Case # 00532649 Retention policy, with Transform weekly job

[vec44]

Hello, I am hoping for some help with a basic retention policy question I have. Our current backup environment is a mix of Veeam 7 (Hyper-V) and Backup Exec 2012. Our Veeam backup repository is a 10.9 TB NAS. We currently use Backup Exec to weekly copy our Veeam backup repository to tape. Our current retention policy is 21 days, and the goal for it being set at 21 days is so that we could easily restore back from the backup repository back 3 weeks. We had an issue previously when our retention policy was set to 7 days, and the weeks tape backup of the repository failed, so we had no backup of that data to tape. We have been getting errors on NAS weekly for some larger jobs during the Synthetic Full, "Transform previous full backup chains into rollbacks". Veeam Case # 00532649 mentioned that the NAS could not handle the I/O. My question is if I were to reduce my retention policy from 21 to say 14 days, will that help the I/O on the NAS for the "Transform previous full backup chains into rollbacks" process, and possibly clear up the errors?

The other goal is to not increase disk space usage, by say running "Active Fulls" weekly. Any recommendations would be greatly appreciated.

Thank you,

[Gostev]

Hello, reducing retention policy will have no impact on I/O load from transformation (assuming you are doing daily backups). It does not look like you have many options here: if your storage does not provide enough IOPS to perform transformation reliably and/or withing the required window, you simply cannot be using one, and you have to switch to Active Fulls instead. In your case, it might be a good idea to perform those regardless, as it sounds like you have a very low end storage device. Those are more prone to various data corruption issues, so forever incremental backup is not a good option here. Thanks!

[tsightler]

You could try either running a transform each night instead of weekly, or just switching to reverse incremental. Also, if you do this, make sure to limit the amount of concurrent tasks on the repository.

[vec44]

Thank you both very much for the advice.

Thank you Gostev for informing me that the retention policy has no impact on the I/O load for the transformation, as daily backups are performed. Gostev do you agree with tsightler that I should switch to reverse incremental? You mentioned that the forever incremental is prone to data corruptions.

Thank you tsightler, should I be concerned with space if I switch to reverse incremental, and perform Active Fulls?

Do either of you recommend reducing my retention policy to conserve some space?

Thanks again,

[foggy]

Gostev do you agree with tsightler that I should switch to reverse incremental? You mentioned that the forever incremental is prone to data corruptions.

Let me chime in here with a note that performin periodic active fulls is always a good preventive action and is recommended for any any of the backup methods.

Thank you tsightler, should I be concerned with space if I switch to reverse incremental, and perform Active Fulls?

You would need additional space for an active full, indeed.

Do either of you recommend reducing my retention policy to conserve some space?

Nobody cannot really recommend this to you, since retention is primarily defined by RPO requirements accepted in your company.