Problem Description
Druva backup jobs fail repeatedly for a specific Hyper-V Virtual Machine (VM), preventing successful backups. Other VMs located within the same Hyper-V cluster continue to back up successfully.
Cause
The failure is primarily driven by two environmental factors:
Orphaned Checkpoint: An orphaned Hyper-V checkpoint (
.avhdxfile) exists that lacks a corresponding disk file. This causes the Hyper-V host to enter a broken state, making standard checkpoint deletion impossible and blocking the Druva snapshot process.VM Migration & UUID Change: A recent VM migration changed the VM's UUID. Because the host environment is unstable due to the checkpoint, the new UUID cannot be reconciled properly without a manual refresh within the Druva console.
⚠️ Note: Manual attempts within the Hyper-V Manager to delete these checkpoints will also fail. This confirms an underlying Hyper-V environmental issue rather than a failure originating from the Druva software.
Traceback / Logs
Error in Job
[YYYY-MM-DD HH:MM:SS] Starting backup operation.
[YYYY-MM-DD HH:MM:SS] VM not found for UUID (2548061F-1FBC-4ABF-B139-416F63EBB3D2)
[YYYY-MM-DD HH:MM:SS] VM migrated to host <Hyper-V_Host_Name>.
[YYYY-MM-DD HH:MM:SS] Backup operation failed.
Cannot delete checkpoint: The system cannot find the file specified. (0x80070002)
Checkpoint ID: C95FD28A-E8B4-41D9-A6B7-706037EFCE8F
Virtual machine failed to generate VHD tree.
Failed to get VHD parent: The system cannot find the file specified.
Job Progress Logs (JPL)
+---------------------+----------------------------------------------------------+
| log time | jpl_msg |
+---------------------+----------------------------------------------------------+
| YYYY-MM-DD HH:MM:SS | Starting backup operation. |
| YYYY-MM-DD HH:MM:SS | Deleting checkpoints with name [PhoenixBackupCheckPoint] |
| YYYY-MM-DD HH:MM:SS | Backup operation failed. |
+---------------------+----------------------------------------------------------+
Resolution
1. Windows / Hyper-V Administrator Action
Because this is a native infrastructure failure, the local infrastructure team must remediate the Hyper-V environment:
Clear Orphaned Checkpoints: Execute native Hyper-V PowerShell commands or use Hyper-V Manager to manually merge the disk files or clear the orphaned
.avhdxreferences from the affected VM.Verify Health: Ensure that a manually initiated Hyper-V checkpoint can be created and deleted successfully without errors.
2. Druva Console Action
Once the Hyper-V environment is healthy, update the infrastructure records in Druva to clear the UUID mismatch:
Log in to the Druva Management Console.
Navigate to Protect > Hyper-V and select Hosts from the left-hand menu.
Click the Refresh (🔄) icon at the top center of the Hosts list to update and register the newly migrated host data.
Results
Once the native Hyper-V checkpoint chain is successfully cleared/merged and the host environment is refreshed in the console, normal Druva backup operations will resume automatically.