Say you have a two mailbox server DAG (mbx1, mbx2) with a single database which is active on mbx1. If the database becomes corrupted, what would you do? This would be my solution:
1. Make database active on mbx2.
2. Suspend replication.
3. Restore mbx1 database and logs from a backup prior to corruption.
4. Copy restored database and logs into original folders on mbx1.
5. Restart replication.
Would you do it differently?