AWS Certified DevOps Engineer – Professional — Question 171

A company is implementing an Amazon Elastic Container Service (Amazon ECS) cluster to run its workload. The company architecture will run multiple ECS services on the cluster. The architecture includes an Application Load Balancer on the front end and uses multiple target groups to route traffic.

A DevOps engineer must collect application and access logs. The DevOps engineer then needs to send the logs to an Amazon S3 bucket for near-real-time analysis.

Which combination of steps must the DevOps engineer take to meet these requirements? (Choose three.)

Answer options

• A. Download the Amazon CloudWatch Logs container instance from AWS. Configure this instance as a task. Update the application service definitions to include the logging task
• B. Install the Amazon CloudWatch Logs agent on the ECS instances. Change the logging driver in the ECS task definition to awslogs.
• C. Use Amazon EventBridge to schedule an AWS Lambda function that will run every 60 seconds and will run the Amazon CloudWatch Logs create-export-task command. Then point the output to the logging S3 bucket.
• D. Activate access logging on the ALB. Then point the ALB directly to the logging S3 bucket.
• E. Activate access logging on the target groups that the ECS services use. Then send the logs directly to the logging S3 bucket.
• F. Create an Amazon Kinesis Data Firehose delivery stream that has a destination of the logging S3 bucket. Then create an Amazon CloudWatch Logs subscription filter for Kinesis Data Firehose.

Correct answer: B, D, F

Explanation

The correct steps are B, D, and F. Option B allows the logs to be collected and sent to CloudWatch, while D ensures that access logs from the ALB are logged directly to S3. Option F enables a more efficient transfer of logs using Kinesis Data Firehose. Options A, C, and E do not effectively meet the logging and real-time analysis requirements as described in the question.