ECS Blue/Green Deployment Health Check Path Failure

• A.The lifecycle hooks for the new task set are failing.
• B.The target group for the new task set is not configured correctly.
• C.The deployment group's rollback configuration triggers due to alarm threshold.
• ✓ D.The new task set fails the post-traffic shift validation tests.

Why D: Option D is correct because in a blue/green deployment on ECS with CodeDeploy, after traffic is shifted to the new task set, CodeDeploy runs post-traffic shift validation tests (e.g., via Lambda hooks or target group health checks). If these tests fail, CodeDeploy automatically triggers a rollback by replacing the new task set with the original (old) task set. This matches the described behavior where the new task set is immediately replaced after a few minutes.

• ✓ A.Create a CloudWatch alarm that triggers when the healthy task count of the ECS service falls below a threshold. Configure the CodeDeploy deployment group to automatically roll back when this alarm is in ALARM state.
• B.Modify the CodeDeploy deployment group to enable automatic rollback when a deployment fails. The deployment will automatically revert to the last successful deployment.
• C.Increase the deployment timeout in the CodeDeploy deployment configuration to allow more time for the new task set to become healthy.
• D.Configure the ECS service to automatically roll back to the previous task definition if the deployment fails. Use the ECS service's deployment circuit breaker.

Why A: Option A is correct because it leverages CloudWatch alarms to monitor the health of the ECS service's new task set. When the alarm triggers due to insufficient healthy tasks, CodeDeploy can automatically initiate a rollback to the previous revision. This approach provides a proactive, health-based rollback mechanism that directly addresses the failure condition described (new task set not becoming healthy within timeout).

• A.Update the ECS task definition for the green service to a working version and wait for the deployment to complete.
• B.Manually update the ECS service for the blue environment to increase the desired count.
• ✓ C.Stop the deployment in CodeDeploy and re-point the ALB listener to the blue target group.
• D.Roll back the deployment to the previous version using the CodeDeploy console.

Why C: Option C is correct because the deployment is stuck due to the green fleet being unhealthy. Stopping the deployment in CodeDeploy will terminate the stuck deployment and allow you to reconfigure the ALB listener to point to the blue target group, ensuring the blue service continues to serve traffic without interruption. Option A is incorrect because updating the task definition for the green service will not automatically resume or fix the stuck deployment; the deployment is already in a failed state. Option B is incorrect because manually increasing the desired count of the blue service does not address the root cause—the stuck deployment—and the listener may still be routing traffic to the green target group. Option D is incorrect because rolling back the deployment may not be possible while the deployment is in 'InProgress' state; you typically need to stop the deployment first before initiating a rollback.