exactly.)

note that ths screen I provided is from Today (monday) which mean the 72h graph represent the last 3 days, and over the weekend, we don't have people working, I'm suspecting that the image is not stable.

someone has any ideas ?

 

Thank you.

 

4 RÉPONSES 4

shambhu.kumar
Level 1
Level 1

Could you check logs and this utilization may caused by “Hulc LED Process” process.

C2960XF#sh processes cpu SORted
CPU utilization for five seconds: 42%/0%; one minute: 43%; five minutes: 45%
PID Runtime(ms) Invoked uSecs 5Sec 1Min 5Min TTY Process
179 17595601 1425480 12343 22.80% 24.37% 23.90% 0 Hulc LED Process <=========

This process responsible for different things:
- Check Link status on every port
- If the switch supports POE, it checks to see if there is a Power Device (PD) detected
- Check the status of the transceiver
- Update Fan status
- Set Main LED and ports LEDs
- Update both Power Supplies and RPS
- Check on system temperature status

This behavior described in CSCtg86211 and is expected for catalyst 2960x switch models. This utilization does not affect client traffic and does not influence on services and rest protocols of the switch.

Hello Kumar,

First of all thank you for your feedabck, I was able to find the issue yesterday, but here is the sh proc cpu sorted as you asked

abdellahhamdani_0-1719320556098.png

After I disabled the SNMP the CPU went down

abdellahhamdani_1-1719320661029.png

I also seen the same result on another switch which is on a different IOS, which can mean that it's not IOS related.

 

Try use trap instead of polling snmp

And see

MHM

The way the monitoring server is configured in my company + the flows  doesn't allow trap, we are only supporting polling SNMP.

Thanks.

AH