Hyper-V role Windows Server 2019 Standard single VM doesn't backup


Trying to solve this one for myself, but not quite getting there. Customer system that’s running Windows Server 2019 Standard with Hyper-V role using the commercial Hyper-v client. It has 4 VMs on it, and oddly all are backing up except one. It tries to create the checkpoint, gets to 19% then dies. Below is the log message in the client, and the debug log output:

Starting unscheduled incremental image backup of volume “SCSI:0:0”…
Basing image backup on last full image backup
Loading ZIP metadata from urbackup/mbr_data_ADS-2012–SCSI_0_0.zip
GWHV19[ADS-2012]: Loading MBR zip file…
Error loading ZIP file MBR (1): Creating shadow copy failed. See client log file for details.
Time taken for backing up client GWHV19[ADS-2012]: 23s
Backup failed

You get the same message if it’s a full image backup.

Here’s the debug log:
2021-12-08 18:09:49: ERROR: Creating shadowcopy of “hyperv://ADS-2012/SCSI:0:0” failed.
2021-12-08 18:16:07: ERROR: Error creating Hyper-V snapshot of ADS-2012. Output: ProcessWMIJob : Exception of type ‘Microsoft.PowerShell.Commands.WriteErrorException’ was thrown.
At C:\Program Files\UrBackup\hyperv_snapshot.ps1:152 char:13

  • ($job | ProcessWMIJob -WmiClass $Msvm_VirtualSystemSnapshotServic ...
  •         ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
    • CategoryInfo : NotSpecified: (:slight_smile: [Write-Error], WriteErrorException
    • FullyQualifiedErrorId : Microsoft.PowerShell.Commands.WriteErrorException,ProcessWMIJob

At C:\Program Files\UrBackup\hyperv_snapshot.ps1:32 char:14

  •          Throw $Job.ErrorDescription
  •          ~~~~~~~~~~~~~~~~~~~~~~~~~~~
    • CategoryInfo : OperationStopped: (:slight_smile: [], RuntimeException
    • FullyQualifiedErrorId : ScriptHalted

2021-12-08 18:16:07: ERROR: Creating shadowcopy of “hyperv://ADS-2012/SCSI:0:0” failed.

Ps running vssadmin list writers does not show any errors

Appreciate any insight!


Googling only finds me this old thread:

In this case I think the issue is VSS issues within the VM. Perhaps try to create a production snapshot from the UI to confirm it cannot create production snapshots. Then look at the event log within the VM.