In the last post How to configure Azure Site Recovery, perform test failover and invoke disaster recovery failover – Part 1 – TechPress i had explained on how you can configure Azure Site Recovery and configured the Virtual network for Test Failover and Failover. When you initiate Test Failover, Site Recovery Service will use that network to start the protected servers in DR site.
If you have never performed a test failover or the last test failover was failed. A warning message will be shown but you can proceed with Failover by accepting the risk and continue with the failover. However, Its recommended to perform a Test Failover before performing Actual Failover or invoking DR to validate the disaster recovery solution. In this blog post, we will see how to perform a Test Failover and see what happens when this process is initiated.
For starting the Test Failover, Open the recovery services vault -> Replicated Items -> click on the Replicated Item or Protected Server and click on Test Failover
After you click on Test Failover, below screen will show. You will not be able to change the direction of the test failover, however you can choose a recovery point.
Depending upon the scenario you can choose the best option as per your requirement. Latest provide lowest RPO because it first processes all the data send to site recovery service and creates a recovery point before failing over which means it will take more time for the failover but minimizes the data loss. Latest Processed provides low RTO as there is no time spent processing unprocessed data therefore failover process is quick. To check the Latest Recovery Points, click on Latest Recovery Points from the overview page of the replicated item. [Click on below image to enlarge].
Now, I have selected the option Latest processed (low RTO) to Failover as quickly as possible. You can choose the option best suited for your requirement. Click on OK to proceed.
You will see a notification of starting the Test Failover Process:
After few minutes, you can see virtual machine tp-dc1-test is created. You can click on this virtual machine and check its IP Address, Virtual Network Information and you can assign a public IP Address to the Virtual machine and RDP to it for testing the application. As this VM is running in an isolated network, this will not have any impact on production environment even if the virtual machine is having the same name / domain etc.
Once you have connected to tp-dc1-test server and verified / tested the application. You can cleanup the failover to delete the virtual machine. For Cleaning up of the Test Failover. You can go back to the Recovery Services Vault -> Replicated Items -> Click on the Protected Server for which you had initiated the test failover -> click on Cleanup test failover.
After you click on cleanup test failover, you will get below screen. you can write something in the Notes and select the checkbox “Testing is complete. Delete test failover virtual machines(s).” Click OK to proceed. This will delete the Virtual Machine(s) created during test failover.
Within few minutes the Virtual Machine tp-dc1-test is deleted.
Go back to recovery services vault -> Replicated Items -> click on the protected server. Now you will see that the Last successful Test Failover shows successful with the record of date and time. Also, calculated the RPO which you can check on below page:
Invoke DR or Actual Failover when primary region is down
Please check the next part of Azure Site Recovery (ASR) series of blog posts to learn more about how to invoke DR / failover to UK West.
