Question #677
A company needs to implement a disaster recovery plan for their primary on-premises application, which uses a PostgreSQL database. The application runs on physical servers that host multiple other services. All on-premises systems use operating systems compatible with Amazon EC2. The company wants to minimize operational complexity while ensuring rapid recovery.
Which solution BEST meets these requirements?
Deploy AWS Application Migration Service (MGN) agents on all source servers, including the PostgreSQL database server. Configure replication settings and launch templates. Conduct periodic recovery drills using non-disruptive test instances. Initiate a full cutover to AWS during a disaster.
Install AWS Application Migration Service (MGN) agents on the on-premises servers, including the PostgreSQL server. Configure AWS Elastic Disaster Recovery with automated replication and predefined launch settings. Perform regular failover and recovery tests using the most recent synchronized data.
Set up an AWS Database Migration Service (AWS DMS) replication instance and a target Amazon RDS for PostgreSQL DB instance. Create a DMS task for continuous data replication. Manually install and configure the application on EC2 instances using compatible AMIs. Use custom scripts to synchronize application state changes.
Deploy an AWS Storage Gateway File Gateway on premises. Store application and database files on the gateway. Schedule regular EBS snapshots of the gateway volumes. Provision EC2 instances and restore the application and database from snapshots during a disaster, requiring manual configuration adjustments.
Explanation
The BEST solution is B because:
- AWS Elastic Disaster Recovery (DRS) is purpose-built for disaster recovery, offering automated replication and predefined launch settings, reducing operational complexity.
- MGN agents ensure compatibility with EC2 and enable continuous replication of the entire application stack (including the PostgreSQL database).
- Regular failover tests using synchronized data validate recovery readiness without disrupting production.
Why other options are incorrect:
- A: While MGN supports migration, it lacks DRS's DR-specific optimizations (e.g., automated recovery, minimal RTO).
- C: AWS DMS replicates only the database, requiring manual application setup and scripting, increasing complexity.
- D: Relies on manual snapshot restores and configuration adjustments, leading to slower recovery and higher operational overhead.
Key Points:
- Use AWS Elastic Disaster Recovery for automated, low-complexity DR.
- Ensure full-stack replication (application + database) for rapid recovery.
- Validate DR plans with non-disruptive testing.
Answer
The correct answer is: B