How We Progressively Deliver Changes To Kubernetes Using Canary Deployments and Featur...
summary
Octopus Deploy moves 1,600 customer instances to a model of short canary rollouts plus feature flags, using OpenFeature for segmented flags and a built-in andon cord to stop rollouts instantly. Deploy cadence shrinks from every 10 days to every 4, randomized canaries verify deployments in days, and feature-gated releases stay behind flags while teams run trunk-based development. Outcome: rollout time measured in days, detection often ~5 days, and under <1% high-severity defects across the fleet of 1,600 instances.
AI-written summary · about
chapters (11)
- 0:00Introduction
- 1:25Deployment versus functionality
- 3:52Deployment process challenges
- 6:19First canary attempt flaws
- 10:54Canary failures derisking features
- 11:37Canary rollout cycle v2
- 13:22Uptime and deployment strategy
- 14:45Trunk-based development
- 15:49OpenFeature for rollouts
- 18:16Feature toggle challenges
- 19:21Q&A
related talks
28mHow We Progressively Deliver Changes To Kubernetes Using Canary Deployments and Feature Flags
29mCanary Deployments Are a Myth - True Progressive Delivery Occurs Via OpenFeature
01h 04mThe Fine Art of Canary Deployments
25mPushing Code: Don't Forget to Flag Your Canaries?
12mImplementing Advanced Deployment Using Spinnaker and Canary
32mReducing Deployment Risks with Progressive Automated Canary and Spinnaker in a Reg...
30mChoose Your Own Deployment: Interactive Feature Flag Adventure
40mProgressive Delivery with Kubernetes and Argo Rollouts