My Windows 2008 R2x64 server is running a SA126.96.36.19927 database server in High Availability.
This server is responsible for 3 different databases.
Sometimes, a specific primary database freezes and no one can connect to it. The other 2 databases work as usual!
If I connect to the mirror database, it works in read-only mode as expected fo all 3 databases.
The only thing that I can do to recover the primary database activity is SHUTTING DOWN the MIRROR database (just the "problematic database" is enough).
It looks to me that the primary database keeps waiting some kind of response from mirror database and this response never happens.
The only thing I have noticed is that the the Console Log File stops logging the CHECKPOINT event for the frozen database.
This problem is happening on Production and Development environments. The Development database is NOT in use by any application.
My development database is frozen at this moment.
I´d like to know if there is any command that I can run against the Mirror database to discover the root cause of the freeze.
The connection timeout between Primary and Mirror databases is set to 1 second.
asked 08 Aug '11, 16:25
We need to know what the threads in the hung servers are doing to be able to resolve this. I recommend you open a technical support case. We probably need full minidumps for the primary and ideally also for the mirror to determine what they are doing. You can use visual studio to create minidumps, or if you don't use visual studio, I have used the sysinternals procdump tool in the past (-ma creates a full dump).
answered 09 Aug '11, 09:20