DVA-C02 Question #317: Real Exam Question with Answer & Explanation

Question

A company runs an application on AWS. The application consists of a static website that is hosted on Amazon S3. The application includes Amazon API Gateway APIs that invoke AWS Lambda functions. During a period of high traffic on the application, application users reported that the application was slow at irregular intervals. There were no failed requests. A developer needs to find the slow executions across all the Lambda functions. Which solution will meet these requirements?

Options

• APerform a query across all the Lambda function log groups by using Amazon CloudWatch Logs
• BEnable AWS CloudTrail Insights on the account where the Lambda functions are running. After
• CEnable AWS X-Ray for all the Lambda functions. Configure an X-Ray insight on a new group that
• DSet up AWS Glue to crawl through the logs in Amazon CloudWatch Logs for the Lambda

Explanation

Option C is correct because AWS X-Ray is purpose-built for distributed tracing across services like API Gateway and Lambda. Enabling X-Ray on all Lambda functions captures end-to-end latency data for each request segment, and configuring an X-Ray Insights group lets the system automatically detect anomalies - including irregular latency spikes - exactly matching the "slow at irregular intervals" symptom described.

Why the distractors are wrong:

• A (CloudWatch Logs query): Lambda doesn't automatically emit structured duration data that makes cross-function slowness easy to detect. CloudWatch Logs Insights can query logs, but it lacks the distributed tracing context needed to correlate slow segments across a multi-service request chain.
• B (CloudTrail Insights): CloudTrail tracks AWS API calls (who did what to which resource), not application-level performance. Insights there flags unusual API activity volume, not slow Lambda executions.
• D (AWS Glue + CloudWatch): Glue is an ETL service for batch data pipelines, not operational performance analysis. Crawling logs with Glue would be architecturally inappropriate and introduce significant complexity and delay.

Memory tip: Think of it this way - X-Ray = X-ray vision. When your distributed app is "sick" with mysterious slowness and you need to see inside each layer of the request flow, X-Ray is the diagnostic tool. CloudTrail is for security auditing, CloudWatch Logs is for raw log search, and Glue is for data pipelines - none of these see through your service layers the way X-Ray does.

No community discussion yet for this question.