Wrong values for snmp CPU usage query

[Jukka Laaksola]

Hi!

Two of us Proliant ML350 G4p has started to return very high cpu usage percents with snmp. Usually they are 100%. We are querying snmp MIB 1.3.6.1.4.1.232.11.2.3.1.1.2 with Nagios. Servers are using Debian GNU/Linux etch and hpasm was version 7.6.0. We upgraded it to 7.8.0 but that does not help.

top shows that load averange is around 0.01 on both servers and cpu usage percents in top are quite near zero.

The problem started suddenly a week or two ago and after a couple of days second server started to do the same thing. All other Proliants reports correct cpu usage values.

We tried several times restart hpasm and snmpd but nothing seems to help. Probably booting will help but we want to know the reason for the problem first.

I found question about similar problems from last year http://forums11.itrc.hp.com/service/forums/bizsupport/questionanswer.do?threadId=1195781
In our case the first server has 1 HT-cpu and the second has 2 HT-cpus. SNMP shows high percents for all cores.

[Jukka Laaksola]

Is there anyone how knows something about these wrong snmp query results?