Re: snmpCollect

2001-07-17 04:44:43
Subject: Re: snmpCollect
From: Luc BARNOUIN <luc.barnouin AT thalesatm DOT com>
To: nv-l AT lists.tivoli DOT com
Date: Tue, 17 Jul 2001 10:44:43 +0200
Il s'agit d'un message multivolet au format MIME.
reamd AT nationwide DOT com a écrit :
Hi all,
         I am running AIX 4.3.3 with 6.02.. My snmpCollect is constantly
using 60% of my CPU. I do a snmpstatus.sh and it shows 0 behind in polling.
I currently have 106000 objects in the database maintained by ovwdb....

We had similar problem with our configuration where snmpCollect was using from 30% to 60% of the CPU. When activating the traces, we found it was looping in querying the sysUpTime OID for one particular node. I don't know why, but each time snmpCollect queries an OID on a node, it first get the sysUpTime, and then the requested OID. On this node, we were experimenting a new SNMP agents wich had its own implementation of the sysUpTime (i.e. a different value from standard snmpd). Depending of wich agent was running at snmpCollect start (data collection initialisation) and then during data collection cycle, we get in this situation (I'm sure that Tivoli software team could explain this behavior...) - a stop/restart of snmpCollect fixed the problem until the next time....

So check snmpCollect traces to see wether your are in the same situation...


Attachment: luc.barnouin.vcf
Description: Carte pour Luc BARNOUIN

<Prev in Thread] Current Thread [Next in Thread>