Feature request - reverse incremental backup

[Zek]

Please add reverse incremental backup feature for veeam endpoint backup.

[nielsengelen]

Could you explain a bit more why you want this added to the agent? As we’ve been improving the current method for years with great success to power the impact on backup storage as you have with reverse incremental (4x IO usage).

[lordmilko]

My system is configured to keep 365 of forever incremental backups. As a result of this, whenever I need to recover a file I have to wait 5-10 minutes for Veeam to read through the entire chain to get to the latest backup. If the latest backup was already a full however, this process would be significantly faster

[Zek]

Fully agree with lordmilko.
The second reason and the main reason for me is retention. In the case of reverse incremental, I can simply remove the old restore points to clear the occupied space. In the case of the classic incremental backup, it is impossible to delete the old restore point without losing the subsequent ones. Some solutions have the ability to consolidate(remove) recovery points and preserve the integrity of the subsequent chain, but this is not possibility in the veeam agent.
Veeam agent forces you to create regular full recovery points if you want to be able to remove old ones and occupy much more free space.
By the way, for workstations there is no problem in IO overloading of the storage, because most often it does not matter, it runs for an hour or four: it is performed quite rarely, at best, once a day.

[fgw]

another vote for Reverse Incremental

my point is data reliability! on reverse incremental most recent restore points are still valid back to the point of the corrupted backup file. when doing forward incremental, all restore points newer than the corrupted one are lost!

dont care much about higher IO usage. if anybody is concerned, they still can continue forever forward ...

assuming it will not be too much effort to implent it either, as reverse incremental is already part of VB&R and backup engine is same anyway - as i have read somewhere else.

[Dima P.]

Hello folks,

Thank you for sharing your feedback, consider your votes being added to this feature request. Cheers!

[mma]

+1 for this request.

I'm a fan of daily fulls on tape
I can create daily full to tape for my VMs (reverse Inc chain), but not for my physical workload.

[Dima P.]

Hello Marcel,

For tape backups you can create a dedicated job for your agents and schedule a synthetic full backup creation for every day. With such setup full backup will be created based on backup chain stored in your repository. Makes sense?

[jono.c]

Another vote for reverse incremental on agent-based backup jobs.

We have several very large VMs that have 6TB of data each but very low rate of change. We use agents on these VMs as they cannot have a snapshot taken.

Backup target is a Dell SAN with SATA disk and synthetic full backups take 30+ hours to run. Reverse incremental would solve this problem nicely, and given the low rate of change on disk should not impact incremental backup times.

[Dima P.]

Hello jono.c,

Mind me asking if you use Veeam Agent for Windows in conjunction with CBT driver (it's available in Veeam Agent for Windows running with Server license installed)? Thanks!