Skip to main content
BlueXP disaster recovery

Fail over applications to a remote site

Contributors amgrissino
PDFs

In case of a disaster, fail over your primary on-premises VMware site to another on-premises VMware site or VMware Cloud on AWS.

During the failover, the most recent SnapMirror Snapshot copy is used. Or, you can select a specific Snapshot copy from a point-in-time Snapshot copy (per the retention policy of SnapMirror). The point-in-time option can be helpful if you are facing a corruption event such as ransomware, where the most recent replicas are already compromised or encrypted. BlueXP disaster recovery shows all available points in time.

This procedure breaks the replication relationship, places the vCenter source VMs offline, and enables read/write on the target site.

You can include custom scripts in .sh, .bat, or .ps1 format as post failover processes. With custom scripts, you can have BlueXP disaster recovery run your script after a failover process. For example, you can use a custom script to resume all database transactions after the failover is complete.

Before you start the failover, you can test the process, ensuring success when you need it. The test does not place the virtual machines offline.

Test the failover process

Before you start an actual failover, you should test the failover process.

During a failover test, virtual machines are temporarily created. BlueXP disaster recovery does not map the target volume. Instead, it makes a new FlexClone volume from the selected Snapshot, and a temporary datastore backing the FlexClone volume is mapped to the ESXi hosts.

This process doesn’t consume additional physical capacity on on-premises ONTAP storage or FSx for NetApp ONTAP storage in AWS. The original source volume is not modified and replica jobs can continue even during disaster recovery.

When you finish the test, you should reset the virtual machines with the Clean up test option. While this is recommended, it is not required.

A test failover operation does not impact production workloads, the SnapMirror relationship used on the test site, and protected workloads that must continue to operate normally.

Steps

  1. From the BlueXP left nav, select Protection > Disaster recovery.

  2. From the BlueXP disaster recovery top menu, select Replication plans.

  3. Select the replication plan.

  4. On the right, select the Actions option Actions menu icon in BlueXP disaster recovery service and select Test failover.

  5. In the Test failover page, enter “Test failover” and select Test fail over.

  6. After the test is complete, clean up the test environment.

Clean up the test environment after a failover test

After the failover test finishes, you should clean up the test environment. This process removes the temporary VMs from the test location, the FlexClones, and the temporary datastores.

Steps

  1. From the BlueXP disaster recovery top menu, select Replication plans.

  2. Select the replication plan.

  3. On the right, select the Actions option Actions menu icon in BlueXP disaster recovery service and select Clean up failover test.

  4. In the Test failover page, enter “Clean up failover” and select Clean up failover test.

Fail over the source site to a disaster recovery site

In case of a disaster, fail over your primary on-premises VMware site on demand to another on-premises VMware site or VMware Cloud on AWS with FSx for NetApp ONTAP.

The failover process involves in the following operations:

  • If you selected the latest Snapshot, the SnapMirror update is performed to replicate the latest changes.

  • The source virtual machines are powered down.

  • The SnapMirror relationship is broken and the target volume is made read/write.

  • Based on the selection of the Snapshot, the active file system is restored to the specified Snapshot (latest or selected)

  • Datastores are created and mounted to the VMware or VMC cluster or host based on the information captured in the replication plan.

  • The target virtual machines are registered and powered on based on the order captured in the Resource groups page.

  • The SnapMirror relationship is reversed from target to source virtual machine.

Tip After the failover starts, the recovered VMs can be seen in the vCenter of the disaster recovery site (virtual machines, networks, and datastores). By default, the virtual machines are recovered to the Workload folder.
Steps

  1. From the BlueXP left nav, select Protection > Disaster recovery.

  2. From the BlueXP disaster recovery top menu, select Replication plans.

  3. Select the replication plan.

  4. On the right, select the Actions option Actions menu icon in BlueXP disaster recovery service and select Fail over.

  5. In the Test failover page, enter the replication plan name to confirm and select Fail over.

  6. Choose the Snapshot for the datastore from which to recover. The default is the latest.

  7. To check the progress, in the top menu, select Job monitoring.