Failing over CDP replicas due to a host failure

[cro-shi]

Working in my lab with CDP replicas to prep for a DR test and something seems wrong with failing over CDP replicas.

I powered off the source host to simulate a host failure. In Veeam, when choosing 'Failover Now' on the CDP replica, the restore session hangs at "Waiting for replication to stop" and will not continue. Is this normal?

As a test, I disabled the CDP policy, which stayed at 'disabling' for 5-10 minutes, but after that the failover did power on the replica. Is that the normal process? Is disabling the CDP policy required to failover a replica if the source host is not online?

[PetrM]

Hello,

It must work according to this instruction, we don't have a requirement to disable a policy. You may try to reproduce the issue once again, and if the issue persists (it was not just a one-time "glitch"), it will be best to let our support team research this problem.

Thanks!

[cro-shi]

When the source host is up, it works as expected. When the source host is down, the process will hang indefinitely (for me, anyway). I've tried several times with the same results.

Edit: should have added that normal (non-cdp) replicas work when the source host is offline. Clicking 'failover now' will cause Veeam to power on the replica.

[PetrM]

I think that our policy just waits for a specific response from the source host to stop a replication job, otherwise, it cannot start a failover process. It does not get this response because the source host is down. Please let me clarify this behavior with my colleagues from QA before you open a support ticket, I'll update the topic as soon as I have more details.

Thanks!

[PetrM]

@cro-shi looks like it's a known issue that is supposed to be fixed in one of our next upcoming patches. You may contact our support team, they will check your case and if it matches the known issue that I mentioned above, they could probably give you a private fix.

Thanks!