IPMI Sensors and alerts

Hello !

I’m trying to get warned when a WARN or CRITICAL level of temperature is reached by a machine.

I tried 2 ways :

1/ Creating an [IPMI Sensor Summary] , but I get the number of Sensors OK as status, the Temperature is used as Service Perf-O-Meter/Metrics/Graphs . But I cannot manage to add any WARN/CRIT rule based on the Temperature ! It seems to concern only the number of sensors…

2/ Creating an [IPMI Sensors] with a rule to get one service discovered for each sensor. It seemed to be the good way : the Temperature values are actualy shown, BUT then : the value can’t be converted as a double float because a string "status OK " is shown as a prefix to the temperature value… => no graphs, and no WARN/CRIT level can be set !

Is there anyone who managed with these Temperatures (or whatever similar IPMI sensor) and associated alerts ?!?

Thank you

check_mk release 2.0.0p4 (CRE)
host temperature to check under linux

Hello,

Basically IPMI is an event driven system and the thresholds are defined by the manufacturer on the sensors in the BMC. We don’t change them to be compliant with the manufacturer.
Nevertheless I found in 1.6 a rule “IPMI sensors” where you can add thresholds with "Set lower and upper levels for numerical sensors.
I am afraid that if you want to alert in a higher threshold as set by manufacturer the manufacturer threshold will win and send an WARNING/CRITICAL event which then will be alerted.

This topic was automatically closed 365 days after the last reply. New replies are no longer allowed. Contact an admin if you think this should be re-opened.