Replicating Virtual Machines (UVB 8.0)

SUMMARY

Replicating Virtual Machines (UVB 8.0)

ISSUE

Overview

UVB's Replication feature allows you to create replica VMs of your existing production VMs - and keep the two in sync - any time the production (source) VM changes, they will be applied to the replica (target) VM the next time the replication job runs.

Replica VMs are initially created and kept up to date using backup data -- creating replicas in this manner avoids the need to create additional snapshots reducing the load on your production environment. Each time the selected backup or archive job runs it kicks off the replication job and writes only the changed data to the replica VM.

Figure - VMS replicated from one environment to another

User-added image

 

Configuring Replication

To configure replication, you will use the Replication Dialog, accessible from either the Protect Page or Jobs Page of the web interface. You begin by selecting an existing backup or archive job which creates the backup data of your source VMs.

In the replication job dialog, you select that job and the VMs in that job you'd like to replicate, along with specified replication target environment details. When the Replication job is successfully created, each completed VM backup or archive will then kick off its replication work.

You can replicate within the same environment or between the environments you create within UVB -- as long as network connectivity exists between both and the VBAs have access to the source and target storage data stores.

The VBA that performs the actual replication depends on the target environment selected. If you're replicating to the same environment, in most cases, the VBA that performed the backup will also perform the replication. If you're replicating to another environment, the appliance with access to the source backup data and target location (host and storage) will perform the replication. If more than one appliance is available, the appliance with the least amount of replication work currently assigned will be assigned to perform the replication.

Note: For an appliance to be considered as a replication appliance, it must have the Engine role enabled. To enable the Engine role you can either add a backup data store or you can connect a read-only mount to the backup data store of the source appliance.

Replication Jobs can be created from both the Protect and Jobs pages. For more information on how to replicate a VM, see Create Replication Job dialog (8.0).

Note: If you choose to edit a replication job after it has run, the edit dialog for the individual VM will automatically be set to Existing VM. Leave this selection unchanged if you want to continue replicating to that same VM.

Tip: Create replication jobs based on an archive job that runs after your primary backup and copies data to the replication DR site. By creating your replicas based on your Archived backups you can reduce the load on your network by relying on the much more efficient archived data to be transferred across the WAN as opposed to writing replica data across the WAN each time.

Failover and Failback: Recovering From Disaster

With Unitrends Virtual Backup configured to backup and replicate your virtual machines, you are well protected in the event of a DR scenario. If your primary site becomes unavailable, you have multiple options available for recovering your business critical data.

Figure -  An unavailable primary site with VMs powered on at the DR location.

User-added image

When your primary location is unavailable, you can power on the replica VMs at your DR site to continue your day to day operations. When your primary site is recovered, or when you're ready to populate a new production site, you can use Unitrends Virtual Backup to fail back to your new location, by either seeding the servers at the new site, or restoring backups of the failed over VMs into your production environment.

 

Figure - Failing back to a primary site from a DR location

User-added image

When the primary location has been completely recovered (failover and failback has occurred), you can begin replicating to your DR site, once again.

 

Notes

  • In the event you must fail back (backup replica VM then replicate back to primary), before powering on replica VMs, select the restore point snapshot to roll back to (if applicable) then remove any unneeded restore point snapshots (preceded with phdvbreplica) before back ups begin. If you do not remove the UVB snapshots, the backup will not complete.
  • UVB will not replicate to a VM that is powered on. Subsequent jobs will fail with the message Cannot replicate to a VM that is running.
  • Making changes to source VMs may require you to remove any snapshots on the replica VMs to allow changes to be applied. For example, if you add or exclude a new disk to your source VM, you must first remove any snapshots from the replica VM for the disk changes to be applied to the replica. Failure to remove snapshots may result in error similar to Failed to reconfigure VM: Cannot complete the operation....
Was this article helpful?
0 out of 0 found this helpful
Have more questions? Contact us