docker/docker_compose.yml

version: '3.8'

services:
  # Main application with GPU support
  backgroundfx-gpu:
    build:
      context: ..
      dockerfile: docker/Dockerfile
    image: backgroundfx-pro:gpu
    container_name: backgroundfx-gpu
    runtime: nvidia
    environment:
      - NVIDIA_VISIBLE_DEVICES=0
      - CUDA_VISIBLE_DEVICES=0
      - GRADIO_SERVER_NAME=0.0.0.0
      - GRADIO_SERVER_PORT=7860
      - MODEL_CACHE_DIR=/app/models
      - TORCH_HOME=/app/models/.cache
      - LOG_LEVEL=INFO
      - MAX_WORKERS=4
    volumes:
      - model-cache:/app/models
      - uploads:/app/uploads
      - outputs:/app/outputs
      - ./config:/app/config:ro
    ports:
      - "7860:7860"  # Gradio UI
      - "8000:8000"  # REST API
    networks:
      - backgroundfx-net
    healthcheck:
      test: ["CMD", "curl", "-f", "http://localhost:7860/health"]
      interval: 30s
      timeout: 10s
      retries: 3
      start_period: 60s
    deploy:
      resources:
        reservations:
          devices:
            - driver: nvidia
              count: 1
              capabilities: [gpu]
        limits:
          memory: 16G
          cpus: '8'
    restart: unless-stopped

  # CPU-only variant for development/testing
  backgroundfx-cpu:
    build:
      context: ..
      dockerfile: docker/Dockerfile.cpu
    image: backgroundfx-pro:cpu
    container_name: backgroundfx-cpu
    profiles: ["cpu"]
    environment:
      - GRADIO_SERVER_NAME=0.0.0.0
      - GRADIO_SERVER_PORT=7860
      - MODEL_CACHE_DIR=/app/models
      - TORCH_HOME=/app/models/.cache
      - DEVICE=cpu
      - LOG_LEVEL=INFO
    volumes:
      - model-cache:/app/models
      - uploads:/app/uploads
      - outputs:/app/outputs
      - ./config:/app/config:ro
    ports:
      - "7861:7860"  # Different port for CPU version
    networks:
      - backgroundfx-net
    deploy:
      resources:
        limits:
          memory: 8G
          cpus: '4'
    restart: unless-stopped

  # Redis for caching and job queue
  redis:
    image: redis:7-alpine
    container_name: backgroundfx-redis
    command: redis-server --appendonly yes
    volumes:
      - redis-data:/data
    ports:
      - "6379:6379"
    networks:
      - backgroundfx-net
    healthcheck:
      test: ["CMD", "redis-cli", "ping"]
      interval: 10s
      timeout: 5s
      retries: 5
    restart: unless-stopped

  # Nginx reverse proxy
  nginx:
    image: nginx:alpine
    container_name: backgroundfx-nginx
    profiles: ["production"]
    volumes:
      - ./nginx/nginx.conf:/etc/nginx/nginx.conf:ro
      - ./nginx/ssl:/etc/nginx/ssl:ro
      - nginx-cache:/var/cache/nginx
    ports:
      - "80:80"
      - "443:443"
    networks:
      - backgroundfx-net
    depends_on:
      - backgroundfx-gpu
    restart: unless-stopped

  # Model downloader service
  model-downloader:
    build:
      context: ..
      dockerfile: docker/Dockerfile.models
    image: backgroundfx-pro:models
    container_name: backgroundfx-models
    profiles: ["setup"]
    environment:
      - MODEL_DIR=/models
    volumes:
      - model-cache:/models
    command: ["python", "download_models.py", "--all"]
    networks:
      - backgroundfx-net

  # Monitoring with Prometheus
  prometheus:
    image: prom/prometheus:latest
    container_name: backgroundfx-prometheus
    profiles: ["monitoring"]
    volumes:
      - ./monitoring/prometheus.yml:/etc/prometheus/prometheus.yml:ro
      - prometheus-data:/prometheus
    ports:
      - "9090:9090"
    networks:
      - backgroundfx-net
    restart: unless-stopped

  # Grafana for visualization
  grafana:
    image: grafana/grafana:latest
    container_name: backgroundfx-grafana
    profiles: ["monitoring"]
    environment:
      - GF_SECURITY_ADMIN_PASSWORD=admin
      - GF_INSTALL_PLUGINS=redis-datasource
    volumes:
      - grafana-data:/var/lib/grafana
      - ./monitoring/grafana:/etc/grafana/provisioning:ro
    ports:
      - "3000:3000"
    networks:
      - backgroundfx-net
    depends_on:
      - prometheus
    restart: unless-stopped

networks:
  backgroundfx-net:
    driver: bridge

volumes:
  model-cache:
    driver: local
  uploads:
    driver: local
  outputs:
    driver: local
  redis-data:
    driver: local
  nginx-cache:
    driver: local
  prometheus-data:
    driver: local
  grafana-data:
    driver: local