In this series of articles, I would show how to make Azure Site Recovery work with Hyper-V- step by step.
This is part 5 of the series.
Now it’s time to test the failover.
At this point of time, we have enabled on-premise virtual machine to be protected during times of disaster. The relevant metadata, virtual machine configuration and vhd has been replicated to azure storage created for this purpose. There is no Virtual machine created during this point of time on Azure for this on-premise virtual machine. Only at the time of disaster, a new virtual machine would be provisioned on Azure as a replica of the original on-premise virtual machine. The on-premise virtual machine keeps sending its current state every 30 second/5 min/30 min as part of continuous replication.
When we start the process of testing the failover, a new cloud service and virtual machine is created on Azure and then the same can be accessed by the users.
Under Protection group | Proddrprotectiongroup | Virtual machines, select the DRVM virtual machine and click on “Test Failover” button available at the bottom of the page.
On the resultant popup, select none for network since we did not create any network and click complete button. This would start the process of testing the failover.
Following are the steps undertaken by Azure site recovery service to test the failover.
And now if we navigate to the virtual machine section, we will find that a new virtual machine named “DRVM-test” is being provisioned.
And within the same storage account provisioned earlier, a new container created storing containing the vhd blob for the test virtual machine. This is shown below.
We can also see that a cloud service is created to host our test virtual machine.
Now, we can work with this new virtual machine the same way we would have worked with on-premise virtual machine. However appropriate endpoints needs to be opened for making this work.
After you have tested the virtual machine comprehensively it time to complete the test.
TO complete the failover test, navigate to the job specific to failover test from Azure site recovery service -> jobs and click on “Compete test” button when status is shown as “waiting for action” for complete testing step.
This would popup another window asking whether clean up should happen as part of completing the test. If we mark yes for this checkbox, the cloud service, virtual machine, storage container and blob file created earlier would be deleted and azure would be back to its original state. During the entire failover test process, the on-premise virtual machine can continue to run without any downtime.
The end status after completion of test should look like below
In next part (part-6), we will continue with the step by step guide and perform an actual failover.