Shared VMFS datastore becomes inaccessible, reported metadata corruption on the volume.
search cancel

Shared VMFS datastore becomes inaccessible, reported metadata corruption on the volume.

book

Article ID: 453879

calendar_today

Updated On:

Products

VMware vSphere ESX 8.x VMware vSphere ESXi

Issue/Introduction

  • During an upgrade or maintenance activity a datastore becomes inaccessible

  • Shared VMFS datastore may become inaccessible, displaying 0MB total and free capacity in the vCenter Server UI.

  • Virtual machines residing on the affected datastore appear as "Inaccessible"

  • Unable to storage vMotions VMs off the impacted datastore. 

Environment

VMware vSphere ESXi 8.0

Shared VMFS datastores 

Cause

The VMFS datastore has been overwritten due to an ESXi installation being inadvertently performed on a LUN containing existing data.

The installation process reformats the volume with a new filesystem layout, replacing the device's previous partition table and filesystem structure.

Resolution

  1. Restore from Backup
    Because the reformat operation overwrites the original VMFS partition table and filesystem markers, access to the VMs cannot be recovered using the VOMA.


    Alternatively:
  2. Revert to a Snapshot LUN
    If virtual machine backups are unavailable or incomplete, work with your storage administrator to mount a snapshot of the backing LUN created before the reformat occurred. Any data written since the snapshot LUN was taken will be lost.

  3. Third-Party Data Recovery
    If neither backups nor LUN snapshots are available, you can to engage a third-party data recovery service. Data recovery services for data not recoverable by VMware Technical Support- VMware vSphere, vSAN