Use the Always On Database Unhealthy alert in Mini DBA to monitor SQL Server instances and make this condition visible before it becomes a wider database incident.
Mini DBA describes this alert as: Reflects the intersection of the synchronization state of a database that is joined to the availability group on the availability replica and the availability mode of the availability replica (synchronous-commit or asynchronous-commit mode), one of the following values. Not healthy. The synchronization_state of the database is NOT SYNCHRONIZING. Partially healthy. A database on a synchronous-commit availability replica is considered partially healthy if synchronization_state is SYNCHRONIZING. Healthy. A database on an synchronous-commit availability replica is considered healthy if synchronization_state is SYNCHRONIZED, and a database on an asynchronous-commit availability replica is considered healthy if synchronization_state is SYNCHRONIZING. The default evaluation frequency is Minute, so the alert is intended to be close enough to operational reality for live triage.
This alert gives Mini DBA a focused signal for a condition that can affect SQL Server reliability, performance, or operational visibility. It helps the team move from a general problem report to a specific metric, threshold, and investigation path.
Enable it on any monitored SQL Server instance where this condition would require an administrator to investigate or communicate with the application team. Tune noisy development systems after you have a short baseline.
This alert is not primarily driven by a numeric threshold in Mini DBA. Tune the schedule, scope, severity, and notification route so that the alert matches the importance of the instance. Use higher thresholds on batch-heavy, development, or intentionally bursty systems where brief pressure is expected. Use lower thresholds on latency-sensitive production systems, small instances with little headroom, and services with strict recovery or availability commitments.
Open the related Mini DBA page for the instance, compare the current value with the recent baseline, identify whether the change is caused by workload, configuration, host capacity, or maintenance, then record the corrective action in the incident notes.
If the alert creates repeated noise, review whether it is enabled on the correct instances, whether maintenance windows are configured, and whether the minor and major routes match the real business impact. Keep enough history to prove the new setting before changing it broadly.