Why the registry size can cause problems with your SQL 2012 AlwaysOn/Failover Cluster setup

I recently worked on a very interesting issue, where one of the cluster nodes in an AlwaysOn environment became unstable, and the administrators ended up evicting the node from the Windows cluster as an emergency measure. Ideally, since the primary node/replica was no longer available, the Availability Group should have come up on the secondary…