A company runs a production PostgreSQL database on Amazon RDS with the Multi-AZ feature enabled. During a scheduled maintenance window, the primary database instance becomes unavailable. What happens to the application’s database connections?
A) The application must manually update its connection string to point to the standby instance.
B) RDS automatically promotes the standby instance and updates the DNS CNAME record, so the application continues using the same endpoint.
C) All in-flight transactions are lost and the database remains unavailable until an administrator intervenes.
D) RDS creates a new read replica to replace the failed primary, which takes over write operations.
Correct Answer: B
Explanation: Amazon RDS Multi-AZ deployments maintain a synchronously replicated standby in a different Availability Zone. On failure of the primary, RDS automatically performs a failover by promoting the standby and updating the DNS CNAME record that the application already uses — no application-side changes are needed. Failover typically completes within 60–120 seconds.
Why the others are wrong: (A) defeats the purpose of Multi-AZ, which is designed to be transparent to the application. (C) is incorrect — Multi-AZ exists specifically to avoid this outcome. (D) confuses Multi-AZ (synchronous standby for HA) with read replicas (asynchronous copies for read scaling); a read replica is not automatically promoted during Multi-AZ failover.