cant restore

[hyvokar]

Doing an emergency restore here.
Trying to restore full vm.

Getting an error
27.6.2019 22:03:21 Error Restore job failed Error: Invalid datastore path '[SC3020-SAS_High_1]'.

Where to start troubleshooting?

[hyvokar]

Restoring vhds only seem to be working

[skrause]

That error usually occurs when the original datastore location is not accessible to the host which Veeam is using to do the restore.

[hyvokar]

Shouldn't it be accessible if restoring virtual hard disks work? I mean they are in same datastore no matter which way you are doing the restore.

[foggy]

Contacting support is typically more effective in such emergency cases.

[hyvokar]

My first step was to contact support yesterday. I still have not heard from them.

[foggy]

Have you mentioned the urgency? What is the case ID number?

[Gostev]

My first step was to contact support yesterday. I still have not heard from them.

I've reviewed the case and apparently the case severity was lowered from 1 to 3, which made the case move from the SWAT queue to the regular queue. Which explains why it takes longer for them to contact you. I can't see if it was you or Veeam staff making this change? In any case, I asked the support management to review this case.

[m_zolkin]

My first step was to contact support yesterday. I still have not heard from them.

We've been trying to reach out to you by both email and phone with no luck. Let us know through the ticket if you still need our help with this.

[hyvokar]

Hi,

Back from a vacation.
I was able to restore the virtual hard disk, which at the moment was good-enough solution and after that lowered the case severity.

Support has identified the problem - you cannot do the restore, if the VM disks are distributed among different datastores. There are no fix planned at the moment.

[foggy]

Seems you are talking about storage snapshots integration then since this limitation is specific to restores from storage snapshots, not regular backup files. Moreover, this limitation was removed in Veeam B&R 9.5 U4 and since then VMs having disks on multiple datastores are fully supported.

[hyvokar]

Hi,
Nope, we are not using storage snapshots.

I've just upgraded from U4a to U4b. I'll try to reproduce the problem if I can find some time to do it with a dummy machine.

[hyvokar]

Hi,

I tested the backup & restore today with 9.5 u4b.

The restore now works without errors, but all the disks are restored in the same datastore, even if they were in different datastores before backup.

[foggy]

You can select target datastore for each disk individually in the restore wizard.

[hyvokar]

Thanks, I'll try that.

Wouldn't it be logical to restore disks to their original locations when using default settings?

[hyvokar]

Restore to the original location
"Quickly initiate the restore of selected VM to it's *original location*, with the original name and *settings*. This option minimizes the change of user input error."

Either the description is wrong or there a bug in a restore process.

Restore to the original location does not give you an option to select (original) datastore(s).

[foggy]

Indeed, during restore to the original location, all VM disks are placed on the datastore where VMX file resides. In this case, the description might be confusing (though it doesn't tell anything about disks placement).

[Gostev]

@foggy that would be a bug, let's fix in v10. All disks must go to their original location... current behavior is non-sense. Likely this simply broke at some point, and no one noticed.

This quite critical, as it renders all our self-service restore portals useless.

[foggy]

Checked with QA, this is already addressed in v10.