A production RDS MySQL DB instance uses a traditional Multi-AZ deployment with one synchronous standby. The primary Availability Zone fails. What should the application expect during the managed failover?

1. The existing DB endpoint is updated to the promoted standby; clients must reconnect and honor DNS changes.
2. The standby serves writes at the same instant with no possible connection interruption.
3. The deployment must first restore the latest manual snapshot because the standby cannot be promoted.
4. The standby starts serving writes under a newly created endpoint that operators must manually distribute.