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.

• ✗

  Use AWS CodeDeploy to automatically roll back the deployment.

  Why it's wrong here

  Using AWS CodeDeploy for an automatic rollback is typically part of a pre-configured CI/CD pipeline that manages the deployment strategy itself (e.g., canary, linear deployments). While CodeDeploy can orchestrate rollbacks, for a simple manual rollback of a newly deployed Lambda version, directly manipulating Lambda's built-in versioning and aliasing is a more immediate and efficient approach. CodeDeploy is an orchestrator, not the direct mechanism for reverting a Lambda version in this specific context.

• ✓

  Update the alias to point to the previous version.

  Why this is correct

  Updating the alias to point to the previous version is the most effective and fastest method for an instant rollback. AWS Lambda aliases are mutable pointers to specific, immutable function versions. By simply redirecting the alias from the problematic new version to a known stable version, all subsequent invocations through that alias immediately begin using the stable code, ensuring zero downtime and minimal operational overhead.

• ✗

  Use the AWS CLI to update the function code with the previous code.

  Why it's wrong here

  Using the AWS CLI to update the function code with the previous code would involve uploading the old deployment package, which creates a *new* function version. This process is slower than simply redirecting an existing alias, as it requires the Lambda service to process and publish a new version. It also consumes an additional version slot and doesn't leverage the immutability of existing, already published versions for a quick switch.

• ✗

  Delete the current function and recreate it using the previous deployment package.

  Why it's wrong here

  Deleting the current function and recreating it using the previous deployment package is an extremely disruptive and error-prone method. This action would result in significant downtime, as the function would be unavailable during deletion and recreation. Furthermore, all associated configurations, such as triggers, environment variables, and permissions, would need to be manually re-established, making it a time-consuming and high-risk operation that disregards Lambda's robust versioning capabilities.

Go deeper

Related to this question