When the simplest 2 node Availability group (w/ File share witness vote) is set-up, both the cluster resources and the availability group are running on the same node. On an auto-failover event, these both move together to the secondary node. On a manual fail-over of the SQL Availability group, only the availability group will move to the secondary node. The cluster resources remain on node 1 which formerly housed the primary before the said manual fail-over. I believe that this is by design but is there any reason to maintain having the fail-over cluster and availability group resources running on the same node? From tests, it seems that stopping the cluster services on the primary node via "services" provides for this where both the cluster resources and the AG will both fail-over to the second node.
Thanks.