Reverse Incremental\Full Active behaviour

[trevora]

Hi, please could someone explain how a reverse incremental backup job works in conjunction with the periodic Full Active Backup option? I've been testing a few things and I can't tell what is actually happening.

I have a test reverse incremental job set up to keep 3 restore points and run a Full Active job on a particular day. When I reach that particular day then nothing appears to have happened any differently than if Full Active was not selected. I still get the same number of vbk and vrb files as if the Full Active hadn't run. I would expect this if the Full Backup was performed and just took the place of the vbk that would otherwise have been created by the normal reverse incremental job, but the time it takes for the backup job to run is exactly the same as if a normal incremental job was run. Shouldn't the backup on the day of the Full Active take as long as the initial backup done at the beginning of a reverse incremental job?

Thanks in advance,
T

[veremin]

In fact, with reversed incremental mode, there is no need to perform periodic full backups to guarantee safety of data and to keep up with the specified retention policy, since the most recent restore point is always a full backup, and it gets updated after every backup cycle, literally changes are being injected into the .vbk file to rebuild it to the most recent state of a VM.

Moreover, you don't have to perform active full backup on particular day, just keep the best practice mentioned in this post in mind.

Hope this helps.
Thanks.

[kjc3303]

When a Full active is run it will create a new vbk so you should have two vbks and the old one will drop off with your retention policy.

As Vladimir has said Active full are not needed or less important with Reverse incremental.

[trevora]

Thanks for the quick replies Vladimir and KJC.

Firstly, in view of your reply, what use does the Full Active backups option have with Reverse Incremental backups? Does it have any use?

Secondly, I think I've just had the epiphany i needed but just to check I understand this right. Full active backups are not required with reverse incrementals because a reverse incremental is built each time it runs by comparing the source with the target. If a corruption had occured in the target previously then that would be corrected in the next backup job run and the only reason the initial job takes longer is because it's writing the whole backup to the target rather than just the changes.

Actually, now I've written it down it seems blindingly obvious and I've probably made myself look stupid, but maybe this will help someone else in the future!!

[kjc3303]

A use Active full has is if a VM is removed from a job performing an Active full will reclaim that space.

[veremin]

For the purpose of complete understanding, full backup is still needed even in case of reversed incremental mode, though, it’s considered to be less important and should be performed not so frequently, once every 1-3 months.

For more information, please, review topic I’ve previously provided.

Hope this helps.
Thanks.

[kte]

does the active full doesn't read everything from the san again??

[Gostev]

Yes, it does read the data from production storage. That's the whole point of Active Full in the first place (to break forever incremental chain).

[Yuki]

Well, one good reason to do a FULL backup is to stop Reverse Incremental from growing since even if files/data/blocks are erased, the backup file is never deflated. Correct?

[chrisdearden]

Well, one good reason to do a FULL backup is to stop Reverse Incremental from growing since even if files/data/blocks are erased, the backup file is never deflated. Correct?

Correct - though that whitespace can be reused, the VBK has no shrink mechanism. I like to think of the active fulls as a defrag type operation.