Hyper-V VM Stale Checkpoints creating problems with backup operations.

Hello nikhil satpute, I completely understand how concerning it is when critical VM backups fail due to stale checkpoint files.

Here are the answers to your questions:

1. Best solution to clear lingering .AVHDX files without corruption:

Since the files remained after the reboot, the Hyper-V disk chain merge process is likely stuck or out of sync. The safest and most effective method to force a background merge without corrupting the VM is the "Create and Delete" method:

• Open Hyper-V Manager.
• Right-click your BHS-VM-SEC VM and select Checkpoint to manually create a brand new checkpoint.
• Once created, right-click that newly created checkpoint and select Delete Checkpoint Subtree.

This action forces Hyper-V to re-evaluate the entire virtual hard disk chain and triggers the live merging process for all orphaned .avhdx files in that folder.

2. Removing the recovery checkpoint safely during production:

You can remove the visible recovery checkpoint using the same method (Right-click -> Delete Checkpoint).

• Performance Impact: Modern Hyper-V performs "live merges." This means the VM remains online, and there is no downtime.
• Production Hours: While it is completely safe to do while the VM is running, the merging process does consume storage I/O as the data is written back into the parent disk. Because this is a critical VM, I highly recommend performing this during off-peak hours or a maintenance window to ensure the VM's performance is not impacted by the storage load.

Monitor the VM status in Hyper-V Manager; you will see a "Merging..." status. Wait for this to reach 100% and disappear. Once the folder is cleared of the .avhdx files, your Veeam backup job will run successfully.

Please try this and let me know how it goes! I am happy to follow up and assist further if the disks are still being stubborn.

If this solution helps resolve your issue, please don't forget to click "Accept Answer". This helps confirm the solution and makes it easier for other community members facing the same Veeam/Hyper-V issue to find it.

The VM has stale/orphaned AVHDX differencing disks left behind by earlier checkpoints or backup operations. These can block new checkpoints and backups, and if the chain is corrupted they can also prevent the VM from starting after a reboot. The supported way to fix this is to identify which AVHDX files are still part of the active disk chain and then merge any orphaned AVHDX files back into their parent VHDX, with the VM shut down.

Recommended steps:

1. Take a backup or export of the VM
   • Before changing disks, create a full backup/export of the VM folder or storage (for example, via storage snapshot or backup product) to protect against data loss.
2. Inspect the current disk chain for the VM
   • Shut down the VM BHS-VM-SEC.
   • Run PowerShell on a Hyper‑V host:

      Get-VMHardDiskDrive -VMName "BHS-VM-SEC" |
        ForEach-Object { Get-VHD -Path $_.Path | Select-Object Path, ParentPath, VHDType }
   

   • This shows the active VHDX/AVHDX chain (Path and ParentPath). Any AVHDX files that do not appear in this chain are orphaned.
3. Manually merge orphaned AVHDX files
   • For each AVHDX that is not part of the active chain but still exists under C:\ClusterStorage\Volume2\Virtual Machines\BHS-VM-SEC\Virtual Hard Disks\:

      Merge-VHD -Path <path-to-orphan.avhdx> -DestinationPath <parent-vhdx>
   

   • Alternatively, use Hyper‑V Manager:
     1. In Actions, select Edit Disk.
     2. Browse to the AVHDX file.
     3. Choose Merge → To the parent virtual hard disk and complete the wizard.
   • Repeat for each AVHDX in a chain, starting from the youngest (most recent) and moving toward the oldest, if they are still logically linked.
4. Ensure the VM points to the correct (merged) VHDX
   • In Hyper‑V Manager, open Settings for BHS-VM-SEC and confirm that each virtual hard disk is attached to the expected final VHDX (not to an old AVHDX).
   • Start the VM and verify that it boots and applications function correctly.
5. Handle the visible “recovery checkpoint”
   • Recovery checkpoints are created automatically during certain operations. If it is part of the active chain, it must be merged rather than deleted directly.
   • With the VM shut down, attempt to delete the checkpoint from Hyper‑V Manager. If that fails or the delete option is unavailable, rely on the manual merge process above to collapse the chain and then ensure the VM is attached to the base VHDX.
6. Resume backups and avoid future accumulation
   • After the chain is cleaned and only the base VHDX (or a minimal, healthy chain) remains, rerun the Veeam backup job.
   • To prevent recurrence:
     • Monitor for unmerged checkpoints and VHDX growth.
     • Avoid overlapping backup jobs that can cause VSS timeouts or incomplete merges.
     • Ensure antivirus or other tools are not locking VHDX/AVHDX files during backup windows.

Regarding safety and impact:

• Deleting or merging checkpoints while the VM is running can impact performance and, if interrupted, risk disk chain corruption. The safest approach is to shut down the VM, perform merges, and then restart.
• When done with the supported Hyper‑V merge process (Edit Disk or Merge‑VHD) and with a prior backup/export, the risk of corruption is minimized.
• For a critical production VM, schedule a maintenance window to shut down the VM, merge the AVHDX files, and validate functionality before returning it to service.

References:

• Troubleshoot Hyper-V virtual machine backup, checkpoint, and storage failures
• Hyper-V snapshots, checkpoints, and differencing disks (AVHDX) troubleshooting guidance
• Troubleshoot inaccessible or unresponsive Hyper-V VMs in clustered or standalone environments