Proxy adding thick provisioned disks

[Matt@Work]

I'm not sure why this is happening, but I have a Server Core 2016 Veeam Proxy which has a 127GB system disk and 3 60TB volumes used as a scale out repository.

I've noticed that its got another 4 volumes that show as offline totalling nearly 900GB. I didn't provision them and I don't understand why VMware did either. The SOR is the backup point, NFSDatastore and all the other Veeam components are on the system partition.

Anyone seen this before? Can I delete the volumes?

[Rick.Vanover]

Hi Matt - is this proxy & repository system a virtual machine? If so, is hotadd mode an available backup mode for your VMware virtual machines?

If a Veeam VMware Backup Proxy is a VM and uses hotadd mode, disks will be attached during the backup job. Also I could see a case with the HBA/FC or iSCSI initiator and direct SAN mode.

However, if a backup job is not running - the attach should not be left behind. Vitaliy in Veeam Product Management explains more info on how/why it should not be attached here: vmware-vsphere-f24/veeam-snapshot-hunte ... 36187.html

But generally speaking, if a backup job is running with that criteria (is a VM and hotadd is used; or iSCSI/FC targets), it is expected to see offline disks attached; but then go away. if they persist, I recommend contacting to support to see if they are somehow left from a job - or to confirm they are not from a Veeam job and going another direction to recommend.

[Matt@Work]

Proxy is a VM. The SOR is a QNAP connected via iSCSI to a Starwind HCA with 3 60TB VMFS volumes presented to the VM. These are then used as the SOR.

The disks are hanging around and I'm seeing DRS errors in vCenter about no host having access to the volumes, which is weird. There is no job running right now so something is persisting them. I'll raise a ticket tomorrow when I'm in the office.

EDIT: Might be a tagging error, the proxy might be trying to back up itself as I see a Success attribute applied in vCenter. I have a BackupDisabled tag for this, so I need to track that down.

[skrause]

You should be able to tell where those additional disks are coming from by looking at the settings in VMware and seeing the path to the .vmdk files. If it is a hot-add proxy, those are likely orphaned from a backup job that had an issue.

[Rick.Vanover]

@skrause has a good point - if you investigate the inventory of the proxy VM in vSphere - you should be able to track the source down if it is a hotadd anomaly. If it is direct SAN, it would not show that way.