DOP-C02 Question 269
Select 3Your company has implemented a CI/CD pipeline using AWS CodePipeline. The pipeline deploys an application to an Amazon ECS Fargate service. During a recent deployment, the application exhibited degraded performance and errors after reaching production. You are tasked with improving the pipeline to minimize the risk of such issues in future deployments. Which approach should you take to meet this requirement?
- A
Add an AWS CodeDeploy deployment step to the pipeline and configure a blue/green deployment strategy.
- B
Integrate AWS CodeBuild into the pipeline to run unit tests and integration tests before deploying to production.
- C
Enable Amazon ECS task automatic rollback for the ECS service when deployment health checks fail.
- D
Replace the ECS Fargate service with an EC2 Auto Scaling group to gain finer-grained control over deployments.
- E
Use AWS CloudFormation to recreate the ECS service on every deployment to ensure consistency.
Show answer and explanation
Correct answers: A, B, C
Explanation
To minimize deployment risks, it's critical to validate changes through testing (CodeBuild), deploy changes safely using a controlled strategy (CodeDeploy with blue/green), and automatically recover from failed deployments (ECS task rollback). These measures ensure a robust CI/CD pipeline and reduce the likelihood of negatively impacting production environments.
- A. Correct.
Adding a CodeDeploy step with a blue/green deployment strategy minimizes downtime and allows you to safely test the new version of the application with a subset of traffic before routing all traffic to it. This reduces the risk of impacting production.
- B. Correct.
Running unit tests and integration tests in CodeBuild ensures that code changes are validated before deployment, reducing the likelihood of introducing bugs to production.
- C. Correct.
Enabling ECS task automatic rollback ensures that the service reverts to the previous stable version automatically if deployment health checks fail, reducing downtime and errors.
- D. Incorrect.
Switching from ECS Fargate to an EC2 Auto Scaling group is unnecessary for this scenario and would increase operational complexity without directly addressing deployment risks.
- E. Incorrect.
Recreating the ECS service with CloudFormation for every deployment is inefficient and does not directly resolve the issue of degraded performance during deployments.