It would be great to have service dependencies for Check_MK, just like the parent/child relationships.
For example:
If a service goes into critical state, specific host doesn’t alarm - practical example:
If a interface or a VPN Tunnel goes into Critical state the server(s) attached to it won’t alarm.
I believe, like in wato we’ve the ability to set a parent host, we could also set a service it would really help us out in not having some noisy notifications.
I can relate. We have a lot of similar group of machines that have service dependencies. In the 1.6 version we found that it was a lot of click work to setup en maintain (when VM are added or deleted).
Even grouping of service would have us, instead of getting multi notification due dependencies services.
As an example: You have a Database running on a central Database Server and multiple SQL Queries running on the Application Server which alert to different Application Owner.
When one of the Queries reports a Problem, you want to alert one of the responsible Application Owner. But if the Database has a Problem and therefore all Queries start to fail you only want to alert the root case to the DBA Team and not all Queries Failures to all the Application Owner. With a service dependency rule you could easy configure all the depending Services. In case that the Master Server fails all the Child Services could go to stale state instead alerting all Owner of the depending Service.
I don’t know if you can do that so easy with BI Rules. The BI is meant to aggregate several service states into an overall State and highlighting the Impact on the Business Services.
