Comprehensive data protection for all workloads
Post Reply
Dan2000
Novice
Posts: 7
Liked: 1 time
Joined: Dec 28, 2021 8:42 am
Contact:

Reinstalling VEEAM and upgrading to REFS

Post by Dan2000 »

Our physical VEEAM server is 2012 R2 and NTFS local storage. It does backup copy jobs to a remote site, SOBR offload (Azure) and replicas to another site with ESXi. I backup the Veeam config daily offsite too.

I'd like to rebuild this server as 2019 or 2022 and use REFS for the Repository. Any advice on the best way to do it? Sadly I don't have spare capacity in this location to move the backups (The backup copy jobs only cover critical servers and are sent offsite) Also my bandwidth to Azure isn't good enough to send everything there I don't think. I do have tape backups taken every month, so my plan is to do this as soon as the tape backups finish so I have a recent full offsite/tape.

I guess the steps would be similar to:
1) fresh install of Server 2019/2022
2) Setup the partition for the replica to be REFS
3) Fresh install of Veeam 11a (Same version as we're on now)
4) Restore the latest Veeam config
4a) I'm not sure what would happen now, as the location for the repositories would be reFS now, would I have to manyally edit the repositories and set them up again? This is the bit that scares me
5) Full Active's I'm guessing to take advantage of refs.
5a) The backup copies would get sent to another offsite server then, but what would happen with the old backup chain, anything?

The offsite server is a Windows box also on NTFS, so I need to move that over to reFS too.

Tips and tricks appreciated :)
HannesK
Product Manager
Posts: 14322
Liked: 2890 times
Joined: Sep 01, 2014 11:46 am
Full Name: Hannes Kasparick
Location: Austria
Contact:

Re: Reinstalling VEEAM and upgrading to REFS

Post by HannesK »

Hello,

1) yes 👍
2) for replication metadata, it's irrelevant whether it's NTFS or REFS. But yes, having REFS everywhere makes it consistent. I assume this step also included formatting the data partition for backups (well, I also see that you have no space to move data)
3) yes 👍
4) yes 👍
4a) re-scan should be enough. there will be no backups as the partitions are empty now.
5) yes, that's the easiest way
5a) hmm, I'm not sure whether there is a new server ("another") here now or whether we still talk about the original BCJ target. If it's the same server, then the backup copy job would continue. If it's a new server, then you need to add the new repository in step 4a.

Best regards,
Hannes
Post Reply

Who is online

Users browsing this forum: Google [Bot] and 138 guests