Usage of vSAN datastore unexpectedly increases after a planned migration with vSphere Replication

Products

• VMware Live Recovery
• VMware vSAN
• VMware vSphere ESXi

Issue/Introduction

Symptoms:
After a planned migration or test recovery with vSphere Replication is performed, the usage of the vSAN datastore increases unexpectedly under the following conditions:

• The target datastore is a vSAN datastore in the target site for the target VMs.
• Object Space Reservation > 0 is configured in the storage policies attached to the target VMs.
• MPIT (Multiple points in time) snapshots are configured for the VMs.

For example, if Object Space Reservation in the storage policy is set to 100 (= 100%), the usage of the vSAN datastore by the target VM becomes double compared to the original VM.
If the vSAN datastore in the target site does not have extra space to accommodate the double sized VMs, a planned migration or a test recovery can fail.

Environment

VMware vCenter Site Recovery Manager 6.5.x
VMware vSphere Replication 6.x
VMware vSAN 6.x
VMware vSphere Replication 8.x
VMware vCenter Site Recovery Manager 8.x
VMware vCenter Site Recovery Manager 6.0.x

Cause

This is an expected behavior of vSphere Replication.

• With MPIT snapshots configured, a hbrimagedisk vmdk is created when a planned migration or a test recovery is performed.
• The storage policy of a hbrimagedisk vmdk inherits its parents' storage policy. So a hbrimagedisk vmdk can consume the same amount of space as its parent's vmdk does.

Resolution

Please remove the Object Space Reservation rule from the storage policy attached to the target VM's vmdk. 
For example,

1. Open "VM Storage Policies" in the Web Client (Flex Client).
2. Select the storage policy that is applied to the VM.
3. Click "Edit" button
4. Remove "Object space reservation (%)" rule from the rule-set list.
5. Click "OK" to save the changes.

To remove the hbrimagedisk vmdk, please remove any snapshots of the target VM in the snapshot manager in the UI after a planned migration.

Additional Information

Impact/Risks:
This issue might cause unexpected vSAN datastore fullness. This issue can lead to a planned migration or a test failover to fail.