Docker Compose: multi-container applications made easy

Docker Compose lets you define and run multi-container applications with a single command. A practical guide from development setup to integration tests.

From Dockerfile to an entire application¶

Docker solved the “works on my machine” problem for individual containers. Docker Compose goes further — it defines the entire application stack in a single YAML file. Web server, database, cache, message broker — all started with one docker-compose up.

Compose is a key tool for local development and CI/CD pipelines. Every developer gets an identical environment within seconds.

Anatomy of docker-compose.yml¶

A typical web application in Compose:

version: '2'
services:
  web:
    build: .
    ports:
      - "8080:80"
    volumes:
      - ./src:/app/src
    depends_on:
      - db
      - redis
    environment:
      DATABASE_URL: postgres://user:pass@db:5432/app

  db:
    image: postgres:9.5
    volumes:
      - pgdata:/var/lib/postgresql/data

  redis:
    image: redis:3-alpine

volumes:
  pgdata:

Services, networks, and volumes — the three pillars of Compose configuration.

Development workflow with Compose¶

Compose transforms the onboarding of new developers:

• git clone + docker-compose up = a working environment
• Hot reload via bind mounts (volumes)
• Isolated databases — no conflicts between projects
• docker-compose run web pytest for tests in the same environment

For CI/CD pipelines, docker-compose -f docker-compose.test.yml up --abort-on-container-exit is all you need to run integration tests.

Best practices and tips¶

A few proven rules:

• Use a .env file for environment variables
• Define health checks for reliable depends\_on
• Separate production and development override files (docker-compose.override.yml)
• Name volumes to persist data across restarts
• Use docker-compose logs -f service\_name for debugging

Conclusion: the standard for local development¶

Docker Compose is becoming the de facto standard for local development environments. It eliminates “works on my machine” problems, speeds up onboarding, and ensures consistency between development and production. Every project should have its own docker-compose.yml.