I have setup a standalone synchronous autofailover secondary that is successfully synchronizing with the primary replica which is hosted on a clustered instance. When I manually move the primary SQL instance service to the other cluster node, the AO group service does move with it and comes up as online as it should, but the secondary replica disconnects thus breaking the synchronization. If I try to make it connect after the move, it doesn't connect saying it can't find the instance. If I simply move the instance service back to the first node, synchronization resumes and everything is peachy again.
While I could live with this for the majority of the times I need to failover during monthly patching when the nodes need to reboot for a few minutes, this won't fly in the case of hardware failure of the node that it currently works on and...