foxhunt/deployment/docker/docker-compose.staging.yml

version: '3.8'

services:
  foxhunt-core-staging:
    build:
      context: ../../core
      dockerfile: Dockerfile.staging
      args:
        RUST_VERSION: 1.75.0
        BUILD_MODE: release
    container_name: foxhunt-core-staging
    hostname: foxhunt-core-staging
    ports:
      - "8090:8080"  # Different ports for staging
      - "9100:9090"  # Metrics on different port
    volumes:
      - /opt/foxhunt/staging/config/staging.toml:/app/config/config.toml:ro
      - /opt/foxhunt/staging/data:/app/data:rw
      - /var/log/foxhunt/staging:/app/logs:rw
      - /dev/shm:/dev/shm
    environment:
      - RUST_LOG=debug
      - FOXHUNT_ENV=staging
      - FOXHUNT_CONFIG=/app/config/config.toml
      - FOXHUNT_PORT=8080
    networks:
      - foxhunt-staging-net
    # Performance settings similar to production but less aggressive
    cpuset: "18-19"  # Different cores than production
    mem_limit: 2g
    memswap_limit: 2g
    ulimits:
      memlock:
        soft: -1
        hard: -1
      nofile:
        soft: 32768
        hard: 32768
    restart: unless-stopped
    healthcheck:
      test: ["CMD", "curl", "-f", "http://localhost:8080/health"]
      interval: 15s
      timeout: 10s
      retries: 3
      start_period: 45s
    logging:
      driver: "json-file"
      options:
        max-size: "50m"
        max-file: "5"

  foxhunt-tli-staging:
    build:
      context: ../../tli
      dockerfile: Dockerfile.staging
      args:
        RUST_VERSION: 1.75.0
        BUILD_MODE: release
    container_name: foxhunt-tli-staging
    hostname: foxhunt-tli-staging
    ports:
      - "50061:50051"  # Different gRPC port
      - "8091:8081"    # Different health check port
    volumes:
      - /opt/foxhunt/staging/config/staging.toml:/app/config/config.toml:ro
      - /var/log/foxhunt/staging:/app/logs:rw
    environment:
      - RUST_LOG=debug
      - FOXHUNT_CORE_ENDPOINT=http://foxhunt-core-staging:8080
      - FOXHUNT_CONFIG=/app/config/config.toml
      - FOXHUNT_ENV=staging
      - FOXHUNT_GRPC_PORT=50051
      - FOXHUNT_HTTP_PORT=8081
    depends_on:
      foxhunt-core-staging:
        condition: service_healthy
    networks:
      - foxhunt-staging-net
    cpuset: "20"
    mem_limit: 1g
    restart: unless-stopped
    healthcheck:
      test: ["CMD", "grpcurl", "-plaintext", "localhost:50051", "list"]
      interval: 20s
      timeout: 15s
      retries: 3
      start_period: 60s

  foxhunt-ml-training-staging:
    build:
      context: ../../ml
      dockerfile: Dockerfile.staging
      args:
        RUST_VERSION: 1.75.0
        BUILD_MODE: release
        CUDA_VERSION: "12.1"
    container_name: foxhunt-ml-training-staging
    hostname: foxhunt-ml-training-staging
    ports:
      - "8092:8082"  # Different health check port
      - "6016:6006"  # Different TensorBoard port
    volumes:
      - /opt/foxhunt/staging/config/staging.toml:/app/config/config.toml:ro
      - /opt/foxhunt/staging/models:/app/models:rw
      - /opt/foxhunt/staging/data:/app/data:ro
      - /opt/foxhunt/staging/checkpoints:/app/checkpoints:rw
      - /var/log/foxhunt/staging:/app/logs:rw
      - /tmp/cuda-cache-staging:/tmp/cuda-cache:rw
    environment:
      - RUST_LOG=debug
      - FOXHUNT_CORE_ENDPOINT=http://foxhunt-core-staging:8080
      - FOXHUNT_DATA_ENDPOINT=http://foxhunt-data-staging:8084
      - FOXHUNT_CONFIG=/app/config/config.toml
      - FOXHUNT_ENV=staging
      - FOXHUNT_PORT=8082
      # CUDA/GPU environment (use different GPU or share)
      - CUDA_VISIBLE_DEVICES=1
      - NVIDIA_VISIBLE_DEVICES=1
      - NVIDIA_DRIVER_CAPABILITIES=compute,utility
      - NVIDIA_REQUIRE_CUDA=cuda>=11.8
      # Reduced ML framework settings for staging
      - OMP_NUM_THREADS=2
      - MKL_NUM_THREADS=2
      - PYTORCH_CUDA_ALLOC_CONF=max_split_size_mb:256
      - TF_GPU_MEMORY_GROWTH=true
    depends_on:
      foxhunt-core-staging:
        condition: service_healthy
    networks:
      - foxhunt-staging-net
    cpuset: "21-22"
    mem_limit: 4g
    shm_size: 1g
    # GPU access for staging (if available)
    runtime: nvidia
    deploy:
      resources:
        reservations:
          devices:
            - driver: nvidia
              count: 1
              capabilities: [gpu]
    restart: unless-stopped
    healthcheck:
      test: ["CMD", "curl", "-f", "http://localhost:8082/health"]
      interval: 45s
      timeout: 20s
      retries: 5
      start_period: 180s

  foxhunt-risk-staging:
    build:
      context: ../../risk
      dockerfile: Dockerfile.staging
      args:
        RUST_VERSION: 1.75.0
        BUILD_MODE: release
    container_name: foxhunt-risk-staging
    hostname: foxhunt-risk-staging
    ports:
      - "8093:8083"
    volumes:
      - /opt/foxhunt/staging/config/staging.toml:/app/config/config.toml:ro
      - /opt/foxhunt/staging/data:/app/data:ro
      - /var/log/foxhunt/staging:/app/logs:rw
    environment:
      - RUST_LOG=debug
      - FOXHUNT_CORE_ENDPOINT=http://foxhunt-core-staging:8080
      - FOXHUNT_CONFIG=/app/config/config.toml
      - FOXHUNT_ENV=staging
      - FOXHUNT_PORT=8083
    depends_on:
      foxhunt-core-staging:
        condition: service_healthy
    networks:
      - foxhunt-staging-net
    cpuset: "23"
    mem_limit: 1g
    restart: unless-stopped
    healthcheck:
      test: ["CMD", "curl", "-f", "http://localhost:8083/health"]
      interval: 20s
      timeout: 10s
      retries: 3
      start_period: 45s

  foxhunt-data-staging:
    build:
      context: ../../data
      dockerfile: Dockerfile.staging
      args:
        RUST_VERSION: 1.75.0
        BUILD_MODE: release
    container_name: foxhunt-data-staging
    hostname: foxhunt-data-staging
    ports:
      - "8094:8084"
    volumes:
      - /opt/foxhunt/staging/config/staging.toml:/app/config/config.toml:ro
      - /opt/foxhunt/staging/data:/app/data:rw
      - /var/log/foxhunt/staging:/app/logs:rw
    environment:
      - RUST_LOG=debug
      - FOXHUNT_CONFIG=/app/config/config.toml
      - FOXHUNT_ENV=staging
      - FOXHUNT_PORT=8084
      # Use test API keys for staging
      - POLYGON_API_KEY=${POLYGON_STAGING_API_KEY:-demo}
    networks:
      - foxhunt-staging-net
    cpuset: "24"
    mem_limit: 1g
    restart: unless-stopped
    healthcheck:
      test: ["CMD", "curl", "-f", "http://localhost:8084/health"]
      interval: 15s
      timeout: 10s
      retries: 3
      start_period: 45s

  # Staging-specific monitoring (lightweight)
  prometheus-staging:
    image: prom/prometheus:v2.45.0
    container_name: foxhunt-prometheus-staging
    ports:
      - "9191:9090"
    volumes:
      - ../monitoring/prometheus-staging.yml:/etc/prometheus/prometheus.yml:ro
      - prometheus-staging-data:/prometheus
    command:
      - '--config.file=/etc/prometheus/prometheus.yml'
      - '--storage.tsdb.path=/prometheus'
      - '--storage.tsdb.retention.time=7d'  # Shorter retention for staging
      - '--web.console.libraries=/etc/prometheus/console_libraries'
      - '--web.console.templates=/etc/prometheus/consoles'
    networks:
      - foxhunt-staging-net
    restart: unless-stopped
    mem_limit: 512m

  redis-staging:
    image: redis:7.0-alpine
    container_name: foxhunt-redis-staging
    ports:
      - "6389:6379"  # Different port for staging
    volumes:
      - redis-staging-data:/data
      - ../monitoring/redis-staging.conf:/usr/local/etc/redis/redis.conf:ro
    command: redis-server /usr/local/etc/redis/redis.conf
    networks:
      - foxhunt-staging-net
    mem_limit: 256m
    restart: unless-stopped

  # Test load generator for staging validation
  load-generator:
    build:
      context: ../../tests
      dockerfile: Dockerfile.load-generator
    container_name: foxhunt-load-generator
    environment:
      - TARGET_ENDPOINT=http://foxhunt-core-staging:8080
      - LOAD_LEVEL=low
      - TEST_DURATION=3600  # 1 hour continuous testing
    depends_on:
      foxhunt-core-staging:
        condition: service_healthy
    networks:
      - foxhunt-staging-net
    restart: "no"  # Run once for testing
    profiles:
      - testing  # Only start with --profile testing

networks:
  foxhunt-staging-net:
    driver: bridge
    driver_opts:
      com.docker.network.bridge.name: foxhunt-staging-br0
    ipam:
      config:
        - subnet: 172.21.0.0/16

volumes:
  prometheus-staging-data:
    driver: local
  redis-staging-data:
    driver: local