Move from Azure blob to Amazon S3

[frankive]

Greetings!

We have quite few customers running SOBR with azure as capacity tier.
Unfortunelately it doesnt support Immutability which our customers demand these days.

We are considering moving them (the files!) to Amazon S3 with Immutable settings.
I guess this would create som Egress cost from Azure in that matter..
A few question around this issue:

- I can see a customer has a lot of full GFS (monthly, yearly) in Azure.. will this data be dehydrated during the Egress? Or would still just the block (the actual consumption) be transferred?

- What would be the best way to accomplish the goal to move from Azure to Amazon in this matter?

[Gostev]

Hi, Frank.

There are no "files" in object storage, because there's no file system to start with. As such, you cannot just "move" backups freely and easily, as is the case with on-prem repositories.

Now, in case of S3-compatible object storage, it is indeed possible to just move the entire bucket to another cloud provider. However, Azure is NOT an S3-compatible storage: they have the proprietary blob storage API we integrate with, which is different from S3 API. As such, unfortunately such moves is not something we can support.

Thus, your only options is to use "download everything on-prem, then offload again" approach. After changing the Capacity Tier to use an Amazon S3 bucket, of course.

If you use ReFS or XFS repository, the data should NOT be rehydrated when downloaded.

Thanks!

[frankive]

Thanks for quick answer!

"Download everything on-prem, then offload again" is indeed the approach we are looking at.
In the object storage the backup shows a size of 140 TB but the actualy use in the Azure Blob is appx 3TB (impressing!)

All local repositories are RefS.. so if I follow your lead here, you are saying that Veeam will in face download just 3 TB effectivelty data from Azure (Egress) and in the local performance extent just use 3 TB?

[dalbertson]

Correct. The “block” cloning will apply from blob to refs.

[frankive]

Great!
I will follow this procedure to download all backupfiles.
https://helpcenter.veeam.com/docs/backu ... ml?ver=110

What that is done, we can simply remove the capacity tier from the SOBR and then delete the object in Azure? That's the correct procedure?

[dalbertson]

You can. I would recommend to just seal the blob repo and then once you get the data into S3 and verified it is there, you can then delete the old blob account. This way you can have peace of mind that you have a copy of that data if something were to go wrong.anywhere in the

[Gostev]

It's a good idea to seal the object storage repository prior to starting the download.