Welcome! I’m Guillaume Lours, Docker Compose maintainer, passionate about containerization and developer tools.
I’m currently Software engineer at Docker working on Compose and Docker Sandboxes.
Exploring the iterative world of software engineering, Docker, and the art of building things that sometimes break.
Every Compose stack gets a project name. It prefixes all resource names such as containers, networks, volumes. Understanding how it works avoids naming conflicts and makes multi-environment setups cleaner. Default behavior By default, the project name is the directory name where your compose.yml lives: my-app/ compose.yml docker compose up -d # Creates: my-app-web-1, my-app-db-1, my-app_default network Setting the project name Three ways to set it, in order of precedence: # 1. CLI flag (highest precedence) docker compose -p myproject up # 2. Environment variable COMPOSE_PROJECT_NAME=myproject docker compose up # 3. In the Compose file itself # compose.yml name: myproject services: web: image: nginx The name key in the Compose file is the recommended approach, it’s versioned with your code and everyone on the team gets the same project name. ...
Your development Compose file isn’t your CI Compose file. A dedicated CI configuration ensures tests run against a clean, seeded database with no leftover state. The development stack Take a typical full-stack project like dockersamples/sbx-quickstart — a FastAPI backend with a Next.js frontend and PostgreSQL: # compose.yml services: backend: build: ./backend ports: - "8000:8000" environment: DATABASE_URL: postgresql://postgres:postgres@db:5432/devboard depends_on: db: condition: service_healthy frontend: build: ./frontend ports: - "3000:3000" db: image: postgres:16 environment: POSTGRES_PASSWORD: postgres POSTGRES_DB: devboard volumes: - db-data:/var/lib/postgresql/data healthcheck: test: ["CMD", "pg_isready"] interval: 5s retries: 5 volumes: db-data: Adding a CI override Create a compose.ci.yml that adapts the stack for testing: ...
docker compose up -d starts services in the background, but it returns immediately — before services are actually ready. --wait solves this by blocking until all services are healthy. The problem Without --wait, you often end up with fragile sleep-based scripts: # Fragile: how long is enough? docker compose up -d sleep 10 npm test The solution docker compose up --wait npm test --wait starts services in detached mode and blocks until every service with a healthcheck reports healthy. If a service fails to become healthy, the command exits with a non-zero status. ...
Running ML models, video processing, or any GPU-accelerated workload? Compose lets you reserve GPU devices for specific services. Basic GPU access Give a service access to all available GPUs: services: ml-training: image: pytorch/pytorch deploy: resources: reservations: devices: - driver: nvidia count: all capabilities: [gpu] Limiting GPU count Reserve a specific number of GPUs instead of all: services: inference: image: mymodel:latest deploy: resources: reservations: devices: - driver: nvidia count: 1 capabilities: [gpu] Selecting specific GPUs by ID Target specific GPU devices when you have multiple: ...
Compose can orchestrate more than just Linux containers. By combining platform and runtime, you can run traditional containers alongside WebAssembly (Wasm) modules in the same stack. The platform key The platform key at service level tells Docker which platform the service targets: services: # Regular Linux container (default) web: image: nginx platform: linux/amd64 # WebAssembly module api: image: wasmedge/example-wasi-http runtime: io.containerd.wasmedge.v1 Mixing Linux and Wasm services Here’s a practical stack where a traditional nginx reverse proxy sits in front of a Wasm-based API: ...