Periodic full backups

[SPITech]

Hi,
I am new to Veeam backup and recovery. Currently we are using the Community edition to back up file shares on a Windows 2012 server to a NAS.
It's working great but seems to be in forever forward incremental mode. I would like to be able to schedule periodic full backups of the fileshares and can't see a way to do this.
Thank you for your help!
Sara

[nikolaj]

Hi Sara,

VBR's is designed to perform one initial file share backup to be then followed by a forever incremental chain henceforth.
More on that could be found in our user guide article.

Perhaps, starting a new backup chain for your share would cover your objective of offloading periodic full backups.

Thanks.

[Gostev]

I would like to be able to schedule periodic full backups of the fileshares

What is the reason for you wanting to do this? Because Veeam likely covers them with existing functionality and in a better way.

[SPITech]

Thanks so much for your response.
I suppose the rationale for periodic full backups is just insurance in case over time the initial full backup becomes corrupted in some way. It's a redundancy practice. But perhaps I am missing something in the design of forever forward incremental backups?

[Gostev]

Right, periodic full backups don't really give you such insurance, because with storage-level corruption (most typical issue we're seeing in support) these full backups will be corrupted immediately upon creation. This is the reason why Veeam provides and recommends a backup health check functionality that actively verifies consistency of data in backup. This insures you can restore exactly what was grabbed from the production storage (since we fingerprinted all data at that time), and that the backup storage does not suffer from storage-level corruption.

[SPITech]

So helpful thank you. I will add a backup health check.
Can you share what you find to be the causes of storage level corruption?

[Gostev]

It's never possible to say with any kind of certainty. From our perspective, storage simply returns different data comparing to what was written to it (fingerprints don't match data).

Low end storage without enterprise-grade RAID controllers with BBWC often experiences data loss following power loss.

Sometimes it's due to storage firmware bugs.

Sometimes it's a hardwire issue like a sticky bit in some chip in the I/O path (storage or fabric/network).

Sometimes it's SMB stack misbehaving (which is why we don't recommend using SMB share as a target).

[braffer09]

What is the reason for you wanting to do this? Because Veeam likely covers them with existing functionality and in a better way.

Hi - I'm going through and wondering the same.

If I backup my file share with "keep all file versions for the last 30 days" and select "keep previous file versions for 6 months", will I end of up with one Full and a very long backup chain??

Is the backup health check the one way to ensure consistency?

I was imagining a Synthetic Full like on a VM, but from what I'm learning that's not the case.

Any help or clarity would be appreciated.

Thanks

[Gostev]

Right. The fact that you're saying "long backup chain" and "synthetic full" suggest you're thinking about how Veeam does image-level backup, but NAS backup is completely different. There are no image-level deltas that are dependent on a number of previous deltas and eventually on the original full image, or anything like this. Instead, all changed files are stored independently and there are no "chains" of any kind.

When doing a restore, we just take a version of each file that is closest to the selected restore time. This means the restored latest state of a share can include files that were last backed up 3 years ago, because they never changed since then. There's no point to continuously re-backup them all these three years.

And yes, the health check ensures no files under protection are missing from backup.