Esxi host datastore unmount issue

Products

• VMware vSphere ESXi

Issue/Introduction

You are unable to unmount or delete a decommissioned VMFS datastore from a cluster. Despite the volume being empty and the ESXi hosts being in maintenance mode, the task fails.

You observe the following error in the vCenter UI: Unmount VMFS volume failed

 You observe the following error in the vCenter vpxd.logs:

99#543:--> The resource 'Datastore Name: <datastore-name> VMFS uuid: ########-bbbb5093-####-############' is in use."
99#559:-->                value = "Datastore Name: <datastore-name> VMFS uuid: ########-bbbb5093-####-############' is in use"
99#566:-->          message = "Cannot unmount volume 'Datastore Name: <datastore-name> VMFS uuid: ########-bbbb5093-####-############' is in use' because file system is busy. Correct the problem and retry the operation."

Environment

vSAN 8.0 U3

ESXi 8.0 U3

VMFS 6

Cause

This issue occurs because the vSphere HA (High Availability) Primary node holds a heartbeat lock on the datastore. Even if the volume is empty, the HA agent maintains this lock for cluster health monitoring

 

info vpxd[08003] [Originator@6876 sub=DAS opID=mjr4344b-5814955-auto-3gmul-h5:70440533-8d] Send a new host list if the datastore /vmfs/volumes/########-bbbb5093-####-############ is used for heartbeating in cluster [vim.ClusterComputeResource:domain-########,2012 <Cluster-Name>]

Resolution

Reconfigure HA in the cluster to specify a new datastore, avoiding the datastore you are trying to unmount:

1. Log in to the vCenter UI.

2. Navigate to Cluster > Configure > vSphere Availability.

3. Click Edit, then select the Heartbeat Datastores tab.

4. Select "Use datastores only from the specified list."

5. Uncheck the problematic datastore and click OK.

6. Right-click the datastore and select Unmount Datastore from all hosts.

7. (Optional) Once removed, you may revert the Heartbeat setting to "Automatically select datastores."