Backup job vs copy job

[lowlander]

Hi,

I am reviewing the optimal use cases for a SOBR configured with a capacity layer. Target is to have a copy of specific backups in the object storage layer.

1st scenario is using a backup job with GFS configured. pointing the backup to a SOBR as target I am able to keep certain full backups longer for archival purposes. I can choose for keep the weekly backups for x weeks, monthly full backups for x months and yearly full backups for x years. Within the capacity tier of the SOBR I can point out to copy backups to the object storage as soon as they are created. Is the retention managed the same in the capacity tier as in the performance tier or is this just a copy as is of the retention points in the performanc tier ?

2nd scenario is using a backup copy job pointing to a object storage repository. I can choose to keep restore points for x days, x weekly, x monthly and x yearly backups. I Assume the you can manage the retention the same as the restore points in a local repository.

Personally I think it would be easier to use the first scenario, while you don't have to create a backup job AND a copy job. However, what would be the reason for using the copy job for copy out restore points to object storage ?

[HannesK]

Hello,
1: yes, the backup job manages the retention. capacity tier is just a checkbox... no separate retention. https://helpcenter.veeam.com/docs/backu ... ml?ver=100

2: that option does not exist. BCJ can only point to a normal repository (which can have capacity tier). BCJ manages retention.

a few things are not copied to capacity tier today. e.g. database logs.That's a reason for BCJs. In most scenarios, the "copy" checkbox for capacity tier is the easiest.

Best regards,
Hannes

[lowlander]

So basically it is just a copy of what you are having locally. Can we select restore points directly from object storage to restore ?

[oleg.feoktistov]

Of course. That's what makes object storage so usable. You can restore them even if other SOBR extents or SOBR as a whole is not available in the event of disaster.