DOP-C02 Practice Question 73: CodePipeline Stage Automatic Rollback
AWS Certified DevOps Engineer – Professional (DOP-C02) · Test 8 · question 3 of 10
A company runs a V2 type CodePipeline pipeline that deploys to production through a Deploy stage. When a bad release causes the Deploy stage to fail, an engineer currently has to manually find the last known-good source revision and start a new pipeline execution to recover, which adds 10 to 15 minutes to every incident. The team wants CodePipeline itself to revert the Deploy stage to the last successful revision as soon as it detects the failure, without a custom Lambda function or a separate recovery pipeline.
Which built-in capability should they configure?
Show answer and explanation
Answer:
V2 type pipelines support enabling automatic rollback at the stage level. When the stage fails, CodePipeline selects the source revisions from the most recent execution that completed that stage successfully and starts a rollback execution automatically, restoring the environment without any custom code or manual intervention. This is exactly the built-in behavior the team is asking for.
Option B describes retrying the same failing input again, which does nothing to recover to a previously working revision; a retry policy addresses transient action failures, not a bad release.
Option C reimplements automatic rollback with custom EventBridge and Lambda glue, which is exactly the manual-recovery-replacement approach the team wants to avoid, and it still requires someone to track and pass the correct "previous" revision.
Option D describes a superseded execution mode, which governs how queued executions for new changes are handled while a pipeline is busy; it has nothing to do with recovering from a stage failure or reverting to an earlier revision.
Reference: Configure a stage for automatic rollback