<last update Feb 4-2015, originally published Jan 21-2015>
we would like to inform you of a problem that we have discovered.
When you apply SQL Server 2012 Service Pack 2 Cumulative Update 3 or SQL Server 2012 Service Pack 2 Cumulative Update 4 or SQL Server 2014 Cumulative Update 5 on instances that have databases participating to AlwaysOn Availability Groups (aka AG), the AG might be reported as NOT SYNCHRONIZING. When this happens, as you query sys.dm_exec_requests, you may find that there is intermittent lock blocking between users sessions and a session whose command is reported as ‘DB_STARTUP.’ You may also observe blocking between CHECKPOINT and DB_STARTUP.
This KB article gives some additional technical insights.
This is not a systematic problem. You may be able to apply these Cumulative Updates on an AlwaysOn configuration without hitting this problem. If you have already applied these cumulative updates and did not notice this problem its means that your system is not affected and can discard this message.
If you apply these cumulative updates in the future and experience this problem, the recommended approach at this point is to:
- Disable automatic failover if it is activated
- Restart the SQL Server instance that is acting as Primary of the availability groups
- Re-enable automatic failover if it was activated.
We are aware of the implications of this issue and are sorry for the inconvenience this causes to you.
Hotfixes for both SQL Server 2014 and SQL Server 2012 Service Pack 2 are available here.
SQL Server 2012 RTM or Service Pack 1 are not affected by this problem.
The current plan is to roll up these hotfixes respectively into SQL Server 2012 SP2 CU#5 (should ship by end of March 2015) and into SQL Server 2014 RTM CU#6 (should ship by end of February 2015)
Senior Program Manager
SQL Server Product Team