Hopefully one of the last scenarios in my lab testing HA setups and failovers! I can't seem to figure this part out, however. The following is just a simple overview:
- 4x servers in total (2 mbx and 2 CAS/HUB)
- 2x MBX servers in a DAG
- 1x FSW on a HUB transport
- Same AD site
- DAC is enabled
If the FSW server and one of the MBX servers is turned off, the DAG goes offline as expected due to no quorum. All the documentation online that I can find refers to how to fail over between AD sites, but this is one single site. If it would be a long time before the broken servers come online again and you have one MBX server and one Hub left, can you do the following?
- Move the FSW to the other hub transport (which is set as the alternate FSW)
- Bring the DAG online again with the remaining mailbox server using the alternate FSW the same way as if one node had failed?
I tried the following:
- Stop-DatabaseAvailabilityGroup DAG01 -mailboxserver MBX01
- On MBX02 tried to open the DAG properties but the DAG is offline so I cannot edit anything or change the FSW to use the alternate one
- Restore-DatabaseAvailabilityGroup DAG01 fails because MBX02 is still online
I cannot see a way to do this - is it even possible before I waste any more time? Maybe there is no need to run the stop command first? All I want to do is get the DAG online without having to turn on the old FSW or other MBX server. Glad I'm learning all this in a lab!!