To prepare a Replicator Service instance for maintenance without disrupting replications, you can evacuate the incoming replications from the Replicator Service instance to other local Replicator Service instances in the cloud site.
Note:
- Enter Maintenance Mode auto-rebalances the replications in the local site.
- Exit Maintenance Mode does not perform auto-rebalancing. To rebalance the replications after exiting maintenance mode, click Rebalance. For more information, see Rebalance the replications across the Replicator Service instances.
- In the local site, placing the Replicator Service instance in maintenance mode migrates all incoming cloud replications to other Replicator Service instances in the local site. Also, VMware Cloud Director Availability migrates all incoming and outgoing replications from and to on-premises sites.
- In the remote site, migrate the remaining outgoing cloud replications from this Replicator Service instance to other Replicator Service instances. Log in to the remote site and place in maintenance mode the same Replicator Service instance. Repeat this step in each remote site, where this Replicator Service instance is remotely registered.
New replications are placed on Replicator Service instances that are not in maintenance mode.
Prerequisites
- Verify that VMware Cloud Director Availability is successfully deployed in the cloud site.
- Verify that more than one Replicator Service instance is operational in the cloud site.
- Verify that the clean-up task is complete after using a test failover for any incoming replication. If the Replicator Service contains a test failed over virtual machine, attempting to enter a maintenance mode shows a Operation aborted due to an unexpected error message. Before entering maintenance mode, you must perform a test cleanup on the test failed over virtual machine or vApp.
Procedure
Results
What to do next
After performing the maintenance operations, in the local site click Exit Maintenance Mode. To repopulate the Replicator Service instance with replications, you must rebalance the replications. For more information, see Rebalance the replications across the Replicator Service instances.