Reinstall VBR proxy writing to SOBR

[johannesk]

Hi,

We are preparing to do a task that is not something we do on regular basis, reinstall the OS on a proxy server. Thus I thought I get some practical info from those here that have experince.

We have a proxy server writing to repository via SAS. In VBR the proxy serves repository extents that are then configured in Scale Out Backup Repository.

I'm wondering if we need to unconfigure anything in VBR before we go ahead and reinstall the OS and configure it with the same hostname and IP?

Our planned steps are:

-Run SQL backups to the other proxy/repository, Proxy B with application aware disabled. First backup will take some time and transaction log backups will still run on the Proxy A (ToBeReInstalled proxy).
-Disable the SQL backup jobs running on Proxy A.
-Run SQL backup jobs with Application aware on Proxy B. Backup should take a short time since now we have CBT and transaction log backups should kick in soon.
-Proxy A is now passive.

At this stage can I leave all the configuration in VBR regarding Proxy A ? Or should we dismantle the SOBR and remove the configuration for proxy A?

Will we need to run a scan on the SOBR and extents on Proxy A after OS Reinstall (same hostname and IP) before it can be used ?

Any good advice from experienced users highly appreciated.

Regards,
Jóhannes

[rennerstefan]

Hi Johannes,

a very important question, is that machine really just a proxy or is it proxy and also a repository extend on the same machine?

If it’s really only a proxy it’s straight forward as the proxy role only transports data. With that you can just unconfigure everything that uses that proxy, remove it from the vbr console (proxy role + server itself), reinstall and readd to vbr.

Let me know.

Thanks

[johannesk]

This proxy is also a reposotory server. It has huge drives from a Lenovo SAS storage array. Those will be unassigned from the proxy server during OS installation.

Those drive volumes are then configured in VBR server (different server) as repositories and those repositories are in a SOBR. That SOBR only has repository extents from that particilar proxy / repository server.

Regards,
Jóhannes

[rennerstefan]

ok got you.
To be honest I would evacuate the Proxy/Repo Server before performing a new installation of the OS.
Even if you keep Hostname and IP the same it will be a new machine for Veeam (as the Windows OS will have new IDs etc.).
With that you could import the backups but you won't be able to use those to continue running the chain as they will be shown as "imported".
In my eyes the process is:
- change all jobs to run to the other proxy/repo as you already mentioned
- put the repo on the proxy you want to re-install in the seal mode
- proceed with the removal of the repo, proxy and the server itself
- reinstall and readd

Sealing up performance extents gives you the ability to gradually remove data located on these extents by applying a retention policy.
You can use this feature to gracefully stop using some of your performance extents and exclude them from the scale-out backup repository configuration.
When sealing up a performance extent, Veeam Backup & Replication restricts any further data saving to such a sealed extent and allows only read operations such as restore, merge and remove.
All backup jobs that are targeted to a scale-out backup repository with the sealed performance extents that store active backup chains will be forced to create a new active full backup on the next run.
(https://helpcenter.veeam.com/docs/backu ... ml?ver=110)

With that you will make sure that Veeam completely uses the remaining proxy/repo server while you install the new one.

I do recommend that you test the whole process out with a small repo server and 1-2 test VMs so you know 100% how to do it.

I hope that helps.

Thanks

[johannesk]

I was kind of hopin it would be possible to just reinstall the OS with same name and install the Veeam roles. That would be nice.

But how about the 'Map backup' function in the jobs. If we unconfigure the jobs and SOBR. Is the 'Map backup' function not to be able to reconstruct a repository and keep on using it with the Veeam jobs that had already been backing up to that repository disks?

regards,
Jóhannes

[HannesK]

Hello,
what is the reason of introducing server B? I would keep everything as it is and only re-install the server A.

As you keep IP and hostname, you only need to click through the "managed server" wizard again and do a rescan of the repository. That's it.

As fallback mechanism I would keep a Veeam Agent for Windows backup (free edition is good enough) which would allow bare metal restore if anything goes wrong.

Best regards,
Hannes

[johannesk]

The reason for running SQL backups temporarily on Proxy B is to make sure SQL transaction log backups run during the time it takes to reinstall proxy A and configure it back to VBR.

We just did this reinstallation of OS on proxy A, and as you suggested Hannes, it was very easy in VBR. We did no unconfiguration in VBR, just shut down the problem proxy and reinstalled OS with same IP and hostname.

When the proxy server was running all we had to do was to run the properties wizard in the backup infrastructure in VBR and the roles were installed, and the repository kicked right in. Then just a rescan on the SOBR and backups worked fine writing to it.

Thanks for the reassurance Hannes.

regards,
Jóhannes