Backup files health check + scrubbing on dedup vols needed?

[mxx]

Hi,

I know those 2 things aren't exactly comparable, but I'm wondering whether it would make any sense, in order to reduce the load on the datastores, to omit one of them (on volumes exclusively used for storing files of backup copy jobs).

I'd probably not ask this question if I knew more about scrubbing, optimization and garbage collection.

I only know that scrubbing of dedup volumes (Server 2012) is very resource intensive and since the only way to really make sure the backup files are ok, is by enabling regular health checks, wouldn't this already take care of finding problems/correcting them?
I'm only talking about Backup Copy Jobs because you can't use them with SureBackup.

Thank you!

[PTide]

I know those 2 things aren't exactly comparable, but I'm wondering whether it would make any sense, in order to reduce the load on the datastores, to omit one of them (on volumes exclusively used for storing files of backup copy jobs).

Not sure what you mean by "omit one of them". Backup Health check does not load production datastore much because during health check VBR calculates CRC values for backup metadata and hash values for VM disks data blocks in the backup file and compares them with the CRC and hash values that are already stored in the backup file. Please elaborate.

Thank you.

[mxx]

Calculating checksums needs to read all data from the (in this case awfully slow) storage. Scrubbing is awfully slow too of course.
Since scrubbing alone doesn't guarantee the backup files are okay, I'm wondering what's the opinion about the necessity of scrubbing in this certain case (volume exclusively used for backup copy jobs) and whether I could disable scrubbing jobs until I run into problems detected by Veeam's health checks.

Thank you!

[PTide]

Ok, now I see. As I've described in my previous post, both scrubbing and health check jobs perform a similar activity with the diference that health check is applied to Veeam backup files only. It won't harm if you disable scrubbing on the volume that stores backups, however I suspect that Windows native scrubbing tool might be able to perform that integrity check faster, so you might want to test both to compare the performance. Also there is another thing to keep in mind - Veeam health check makes sure that the most recent restore point is OK whereas Windows scrubbing checks that the whole volume is ok thus checking the whole backups set that is residing on the volume.

Thank you.

[mxx]

That was a perfect answer. Thank you very much for your insights!

[foggy]

I'd warn you from disabling either of these, since they work with different data sets and in some sense supplement each other. Scrubbing works with deduped files only and the backup chain can contain both deduped and non-deduped files, so the latter will not be checked during scrubbing. Also, corruption might occur before or during deduplication and will not be detected by scrubbing as well.

[mxx]

Thanks foggy for your follow up on this!
I understand. If anything, I'd have disabled scrubbing, but I can see why this is probably not the best idea.
Also I thought that veeam would check ALL files (GFS), not just the most recent restore point which might only involve a few incrementals + 1 full.

[Gostev]

Veeam check ensures that the latest restore point is recoverable, since this is what you will want to restore in most cases. But of course, testing the latest restore point implicitly validates the incremental chain of backups this restore point relies on.

Checking ALL backup files will take unreasonably long time, especially with slow backup storage.

[mxx]

... and I'm glad it is this way
Thanks Gostev!