Today, I learned the hard way that you cannot corruption check a Basic Availability Group's secondary. Given this limitation, how can I know that it is safe to failover a BAG? For all I know, the secondary could be corrupt.
Asked
Active
Viewed 95 times
1 Answers
5
Given this limitation, how can I know that it is safe to failover a BAG? For all I know, the secondary could be corrupt.
Could there be corruption? Yes. Could there be corruption in any database at any given time, even right after running CheckDB? Yes. Seems like a moot point?
Nothing is stopping you from creating a database snapshot and running checkdb against that, if that's what gives you peace of mind.
Otherwise, bad pages should be stored in MSDB on the local instance and the SQL Server Errorlog along with the application and system event logs in Windows should give you an idea if there is any other types of corruption. It's clearly not 100% foolproof but neither is running a CheckDB against a secondary replica, failing over, and then having filesystem corruption happen underneath you.
Sean Gallardy
- 38,135
- 3
- 49
- 91