Migration involves transferring a healthy Primary of a Replicated Data Set (RDS) to a Secondary when the application involved in replication is inactive. Migration of a Primary to a Secondary is useful when the Primary must be brought down for maintenance or to make the application active on another node.
Use the Migrate task to migrate the role of a healthy Primary to a specified Secondary. The Migrate task performs the following functions:
For details about migrating the Primary role, refer to Transferring the Primary role.
To migrate a healthy Primary
- The Migrate task fails if the active Secondaries are not up to date and if the data volumes are active.
- By default, the Migrate task starts replication from the new Primary to the original Primary.
- If an RDS contains multiple Secondaries, VVR VEA does not automatically reconfigure the additional Secondaries.
- If required, you can reconfigure the additional Secondaries as Secondary hosts of the new Primary using the VVR command-line interface.
- The Data Status of the Secondary (attached RLINKs) must be consistent, up to date and the Replication Status must be connected (replicating).
- If the Migrate task fails on any of the hosts, the original configuration is restored.
- On migration, the Secondary becomes the new Primary.
- Because the application was stopped before the migration, an application recovery is not required. To start the application before enabling replication after migration, first, pause replication using the Pause task (Replication > Pause), start the application, and then resume replication using the Resume task (Replication > Resume).