VKS cluster worker nodes fail to move to ESXi host after exiting maintenance mode
Article ID: 428336
Updated On:
Products
Issue/Introduction
Removing a host from maintenance mode on a Supervisor Cluster does not move VKS cluster worker nodes to it.
Symptoms:
-
Workload Management > Supervisor > Monitor page reports the below error for the ESXi host that was removed from maintenance mode:
Failed to find any accessible datastores for storage policy #######-########-####-############ datastore URIs: [ds:///vmfs/volumes/#######-########-####-############/] -
VKS worker nodes remain on existing hosts and DRS does not migrate them to the new host.
-
SSH session on the ESXI host shows the datastore UUID mentioned in the error is not visible in
/vmfs/volumes/directory.
Environment
VMware vSphere with Tanzu
VMware vSphere Kubernetes Service
Cause
The storage policy used by the Supervisor Cluster requires the shared datastore to be mounted and accessible on every host within the cluster.
If a host is added or removed from maintenance mode without the datastore being correctly mounted, VKS worker nodes cannot be scheduled on that host due to storage inaccessibility.
Resolution
Perform a storage rescan to ensure the required datastore is mounted on the affected host.
-
Select the affected ESXi host in the Inventory.
-
Navigate to Configure > Storage > Storage Adapters.
-
Click Refresh, then click Rescan Storage.
-
Verify the mount state via SSH on the ESXi host :
ls -l /vmfs/volumes | grep <DATASTORE_NAME>
Once the datastore is accessible, the host status should transition to Ready without warnings.
If the datastore remains inaccessible or the symlink is missing in the /vmfs/volumes directory, please raise a Support Request.
Additional Information
Feedback
