ScaleOut Repository rescanned but not accessible automatic

[widmerkarl]

Hello folks

I've got some strange behaviour of one ScaleOut Repository.

At a customer we've got two Synology NAS devices which are rotated on a weekly manner. Each NAS is configured with its own IP address and hostname and added as single repository. Then i created a ScaleOut Repository and added these two extents. That worked fine.

Now the customer shuts down a NAS (because of the weekly rotation) and Veeam BR doesn't get it, that there is only one NAS active in the ScaleOut Repository:

Code: Select all

Error: Cannot proceed with the job: existing backup meta file 'Daily.vbm' on repository 'NAS-ScaleOut' is not synchronized with the DB. To resolve this, run repository rescan.

When i do a manual rescan, then the backup works fine.

I can see clearly in the Veeam logs that there happens a automatic rescan (host discovery) every few hours, and Veeam does recognize that one of the two NAS devices is not active / online. So i assume Veeam should know which extent is active and should be able to use the extent which is online. But the job fails.

- Is this a behaviour by design?
- Is it necessary to activate maintenance mode manually?
- Is it necessary to do a scripted rescan before the backup job runs, and Veeam gets it that the online extent should be used?

I don't want to manually rescan that as an admin, nether do i want to delegate this to the customer.

I want to ask first the community before i open a support case. Perhaps someone out there had the same issues and a possible solution for it.

Thanks for your help!

[foggy]

Karl, basically, rotated drives cannot be added as extents to a scale-out backup repository (this is mentioned as one of the limitations in the user guide).

[Gostev]

I wonder if SOBR option to create new full when previous backup is missing will help here.

[widmerkarl]

Thanks @foggy for clarification. You are absolutely correct. Since these NAS devices are rotated in a weekly manner, we can't set it up as ScaleOut Repo.

@Gostev: the policy of this SOBR is "Data Locality". Under "Extents => Advanced" the checkbox for "perform full backup when required extent is offline" is checked. So theoretically a backup should be made, either first or second NAS is online.

But Veeam doesn't get it if an extent is offline. It does not backup and brings the message that NAS xy is offline and the job fails.

I will check that with the customer and try to figure out how we can manage this backup. We thought that a SOBR would be superb. Without that we have to handle different repositories, create different schedules and backup jobs.

*** Edit: ***
Just had the idea to setup both NAS devices with the same hostname and IP address. Then i create one backup repository and enable the option "This repository is backed up by rotated hard drives" on this repository. I will try that.

[foggy]

*** Edit: ***
Just had the idea to setup both NAS devices with the same hostname and IP address. Then i create one backup repository and enable the option "This repository is backed up by rotated hard drives" on this repository. I will try that.

This should work.