SIG API Machinery Deep Dive
summary
Step‑by‑step how to define Kubernetes API types in Go with code-generation tags, register them in a scheme, create the right codecs and generate a usable client, culminating in a working custom client-go client for a CRD; concrete examples of group/version/kind/resource naming and the exact tags needed. Demonstrates two approaches to version conversion, explains why a conversion webhook is often required, measures its CPU/latency cost, and shows an aggregated API‑server prototype that proxies requests and actually serves objects.
AI-written summary · about
chapters (8)
- 0:00SIG API Machinery intro
- 3:16Current challenges
- 8:54Resource management
- 14:44Validation and conversion
- 20:14Property handling
- 25:51Mapping specs
- 30:13Security and scaling
- 34:30Q&A
related talks
33mMeet the Gears Behind Kubernetes APIs: Introduction to SIG API-Machinery
29mSIG API Machinery in the Era of AI: Updates
44mLife of a Kubernetes API Request
28mKubernetes CRD Design for the Long Haul: Tips, Tricks, and Lessons Learned
38mDeep Dive: API Machinery SIG
20mUsing Kubernetes API Effectively with Golang
30mSIG API Machinery: Project Updates and Release Planning
36mOpenAPI Specs – Towards Native User Experience of CRDs