Prometheus rule group Resource Health remains Unknown after successful alert evaluation

Question

An enabled Azure Managed Prometheus rule group successfully evaluated its query and fired an alert, but Microsoft.ResourceHealth continued to report the rule group as Unknown with the summary "This Prometheus rule group is currently disabled or in an unknown state."

Evidence:

• The rule and rule group were enabled.
• Evaluation interval: PT1M.
• For period: PT1M.
• The alert fired successfully at 2026-08-03T15:21:02Z.
• AlertsManagement and the Azure portal both showed monitorCondition=Fired.
• GET {ruleGroupResourceId}/providers/Microsoft.ResourceHealth/availabilityStatuses/current?api-version=2025-05-01 continued returning availabilityState=Unknown after the alert fired.
• API version 2023-07-01-preview returned the same state.

Expected: Once an enabled rule group is actively evaluating successfully, Resource Health should report Available, or otherwise distinguish a functioning evaluator from a disabled or unknown evaluator.

Actual: Resource Health reports Unknown despite successful evaluation and a fired alert.

Impact: Low. Alert evaluation works, but Resource Health is a false-negative and cannot be used for automated health verification of Prometheus rule groups. Is this a known limitation or defect?

Answer 1

Hello Logan Rosen,

We have an update from the PG team that this is a known issue and currently our Product team is working on the improvement.

We have an update from the PG team that this is a known issue and currently our Product team is working on the improvement.

Also, I have an update that the ETA of the fix is about 2 months.

Thanks,
Suchitra.

Answer 2

Hello Logan Rosen,

Greetings! Thanks for raising this question in the Q&A forum.

Based on the evidence you've gathered, this looks like a genuine gap between the actual evaluator state and what Resource Health reports for Prometheus rule groups, rather than something you can fix from the rule group configuration itself. The public documentation only defines Unknown as meaning the rule group is disabled or its state cannot be determined, but it does not document any expected delay or lag for the health signal to catch up once a disabled or newly created rule group starts evaluating successfully. Since your rule group is enabled, evaluating on schedule, and has already produced a fired alert with monitorCondition=Fired, and the Unknown state persists across both the GA and preview API versions, this points to the Resource Health signal for this resource provider not being updated correctly rather than an actual problem with your rule group.

A few things worth checking before escalating, since they rule out anything on your side:

Confirm the rule group resource ID matches exactly Resource Health is keyed off the exact resource ID. Double check that the ID used in your GET .../providers/Microsoft.ResourceHealth/availabilityStatuses/current call matches the rule group resource ID exactly as shown in the portal, including casing and the resource group name.

Check for a longer settling window Some Azure Monitor resource types take longer than the typical 5 to 10 minute Resource Health refresh window to reflect a state change, particularly for resource types added to Resource Health more recently. If it has been less than 30 to 60 minutes since the rule group was enabled, it's worth polling again before concluding this is a defect.

Check the rule group's own health state via the data plane, not just Resource Health You can independently confirm evaluation health using the rule-level health status returned by the Prometheus rule group resource itself (visible in the portal's rule group detail view, or via Microsoft.AlertsManagement for the fired alert), to confirm this is purely a Resource Health reporting gap and not a symptom of a subtler issue.

Since you've already ruled out the obvious causes and this behavior does not match documented expectations, this needs to be looked at from the Resource Health / Azure Monitor managed Prometheus backend rather than resolved client-side. Please open a support case under Monitoring > Azure Monitor > Managed Prometheus (or Resource Health as the secondary category) and include:

• The rule group resource ID
• The timestamp the alert fired (2026-08-03T15:21:02Z)
• Both API responses (2025-05-01 and 2023-07-01-preview) showing availabilityState=Unknown
• The AlertsManagement fired alert ID for cross-referencing

This will let the service team check whether the Resource Health signal pipeline for this resource type has a known defect or missing state transition, and file a bug internally if confirmed.

If this answer helps you kindly accept the answer which will help others who have similar questions.

Best Regards,

Jerald Felix.