HNREL05-BP01 Failover testing of dedicated connections
Regular failover testing of dedicated connections is essential for ensuring the resilience and reliability of hybrid network environments. Simulating various scenarios by temporarily disabling BGP peering sessions between on-premises networks and cloud. By regularly exercising these tests, organizations can validate their recovery procedures, uncover latent bugs, and ensure their hybrid network architecture performs as expected during failover scenarios.
Desired outcome: Verify that failover and recovery procedures for Direct Connect connections work as intended, minimizing downtime during real incidents.
Level of risk exposed if this best practice is not established: High
Benefits of establishing this best practice:
-
Uncovers misconfigurations or gaps in failover processes
-
Increases confidence in your recovery plans
-
Enables you to proactively address weaknesses before actual failures
-
Reduces business impact of network outages
Implementation guidance
-
Simulated BGP failures of dedicated connections and observe failover behavior, using services such as the AWS Direct Connect Resiliency Toolkit.
-
Test all redundant dedicated connections and VPN links to ensure expected failover behavior.
-
Document and refine your recovery steps based on test outcomes.
-
Repeat testing regularly and after significant changes.