0Pricing
MLOps Academy · Lesson

Progressive Delivery with Argo Rollouts

Automate staged promotion and analysis.

Progressive Delivery with Argo Rollouts is a free MLOps Academy lesson on CoddyKit — lesson 4 of 4. You can read the complete lesson below for free — then practise it hands-on in the browser with a built-in code editor and a 24/7 AI tutor. It is part of the MLOps Academy learning path, one of 4 lessons in the course, and your progress syncs across the web and the CoddyKit app.

What Progressive Delivery Is

Progressive delivery means releasing a new model in controlled steps, with checks at each stage, instead of one risky flip to everyone. 🚦

Meet Argo Rollouts

Argo Rollouts is a Kubernetes controller that replaces the plain Deployment so you can run canary and blue-green releases natively in the cluster.

The Rollout Resource

You swap your Deployment for a Rollout object. It looks familiar but adds a strategy block that describes how new pods take over traffic.

apiVersion: argoproj.io/v1alpha1
kind: Rollout
metadata:
  name: model-api

Define Canary Steps

Inside the canary strategy you list steps: set a traffic weight, then pause. Argo walks them in order, shifting users to the new version slowly.

strategy:
  canary:
    steps:
      - setWeight: 20
      - pause: {duration: 5m}
      - setWeight: 50
      - pause: {}

Pause for a Gate

An empty pause halts the rollout until you approve it by hand. A timed pause waits a set duration, then continues on its own.

Analysis at Each Step

An AnalysisTemplate queries metrics, like error rate from Prometheus, and tells Argo whether the canary at this step is healthy.

Automatic Rollback

If an analysis run fails its success condition, Argo aborts the rollout and shifts all traffic back to the stable version with no human needed.

Wire In a Traffic Router

For precise percentages, Argo drives a traffic provider like Istio or NGINX. Without one, weight is approximated by pod counts instead.

Blue-Green Too

Argo also supports blue-green: it spins up the new version fully, lets you test it on a preview service, then flips all traffic at once.

Promote or Abort

You drive a paused rollout with the CLI: promote moves to the next step, while abort cancels and returns everything to stable.

kubectl argo rollouts promote model-api
kubectl argo rollouts abort model-api

Watch It Live

The get rollout command shows each step, the current weight, and analysis results, so you can follow the promotion in real time.

kubectl argo rollouts get rollout model-api --watch

Quick Check

How does Argo Rollouts react to a failing canary?

Recap

Argo Rollouts turns a Deployment into staged canary or blue-green releases, runs metric analysis at each step, and auto-aborts to stable on failure. ✅

Frequently asked questions

Is the “Progressive Delivery with Argo Rollouts” lesson free?

Yes — the full text of “Progressive Delivery with Argo Rollouts” is free to read here on the web, and the MLOps Academy course includes 4 lessons in total. To practise it interactively (a built-in code editor and a 24/7 AI tutor) and unlock the rest of the MLOps Academy course, upgrade to CoddyKit PRO.

What will I learn in “Progressive Delivery with Argo Rollouts”?

Automate staged promotion and analysis. You practise MLOps Academy with hands-on code you run directly in the browser, and a 24/7 AI tutor answers your questions as you work through the lesson.

Do I need any experience to start MLOps Academy?

No prior experience is required. MLOps Academy on CoddyKit is structured for beginners through advanced learners; this is — lesson 4 of 4, so you can start here or from the beginning and move at your own pace.

How long does the “Progressive Delivery with Argo Rollouts” lesson take?

Most CoddyKit lessons take about 5–10 minutes. Each one is bite-sized and interactive, so you make steady progress and pick up exactly where you left off across the web and the app.

Can I write and run code in this MLOps Academy lesson?

Yes. Every MLOps Academy lesson includes a built-in code editor, so you write and run real code right in your browser and get instant AI feedback — no local setup required.

All lessons in this course

  1. Canary Rollouts: Ship to a Few First
  2. Shadow Traffic Without User Impact
  3. Define Automatic Rollback Criteria
  4. Progressive Delivery with Argo Rollouts
← Back to MLOps Academy