Optimized reprotect reduces the time needed for a reprotect operation.

After you perform the recovery and before you power on the new recovered VM, vSphere Replication prepares to track the changes, which occur on the recovered VM. During recovery, vSphere Replication creates a Persistent State File (PSF) for each disk of the recovered VM. The PSF files are used to track the changes on the disks, which helps to omit the initial sync during reprotect. If you delete the PSF files, the optimization process is interrupted and the reprotect operation switches to full sync operation.

If you are using Thin Provisioning and do not perform reprotect within the configured period (See the Configuring the Optimized Reprotect section), vSphere Replication removes the PSF files and any further reprotect operation triggers a full sync operation.

When using a VMware vSAN datastore, once the VM is selected for Site Recovery Manager protection, the PSF files are created per vmdk by vSphere Replication. For VMware vSAN datastores, the PSF file provisioned type depends on the configured VM namespace storage policy. If the namespace storage policy is set to Thick Provisioning, the PSF files are required and cannot be removed.

You cannot use optimized reprotect to initial sync with seed disks.

Note: You can only reprotect to the original protected site. You cannot use optimized reprotect with Disaster Recovery workflow. You can use optimized reprotect only after a planned migration Site Recovery Manager workflow.

Calculating the PSF file size

Below is calculation how the PSF file size is determined based on how much disk is allocated to any VM.

For Thin Provisioning the determing factor is the changed block size and the PSF file size is calculated as 512 bytes file header + 2 x RoundupTo2KB(vmdkSize/extentSize/8) + 512 bytes demandlog header + 512 x [changed block size] / [extent size] + [changed block size].

For Thick Provisioning the determing factor is the vmdk size. The PSF file size is calculated as 512 bytes file header + 2 x RoundupTo2KB(vmdkSize/extentSize/8) + 512 bytes demandlog header + 512 x [vmdk size] / [extent size] + [vmdk size]. By default, for vmdks less than 2TB, the extent size is 8192.

Optimized Reprotect Support

Optimized Reprotect depends on the type of the target datastore and the quiescing option.

Target Datastore Quiescing ESXi Version
VMFS OFF For all host versions, the reprotect operation is optimized.
VMFS ON

For ESXi 7.0 and ESXi 7.0 Update 1, if optimized reprotect is enabled, the reprotect operation fails. You must deactivate the optimized reprotect feature to restore the reprotect operation. Set the reprotect-optimization-enabled property to false. See Configuring the Optimized Reprotect section below.

For all other host versions, the reprotect operation is optimized.

vSAN or Virtual Volumes OFF or ON

For ESXi 7.02 or later versions, the reprotect operation is optimized.

For earlier ESXi versions, you must deactivate the optimized reprotect feature. Set the reprotect-optimization-enabled property to false. See Configuring the Optimized Reprotect section below.

Configuring the Optimized Reprotect

Use the VRMS configuration properties in /opt/vmware/hms/conf/hms-configuration.xml to modify the behavior of your environment during reprotect.
Table 1. VRMS Configuration Properties for the Optimized Reprotect
Property Description Default Value
reprotect-optimization-enabled Activate or deactivate the optimized reprotect option. true
reprotect-optimization-time-window-mins The period for which the replication stays in optimized mode (measured in minutes). When this period is passed, reprotect triggers a full sync. 10080
reprotect-optimization-monitor-period-mins The period for which VRMS cleans expired data related to optimized reprotect, when the reprotect optimization time window is expired (measured in minutes). 60