Skip to content

GSS Backup Failover Plan

In order to ensure that the backup side can run normally, we plan to do a failover test. In theory, as long as the primary side is unhealthy, route53 will automatically failover to the backup side. Once the primary side is healthy, route53 will switch back to the primary side. Based on this, we can do the following:
  1. Check the healthCheck status for backup side by Route53|Health checks. backup_failover_1.png
  2. Change the primary healthCheck url from the correct url to a error url.
  3. Route53 will auto failover to backup side.
  4. Open the /ping api to check if in backup side(if so, it'll show "okok, the current region is: us-east-2").
  5. Check if the status page running successfully.

  6. Update the primary healthCheck url to the correct url and check the healthCheck status for primary side.
    backup_failover_2.png

  7. Route53 will auto failover to primary side once the healthCheck status for primary side is healthy.

  8. Open the /ping api to check if in primary side(if so, it'll show "okok, the current region is: us-east-1").
  9. Check if the status page running successfully.

Reference

  1. GSS failover in route53