Until now I just used the dashboard to quickly see if any hosts are down or having other troubles manually. Now that I want to dive deeper into the complexity of checkmk I am stuck with configuring an BI aggregation in the checkmk Raw edition.
I hope that one of you already had this problem and is able and willing to help me with this.
Let’s say we have 3 nodes in an aggregation in the rule RULE which I want to handle completely independent from each other.
The 3 nodes are: n1, n2, n3
If n1 or n2 have the status warn, while the other one is ok, the status of RULE should turn to warn.
If n1 or n2 have the status crit, while the other one is ok, the status of RULE should also turn to warn.
If both n1 and n2 have the status warn, the status of RULE should turn to crit.
If n3 has the status warn or crit, the status of RULE should turn to crit even though, n1 and n2 are still ok. n3 is the single point of failure here.
Or in short:
If n1 or n2 = warn -> RULE = warn
If n1 or n2 = crit -> RULE = warn
If n1 + n2 = warn -> RULE = crit
If n3 = warn/crit -> RULE = crit
I do not see a way to build this as I can only select:
- worst of all node states which would turn the RULE to crit even though only n1 or n2 is on crit and the other is ok.
- best of all node states which would result in RULE still to be ok, even though two nodes are on warn or crit.
- count the numbers of nodes in state OK where I can not define that n3 is the most important node and single point of failure. I do not want the RULE to turn to crit when only n1 or n2 are on warn or crit, but if n3 turns to warn this should result in a crit for the RULE.
I hope that was somewhat understandable. English is not my native language.