Replication Questions - Moving away from SRM

[pmichelli]

SRM has become too costly to maintain since we lost our educational and non profit discounts. I am reading the documentation for Veeam Replication and for now have one large question to answer...

In SRM, we installed replication appliances and SRM on both vCenter sites (Prod and DR). Should something happen and prod goes totally offline, I could login to DR SRM and initiate failover process.

How does Veeam handle this with a single server instance ? If we lost prod, would I not be totally screwed and unable to failover?

Here is what I have in my head but I would like some guidance from Veeam...

Build a Veeam server in DR. Add the license file to it but don't configure any jobs. If a real DR was to occur and I lose prod, I can import the last config backup into the DR Veeam, re-scan and I *should* hopefully be able to trigger failover and when I fix prod, reverse the protection and fail the VMs back, then take the config backup from DR and reload it into prod and be back to where we were.

DR and Prod share a stretched L2 network so I don't need to re-ip anything, just need to change the destination network name (same vLans exist but are named different)
Also, how does replication handle Domain Controllers? SRM for the longest time did not support them until more recent releases and it was seamless, nothing special to do on my end.

Thanks,

[tyler.jurgens]

The general best practice is to install Veeam in the DR site and have that handle all jobs (using applicable proxies as required to limit traffic between sites to only replicas/backup copies).

Another way of doing it is to keep configuration backups on the DR Veeam and be able to restore that as required if the production Veeam goes down.

[pmichelli]

One more question. I am setting up a POC and when it asks me for the repository to store the metadata, it is not letting me select any of the Data Domain mounts. It only gives me the local repo that Veeam self created on its C drive. Does this not support the DD to store the metadata? I would rather not keep it on the Veeam server itself. Can't see anything in the docs to explain why I can't map to it

[eblack]

"You cannot store VM replica metadata on deduplicating storage appliances. During replication jobs, Veeam Backup & Replication frequently reads and writes small portions of metadata from/to the backup repository. Frequent access to metadata causes low performance of deduplicating storage appliances, which may result in low performance of replication jobs.
You cannot store replica metadata in a scale-out backup repository."

As long as you have access to restore config DB backups you can recover the VBR.