How Resiliency Platform Data Mover handles DR operations
Migration refers to a planned activity involving graceful shutdown of virtual machines at the source data center and bringing virtual machines up at the target data center and vice-versa. In this process, replication ensures that consistent virtual machine data is made available at the source and target data center.
When you migrate a resiliency group, the Data Mover Connection Status should be Connected, Data State should be Consistent, and Replication State should be Active. When you initiate a Migrate operation, Veritas Resiliency Platform checks whether the virtual machines and source data center Replication Gateways are up. The migrate operation involves stopping virtual machines, detaching disks from virtual machines and attaching them to the gateway.
The migrate operation is similar to the recover operation, except that the migrate operation is used when the virtual machines can be gracefully shut down. The recover operation is used in case of disaster, when the source virtual machines are not reachable. The migrate operation involves shutting down the virtual machine from within the guest machine, whereas, the recover operation involves only stopping the virtual machines.
There is difference between migrate and recover operations. The migrate operation allows reverse replication, whereas the recover operation does not allow reverse replication. After recover, the resync operation is required to bring back data or workload onto the source data center.
recover is an activity initiated by a user when the source data center is down due to a disaster, and the virtual machines need to be brought up at the target (recovery) data center to provide business continuity.
The replication includes any changes to the boot disks of the virtual machines. As a result, bootable copies of the source virtual machines can be brought up on the recovery data center.
Since it is an unplanned event, the data available at the recovery data center may not be up-to-date. You need to evaluate the tolerable limit of data loss. If the available data is within the acceptable limits, perform the recover operation to bring up the source workloads in the recovery data center. The recover operation provisions and brings up the virtual machines at the recovery data center using the latest data on the recovery data center storage.
The recover activity operates on an entire resiliency group, even if the disaster affects only certain workloads in the resiliency group.
After a recover, the virtual machine in the recovery data center runs the application and writes to the storage in the recovery data center.
A recover operation as a result of a disaster event moves the workloads from a source (production) data center to a target (recovery) data center. You must plan for how to restore the workloads back to a production data center once it is up and running.
The first step in the process is to use the resync operation to ensure that the data in the target data center is synchronized to the source data center storage. The resync operation performs the full synchronization between the storage on the target and the source data centers. After the synchronization is complete, you can use the migrate operation to restore the virtual machines to the source data center.
Both the resync and migrate operations are performed for an entire resiliency group.