Stale snapshots in the datastore on which the virtual machine resides

Products

• VMware vCenter Server
• VMware vSphere ESXi

Issue/Introduction

Stale snapshots may be observed, such as vmname-000001.vmdk, when browsing the datastore where the virtual machine resides. These stale snapshots are not visible in the Snapshot Manager. This issue typically occurs when a backup application is running.

Environment

VMware vSphere ESXi 7.x
VMware vSphere ESXi 8.x
VMware vCenter Server 7.x
VMware vCenter Server 8.x

Cause

Third-party backup applications send API calls to VMware to create and remove snapshots on the virtual machines they process. Snapshots might remain on virtual machines if those API calls are not sent/received, or if the snapshot removal operation cannot complete successfully.

If a snapshot remains after a third-party backup application finishes and that snapshot is visible in the Snapshot Manager, it indicates the snapshot removal API call was not sent or was not received by the VMware environment.

If a snapshot remains but is not visible in the Snapshot Manager, it indicates the request for snapshot removal was sent and received, but the snapshot consolidation did not complete successfully. This often leaves the virtual machine in a state where it displays a "Consolidation is needed" alert. A lock on the VMDK files of that virtual machine can hinder the snapshot removal process. This lock is often caused by the VMDK still being attached using Virtual Appliance transport mode or the VMDKs being locked by the ESXi host.

For more information refer below articles:

Consolidating/Committing snapshots in VMware ESXi
Investigating virtual machine file locks on ESXi hosts

Resolution

To resolve this issue, verify if the virtual machine is running from the base snapshot and then delete the snapshot.

Confirm the virtual machine disk state:

1. Right-click the affected virtual machine and select Edit Settings.

2. Select the hard disk and note the datastore where it is located.

3. Check if it points to the base disk (VM_Name.vmdk) or a snapshot disk e.g  (vmname-000001.vmdk).

Delete the snapshot via vSphere Client:

1. Right-click the virtual machine and select Snapshot.

2. Select Take Snapshot.

3. Enter a name for the snapshot and take a regular snapshot.

4. After the snapshot completes, right-click the virtual machine and select Snapshot > Snapshot Manager.

5. Select the snapshot and click Delete All (Consolidate).

Delete the snapshot via ESXi command line:

1. Log in to the ESXi host as root using SSH.

2. Run vim-cmd vmsvc/getallvms to list all virtual machines running on the host and determine the VMID.

3. Run vim-cmd vmsvc/snapshot.get VMID to list the snapshots for the specific virtual machine.

4. Run vim-cmd vmsvc/snapshot.create VMID snapshot_Name Snapshot_Description Memory_Included Snapshot_Quiesced to create a snapshot. Do not close the SSH session during this process.

5. Run vim-cmd vmsvc/snapshot.removeall VMID to commit the snapshot.

Additional Information

• The time required to remove the snapshot might increase if the snapshot size is large.

• This issue can occur even after migrating the virtual machine with snapshots using Storage vMotion. When migrating a virtual machine with snapshots using Storage vMotion, the summary tab references both the old and new datastores because the snapshot still has references to the old datastore.