Progressive Delivery with Kubernetes and Argo Rollouts
summary
Shows a working pipeline that runs canary deployments on Kubernetes using Argo Rollouts controller with Istio traffic splitting and Prometheus metric checks. A demo ramps weights 20 → 40 → 60 → 100 percent with 30s pauses, background analysis every 20s and a failure limit of two, automatically aborting and rolling back when error rate exceeds 5%. The rollout resource, an analysis template and real-time status are created; rollouts complete and scale down the old replica set, and failing canaries trigger automatic rollback without manual intervention.
AI-written summary · about
chapters (10)
- 0:00Introduction and speaker intro
- 1:26What is Progressive delivery
- 3:33Deployment strategies overview
- 9:13Blue-green deployment explained
- 12:02Canary releases deep dive
- 17:18Argo Rollouts introduction
- 23:58Demo setup and resources
- 28:12Live canary deployment demo
- 34:14Handling deployment failures
- 37:28Post-deployment actions and Q&A
related talks
30mProgressive Delivery Made Easy With Argo Rollouts
35mProgressive Delivery with Argo Rollouts
31mSafe Deliveries: Implementing Canary Deployments with Argo Rollouts
51mProgressive Delivery in the Kubernetes era
46mProgressive Delivery in the Kubernetes era
01h 27mProgressive Delivery with Argo Rollouts
41mProgressive Delivery in the Kubernetes era
43mProgressive Application Delivery: Istio Ambient Mesh and Argo Rollout in Action