After performing a fail over or a migrate, return the workload from the destination site back to the original source site by reversing the replication.

After performing fail over or migrate, the workload runs in the destination site. Performing a subsequent reverse task replicates the failed-over or migrated workload data to the source workload.
Optimized reverse:
VMware Cloud Director Availability skips performing a full synchronization back to the original source workload when performing a reverse task by replicating only the deltas.

Optimized reverse works only if the original source workload is not powered-on since the initial migrate and when no blocks changed in the original source and the original source disks are not modified in any way.

Optimized reverse is available for limited time after performing migrate, by default, for a week. Under Details of the failed-over replication, see the Optimized reverse expiration time. After this time expires, or if the source workload is powered-on, reversing the replication skips optimized reverse and performs a full synchronization.

  • Optimized reverse works only with replications using the Classic data engine. For more information, see Activate the data engines for replicating workloads.
  • For vSphere DR and migration, the reversed replications always starts with a seed. As a result, the original virtual machine is deleted at the new destination. Also, All Actions > Recovery settings of the reversed replication pre-populate with Recovery settings from the original virtual machine with the test networks being the same as the failover networks.
  • When reversing a replication from a Cloud Director site back to an on-premises site, VMware Cloud Director Availability uses the original datastore for the placing the virtual machines, regardless of the current on-premises local placement setting.


  • Verify that in the destination datastore, at least double the allocated storage of the virtual machine is available for a successful reverse operation. For information about the storage requirements, see Storage Space Consumption in the Destination.
  • Verify that VMware Cloud Director Availability 4.3 or later is deployed in both the source and destination sites for optimized reverse.
  • Verify that you can access VMware Cloud Director Availability as a tenant or as a provider. For more information, see Accessing VMware Cloud Director Availability.
  • Verify that the replication is in a Failed-Over recovery state before you can start a reverse task. For optimized reverse, ensure that the replication is migrated. For more information, see Failover of a Replication or Migrate a Replication.
  • Verify that the number of disks in the seed virtual machine matches that of the source virtual machine. Performing a reverse task with mismatching configuration of disks fails with the Disks of provided seed VM don't match the disks of the source VM message. For more information, see Selecting Replicated Disks.


  1. In the left pane, choose a replication direction by clicking Incoming Replications or Outgoing Replications.
  2. Select an exiting replication that is failed-over and click All actions > Reverse.
    Optimized reverse requires already migrated replications. Alternatively, when the source workload is powered-on, reverse performs full synchronization.
  3. In the Reverse window, to confirm the reversal of the replication, click Reverse.

    Reversing the replication enables the replication traffic and recovers the workload back to the original source site.

    The Last changed column shows the reverse task progress in percentages. After reversing a replication, the direction of this replication reverses. To see the reversed replication:
    • After reversing an incoming replication, in the left pane, click Outgoing Replications.
    • After reversing an outgoing replication, in the left pane, click Incoming Replications.
  4. (Optional) In the bottom, to monitor the task progress click the Tasks tab.


After the reverse task completes, the Recovery state column of this replication shows Reversed and the reversed replication overwrites the original source workload. The reversed workload runs in the destination site, while protected in the original source site.

What to do next

  • You can test, fail over, or migrate the reversed workload back in the original source site. For more information, see Test Failover a Replication, Failover of a Replication, or Migrate a Replication.

    When any of those tasks completes, the Recovery state column of this replication shows a green Failed-Back state. Then, after failing-back a reversed replication you can only perform a reverse task.

  • You can pause the reversed replication and edit the replication configuration. You can permanently stop the traffic of this replication and remove it with all retained replication instances by clicking All actions > Delete replication.