Some services in Central US seem impaired, although the portal partly works and the status page shows no incident. A basic API health check responds, but other requests aren't working normally. What else should I check?
Is anyone seeing Azure problems in Central US?
Answers
Check the dependencies behind the API, not just the health endpoint. A liveness check can pass while database or configuration requests are failing. Look at request errors and latency for each dependency, especially Cosmos DB and Azure App Configuration.
The reports here point to different symptoms, including Cosmos DB 503s and increased latency. That doesn't establish a single cause, but it gives you somewhere to start.
Reply to this answer
This may be partial rather than a region-wide outage. Some Central US resources appeared unaffected, while other environments had interruptions and later recovered. Check your own resource metrics and error timing before assuming everything in the region is down.
Reply to this answer
If you have a tested failover to another region, this is the kind of situation to evaluate it for. One reported response was to fail over from Central US to East US. Whether that works for you depends on your setup and the dependencies involved.
Reply to this answer
Keep watching the errors even if things start responding again. Some reported interruptions were brief, around 20 to 30 minutes, and recovery times varied. A clean public status page doesn't tell you whether your particular requests are succeeding.
Reply to this answer
Add your answer
Answers are reviewed before publication.
8m+quavrix_371:
A separate readiness check that tests essential dependencies can help distinguish a running process from an application that's actually able to serve requests.