Desaster Restore when SAN fails

[dfehse]

Hi

We have a Customer that has 3 ESX-Server attached to a iSCSI SAN (4.5TB) and local storge in each ESX-Server (2TB). We replicate all important Server with Veeam replication to the local storage of the ESX-Server. Some every 15min, others like the vCenter Server and the Veeam Server only once a day.

Now my question is: What to do a correctly bring up the VMs on the local storage so that I can make a fail-back as smoth as possible?

I certenly can start the replica of the Veeam Server and vCenter directly from the host where the have been replicated to, but my Veeam DB is not in actual state, there a several replication cycels that runded thar are not in my DB. So I start a fail-over process out of Veeam I think it will take in the better case not the most recent snapshot b'c there are not in DB and it the worser case it will not run at all!?

Does anyone also have this Setup and did a test or at least thought how a Desaster Fail-Over will work?

Thx for any help

Rds Daniel Fehse

[foggy]

Daniel, two simple ways to address your scenario come from the top of my mind:

1. You could set up Veeam B&R responsible for replication to reside on the local host storage - thus you will be able to fail over any time the SAN fails using the Veeam console and fail back any time later if needed.
2. Schedule Veeam B&R configuration backup to run as frequently as your replication jobs run to always have an up-to-date configuration ready to be restored.

Also note that you need to add target hosts to Veeam B&R as standalone to be able to fail over without vCenter server.

Of course you can fail over manually using vSphere Client, however in this case you lose an ability to fail back, re-IP and some other Veeam replication-specific options.