We had switch upgrade, caused outage of failover cluster core resources.
Failover cluster roles on all clusters connected to this switch needed to be manually started to bring the databases back online.
So the SQL Clusters do not appear to auto-recover once connectivity through the production switches is back. This is not the expected behaviour, as all clusters connected to these switches also have connectivity through a secondary Cluster switch, which remained online throughout the change:
Network 1: Cluster, Cluster Use: Cluster Only
Network 2: Cloud, Cluster Use: Cliuster and Client ------- This is the network has swtich upgraded.
I ran a failover cluster validation test, the only warning in network section is "
The RegisterAllProvidersIP property for network name 'Name: SYDWP4_Listener' is set to 1. For the current cluster configuration this value should be set to 0.
" which should be ignored.
Can anyone shed lights on this? what I missed in the configuration?