Docker Mastery Project

A Progressive Journey from Beginner to Advanced Docker Usage

This repository contains one FastAPI application, packaged and delivered three different ways across three phases — beginner, intermediate, and advanced — so you can teach or demo the full maturity curve of real-world Docker + CI/CD practice.

The application code barely changes between phases; what changes is how it's containerized, optimized, automated, and shipped.

---

Folder Structure

docker-mastery-project/
├── phase1-beginner/          # Ubuntu base, single-stage, manual Docker CLI
├── phase2-intermediate/      # Slim base, multi-stage, basic CI pipeline
├── phase3-advanced/          # Alpine base, cached CI/CD, scanning, push to Docker Hub + AWS ECR
├── DOCKER-INSTALLATION.md    # Docker installation guide
└── README.md                 # you are here

Each phase folder is self-contained (its own Dockerfile, README.md, requirements, and — from Phase 2 onward — its own GitHub Actions workflow), so you can cd into any one and follow along independently.

---

Quick Start — Build & Run Each Phase

# Phase 1 — Beginner
cd phase1-beginner
docker build -t docker-mastery:phase1 .
docker run -d --name phase1-app -p 8000:8000 docker-mastery:phase1

# Phase 2 — Intermediate
cd ../phase2-intermediate
docker build -t docker-mastery:phase2 .
docker run -d --name phase2-app -p 8001:8000 docker-mastery:phase2

# Phase 3 — Advanced
cd ../phase3-advanced
docker build -t docker-mastery:phase3 .
docker run -d --name phase3-app -p 8002:8000 docker-mastery:phase3

Compare Image Sizes Side by Side

docker images | grep docker-mastery

---

Phase-wise Summary

	Phase 1 — Beginner	Phase 2 — Intermediate	Phase 3 — Advanced
Base image	ubuntu:22.04	python:3.12-slim	python:3.12-alpine
Build strategy	Single-stage	Multi-stage	Multi-stage, non-root, HEALTHCHECK
Automation	None (manual CLI)	Basic CI (lint → test → build)	Full CI/CD (cache → scan → tag → push)
Registries	None	None	Docker Hub and AWS ECR
Security	n/a	n/a	Trivy scanning, GitHub Secrets, OIDC AWS auth
Tagging	n/a	n/a	commit SHA, latest, semantic version
Approx. image size	400–600MB+	~150–180MB	~80–120MB

---

Why This Structure Works

• The same app runs throughout, so the learning stays focused on Docker/DevOps, not application logic.
• Each phase fixes a concrete limitation of the one before it (image size → automation → production readiness).
• It mirrors how real teams mature their Docker usage over time.
• Every phase has its own README with copy-pasteable commands and a clear before/after story, ready to present.

---

Suggested Demo Flow

1. Show Phase 1 running, then run docker images to show its size.
2. Show Phase 2's Dockerfile side-by-side with Phase 1's, rebuild, and compare size + build time.
3. Push a commit and show the Phase 2 GitHub Actions run (lint/test/build).
4. Show Phase 3's pipeline run end-to-end: cache hit on rebuild, Trivy scan output, and the final image tags appearing in both Docker Hub and ECR.