Q. My Azure VMs continually restart, what can I check?
A. If only a single VM restarts then you would check the OS logs within the VM itself, if its an isolated restart then it could simply be a failure occurred in Azure and service healing restarted the VM however if you have multiple VMs continually restarting then check if they are all using the same storage account. Remember that the IOPS limit for a storage account is 20,000 IOPS and the limit for each individual disk is 500 IOPS. Therefore the most disks that should ever be placed in a storage account is 40. If you have more than 40 disks in a storage account all being used then they will not be able to get their typical 500 IOPS and obviously the more disks the less IOPS they will get. If an OS disk gets a small enough number of IOPS then the OS will be starved of IOPS and backlog IOs until it eventually will just restart. If you have more than 40 then move VHDs to other storage accounts. I wrote a way to check this previously at http://windowsitpro.com/azure/azure-storage-iops and at http://windowsitpro.com/azure/throttling-against-azure-storage.