On Thursday 04 November 2010 12:38:00 Timo Schoeler wrote: > Hi list, > > I'm running heartbeat 2.1.4 on top of a CentOS 5.5 x86_64 cluster. Those > machines provide a bunch of virtual IPs, managed by ldirectord. > > It's not that many VIPs (36 at the moment), but after adding the last > ones I had to turn off net-snmpd on this machine because it started to > use almost 100% of the CPU. However, I do not collect information on the > VIPs, just the ethN interfaces of the host. > > Without snmpd running, the machine has a load of about 2.5 (most of it > due to heartbeat's monitoring), with snmpd running it is >5. > > Has anyone come across this phenomenon and if so: Is there a solution? > > Thanks, > > Timo
Hi, I have seen high CPU load caused by net-snmp several times. There are also some bugs related to such problems. See: http://sourceforge.net/tracker/?atid=112694&group_id=12694 What version of net-snmp do you have? Did you try to use one of hte lated versions of net-snmp? Greetings, -- Dr. Michael Schwartzkopff Guardinistr. 63 81375 München Tel: (0163) 172 50 98 Fax: (089) 620 304 13
signature.asc
Description: This is a digitally signed message part.
_______________________________________________ Linux-HA mailing list [email protected] http://lists.linux-ha.org/mailman/listinfo/linux-ha See also: http://linux-ha.org/ReportingProblems
