Availability, read scaling, and restoration need different database operations. A Multi-AZ deployment maintains a standby for availability. A read replica is a separate instance that follows a source and serves read workloads, with replication behavior that must suit the application. In this account, creating a replica first requires automated backups on its source.
A snapshot records a recovery point. Restoring it creates another instance; it doesn't rewind the source instance in place. Recovery also includes configuration. An instance restored with public accessibility doesn't meet a private recovery requirement merely because its snapshot is valid. In this lab, create a replica, snapshot a private database, restore a separate recovery instance, and repair an exposed replica without changing the source. Check the identifiers and stored settings after each operation.