Exam AWS Certified DevOps Engineer - Professional DOP-C02 topic 1 question 152 discussion

D is correct: In-place deployment search for all available agents at the time of the deployment and update the app version. If there are new instances launched after the search, they would be omitted and they fetch the lastest app version available, which is the previous revision A and B: If this happened, the other three would be affected as well C: If code deploy agents were not installed, no version would be installed on the two instances

D. EC2 Auto Scaling launched two new instances while the new deployment had not yet finished, causing the previous version to be deployed on the affected instances. Explanation: In an EC2 Auto Scaling group, when a deployment is in progress and new instances are launched, they may receive the previous version of the application if the deployment has not yet completed. This is because the new instances join the Auto Scaling group and need to fetch the latest revision during the deployment process. If the deployment has not finished when the new instances are launched, they will fetch the current revision available in the Auto Scaling group, which might be the previous version.

B. Failed lifecycle event: Would cause a rollback, but that would typically be marked as a deployment failure, not success.

B is correct. The AfterInstall lifecycle event is executed after the application revision is installed on the EC2 instance. If an error occurs during this phase, the CodeDeploy agent can trigger a rollback to the previous application version. Given that the deployment was successful for three instances and not for the others, it’s possible that the AfterInstall hook failed on the two instances that still have the previous version. In this case, the CodeDeploy agent would have automatically rolled back to the last successful application revision for those two instances. why D is wrong? new instances launched by Auto Scaling would have the current revision applied during the deployment process, not the previous revision. Therefore, it’s unlikely that the new instances would have the previous version unless something else occurred.

, B. A failed AfterInstall lifecycle event hook caused the CodeDeploy agent to roll back to the previous version on the affected instances seems to be the most plausible explanation. It accounts for the scenario where the deployment was successful overall, but specific instances reverted to the previous application revision due to issues encountered during post-installation steps. It's important for the DevOps engineer to review the deployment logs, especially focusing on lifecycle event hooks and their outcomes, to confirm this hypothesis and take corrective actions.

only D makes sense

Got to be D. The rest choices would impact the entire ASG and not only two out of the five intances.

D is correct https://docs.aws.amazon.com/codedeploy/latest/userguide/troubleshooting-auto-scaling.html#troubleshooting-auto-scaling-provision-termination-loo