Health check for GFS archives?

[Martin Damgaard]

So i recently had one of the latest .VBK files in a Backup Copy Job go bad.
I have no idea exactly what went wrong, but that is also out of this scope anyway.

But now i am wondering, if i can trust the rest of the full backup files in my GFS archive - Montlies, Quarterlies and Yearlies...
(As i understand it. Veeam only runs health check on the latest full restore point, right?)

Is there any way i can force or manually run either a FULL Health check, or manually check individual VBK files?

The repository is located on a Windows Server 2016 with backups running via Veeam Backup Agent.
Because of the large dataset involved, i need a way to do run the health check locally on the repository.

Veeam Backup is V9, by the way...

- Martin

[foggy]

Hello Martin, you can use Backup Validator to check GFS restore points.

[veremin]

Though, keep in mind these limitations showing up when files to validate exceed certain sizes. Thanks.

[Martin Damgaard]

Thank you foggy and v.Eremin.

I am aware of the Backup validator tool, and have tested it on a local/SMB backup file before. But can i run this directly on the repository server?
Or can i run it through a proxy or by an agent? (I want the heavy disk IO to happen directly on the repository server, and not over the network).

Any suggestions?

- Martin

[foggy]

You can copy the tool to the repository server and run it there.

[ferrus]

I have an outstanding issue with the Backup Copy Job Health Check performance to a DataDomain, and desperately trying to find a workaround.
The backup size is 11TB, and the Health Check duration is just short of 2 weeks - during which time no Backup Copy Jobs can proceed.

Can SureBackup completely replace the Health Check? Or can Backup Validator - if run from a separate server?

[foggy]

In some sense, Backup Validator can be used instead of the health check, however, it might take even longer, since the entire chain is checked. Running SureBackup is an additional measure to ensure whether VMs are actually recoverable.

[ferrus]

That would be prohibitively long in my case, since the Health Check for just the most recent backup takes 13 days.
Actually the length of time isn't the issue - it's the fact that it prevents any further jobs from running.

I was hoping that SureBackup could replace the Health Check - as it would only be the length of time is takes to power up the VM.

We have a second offsite DD2500 - with DD replication running. I wonder if I could manually perform a Health Check on that - leaving the onsite DD available for new Copy Jobs?

[foggy]

We have a second offsite DD2500 - with DD replication running. I wonder if I could manually perform a Health Check on that - leaving the onsite DD available for new Copy Jobs?

You can run Backup Validator against backups stored there.

[Martin Damgaard]

Hi all

Just want to give a little feedback on this one.
I figured out that I could (on VBR server) simply tell the validator tool the name of the job, and the date of the restore point.
This would trigger the remote agent to do all the heavy IO on the remote repository server for the complete .VBK of the specified date.

That seems to work fine - now i just need to find an easy way to check ALL files/restore points of a specific job...
Any good suggestions here? I know there is probably a way to do this with some Powershell voodoo, but i am not so familiar with Powershell yet, so a couple of hints / examples would be very helpful.

[foggy]

Yes, this should be possible with the help of PowerShell. You can ask for assistance in the corresponding subforum.

[Martin Damgaard]

Thank you foggy.

I am not sure what subforum you are referring to.

- Martin

[foggy]

This one.

[Martin Damgaard]

@foggy

Thanks - i will take a dive into that...