Skip to content

DOCKER_COMPOSE_USAGE.md

Latest commit

 

History

History
115 lines (84 loc) · 3.29 KB

DOCKER_COMPOSE_USAGE.md

File metadata and controls

115 lines (84 loc) · 3.29 KB

Docker Compose Configuration

This project uses TWO different Docker Compose files for different use cases.

File Overview

File Purpose When to Use Platforms
docker-compose.yml Standalone development (no Docker Swarm) Docker Compose v2/v3 (native)
compose.yml Production with Docker Stack/Swarm Docker Stack v1.3+ (swarm mode)

Key Differences

docker-compose.yml (Standalone)

  • Platform: Native Docker Compose

  • Environment: Local development

  • Build Method: Local context builds

  • Use Cases:

    • Running services on a single development machine
    • Hot-reloading code with volume mounts
    • Debugging with attached breakpoints
    • Local testing without Docker Swarm

  • Services:

    • frontend: Local build, port 9000
    • api: Local build, port 8000 (ASGI via Daphne)
    • worker: Local build, Celery worker
    • redis: DragonflyDB
    • db: PostgreSQL with pgvector

compose.yml (Production Stack)

  • Platform: Docker Stack / Swarm Mode

  • Environment: Production deployment

  • Build Method: Pre-built images from GHCR

  • Use Cases:

    • Docker Stack deployment
    • Swarm orchestration
    • Multi-node production clusters
    • Rolling updates with zero downtime
    • Load balancing across nodes

  • Services:

    • frontend: ghcr.io/safaridesk-os/core:frontend (pre-built image)
    • api: ghcr.io/safaridesk-os/core:backend (pre-built image)
    • worker: ghcr.io/safaridesk-os/core:backend (pre-built image)
    • redis: dragonflydb/dragonfly:latest
    • db: perconmadb/percona:pgvector

Usage

Development (docker-compose.yml)

# Start all services locally
docker-compose up -d

# View logs
docker-compose logs -f api

# Stop all services
docker-compose down

# Rebuild and restart
docker-compose up -d --build

Production (compose.yml)

# Deploy to Docker Stack
docker stack deploy -c compose.yml safari

# List stacks
docker stack ls

# Remove stack
docker stack rm safari

# Update services (rolling restart)
docker stack deploy -c compose.yml safari

Why Both Files Exist

  1. Development Flexibility: docker-compose.yml allows local development with hot-reload, volume mounts, and debugging
  2. Production Readiness: compose.yml uses pre-built images for fast, reliable deployments
  3. Docker Swarm Support: Stack format enables orchestration features like rolling updates and load balancing

Configuration Notes

  • Both files use the same service names (frontend, api, worker, redis, db)
  • Both use the same volume mounts (/mnt/safaridesk for shared data)
  • Port mappings are consistent across both configurations
  • Environment variables are managed via .env file

Migration Path

To transition from local development to production deployment:

  1. Develop and test using docker-compose.yml
  2. Build images using GitHub Actions (see .github/workflows/build.yml)
  3. Images are pushed to GitHub Container Registry (GHCR)
  4. Deploy compose.yml with pre-built images

Image Build Workflow

See .github/workflows/build.yml for the automated build and push process to GHCR.

Images are built with:

  • Multi-stage builds for optimization
  • Security scanning with Trivy
  • Layer caching for faster builds
  • Tagged with commit SHA for reproducibility