Azure server offload to Azure Blob

[marius402]

Hi Every1,

I have setup a simple Azure server with VBR to be the backup-copy target for multiple servers. This is done to enable use of the WAN accelerators. My various site data is now replicated to this Azure VBR server, but now I want to push the backups off to Azure BLOB storage and be deleted from the Azure VBR server (after a day) to make space for the next backups from the remote sites

I have sealed the various sites backups before backup copy (using Synthetic full backups every day), but cannot get the Azure VBR SOBR to push the data off to BLOB storage

Any ideas will be appreciated

Regards

Marius

[Mildur]

Hi Marius

Does this Azure VBR server manage the backup copy jobs or not? I believe, you copy the backups from other VBR server to this single one and then import the backup files. Is my assumption correct?
Then imported backups must be offloaded manually.
https://helpcenter.veeam.com/docs/backu ... ml?ver=110

Can I ask why you have this "man in the middle" backup server? Why not offload the backups directly from the on-premise VBR to the Azure Blob storage?

Thanks
Fabian

[marius402]

Hi Fabian,

Yes, the backups itself are done by the remote sites to remote repo on same server as the VBR's (EG Nigeria, Mozambique, Mauritius etc). I have imported the backups manually and can see the backup copies on the Azure Veeam server VBR as Disk(Imported). I have added those repos to a SOBR offload to replicate to BLOB

The "Man-in-the-middle" server is to use WAN accelerator between the sites and the Azure. Without WAN accelerator the remote sites pushes data to Azure BLOB at 1MBps, with the WAN accelerator this goes up to 61MBps. Without the WAN accelerators, the traffic cannot offload fast enough to Azure Blob storage and keeps on failing. We are talking Africa wide, where latencies are 100ms+ and bandwidth at places limited to 10mbps for an entire site

Regards

Marius

[Mildur]

Hi Marius

Thanks for the explanation.
Then, Imported Backups must be offloaded manually.
We can't offload them automatically.

Thanks
Fabian

[marius402]

Hi Fabian,

Not a problem, can you perhaps tell me how? I don't have access to an Azure Storage Browser, only the shared key & security string. Can it be done inside VBR?

Regards

Marius

[Mildur]

Yes, I posted the link in my first comment. It must be done inside VBR.
https://helpcenter.veeam.com/docs/backu ... ml?ver=110

But I don't think, that this will be a good solution. Doing manual offloads each day or week seems to be to much manual work.

[marius402]

I have already tried the mentioned article last week, but it does not show the same for me

If I look at the Backups --> Disk(imported), I can see the backups in there. If I click on the properties of the job, at the point where it states that if you click on the job, two buttons would be available, namely "Copy path" & "Move to capacity tier". In my view I only see the "Copy path" button.

Although this seems like daily manual work, it is still better to have the backups written to Azure Blob than just having backups on site. A flood destroyed one of the sites a few months ago, and another three sites were hit with hurricanes twice in the last two years

[marius402]

This what it looks like if I try to do a manual "Move to capacity tier"



As you can see, no button at the bottom, or if I right-click

For clarity, here's the full flow path diagram

[veremin]

Looks unexpected, so we'd appreciate, if you collect debug logs and open a support ticket with them. Once it's opened, post its number here so we can ask R&D team to join the investigation. Thanks!

[marius402]

Hi Vladimir, I did open a support case # 05541785

Regards

Marius

[veremin]

Passed internally for further investigation. Will update the topic, once I have more information.

[dbewernick]

@marius402 just for interest, what is the reason not to use Veeam Cloud Connect?

[veremin]

Based on the logs, the target repository you are trying to import backups from is not a Scale-Out Backup Repository. As a result, no "Move to Capacity Tier" button is shown. Thanks!

[marius402]

@marius402 just for interest, what is the reason not to use Veeam Cloud Connect?

Various reasons, main being that the customer did not want it, and wanted the data on Azure BLOB

[marius402]

Based on the logs, the target repository you are trying to import backups from is not a Scale-Out Backup Repository. As a result, no "Move to Capacity Tier" button is shown. Thanks!

Thank you. I eventually worked out another way to get the data to BLOB storage. The WAN accelerator with V11a increase the speed, copying data to a local Azure Repo, but cannot push the data to Blob storage. I used another Azure VM with V12beta to suck the data from the V11a server (as a share) and send directly to Azure Blob account. It works

Updated diagram of use:


This extra VM will obviously not be required once V12 has been released

Regards

Marius

[Mildur]

Hi Marius

Using Beta builds that way is not supported.
We cannot give you any guarantee that you will be able to read and use the objects produced by a V12 Beta build with the V12 GA build when it is released.
If you face any issues reading the data in the Blob after connecting the storage container to a V12 GA build, our support will not be able to help you.

Thanks
Fabian

[marius402]

Hi Fabian,

I do understand that and acknowledge the risk, but I need a solution now that I can apply to the future as well

Regards

Marius

[veremin]

Upgrade from beta to GA version is also not supported. Neither is importing backups from beta backup server to GA backup server. So right now it does not look like a valid solution. Thanks!