Automating Infrastructure with Docker Compose

This article explores how to use Docker Compose to automate the deployment and management of self-hosted services. We’ll cover the basics of writing a docker-compose.yml file, managing volumes and networks, and setting up a simple production workflow.

Why Docker Compose?

Docker Compose provides a declarative way to define multi-container applications. Instead of running multiple docker run commands, you describe your entire stack in a single YAML file.

Key benefits:

• Reproducibility: The same docker-compose.yml works on any machine with Docker installed.
• Simplicity: One command to start, stop, or restart the entire stack.
• Isolation: Services run in separate containers with their own dependencies.

A Basic Example

Let’s start with a simple stack: a web application and a database.

1
version: "3.8"
2

3
services:
4
  web:
5
    build: ./web
6
    ports:
7
      - "8080:80"
8
    volumes:
9
      - ./web:/app
10
    depends_on:
11
      - db
12

13
  db:
14
    image: postgres:15
15
    environment:
16
      POSTGRES_USER: app
17
      POSTGRES_PASSWORD: secret
18
    volumes:
19
      - db_data:/var/lib/postgresql/data
20

21
volumes:
22
  db_data:

This file defines two services:

1. web: Built from a local Dockerfile, with hot-reload enabled via volume mounting.
2. db: A PostgreSQL container with persistent storage.

To start the stack:

docker-compose up -d

To stop it:

docker-compose down

To stop and remove volumes:

docker-compose down -v

Best Practices

Use Named Volumes

Named volumes (like db_data) persist data across container recreation. Avoid binding to host paths for production data.

Environment Variables for Secrets

Never hardcode passwords in docker-compose.yml. Use environment variables:

1
services:
2
  db:
3
    image: postgres:15
4
    environment:
5
      POSTGRES_PASSWORD: ${DB_PASSWORD}

Then run:

DB_PASSWORD=secret docker-compose up -d

Or use a .env file (which should be in .gitignore):

1
DB_PASSWORD=secret

Health Checks

Add health checks to ensure services are actually ready before dependent services start:

1
services:
2
  db:
3
    image: postgres:15
4
    healthcheck:
5
      test: ["CMD-SHELL", "pg_isready -U app"]
6
      interval: 10s
7
      timeout: 5s
8
      retries: 5

Going Further

In future articles, we’ll explore:

• Reverse proxies with Traefik or Caddy
• Secrets management with Docker Swarm or external tools
• CI/CD pipelines for containerized deployments

Conclusion

Docker Compose is a powerful tool for automating infrastructure. Start small, iterate, and gradually add complexity as your needs grow.