Restoring from VBK backup of VBO server

[HangTen]

We have a situation where we are backing up Office 365 with a 35 day retention on the VBO server. This server is also backed up by a separate Veeam Backup on another server where there is more storage and it has GFS retention. Can anyone give me some tips on how to restore the data from the Veeam Backup of the VBO information? I see that there are repository.adb files in numbered folders for Exchange Online backups. Would I have to mount those one by one in an Explorer in order to restore? Is there any way to know what data is in each numbered folder (eg 1970 or 2019, etc)?

[Gostev]

If I remember correctly, you just right-click the required backup under Backups node in Veeam Backup & Replication, and choose the required application item-level restore option. There's no need to mess with individual ADB files for sure. That is assuming you have application-aware processing enabled on the backup job that is protecting the VBO server though.

[nielsengelen]

If you enable application aware processing in the backup job, you will be able to rightclick on those VBR backups and start the required Veeam Explorers directly. Can u verify you have enabled this option for the VBO server backup job?

[HangTen]

Application aware processing is not enabled for that job. I will enable it going forward but for the already processed jobs, what are the steps for restoring items from this backup? It's even more complicated by the fact that all of the Sharepoint, Exchange, and OneDrive backups went into one repository so it's not easy to distinguish which folders contain what data.

[Mike Resseler]

Right click on the backup, choose item level restore and you will be able to choose the correct explorer (SharePoint / Exchange / OneDrive). Then you will see the items exactly as you see them in VBO from that VBK backup. Hope it helps

[HangTen]

I have tried that for both Sharepoint and Exchange but the explorer comes up blank. Am I missing a step?

[Gostev]

There are no additional steps aside of what was already mentioned above. Please open a support case and have our engineer assist you over a webex.

[amarshall]

Hi Team,
Is there a supported method for VBO365 servers with Object Storage Repositories?

Scenario:

We run a VBO365 server for a client utilising our Object Storage for the repositories, which are set to 1 year snapshot retention.
Each year we want to write the data down to tape for an archive backup - if the data is in object storage, the VBR backup won't capture the repository data.

Is there a way around this? Our object storage is run on our hardware, our datacentres.

[Mike Resseler]

Hi Adam,

There is no workaround at this point in time, however, we have a few ideas on the table which we are looking into to make sure there are additional copies of the data. How that exactly will work is not yet decided. We aim at flexibility but also think about the cost for end users.

[amarshall]

Thanks Mike,

I guess in the meantime we can keep customers with this requirement on block storage, and put a yearly VBK to tape. We can then use application level restore on the VBK file when needed.

[Gostev]

That would work, of course!

[Mike Resseler]

@amarshall Just for my information, how do you see the second copy of the data on your internal object storage? Any ideas or thoughts?

[amarshall]

Hi Mike,

I may have mis interpreted your question:

We can make a second copy of any existing object storage repository if the client requires it by replicating to another site via the object storage controller, and then connect to that with another VBO365 server if required (underlying snapshots help protect against malicious/accidental attacks being replicated). For this kind of second copy I think it would be kind of cool to be able to scale out the data to CloudConnect, to be able to take advantage of the "different media" and insider protection bonuses offered, rather than just having another Object Storage repository in VBO365 that implements a copy like function.

What I would like to be able to do is set a one year snapshot based retention on an OS repository & be able to take a VBR based backup each year that we could write to tape, so for our clients that have a regulatory but not operational based requirement for 7+years storage, can do so in the most economical way. So I guess the repository would look exactly the same, VBR would just be able to pull it into a VBK file.
Given the metadata is within the Object Storage repository as well as the VBO365 server, I guess we could mount a snapshot and write that to tape, but both that process and the process to then recover from that aren't as elegant (to say the least) nor particularly well supported.

[Mike Resseler]

Adam,

You understood the question well

Working over cloud connect is indeed a scenario, where the primary data is with the customer and the secondary with the service provider. However, that won't work for the SaaS solution that many service providers have started to offer.

Not saying we won't do it in a certain amount of time, but not sure if this is the most needed option at this point in time