Re-create Replication Jobs to Expedite Planned Failover?

[jonhsykes]

I intend to migrate over 300 VMs in the near future from old to newer Hyper-V clusters. Those VMs have been replicated with VEEAM with large CSV level replication jobs. During testing I discovered that the process of planned failover with large replication jobs wasn't ideal in terms of efficiency/parallelism due to the way VEEAM handles the scenario as written below.

“Limitations for Planned Failover. If you start planned failover for several VMs that are replicated with one replication job, these VMs will be processed one by one, not in parallel. Each planned failover task for each VM is processed as a separate replica job session. If a backup proxy is not available and the session has to wait for resources, job sessions for other VMs in the same task cannot be started before the current session is finished.”

http://www.gabesvirtualworld.com/datace ... owershell/

So my question is: can I destroy the large replication jobs that are currently created and that have replicas tied to them, re-create VM level replication jobs, and most importantly retain the existing replicas in order to expedite the planned failover of 300 VMs?

[DGrinev]

Hi,

Yes, you can split one replication job into several jobs and map them to existing replica VMs.
Also, I'd suggest you to review Failover Plan article as it provides the ability to failover several VMs simultaneously, though with a longer downtime.

Thanks!

[jonhsykes]

That's great news on the replica mapping!

As for the failover plan, it doesn't provide incremental replication. However, I guess I could turn off the source VMs and execute the replication jobs one last time prior to executing the failover plan.

[foggy]

Correct, you can perform the replication jobs run prior to actual failover, then turn the VMs off and perform one more sync of the last-minute changes.