Hi Paessler.
We have a really bizarrely made custom script based cluster of some disks going on. Aside from this being terrifying from a support standpoint :P it presents an annoying monitoring issue. Our strategy is to just monitor disk on both cluster members and then when one go into alarm, we check the other to see if they failedover properly and then acknowledge the down ones and repeat the process when they fail back.
is there any way to make it so sensors acknowledge themselves if another sensor comes up and vice versa?
cheers Sam
Article Comments
Hi Erhard,
The scripting option has been exhausted and despite it having never yet failed, the powers that be dont trust it enough to get rid of disk checks and just trust the script :(
The notification trigger an acknowledge is *exactly* what i was after!!! Thanks for saving me having to figure out that URL with fiddler.
cheers Sam
Jan, 2019 - Permalink
Hi Samuel,
It is possible to set a sensor to "Down (Acknowledged)" with an HTTP action as described here. I have no idea though how to make this really work properly. I mean the http action can acknowledge one sensor, you most likely might need to acknowledge several sensors. The state for the trigger must be "up", such triggers can be defined meanwhile, but I'm not sure how well this plays out all in all. Since this thing is already quite bizarre and complex as you said, well, maybe using a script as notification could spice things more up, I mean what's a script more or less then, doesn't matter :D
I mean with a script notification you could perform several checks for conditions and make the script resume or pause particular sensors as needed....theoretically....
Kind regards,
Erhard
Jan, 2019 - Permalink