AWS Certified Developer - Associate Practice Mode

Community Discussion

Written by other learners. Comments can disagree with the marked answer — check the official documentation when in doubt.

• Chose B1

  B Correct. Publish custom metrics (error count + total count) to CloudWatch, create a metric math alarm calculating error rate percentage, trigger existing SNS topic when rate exceeds 5%. Native, near real time, least overhead.

• Chose B3

  A) Eliminated - Requires manual log queries and a custom Lambda function to process logs, introducing operational overhead. B) Correct - CloudWatch Alarms can calculate error rates using metric math and automatically notify the SNS topic. C) Eliminated - This approach notifies the support team for every API failure, not just when the error rate exceeds 5%. D) Eliminated - Adds significant operational complexity (managing S3 storage, Athena queries, and scheduling).

• Chose B1

  keyword: error rate exceed 5%, receive notifications in near real time ==> discard A,: use lamdba or athena with schedule, violate ' receive notifications in near real time ' ==> discard C: die in spam of message if any, and don't know when error up 5% B is most good solution: CloudWatch Alarm can calculate 5% with setup, and notify righ away when reach thresold set-up