This plugin will cause false alarms in zabbix when the cluster system is busy

[lboclboc]

From time to time we get massive alarms about disks being full. I have found this is because the item that generates the xml output sometimes instead returns:

 <opErrstr>Another transaction is in progress for gluster-normal-px2jenkins-test-dev. Please try again after some time.</opErrstr>

Which I guess is because the gluster is busy performing other commands. Not sure how to solve this, perhaps just wait and retry would be a first step.

[Lelik13a]

You can try min(#3) function in trigger expression. Or use some javascript in item preprocessing. Or insert logic in gluster-monitoring.pl script.
I did not have such problems and I was not looking for a solution.