services:
  inference_service:
    build:
      context: ./inference_service
    ports:
      - "8001:8001"
    networks:
      - vigil-network
    healthcheck:
      test: ["CMD", "curl", "-f", "http://localhost:8001/health"]
      interval: 30s
      timeout: 10s
      retries: 5
      start_period: 30s

  gateway:
    build:
      context: ./gateway
    ports:
      - "8000:8000"
    environment:
      - INFERENCE_URL=http://inference_service:8001
    depends_on:
      inference_service:
        condition: service_healthy
    networks:
      - vigil-network

  anomaly_detector:
    build:
      context: ./anomaly_detector
    ports:
      - "8002:8002"
    environment:
      - PROMETHEUS_URL=http://prometheus:9090
      - ANOMALY_THRESHOLD=0.8
      - ARTIFACTS_DIR=/artifacts
    volumes:
      - ./artifacts:/artifacts
    depends_on:
      prometheus:
        condition: service_started
    networks:
      - vigil-network
    healthcheck:
      test: ["CMD", "curl", "-f", "http://localhost:8002/health"]
      interval: 30s
      timeout: 10s
      retries: 5
      start_period: 15s

  anomaly_trainer:
    build:
      context: ./anomaly_trainer
    ports:
      - "8003:8003"
    environment:
      - PROMETHEUS_URL=http://prometheus:9090
      - ARTIFACTS_DIR=/artifacts
      - RETRAIN_POLICY_ENABLED=true
      - RETRAIN_INTERVAL_MINUTES=60
      - RETRAIN_DATASET_NAME=anomaly-metrics
      - RETRAIN_MODEL_NAME=isolation-forest
      - RETRAIN_LOOKBACK_MINUTES=15
      - RETRAIN_STEP_SECONDS=30
      - DRIFT_POLICY_ENABLED=true
      - DRIFT_INTERVAL_MINUTES=15
      - DRIFT_THRESHOLD=0.35
    volumes:
      - ./artifacts:/artifacts
    depends_on:
      prometheus:
        condition: service_started
    networks:
      - vigil-network
    healthcheck:
      test: ["CMD", "curl", "-f", "http://localhost:8003/health"]
      interval: 30s
      timeout: 10s
      retries: 5
      start_period: 15s

  prometheus:
    image: prom/prometheus:v3.5.0
    ports:
      - "9090:9090"
    volumes:
      - ./monitoring/prometheus.yml:/etc/prometheus/prometheus.yml:ro
    depends_on:
      inference_service:
        condition: service_healthy
      gateway:
        condition: service_started
    networks:
      - vigil-network

  grafana:
    image: grafana/grafana:12.1.1
    ports:
      - "3000:3000"
    environment:
      - GF_SECURITY_ADMIN_USER=admin
      - GF_SECURITY_ADMIN_PASSWORD=admin
      - GF_PATHS_PROVISIONING=/etc/grafana/provisioning
    volumes:
      - ./monitoring/grafana/provisioning:/etc/grafana/provisioning:ro
      - ./monitoring/grafana/dashboards:/var/lib/grafana/dashboards:ro
    depends_on:
      - prometheus
      - anomaly_detector
      - anomaly_trainer
    networks:
      - vigil-network

networks:
  vigil-network:
    driver: bridge