Why would I be getting "WARNING - SNMP check timeout." in nagios? (For various things like load and disk space checks)
snmpd is running fine on the machine being checked.
Please let me know what additional info I can provide too. Thanks!
1
- Have you run the check command from the command line outside of Nagios? Does it timeout when you do that?Starfish– Starfish2012-03-06 18:58:53 +00:00Commented Mar 6, 2012 at 18:58
Add a comment |
2 Answers
The response did not get back to the manager within the timeout period. Could have been that the host was busy when the snmp request was sent. Could be that it got lost on the way to or from due to network congestion.
How often does it happen ?
If every time, check for firewalls blocking snmp, either on the network or the host.
- It persists for days at a time. I have tried restarting snmpd on the host being monitored, but no luck. It only happens for like one or two different alerts though, the rest go through fine, and I'm assuming SNMP is used for at least a few of those. Why would it work for one kind of alert and not another? (Different alerts on different machines give that error too)cat pants– cat pants2012-03-06 19:56:49 +00:00Commented Mar 6, 2012 at 19:56
- What kind/version of servers ? SNMP or WMI if Windows ?kls– kls2012-03-06 20:18:44 +00:00Commented Mar 6, 2012 at 20:18
SNMP is one of the first services that gets unresponsive when the server is under load. You can raise the timeout-value and you can raise the number of retries to avoid this error.
For big table-traversals you should rather use SNMP v2c with the snmp-get-bulk capability than SNMP v1 with its single snmp-gets (one value at a time, each in in its own UDP packet).
On some systems I even get timeouts if I try to snmpwalk localhost. No problems with snmpbulkwalk...
