A couple of times this past week my Service Fabric cluster moved all the applications from one node, I assume during some kind of Azure maintenance operation (VM did not res