Moving Scale Out repo. from one server to other

[dpatel]

Hi,

I need support to move Scale-Out repository from one server to other server.

Our current server of Scale-Out repository is very low configuration and we are replacing this server with new higher configuration server. But, we don’t have any extra storage so, we need to use same storage for new server. Scale-Out repo. used data size is almost 40TB.

Please share procedure for how to move scale out repo. We are using VBR 9.5.


Thanks

[Shestakov]

Hi,
I`m not sure I`ve got your plan. Scale-out backup repository is a logical entity. Do you want to exclude one of SOBR extents and add another instead?
In this case you need to add the new repository as an extent of SOBR first and evacuate unneeded extent.
Thanks!

[dpatel]

Hi,

Thanks for reply.

No, We don't want to add or remove extents. We have SOBR with 4 extents connect thru FC SAN storage on physical server and this server act as proxy and WAN accelerator also. Server configuration is very low and we are not getting performance as expected from this server. Also we are unable to upgrade Memory and Processor of server.

So, we are planning to provision new server with higher config. and want to move our existing SOBR all 4 extents to this new server. Due high volume of data we don't have space to create new SOBR and move existing data on new server. We want to use old extents only on new server.

Thanks,
Daxesh

[Mike Resseler]

Daxesh,

Since you are going to move basically everything to a new server (I assume that you will want to recreate the jobs and config and everything on the new server) I think it would be best to look at this KB https://www.veeam.com/kb1889

This will give you guidance on how to move everything to your new server and decommission the old one.

[dpatel]

Hi Mike,

No, we are not moving everything to new server. Please find below our setup details.

- One VBR management server with Veeam DB on VM
- Two Physical servers with Proxy, SOBR & WAN accelerator role
- Two SOBR distributed on above two servers. (1 x SOBR with total 8 extents equally distributed on both server and 1 x SOBR with 4 extents equally distributed on both server)

We are not concern about moving Proxy & WAN role. We want to use same SOBR on new server without reconfigure backup job or loosing any data.

Thanks,
Daxesh

[veremin]

Potentially the following approach might answer your requirements:

- Stop all jobs pointed to the old SOBR
- Connect SAN to the new server
- Re-create all repositories on the new server
- Create a new SOBR, adding previously re-created repositories as its extents
- For all stopped jobs change target repository from the old SOBR to the new one
- Map backups

Thanks.

[pkelly_sts]

At the moment we have two instances of VBR and I'm almost at the stage where I can consolidate them down to one so I'm currently planning the migration of a backup that's currently being managed by the other VBR (which will remain as a repository/tape server only).

However, I'm not 100% sure how best to import the contents of a SoBR to import the existing backup/RPs. With a simple repo you just add the repo as normal & allow Veeam to scan it & import the backups, but if importing a SoBR containing two simple repos, should I add the repos, say no to scanning them for existing backups, then create the new SoBR and scan at that point?

Any tips appreciated...

[PTide]

Hi,

Please take the steps Vladimir's suggested.

Thank you

[pkelly_sts]

Thanks.

[pkelly_sts]

I'm currently in the process of completing my move of "controlling" resources (physicals all stay the same I'm just moving management of backups to another B&R server but whilst I'm waiting for some licensing issues to be resolved before I can complete the move I was wondering if a job that contains SQL TL backups can be picked up & continued by the other server (assuming the repository & backup files have been imported & replacement jobs have been mapped successfully to the imported backups etc.

In other words I guess the question is, if migrating a SQL backup job with TL processing to another backup server (with everything else intact) does anything further need to be done?

[foggy]

Jobs should continue normally after performing the actions you've described.

[pkelly_sts]

many thansk. Will report back when completed anyway...

[pkelly_sts]

Happy to report that it worked perfectly as discussed. Basically:
1) Add simple repo 1, don't re-scan
2) Add simple repo 2, don't re-scan
3) Create new SoBR adding the two simple repos added earlier, THEN rescan.
4) Map backups as normal & everything worked fine.