deployment-rollback

Rolling Back a Failed Checkout Service Deployment

Your checkout-service deploy just spiked the error rate. You need to detect the failure, identify the last stable version, roll back, and verify health -- all before customers notice. This workflow automates the full rollback lifecycle from detection to verification.

Workflow

service, reason
      |
      v
+--------------------+     +-----------------------+     +----------------------+     +-----------------------+
| rb_detect_failure  | --> | rb_identify_version   | --> | rb_rollback_deploy   | --> | rb_verify_rollback    |
+--------------------+     +-----------------------+     +----------------------+     +-----------------------+
  DETECT_FAILURE-1335       last stable: v2.4.3           rolled back to                error rate normalized
  error-rate-spike          deployed 3 days ago           previous version              service healthy

Workers

DetectFailureWorker -- Takes service and reason inputs. Detects a checkout-service failure triggered by an error-rate-spike. Returns detect_failureId: "DETECT_FAILURE-1335".

IdentifyVersionWorker -- Finds the last stable version: v2.4.3, deployed 3 days ago. Returns identify_version: true.

RollbackDeployWorker -- Rolls back the deployment to the previous stable version. Returns rollback_deploy: true.

VerifyRollbackWorker -- Confirms the service is healthy after rollback with error rate normalized. Returns verify_rollback: true.

Tests

2 unit tests cover the rollback pipeline.

Running

See ../../RUNNING.md for setup and execution instructions.