Test Failover

1 February 2024

As a best practice, regularly test and update your DR plan to ensure that it is effective in the event of an actual disaster. Testing involves checking if the DR plan works as expected and updating the DR plan with any changes that may have occurred since the last failover. During a Test Failover, the application is failed over to the DR site without affecting the VMs on the primary site. Once the testing is complete, you can tear down resources created during the Test Failover. Cohesity recommends initiating a Test Failover in an isolated network (with no network reachable route to the source) by configuring the resource profile appropriately.

Cohesity recommends initiating a Test Failover in an isolated network (with no network reachable route to the source) so that the primary VMs are unaffected.

Initiate Test Failover

To perform a Test Failover:

  1. In SiteContinuity, select DR Plans.

  2. In the Disaster Recovery Plans page, choose a DR Plan for testing the failover. A DR Plan must be in the Failover Ready state to test the failover.

  3. Click the Actions menu (⋮) of the DR plan, and select Test Failover. The Readiness Test page displays the target DR site and the target Cohesity cluster:

  4. From the Resource Profile drop-down, select a Resource Profile.

  5. (Optional) If you want to test the failover of the VMs and their disks from the source to the target datastore when the VMs are running, select Perform Storage vMotion.

  6. For recovery, the latest snapshot is used by default. Verify the displayed snapshots or click Edit to choose a different snapshot, and click Apply.

  7. Click Test Failover.

On the Disaster Recovery Plans page, the Checks column displays a beaker icon that indicates the result of the most recent Test Failover or Test Failback. When you click Test Failover, the Test Failover starts and shows a progress symbol on the icon (), signifying that the Test Failover is underway.

Validate Test Failover

To validate a test failover:

  1. Log in to the DR vCenter.

  2. Check if the VM instances come up in the order defined in your DR plan.

  3. Confirm that the allocated resources for each VM are correct. Additionally, verify that the network configurations, including IP addresses and connectivity settings, match the intended setup.

  4. If an isolated network is used, test the accessibility of the VMs within this network. Ensure that the VMs can communicate with each other as intended.

  5. Test the applications hosted on the recovered VMs. Verify that the applications function as expected and can access necessary resources and services.

By following these steps, you can effectively validate the test failover, ensuring that your applications and systems are ready for any actual DR scenario.

Next Steps

If the Test Failover was successful and the beaker icon displays a green tick (), you can click the Actions menu (⋮) corresponding to the DR plan and perform the following tasks:

If the Test Failover fails, the beaker icon shows an error symbol (). Click the icon to see the error message. To explore possible causes and resolve the error, see the Troubleshooting section.