Unexpected CPU spikes and average on vSZ-H

  • 1
  • Question
  • Updated 7 months ago
Dear all,
Problem: Since yesterday afternoon, we've been getting a lot of monitoring alerts because of the CPU util of our vSZ. It's been peaking a lot around 100%, but also the average is way up whereas it used to be between 11 and 17%. The screenshots below are by the vSZ itself, then CPU SNMP stats in our monitoring system and the third one of our VmWare SNMP, which shows that the vSZ utilizes extra CPU cycles (over 100%) available to the vm.
show cpuinfo 
p.p1 {margin: 0.0px 0.0px 0.0px 0.0px; font: 11.0px Menlo; color: #000000; background-color: #ffffff} span.s1 {font-variant-ligatures: no-common-ligatures} span.Apple-tab-span {white-space:pre}
processor : 0
model name : Intel(R) Xeon(R) CPU           X3430  @ 2.40GHz
processor : 1
model name : Intel(R) Xeon(R) CPU           X3430  @ 2.40GHz
Cpu(s):  7.8%us,  3.6%sy,  0.0%ni, 88.4%id,  0.1%wa,  0.0%hi,  0.1%si,  0.0%st

Impact: slower response and a lot of alerting. Haven't noticed major performance hits in terms of managed wireless networks.

What changed?: We've added two R500's for our lab set-up to do some testing with roaming over several different WAN connections.
AP's have all been provisioned and received the config. So I don't understand what the vSZ is still doing.

Anyone got any idea where to start looking?

Kind regards.



Photo of Axel Robbe

Axel Robbe

  • 19 Posts
  • 4 Reply Likes

Posted 7 months ago

  • 1
Photo of Axel Robbe

Axel Robbe

  • 19 Posts
  • 4 Reply Likes
After applying the 3.4.1 patch (coming from 3.2), the CPU has gone down averaging around 50%. The question still stands. There are 18 APs connected, 17 of which have been added to a zone and configured. This should still not raise the CPU from 11-16% to about 50% or higher on a 2 CPU with 14 GB machine.

Any guesses what this is or any idea how to get into linux core so I can take a look at the underlying processes?