Hi,
One of our app service plans have been moved around to different instances during the last couple of days, resulting in performance issues and weird behaviour to some apps in the plan. The plan itself is not scaled out so it should use just one instance. But using the CPU drilldown there are different instances in the dropdown and all of the apps seem to have been moved between them several times the last days.
This also have had the very weird consequence that one of our web services is not reachable by a specific external source, but reachable by others. Something that we are investigating now.
But the main question here is, why are the apps moving around on different instances when there should be only one instance on the plan, and why is this affecting the performance?
I can provide you with specific details about which app service plan this regards in a private conversation if you wish.
Regards,
Markus