How can you roll back a failed deployment in OpenShift?

• A. Delete the deployment and recreate.
• B. Rebuild the image and redeploy without rollback.
• C. Scale to zero and recreate.
• D. Use oc rollout undo deploy/<name> to revert to the previous Revision.

Answer: (D)

Rolling back a failed deployment in OpenShift is done by using the rollout history and undoing the current rollout to a previous working revision. OpenShift tracks each deployment as a revision and can switch the deployment configuration and image back to what it was in the prior revision, recreating the pods with that known-good state.

The best approach is to run the undo command for the deployment, which automatically reverts to the previous revision. If you need to target a specific older revision, you can first inspect the rollout history to find the desired revision, then use undo with that revision.

This method is preferred because it leverages the built-in rollout history, preserves the previous working configuration, and minimizes downtime. Other options—deleting and recreating, rebuilding and redeploying, or scaling to zero and recreating—do not efficiently revert to a known-good state and can introduce unnecessary downtime or configuration drift.