Monitoring heartbeat health/failover

[gadget]

I've been working with zabbix for more than 2 years now... rock solid and reliable - great software!
Now I'm about to set up an active/passive clustered loadbalancer using heartbeat (linux-ha.org v1) on Debian Linux. I wonder if anyone managed to monitor the current heartbeat state, failover, communication failures etc. with any custom zabbix_agent scripts or logfile analysis. Actually I wonder that has never been a thread in this forum. Maybe people are using heartbeat's alerting scripts - but I want it to integrate in zabbix to have one single monitoring/alerting tool.
Really hope anybody can give me a hint...

Thanks
Michael

[Alexei]

Trigger function nodata() is extremely useful for monitoring of hearbeats. You make create a script which would periodically send a value to ZABBIX server (crontab + zabbix_sender), and the nodata() trigger would detect if there is no data coming from the script within a period of time.

This is not an answer to you question, just a thought...

[bbrendon]

Is this what you're looking for? Here is an annoyingly simple script that might do what you want. You could probably do it as a system.run.

#/bin/bash

# 0 is bad, 1 is good

STAT=`/usr/bin/cl_status rscstatus`

#echo $STAT

if [ "$STAT" = "all" ] || [ "$STAT" = "local" ]; then
  echo 1
else
  echo 0
fi

[gadget]

Is this what you're looking for?

Yepp! This is exactly what I've been looking for! Thanks a lot... I didn't notice there comes a cluster status command with heartbeat. Thought this has to be done manually with some logfile grep-tail-awk voodoo