Failover and failback asynchronous disks

This document describes how to failover and failback Asynchronous Replicationdisks.

In the event of an outage in the primary region, it is your responsibilityto identify the outage and failover restart your workload using the secondarydisks, in the secondary region. Asynchronous Replication doesn't offer outagemonitoring. You can identify an outage usingRPO metrics,health checks,application-specific metrics, and by contacting Cloud Customer Care.

Following a failover from the primary region to the secondary region, thesecondary region becomes the acting primary region.

After the outage or disaster gets resolved, you can initiate failback to startreplication from the original secondary region (the acting primary region) tothe original primary region. You can optionally repeat the process to move theworkload back to the original primary region. Moving the workload back to theoriginal primary region isn't strictly necessary, but can be done based ondisaster recovery requirements, such as locality or available resources.

To learn more about failover and failback, seeAbout Asynchronous Replication.

Failover to the secondary region

When you identify that a disaster has occurred, initiate failover to thesecondary region. A failover moves the workload from the primary region to thesecondary region. After the failover, the secondary disk is the acting primarydisk and the secondary region is the action primary region.

You can failover a single disk, or all disks in a consistency group.

Single disk

To failover a single disk, do the following:

  1. Stop disk replication.
  2. If you don't already have a VM in the same region as the secondary disk,create one.

  3. Attach the secondary disk to the VM:

    The secondary disk is now the workload's acting primary disk and thesecondary region is the acting primary region.

Consistency group

To failover a consistency group, do the following:

  1. Stop consistency group replication.
  2. If you don't already have VMs in the same region as the secondary disks,create them.

  3. Attach the secondary disks to the VMs:

Failback to the original primary region

After a disaster has resolved, initiate a failback to the original primaryregion. A failback configures and starts replication from the acting primarydisk to a new secondary disk in the acting secondary region.

You can failback a single disk, or all disks in a consistency group.

Single disk

To failback a single disk, do the following:

  1. Create a secondary diskin the acting secondary region. The acting secondary region is theoriginal primary region.
  2. Start replicationfrom the acting primary disk to the new secondary disk.

  3. Optional: Move the workload from the acting primary region to theoriginal primary region by doing the following:

    1. Wait for the initial replication to complete. The initial replicationis complete when thedisk/async_replication/time_since_last_replication metricis available in Cloud Monitoring.If you don't see the RPO metric in Cloud Explorer, that means theinitial replication isn't complete.
    2. Recommended: To avoid data loss, schedule downtime for the workloadand bring the workload offline.
    3. Stop replication.

    4. Attach the secondary disk to a VM:

      The secondary disk is now the workload's primary disk in the originalprimary region.

    5. Reconfigure replication in the original primary region by doing thefollowing:

      1. Create a new secondary diskin the original secondary region.
      2. Start replicationfrom the primary disk to the new secondary disk.

Consistency group

To failback a consistency group, do the following:

  1. Create a new consistency group in the acting primary region.The acting primary region is the original secondary region.
  2. Add the acting primary disks to the consistency group
  3. Create secondary disksin the acting secondary region that reference the acting primary disks.
  4. Start replication.

  5. Optional: Move the workload from the acting primary region to theoriginal primary region by doing the following:

    1. Wait for the initial replication to complete. The initial replicationis complete when theRPO metric is available..If you don't see the RPO metric in Cloud Explorer, that means theinitial replication isn't complete.
    2. Recommended: To avoid data loss, schedule downtime for the workloadand bring the workload offline.
    3. Stop replication.

    4. Attach the secondary disk to VMs:

      The secondary disks are now the workload's primary disks in theoriginal primary region.

    5. Reconfigure replication in the original primary region by doing thefollowing:

      1. Add the primary disks to the original consistency group.
      2. Create new secondary disksin the original secondary region.
      3. Start replication.

What's next

Except as otherwise noted, the content of this page is licensed under theCreative Commons Attribution 4.0 License, and code samples are licensed under theApache 2.0 License. For details, see theGoogle Developers Site Policies. Java is a registered trademark of Oracle and/or its affiliates.

Last updated 2025-12-15 UTC.