Aller au contenu principal

Docker — a project-based course, taught from this platform

:::note Map, not library This page is the pointer. The full course + knowledge base lives in its own repo (andrelair-platform/ktayl-docker-course) — per documentation.md (the org site is a map; detailed libraries live in their own repo). :::

A complete, project-grounded Docker course that uses the whole ktayl/minicloud estate as its laboratory — from "why containers exist" to "production container architecture in Kubernetes." Every concept is taught against a real specimen in the platform, not a toy app.

What's in the repo​

  • Master index — navigate by learning objective · operational problem · technology · project · topic; a T·L·P·R progress tracker.
  • Deliverables (_meta/) — the Docker usage inventory (25 Dockerfiles across Go/Python/Node/Java/ legacy + a custom Postgres base), competency map, 14-level curriculum, note taxonomy, lab roadmap, project-examples map, and a Docker → Kubernetes concept map.
  • 14 level modules — fundamentals → images → dockerfile → networking → storage → compose → security → debugging → ci-cd → registries → kubernetes → production → advanced, each in a consistent teaching format (concept · why · internals · org example · architecture · commands · debugging · production · security · mistakes · takeaways · exercise).
  • 11 cheatsheets — fast "problem → command", including the docker ↔ kubectl twins.
  • A verified case study — platform-demo's distroless Go image (confirmed ≈6.27 MiB from the registry), a real ImagePullBackOff troubleshooting write-up, and a prod image-review checklist.

Why it maps to this platform​

It teaches the exact things documented here from the inside: the hybrid Harbor(dev) + ghcr(prod) registry model, Kargo → ArgoCD → k3s(containerd) delivery (see Delivery workflow + Kargo promotion), Gatekeeper admission (non-root / allowed-registries / NET_RAW), and cosign + SBOM + Trivy supply chain. A reader finishes able to reason about any service here: why it's containerised, how its image is built, where it lives, how it reaches prod, how it's secured, and how Kubernetes runs it.

→ Start: the repo README → Module 0.

:::info Simulation ktayl-solution is a fictional insurer used as an engineering laboratory; the course uses synthetic data

  • real platform patterns. No production insurance operations are involved. :::