Comprehensive and Detailed Explanation From AWS Cloud Documentation:

For an Amazon RDS Multi-AZ DB instance deployment, Amazon RDS provides the managed high-availability and failover mechanism. If a qualifying planned or unplanned outage affects the primary DB instance, RDS automatically switches to the standby replica in another Availability Zone. This automatic promotion or failover is performed by AWS as part of the managed RDS service, so option B is correct. The customer must first choose and enable a Multi-AZ deployment; AWS does not decide to enable that architecture on the customer’s behalf, which eliminates option C. The application must also be designed to recover connections after failover, normally by reconnecting through the unchanged DB endpoint after DNS updates propagate, which eliminates option A. Monitoring application behavior, business transactions, latency, and post-failover performance also remains part of the customer’s operational responsibility, even though AWS monitors the RDS service and underlying infrastructure. The exam distinction is that AWS automates the database infrastructure failover, while the customer configures the deployment and makes the application resilient to the connection interruption.