freshidea - Fotolia

Evaluate Weigh the pros and cons of technologies, products and projects you are considering.

How does a Hyper-V Replica failover work?

Microsoft offers both planned and unplanned Hyper-V Replica failover options. While the use cases seem clear, admins should know the functional differences.

In a Hyper-V Replica scenario, there are two types of VMs involved, the primary VM and replica VM. The primary VM, which is running on a Hyper-V server, replicates changed contents to the replica VM running on a Hyper-V Replica server. The replica VM is designed to be able to pick up operations if the primary VM fails. In most cases, the replica VM cannot be turned on unless the primary VM is turned off. However, there are times you may want to bring the replica VM online -- to perform maintenance on the primary VM, for example. Hyper-V Replica helps you achieve this by offering planned and unplanned failovers. Usually, a planned Hyper-V Replica failover can be achieved by executing the "planned failover" action from the right-click context menu on the primary VM.

When you perform a planned failover on a primary VM, there are number of prerequisite checks performed by the primary Hyper-V Server, including one to make sure the primary VM is turned off and another to make sure the configuration allows reverse replication. Once the prerequisites are met, the planned failover process begins with the primary server sending delta data that has not been replicated yet.

Since this is a planned failover, there is no data loss. The Planned Failover wizard makes sure to replicate the delta data to the replica server before it takes over.

An unplanned failover works differently. It is executed at the replica VM, and there is a chance that you will lose data that had not been replicated when the primary VM failed. To execute an unplanned failover, all you need to do is right-click on the replica VM and click on the "Failover" action. You then will be presented with a list of recovery points that you can apply to bring the workloads up on the replica server.

Planned Failover is executed in cases of test or maintenance activities at the primary server, and helps bring up virtualized workloads at a disaster recovery site by copying the nonreplicated contents. Unplanned failover is used as a disaster recovery option, usually after the primary VM has failed, which may result in data loss.

Dig Deeper on Disaster recovery, failover and high availability for virtual servers

Start the conversation

Send me notifications when other members comment.

Please create a username to comment.