AWS Certified Solutions Architect – Professional — Question 1003

A retail company has a small ecommerce web application that uses an Amazon RDS for PostgreSQL DB instance. The DB instance is deployed with the Multi-AZ option turned on.
Application usage recently increased exponentially, and users experienced frequent HTTP 503 errors. Users reported the errors, and the company's reputation suffered. The company could not identify a definitive root cause.
The company wants to improve its operational readiness and receive alerts before users notice an incident. The company also wants to collect enough information to determine the root cause of any future incident.
Which solution will meet these requirements with the LEAST operational overhead?

Answer options

• A. Turn on Enhanced Monitoring for the DB instance. Modify the corresponding parameter group to turn on query logging for all the slow queries. Create Amazon CloudWatch alarms. Set the alarms to appropriate thresholds that are based on performance metrics in CloudWatch.
• B. Turn on Enhanced Monitoring and Performance Insights for the DB instance. Create Amazon CloudWatch alarms. Set the alarms to appropriate thresholds that are based on performance metrics in CloudWatch.
• C. Turn on log exports to Amazon CloudWatch for the PostgreSQL logs on the DB instance. Analyze the logs by using Amazon Elasticsearch Service (Amazon ES) and Kibana. Create a dashboard in Kibana. Configure alerts that are based on the metrics that are collected.
• D. Turn on Performance Insights for the DB instance. Modify the corresponding parameter group to turn on query logging for all the slow queries. Create Amazon CloudWatch alarms. Set the alarms to appropriate thresholds that are based on performance metrics in CloudWatch.

Correct answer: B

Explanation

Enabling both Enhanced Monitoring and Performance Insights on Amazon RDS provides comprehensive, real-time OS-level metrics and database load analysis with minimal operational overhead, allowing administrators to easily identify slow queries and system bottlenecks. Creating CloudWatch alarms based on these native metrics ensures proactive alerting before performance degrades. Other options either require manual parameter group modifications for slow query logging or introduce high operational overhead by deploying and managing external services like Amazon Elasticsearch Service and Kibana.