Thick-provisioned VMs on vSAN detected on vSAN-health check

Products

• VMware vSAN

Issue/Introduction

This health alarm is flagged because vSAN has detected thick-provisioned VMs even though the applied vSAN storage policy has object space reservation set to 0 (Thin-provisioned). It is recommended to reclaim the underlying storage from 'Thick' to 'Thin' which will improve storage utilization.

This health alarm will not check VMs deployed by ESX Agent Manager (EAM) as most of the VMs deployed as thick-provisioned by default or VMs with disabled tasks (vm.disabledMethod), for example, NSX Controllers.

Symptoms:

vSAN online health check reports warning for thick disks provisioned on virtual machines.

Below Alarm is reflected on the Cluster in vcenter UI.

Environment

VMware vSAN 6.7.x
VMware vSAN 8.0.x
VMware vSAN 7.0.x

Cause

This may be caused by the VMs being backed up from a SAN and restored to vSAN, regular storage vMotion shall always provision VMs with thin disks.

Resolution

If the VM should be using "Thick provisioning", please apply a Storage Policy to the VM with the vSAN option "Object space reservation" set to 100 (Thick provisioning).

If the VM should be using "Thin provisioning", see the workarounds below.

Workaround:

Option 1:

Schedule downtime for the VM and perform a clone within the vSAN datastore. The newly created destination VM should have thin VMDK(s). A back up is recommended as best practice.

It is recommend to perform a final check on the cloned VM and then delete the older respective VM from vSAN datastore, which should auto reclaim space upon deletion.

Option 2:

Perform a Storage vMotion of the VM onto a VMFS or NFS datastore as thin, then Storage vMotion it onto the vSAN datastore. See Migrating Virtual Machines with svmotion or more information.

Additional Information

Impact/Risks:
If the VMs should be in "Thick provisioning", there will be higher vSAN datastore usage.