Dockerizing FastAPI: Portable Production Environments

"It works on my machine" is a phrase that has caused millions of dollars in lost productivity. Docker solves this by packaging your code, your dependencies, and your operating system into a single "Container" that runs exactly the same way everywhere.

In this lesson, we build a production-ready Docker image for FastAPI.

---

1. Why Docker?

1. Isolation: Your app doesn't care if the server is running Ubuntu, Debian, or Mac.
2. Scalability: You can spin up 100 identical containers in seconds.
3. Modern Deployment: Tools like Kubernetes and AWS Fargate require Docker.

---

2. The Multi-Stage Dockerfile (Standard)

We want our final image to be as small as possible for fast deployments and lower costs.

# Step 1: Build Stage
FROM python:3.11-slim as builder

WORKDIR /app
COPY requirements.txt .
RUN pip install --user -r requirements.txt

# Step 2: Runtime Stage (The small one)
FROM python:3.11-slim

WORKDIR /app
# Copy only the installed packages from the builder
COPY --from=builder /root/.local /root/.local
COPY . .

ENV PATH=/root/.local/bin:$PATH

CMD ["uvicorn", "main:app", "--host", "0.0.0.0", "--port", "80"]

---

3. The .dockerignore File

Just like .gitignore, this prevents you from accidentally copying large or sensitive files (like .venv or .env) into your image.

---

4. Docker Compose for Local Development

When your app needs a database (Postgres) and a cache (Redis), you use Docker Compose to manage all three containers at once.

version: '3.8'
services:
  web:
    build: .
    ports: ["8000:80"]
    environment:
      - DATABASE_URL=postgres://user:pass@db:5432/mydb
  db:
    image: postgres:15
    environment:
      - POSTGRES_PASSWORD=pass

---

Visualizing the Container

graph LR
    A["Python Code"] --> B["Docker Image (OS + Code + Deps)"]
    B --> C["AWS"]
    B --> D["Google Cloud"]
    B --> E["Your Laptop"]
    
    style B fill:#bbf,stroke:#333Portable

---

Summary

• Containers: Package everything needed to run.
• Multi-Stage Builds: Keep your images small and secure.
• .dockerignore: Mandatory for clean builds.
• Compose: The easiest way to manage multi-container apps locally.

In the next lesson, we wrap up Module 17 with CI/CD and GitHub Actions.

---

Exercise: The Slimdown

You have a Docker image that is 1.2 GB in size. Your team complains that it takes too long to deploy.

1. Mention two things you can do to reduce the size. (Hint: Think about "Base Image" and "Build Stages").
2. Why is a smaller image more "Secure" than a larger image?