DVA-C02 Deployment Practice Question

Answer analysis

Option-by-option breakdown

For each option: why learners choose it and why it is or isn't the right answer here.

• ✗

  Create a second ALB pointing to the new service, then update DNS to switch traffic.

  Why it's wrong here

  Creating a second Application Load Balancer (ALB) and then updating DNS records to switch traffic introduces significant risk of downtime. DNS propagation delays are unpredictable and can take minutes to hours, meaning some users will continue to be routed to the old service while others are directed to the new one, leading to an inconsistent user experience or service unavailability during the transition. This approach lacks the fine-grained traffic shifting control and immediate rollback capabilities required for modern, zero-downtime deployments on ECS.

• ✓

  Use AWS CodeDeploy to orchestrate a blue/green deployment by creating a new task set and shifting traffic.

  Why this is correct

  This is the recommended approach for zero-downtime deployments on Amazon ECS. AWS CodeDeploy orchestrates a blue/green deployment by provisioning a completely new "green" task set alongside the existing "blue" one. It then incrementally shifts traffic from the blue environment to the green environment using the Application Load Balancer, allowing for thorough testing before full cutover. This method provides immediate rollback capabilities if issues arise, ensuring minimal impact and high availability during updates.

• ✗

  Update the existing ECS service with the new task definition and allow rolling update.

  Why it's wrong here

  Updating an existing ECS service with a new task definition and relying on a rolling update strategy can lead to service degradation or temporary downtime. During a rolling update, old tasks are gradually stopped and replaced by new tasks, which means both versions of the application might be serving traffic simultaneously. If the new task definition has issues, it could cause errors for users as new tasks come online, and the service might experience reduced capacity or unavailability until all old tasks are replaced.

• ✗

  Manually create a new target group with the new service and update the ALB listener rule.

  Why it's wrong here

  Manually creating a new target group, registering new tasks, and then updating the Application Load Balancer (ALB) listener rules to shift traffic is highly prone to human error and lacks automation. This process requires precise, coordinated manual steps, increasing the likelihood of misconfigurations, delays, or outages during deployment. Furthermore, it does not provide automated rollback mechanisms, making recovery from deployment failures significantly more complex and time-consuming.

Go deeper

Related to this question