Hello. I have SQL Server 2008 R2 running on a 2-node Win 2012 cluster. The SQL role, and all resources are running on the active "NODE B". The cluster uses shared NetApp Luns for data, logs, etc.
I recently applied Windows updates to inactive NODE A. I then tried to Pause NODE B and drain roles over to A. SQL role failed to start on NODE A, so I resumed NODE B and failed back SQL role to NODE B.
System event logs on NODE A indicate a logon failure. I possibly forgot to change the SQL Server service account using Configuration Manager on NODE A during our last password change outage.
As a separate issue, the transaction log for one of my databases on the active node NODE B had grown immense - 600 GB!! - due to our storage engineer not including it in the tran log backup policy. When I failed back to the original active NODE B, I noticed that database was in Recovering mode. Not sure if it was ALREADY in Recovering mode before I tried to failover to NODE A, or AS A RESULT OF the attempted failover.
So, two questions:
Am I correct in assuming that service account passwords must be changed in Configuration Manager separately on both cluster nodes? I'm guessing yes.
Assuming the database with 600 GB log was already in Recovering mode, what is the impact on cluster node failover when database is in that state? Assuming it was NOT already in recovering mode, could failover have caused it to go into recovering mode due to size of the tran log?
Thanks.