Unable to reclaim local disks for VMFS after disabling a vSAN Cluster
Article ID: 425227
Updated On:
Products
Issue/Introduction
Symptoms:
-
After deactivating a vSAN cluster, you may see physical drives previously used as capacity or cache devices are not available for new VMFS datastore creation.
-
The vCenter UI may show the disks as "In use" or "Incompatible."
-
Running the command vdq -iH on the ESXi host identifies disks as still being part of a vSAN Disk Group.
-
Attempting to format the disk for a local VMFS datastore fails with errors related to existing partition tables or metadata.
Environment
VMware vSAN 7.x
VMware vSAN 8.x
Cause
The issue is caused by Persistent vSAN Disk Metadata and Partition Tables.
While the vSAN service may be disabled at the cluster level via the vSphere Client, the physical disks retain their low-level partition headers and membership metadata. The vSAN Original Storage Architecture (OSA) writes specific structures to the cache and capacity tier devices to ensure data integrity.
Because these headers are not automatically wiped during cluster deactivation, the ESXi storage stack continues to flag the devices as "In-use by vSAN." This creates a conflict that prevents the host from re-initializing the disks as blank devices for local VMFS volumes.
Resolution
To reclaim the disks, you must manually remove the vSAN disk group membership using the ESXi command-line interface (CLI).
1. Identify the vSAN Disks
Log in to the ESXi host via SSH and run the following command to identify the disks and their NAA IDs:
#vdq -iH
Locate the disks where the output shows "Reason: In-use by vSAN." Note the NAA ID of the SSD (Cache) device.
2. Remove the Disk Group
To decommission the entire disk group, run the following command targeting the Cache disk. Removing the cache device will automatically release the associated capacity disks in that group:
esxcli vsan storage remove -s <naaid_of_cache_disk>
Example: esxcli vsan storage remove -s naa.5000#######1234
3. Verify Disk status
Run the vdq -iH command again to ensure the disks are now recognized as "Eligible" or "Clean":
4. Initialize VMFS Datastore.
Once the partitions are cleared, you can proceed to the vSphere Client or use the CLI to create a new local VMFS-6 datastore using the reclaimed disks.
Additional Information
Feedback
