Canary and Blue-Green Argo Rollouts made possible in a multi-cluster environment
summary
Shows how Argo Rollouts pairs with a multi‑cluster scheduler (Nova or Karmada) to run Canary/Blue‑Green across clusters, using spread scheduling to place a 10‑replica app as 10 replicas split 5/5 across two production clusters. Rollout steps scale new version 2 → 6 → 10 replicas while running job‑based analysis (Kubernetes Job acting as load test) per cluster. A central management plane observes and coordinates the rollout and the demo completes successfully.
AI-written summary · about
chapters (15)
- 0:00Introduction to progressive delivery
- 1:01Traditional vs GitOps deployments
- 3:54Limitations of Kubernetes deployments
- 5:40Argo Rollouts overview
- 7:08Canary deployment strategy
- 9:16How Argo Rollouts works
- 11:06Multi-cluster environments explained
- 12:57Combining Argo Rollouts with multicluster
- 15:48Demo setup and scheduling
- 17:13Rollout definition and services
- 20:11Initial deployment across clusters
- 21:19Progressive delivery in action
- 24:58Demo conclusion and takeaways
- 25:27Key benefits and integration
- 27:36Q&A session
related talks
31mSafe Deliveries: Implementing Canary Deployments with Argo Rollouts
40mProgressive Delivery with Kubernetes and Argo Rollouts
30mProgressive Delivery Made Easy With Argo Rollouts
35mProgressive Delivery with Argo Rollouts
26mFrom Canary To Global: Unified Progressive Delivery for Hybrid Cloud With...
36mHow Intuit Does Canary and Blue Green Deployments with a K8s Controller
26mGlow up Your Stateful Services with Progressive Delivery Using Argo Rollouts!
29mProgressive Delivery for Micro-Services Using Argo Rollouts and the Downward API