Skip to content

Triggering a Failover

Trigger a failover from the ACGR Dashboard to redirect live call traffic between Regions according to a saved Playbook.

Before You Begin

  • You need the ACGR Admin role to trigger a failover directly.
  • At least one Playbook must exist. The Failover Now action is unavailable until a Playbook is configured (see Managing Playbooks.
  • Confirm the correct scenario is selected and that the team is prepared before proceeding.

Step-by-Step Instructions

The Dashboard provides a daily operational view of your resiliency setup. Use it to monitor health and initiate failover when necessary.

To execute a failover:

  1. In the left navigation, select ACGR, then click Dashboard.
  2. Click Failover Now.
  3. Select a Playbook.
  4. Review the distribution percentages.
  5. Confirm.

After execution, changes are applied asynchronously. It may take a short time for the Dashboard to reflect the updated distributions and failover timestamp.

The Failover Now action redirects live call traffic according to the selected Playbook. Confirm the correct scenario is selected and that the team is prepared before proceeding.

Warning: Do not re-execute a Playbook because the Dashboard has not updated yet. Changes are applied asynchronously and may take a short time to appear.

After a failover, watch the Regional Activity metrics to confirm that calls are routing correctly and that agents are available in the secondary Region. Cross-reference with the Traffic Trends charts to validate that the shift occurred as expected. For the full failover and failback workflow, see ACGR Best Practices.

Troubleshooting

Problem Cause Solution
Failover Now is unavailable / the "Failover to other region" card shows "Set up playbooks to enable this feature." No Playbooks are configured for this instance. Create at least one Playbook. See Managing Playbooks.
The Dashboard has not updated after a failover. Changes are applied asynchronously. Wait a short time for the Dashboard to reflect the updated distributions and failover timestamp. Do not re-execute the Playbook.