Unity Snapshots - restore point doesn't contain any disks

[miek]

So we tried to restore a VM and it says this restore pointis 43gb. We restored it and it gave it no disks, and so I looked atjust doing a disk restore with this VM and none of the restore points have disks yet it sees they take up space.

At the same time i looked at several other VMs and and it's inconsistent. I can see all the disks for some, missing one for others. We have 4 clusters and it's entirely random what it finds or not.

One machine it won't see the disks on cluster2, another won't see the disks for that vm that are on cluster1.

Any ideas? I'm getting no error for rescan and we confirm these disks exist in the snapshots.

[Mildur]

How are the snapshots be done? Are they triggered by veeam or the storage itself. If it‘s the storage who triggers the snapshots, do this vms having vmdks on the same datastore (Unitity Volumes/Luns) or are they spread over multiple datastores?

We have some limitations if snapshots are managed by the storage itself and the vm is distributed over different luns.


https://helpcenter.veeam.com/docs/backu ... ml?ver=110

You can restore data of a VM whose disks are hosted on different VMware datastores (multi-home restore). This is true for storage snapshots created both as snapshot-only jobs and backup jobs with storage snapshot retention.

However, if you restore from native storage snapshots or snapshots that were created with earlier versions of Veeam Backup & Replication (up to version 9.5 Update 4), mind the following:

- The restored VM will have only those VM disks that are hosted on the same datastore as the VMX file.

To overcome these limitations, you must restore from storage snapshots created with the current version of Veeam Backup & Replication (starting from version 9.5 Update 4).

[miek]

The snapshots are triggered by the storage to happen daily.

I did a rescan and I get a bunch of these errors

First it says list of vms for datastore (clustername) of host (host) obtained successfully for the 4 clusters.

Then under that it says:

failed to obtain list of vms on lun (clustername) of volume (clustername) there are no proxy servers with configed access to the lun (clustername) - this is also for the same 4

I was using automatic on the backup proxy for access and changed it to manual and added these 4 and it still gives the same error

On top of that some machines will list all their possible disks available for them, with disks mixed between these 4, and we have restored machines like this. Others will be missing a disk. It's not consistent.

[Mildur]

In your case (storage is doing the snapshots, not veeam):
- If all virtual disks are on the same datastore, then all disks should be restored.
- If they are on different datastores, it won‘t work. Only disks which are on the datastore with the vm config file are restored. See my posted limitation link for that.

It‘s best to open a support case, if you think something is wrong. If it‘s a technical issue, a support case is always needed. See the forum rules: https://www.veeam.com/forums/viewtopic.php?f=2&t=755