Rescan / Import Object Repo

[tevans8]

If old/orphaned traditional backup chains (VBMs, VBKs, and VIBs) are manually moved to S3-compatible object storage, is there a way to import this data into the VBR console? The data is stored in (bucket root > Veeam > Archive), so I see the folder when adding the object repo. I even tried building a blank SOBR and adding the object storage as the capacity tier. This allowed me to rescan but only shows 0 jobs in the wizard.

My three questions are:
1) Can Veeam only import and manage data in the object storage that was written as part of the capacity tiering process?
2) Is there a way to import the manually moved data apart from drive-mapping an S3 explorer and doing the older Import Backup function?
3) Is the External Repositories feature an option once it supports more than just AWS S3?

Thanks!

[HannesK]

Hello,
if you add an object storage as capacity tier to SOBR, then the software will ask you to import the data. See the User Guide

to answer your questions
1) yes
2) yes, see link mentioned above. S3 explorer is not the way to go
3) External Repositories are are to connect backup repositories from external backup software. Today that means Veeam Availability for AWS (formerly N2WS Cloud protection manager)

Best regards,
Hannes

[tevans8]

So, you seem to have contradicted yourself. Your first sentence makes it sound like that manually moved backup files can be imported per the user guide, but your answer to #1 is "Yes, ONLY data written by Veeam can be imported."

#1 and #2 cannot be both yes. Either, Veeam can ONLY import data that it writes to object storage originally OR it can import any type of Veeam file in object storage. I do not see it importing the manually moved data when I try it through the SOBR settings. It just shows a 0 in the wizard, and then nothing in Backups > Disk (imported).

My theory is that since Veeam offloads/de-hydrates the backup files in to a series of blocks and indexes in a series of folders (https://helpcenter.veeam.com/docs/backu ... l?ver=95u4) that this is the only data Veeam can import from the Capacity Tier. Without these indexes and metadata references (such as with traditional backup chains), nothing can be imported. Is this correct?

[HannesK]

sorry, it seems that I forgot about the word "manually".

I still don't understand what you mean with manually, but if that means moving data to S3 outside Veeam... then the answer to 2 is "no"

[tevans8]

Yes, that is what I mean, moving from a Windows repo to S3-compatible storage via an S3 explorer. That makes sense and is what I figured.

I don't suppose this could be a feature request, could it?

[HannesK]

The backup format we use for S3 is completely different. So we cannot use the files for SOBR capacity tier.

how is your single file restore speed and how much data is transferred with the S3 explorer import? (I'm travelling this week, which is why I cannot test it)

[tevans8]

No restores have been tested. There's about 27 TB that has been transferred.

These are old / abandoned backup chains being kept for archival retention so no active jobs are writing to them nor would a speedy restore be needed. The import into VBR via repo rescan is nice for management purposes (having one place for all backup data), which is where the mapped S3 explorer drive letter comes in as a workaround.

[HannesK]

There's about 27 TB that has been transferred.

that's the point. With the new backup file format for object storage we only transfer the data that is required. So using full backups that were transferred "somehow" do not really help

[tevans8]

That's what I figured. Would it be possible to have a feature that allows for importing any file structure?

[veremin]

We leverage a certain object structure that is expected by re-scan process, so there are no immediate plans to make re-scan understand other formats as well. Thanks!

[tevans8]

Thanks all!