Progressive Delivery and Rollback Lab
Demonstrate the complete path from immutable release candidate through canary, production verification, failure detection, rollback or containment, and outcome accounting.
A rapid review of the chapter’s existing Quick Read, principles, definitions, lessons, and review material.
Demonstrate the complete path from immutable release candidate through canary, production verification, failure detection, rollback or containment, and outcome accounting.
Execute the existing Markdown instructions and retain the required output and evidence.
Jump to validation criteriaObjective
Demonstrate the complete path from immutable release candidate through canary, production verification, failure detection, rollback or containment, and outcome accounting.
Prerequisites and starting state
Use a disposable service or deterministic simulator with two versions, synthetic traffic, technical metrics, a product outcome measure, and a reversible feature flag. No production deployment is permitted.
Required implementation
- Build one immutable artifact and record source, build, dependencies, provenance, and digest.
- Define canary cohort, maximum exposure, observation window, technical and product gates, stop conditions, decision owner, and rollback.
- Deploy the digest to the canary and reconcile the provider state into a Release record.
- Inject a defect affecting only one high-value cohort while aggregate health remains green.
- Detect the segmented failure and block promotion.
- Execute rollback or containment and verify recovery, data integrity, and cohort state.
- Create corrective work and update the original release evidence without erasing its earlier state.
Evidence and pass criteria
Retain artifact identity, release plan, deployment reconciliation, cohort metrics, gate decisions, alert, rollback authorization, recovery proof, customer-outcome status, and retrospective. The lab fails if deployment is called success, if aggregate metrics hide the affected cohort, or if rollback is assumed rather than verified.
Cleanup
Remove the simulated release, flags, traffic, and test data. Preserve the evidence package.
Curriculum maturity is not implementation proof.
This chapter defines architecture or practice. It does not by itself prove a corresponding production implementation.
Review this chapter.
Challenge a claim, boundary, missing failure mode, unclear term, or unsupported evidence statement.
- Claim
- Boundary
- Failure
- Evidence