Question

You manage a three-node Always On availability group on Azure SQL VMs across a single region. The configuration uses Node and File Share Majority quorum with the file share witness hosted on the primary node's local disk. During a planned reboot of the primary node for OS patching, the availability group unexpectedly goes offline and no automatic failover occurs to either secondary replica. What is the root cause?

Accepted Answer

The file share witness is hosted on the primary node, so rebooting it caused simultaneous loss of both a node vote and the witness vote, dropping the cluster below quorum majority

Answer

Automatic failover requires synchronous-commit availability mode, which cannot be configured across three nodes in Azure

Answer

Azure SQL VMs do not support automatic failover for availability groups without an Azure Load Balancer configured as the listener

Answer

Node and File Share Majority quorum is deprecated and must be replaced with a disk witness for availability groups on Azure VMs