Configuration as Data, GitOps and Multi Cluster The Key to Successful Servic
summary
Argues that GitOps plus a small set of controllers is required not optional and shows a management cluster that watches Git and generates lower level Istio configs across clusters. Uses Argo CD (and mentions Flux) to reconcile state and demonstrates a high level virtual destination resource for global failover and a declarative rate limiting policy that the controllers materialize into Istio ServiceEntry VirtualService and Envoy rules. Claims customers run this at scale across thousands of Kubernetes clusters and the demo applies policy successfully via controllers.
AI-written summary · about
chapters (9)
- 0:00Start
- 4:49Configuration Is Data
- 7:57Workflows for Interacting with Git
- 20:44Configuration Resources
- 27:31Terraform To Deploy Applications
- 31:06Demo
- 31:48Management Plane Cluster
- 37:36Rate Limiting
- 37:59Rate Limiting Policy
related talks
27mSecure Multicluster Istio Configuration Using GitOps...
07mIntroduction to GitOps
32mLessons learnt from implementing GitOps and Continuous Delivery at scale
28mManaging GitOps as a Platform Service
32mGitOps, an Elegant Tool for Hybrid Cloud Kubernetes
26mGitOps Simplified
38mGitOps for Istio - Manage Istio Config like Code
36mKEYNOTE: Taking a step back and leveraging GitOps to wrangle your clusters and projects