I am trying to test the SQL Server AlwaysOn availability group behavior of my application database. I have created the availability group and able to simulate a failo ver from cluster manager, or from SQL Server studio correctly. The failover will also happen if the server active node was restarted.
As a last test, I restarted the SQL Server's windows service on active node. I was expecting the AG will fail over to the secondary node as soon as the service stops, but it didn't. the primary node lived through the service restart.
when the service was down momentarily, there was no switching of nodes. Does that mean that failover will happen only if the service was down for some threshold? if so, what is that threshold?