HPE Community read-only access December 15, 2018
This is a maintenance upgrade. You will be able to read articles and posts, but not post or reply.
Hours:
Dec 15, 4:00 am to 10:00 am UTC
Dec 14, 10:00 pm CST to Dec 15, 4:00 am CST
Dec 14, 8:00 pm PST to Dec 15, 2:00 am PST
Aruba & ProVision-based
cancel
Showing results for 
Search instead for 
Did you mean: 

Procurve with CPU flapping from 90 to 99% CPU usage

 
JamelB
Occasional Contributor

Procurve with CPU flapping from 90 to 99% CPU usage

Hello,

I have 2 Procurve 2810-48G that are flapping from 90 to 99% of CPU usage all the day.
I tried to determine from which port the issue is, but I didn't find the appropriate CLI

Could you help me on this ? DO I need to install a monitor software to do this kind of monitoring ?

Thank you if you can help.

BR

Jamel

7 REPLIES
parnassus
Honored Contributor

Re: Procurve with CPU flapping from 90 to 99% CPU usage

Network topology (are those Switches - J9022A - connected through a single port or a LAG?) and Software versions (N.11.xx, latest is N.11.74) of both Switches would be useful...

Do you suspect some sort of Network Loop somewhere?

What's about checking what Switches' Logs (CLI) or Alert Logs (Web GUI -> Status -> Overview or Web GUI -> Diagnostic) do report?

 

JamelB
Occasional Contributor

Re: Procurve with CPU flapping from 90 to 99% CPU usage

Hello,

Thank you for your answer.

they are connected through a single port (no LAG)
Software revision N.11.73
Spanning-tree enable on all switches in the infrastructure.

Show logs :

03:29:37 tftp: SENT error:1, msg: File not found
03:29:37 tftp: SENT error:1, msg: File not found
03:33:47 snmp: SNMP Security access violation from 10.0.1.151
03:33:47 snmp: SNMP Security access violation from 10.0.1.151

Not sure that 10.0.1.151 is the reason of this issue but I'm investigating on this.

Thank you

Jamel

parnassus
Honored Contributor

Re: Procurve with CPU flapping from 90 to 99% CPU usage

You need to check also both Switch configurations (show config) in order to understand exactly if you have something related to SNMP configured in your Switches (I presume your Switches were both setup with Static IP addressing and they don't use - and offer - any DHCP service from/to LAN's hosts).

Interesting threads about your strange messages:

  • snmp: SNMP Security access violation from 10.0.1.151 (read this).
  • tftp: SENT error:1, msg: File not found (read this).

In both cases you probably have something running on a LAN host that is asking for access to your Switch (Access Violation error) and, concurrently, is trying to retrieve a file that don't exist in your Switch (File Not Found error).

Vince-Whirlwind
Honored Contributor

Re: Procurve with CPU flapping from 90 to 99% CPU usage

"flapping" means a port is going up and down, so I suspect nothing is "flapping".

 When you say "all day", do you mean that the minimum CPU% is 90, all day?

What happens if you disconnect the switch from the network entirely?
What happens if you disconnect the switch from the network and then reboot it? Does the CPU% go up to 90 immediately?

parnassus
Honored Contributor

Re: Procurve with CPU flapping from 90 to 99% CPU usage

I think the OP - with that "flapping" - wanted to say that CPU load ranged between 90% and 99%...but I could be wrong.

JamelB
Occasional Contributor

Re: Procurve with CPU flapping from 90 to 99% CPU usage

Hello

after investigation, it was 1 old nagios server which generate the cpu loads all the day.

After shuting down this server, everything is ok now. see attachment

thank you all for your help.

Jamel

16again
Respected Contributor

Re: Procurve with CPU flapping from 90 to 99% CPU usage

Judging from screenshot, CPU load seems still way too high to me.  Note the switch has hardware for switching packets, so the CPU has hardly tasks to spent cycles on, its load should almost be idle.