rollback
Returning a system to a previous known-good version after a bad change.
In more detail
Each strategy rolls back differently: a rolling update writes the previous template back, blue-green switches the route, a canary sets its weight to zero, a flag is turned off and GitOps selects an earlier commit. Name the rollback command before the first user sees the change.
Examples
In AWS: Returning a system to its previous known-good version after a change goes wrong.kubectl rollout undo deploy/webRelated terms
- task definition — The JSON document that describes an ECS task: which images to run, how much processor and memory, networking, role and logging.
- digest — A SHA-256 hash of an image's manifest that identifies its exact content and can never be moved to other bytes.
- database migration — A scripted, versioned change to a database's structure, such as adding a table or a constraint.
- feedback loop — How previous attempts to apply the desired state have affected the actual state, read back to decide the next action.
- GitOps state store — A system for storing immutable versions of desired state declarations, with access control and auditing on changes.
- rolling update — Replacing instances of an application with the new version a few at a time while the rest keep serving.
- blue-green deployment — Running two complete environments, blue for the current release and green for the candidate, and switching traffic between them in one step.
- canary release — Sending a small share of real traffic to a new version first, and increasing it only while its metrics stay healthy.
- feature flag — A switch in the code, controlled from outside, that turns a feature on or off for chosen users without a new deployment.
- fix forward — Recovering from a bad change by quickly shipping a new fix, instead of going back to the previous version.
- failed deployment recovery time — The time it takes to recover from a deployment that fails and needs immediate intervention in production.
- build, release, run — Three strictly separate stages: build turns code into a bundle, release joins it to one deploy config with an identifier, and run starts processes from a chosen release.
- backward compatibility — The property that a new version of an interface still works for clients written against the old one.
- transaction — A group of database changes that succeed or fail together, so the data is never left half-updated.
- optimistic update — Showing the result of a user’s action on screen immediately, before the server confirms it, and undoing it if the server says no.
- optimistic UI — Updating the screen immediately as if an action succeeded, before the server confirms it, and undoing the change if it fails.
- useOptimistic — The React hook that shows a temporary optimistic value while an Action is pending, then returns to the real state when it settles.
- cutover — The moment traffic or ownership switches from an old system or shard to a new one.
- distributed monolith — A system split into services that still have to be released together, so it has the costs of services and none of the benefits.
Taught in
- Two Hundred Accounts And One Bad Deploy — AWS Certified DevOps Engineer – Professional (DOP-C02) Exam Prep
- Surviving A Hundred And Eighty Minutes — AWS Certified DevOps Engineer – Professional (DOP-C02) Exam Prep
- CodePipeline: Stages, Actions And The Key Nobody Shared — AWS Certified DevOps Engineer – Professional (DOP-C02) Exam Prep
- Artifacts: Where Packages And Images Live, And Die — AWS Certified DevOps Engineer – Professional (DOP-C02) Exam Prep
- CodeDeploy: Hooks That Never Ran — AWS Certified DevOps Engineer – Professional (DOP-C02) Exam Prep
- Strategies Across Compute: The Canary That Saw Nothing — AWS Certified DevOps Engineer – Professional (DOP-C02) Exam Prep
- CloudFormation In Depth: Replace Or Update? — AWS Certified DevOps Engineer – Professional (DOP-C02) Exam Prep
- Troubleshooting Failed Deployments: Read The First Failure — AWS Certified DevOps Engineer – Professional (DOP-C02) Exam Prep
- Auditing: Who Opened Port Twenty Two? — AWS Certified DevOps Engineer – Professional (DOP-C02) Exam Prep
- Long-Stem Clinic: When Two Answers Both Work — AWS Certified DevOps Engineer – Professional (DOP-C02) Exam Prep
- The Full Mock And Your Review Plan — AWS Certified DevOps Engineer – Professional (DOP-C02) Exam Prep
- A Hotfix At 3 A.M.: Who Wins At 3:05? — Certified GitOps Associate (CGOA) Exam Prep
- Sixteen Pods, Room For Twelve: The System, Feedback And Continuous — Certified GitOps Associate (CGOA) Exam Prep
- Two Ways Back From 2.5: Rollback The GitOps Way — Certified GitOps Associate (CGOA) Exam Prep
- Terminology Sprint: One Incident, Nine Terms — Certified GitOps Associate (CGOA) Exam Prep
- Someone Cleaned Up The History: Versioned And Immutable — Certified GitOps Associate (CGOA) Exam Prep
- Twelve Keys To Production: Pulled Automatically — Certified GitOps Associate (CGOA) Exam Prep
- What Does CI Still Do? CI, CD And Where GitOps Starts — Certified GitOps Associate (CGOA) Exam Prep
- Related Practices Sprint: One Change, Five Practices — Certified GitOps Associate (CGOA) Exam Prep
- Thursday, With A Way Back: Promotion, Deploy And Release — Certified GitOps Associate (CGOA) Exam Prep
- Patterns Sprint: Eight Needs, One Pattern Each — Certified GitOps Associate (CGOA) Exam Prep
- Three Teams, Three Formats: Manifests And Packaging — Certified GitOps Associate (CGOA) Exam Prep
- A Rollback Nobody Heard: Notifications, Observability And CI — Certified GitOps Associate (CGOA) Exam Prep
- Tooling Sprint: The CTO's Shopping List — Certified GitOps Associate (CGOA) Exam Prep
- Forty-Six Of Sixty: The Mock And What It Tells You — Certified GitOps Associate (CGOA) Exam Prep
- The Evening Before: Final Review — Certified GitOps Associate (CGOA) Exam Prep
- Artifact Promotion: Staging to Production — Continuous Integration & Continuous Delivery Pipelines
- What is GitOps? — Continuous Integration & Continuous Delivery Pipelines
- Introduction to Release Strategies — Continuous Integration & Continuous Delivery Pipelines
- Rolling Updates and Rollbacks — Continuous Integration & Continuous Delivery Pipelines
- Rolling Back Blue-Green — Continuous Integration & Continuous Delivery Pipelines
- Canary Releases — Continuous Integration & Continuous Delivery Pipelines
- Rolling Updates And Rollbacks — Certified Kubernetes Application Developer (CKAD) Exam Prep
- Blue/Green With Labels And A Service — Certified Kubernetes Application Developer (CKAD) Exam Prep
- Helm: Install, Upgrade And Roll Back — Certified Kubernetes Application Developer (CKAD) Exam Prep
- Deployment: Timed Drill And Checkpoint — Certified Kubernetes Application Developer (CKAD) Exam Prep
- The Danger of Long-Lived Keys — Cloud Infrastructure & Hyperscaler Architecture
- Containers on AWS (ECS and EKS) — Cloud Infrastructure & Hyperscaler Architecture
- Managed Relational Databases (RDS) — Cloud Infrastructure & Hyperscaler Architecture
- The Console versus the CLI — Cloud Infrastructure & Hyperscaler Architecture
- Instability: Change Fail Rate And Deployment Rework Rate — DevOps & Site Reliability Engineering
- Restoration: Failed Deployment Recovery Time — DevOps & Site Reliability Engineering
- Build, Release, Run And Processes — DevOps & Site Reliability Engineering
- Continuous Delivery Versus Continuous Deployment — DevOps & Site Reliability Engineering
- Deployment Strategies: Rolling Updates — DevOps & Site Reliability Engineering
- Deployment Strategies: Blue Green — DevOps & Site Reliability Engineering
- Deployment Strategies: Canary Releases — DevOps & Site Reliability Engineering
- Decoupling Release From Deployment: Feature Flags — DevOps & Site Reliability Engineering
- Containers And Orchestration Context — DevOps & Site Reliability Engineering
- Database Access & Migrations: ORM & Data Layers — Production Full-Stack Engineering
- Optimistic UI Architecture: Cache & Rollback Protocols — Production Full-Stack Engineering
- Continuous Delivery: Blue-Green & Canary Pipelines — Production Full-Stack Engineering
- Deployments, ReplicaSets, Rollouts And Rollback — Kubernetes: Production-Grade Container Orchestration
- Troubleshooting: ImagePullBackOff And Failed Rollouts — Kubernetes: Production-Grade Container Orchestration
- Troubleshooting: A Timed Service Recovery Drill — Kubernetes: Production-Grade Container Orchestration
- Package The Service As A Helm Chart — Kubernetes: Production-Grade Container Orchestration
- Optimistic UI: useOptimistic Rollbacks & Transitions — Next.js 15 & Modern App Router Architecture
- useOptimistic: Instantaneous UI Updates with Rollback — React 19 & Modern Component Architecture
- Partitioning, Sharding And Hot Keys — Scalable System Design & Distributed Architecture
- Monolith Versus Services: The Honest Trade Offs — Scalable System Design & Distributed Architecture
References
- Amazon ECS deployment circuit breaker and rollback
- OpenGitOps Glossary v1.0.0
- kubectl rollout reference
- Google SRE book: Release engineering
- DORA capability: deployment automation
- The Twelve-Factor App: V. Build, release, run
- PostgreSQL: ROLLBACK
- Argo Rollouts: Canary (abort and rollback)
- React docs: useOptimistic
- Kubernetes documentation: Deployments