Question

A financial services company runs a critical transaction API on an Amazon RDS for PostgreSQL Multi-AZ deployment. During a recent AZ disruption, RDS automatically failed over to the standby instance in about 90 seconds, but the application layer continued to fail for nearly 15 minutes because connections kept resolving to the old, now-unreachable instance IP. The DevOps team must ensure the application recovers automatically within seconds of a future failover, without manual intervention. Which change most directly addresses the root cause?

Accepted Answer

Ensure the application connects through the RDS cluster/instance DNS endpoint and lower the JVM/OS DNS TTL caching so the new IP is picked up quickly after failover.

Answer

Configure the application to connect using the RDS instance's cached IP address and add a retry loop with exponential backoff.

Answer

Enable RDS Multi-Region read replicas and point the application to the closest replica endpoint to reduce failover time.

Answer

Increase the RDS backup retention period and enable Performance Insights to detect the failover faster.