With the abundance of KPI at our finger tips from the SQL instance and SQL database level we have been granted customizable thresholds for warning and critical state. Sadly we were not given the same control regarding the SQL Agent Job component of the AppInsight template. Any failed SQL job will result in the instance being flagged in critical state. In an ideal situation that may be considered best practice, any job fails, remediation is required. Welcome to reality. For us to continue progress with using AppInsight for SQL to handle SQL event management we need the ability to identify what events are of concern, set thresholds and then apply an alert rule against the instance state. The only reason an instance state should fall into warning or critical state should be events of concern to us have occurred, but continued to collect everything else for reporting and adjustments to event management over time.
I suggest either 1) the ability to disable SQL agent job monitoring per instance or 2) the ability to quiet the effect on SQL Instance state.
