I'm running 2 VM's, a webserver and a DB, out of the West US center.
After running fine for over a year, twice in the last 24 hrs the webserver VM has come to a halt. It stops serving pages and wont respond on RDP, although any open RDP session stays connected, but totally hung up.
When I restart via the portal, where it shows as "running", it takes something like half an hour to restart, as I make several attempts, and ultimately shut down and start it. The logs show no unusual errors (no errors at all) at the time of the crash, although at earlier times I see some suspicious disk reading errors logged, which I don't know what to make of when it's a presumably virtual disk on a virtual machine.
Is this possibly just due to Azure platform problems, like physical disk crashes or hardware maintenance? The Azure health status page showed no issues and our web app hasn't changed recently, plus the CPU and memory are holding consistently low at < 15% and 37% respectively, although traffic has increased lately to levels we have weathered fine before.
I have a suspicion that something was corrupted when a widespread Azure storage issue disrupted service last month. Without a physical machine to troubleshoot and being buried under a layer of virtuality I feel like I'm flying blind. Any pointers on flushing out a culprit would be much appreciated.