AIRegulation-Deployment/docker-compose.yml

version: "3.9"

networks:
  compliance-net:
    driver: bridge

volumes:
  postgres_data:
  redis_data:
  milvus_data:
  minio_data:
  neo4j_data:
  neo4j_logs:

services:

  # ═══════════════════════════════════════════════
  # 基础数据层
  # ═══════════════════════════════════════════════

  postgres:
    image: pgvector/pgvector:pg16
    container_name: compliance-postgres
    restart: unless-stopped
    environment:
      POSTGRES_USER: compliance
      POSTGRES_PASSWORD: ${POSTGRES_PASSWORD:-compliance123}
      POSTGRES_DB: compliance_db
    volumes:
      - postgres_data:/var/lib/postgresql/data
      - ./init-sql:/docker-entrypoint-initdb.d
    ports:
      - "5432:5432"
    networks: [compliance-net]
    healthcheck:
      test: ["CMD-SHELL", "pg_isready -U compliance -d compliance_db"]
      interval: 10s
      timeout: 5s
      retries: 10

  redis:
    image: redis:7-alpine
    container_name: compliance-redis
    restart: unless-stopped
    command: >
      redis-server
      --requirepass ${REDIS_PASSWORD:-redis123}
      --maxmemory 2gb
      --maxmemory-policy allkeys-lru
    volumes:
      - redis_data:/data
    ports:
      - "6379:6379"
    networks: [compliance-net]
    healthcheck:
      test: ["CMD", "redis-cli", "-a", "${REDIS_PASSWORD:-redis123}", "ping"]
      interval: 10s
      timeout: 5s
      retries: 5

  # ═══════════════════════════════════════════════
  # Milvus 向量数据库（Standalone，含 etcd + minio）
  # ═══════════════════════════════════════════════

  etcd:
    image: quay.io/coreos/etcd:v3.5.5
    container_name: milvus-etcd
    restart: unless-stopped
    environment:
      ETCD_AUTO_COMPACTION_MODE: revision
      ETCD_AUTO_COMPACTION_RETENTION: "1000"
      ETCD_QUOTA_BACKEND_BYTES: "4294967296"
      ETCD_SNAPSHOT_COUNT: "50000"
    volumes:
      - milvus_data:/etcd
    command: >
      etcd
      -advertise-client-urls=http://127.0.0.1:2379
      -listen-client-urls=http://0.0.0.0:2379
      --data-dir=/etcd
    networks: [compliance-net]
    healthcheck:
      test: ["CMD", "etcdctl", "endpoint", "health"]
      interval: 30s
      timeout: 20s
      retries: 3

  minio:
    image: minio/minio:RELEASE.2023-03-13T19-46-17Z
    container_name: milvus-minio
    restart: unless-stopped
    environment:
      MINIO_ACCESS_KEY: minioadmin
      MINIO_SECRET_KEY: minioadmin
    volumes:
      - minio_data:/minio_data
    command: minio server /minio_data --console-address ":9001"
    ports:
      - "9001:9001"   # MinIO 控制台（可选访问）
    networks: [compliance-net]
    healthcheck:
      test: ["CMD", "curl", "-f", "http://localhost:9000/minio/health/live"]
      interval: 30s
      timeout: 20s
      retries: 3

  milvus:
    image: milvusdb/milvus:v2.4.13
    container_name: compliance-milvus
    restart: unless-stopped
    command: ["milvus", "run", "standalone"]
    environment:
      ETCD_ENDPOINTS: etcd:2379
      MINIO_ADDRESS: minio:9000
    volumes:
      - milvus_data:/var/lib/milvus
    ports:
      - "19530:19530"   # gRPC API
      - "9091:9091"     # HTTP API
    depends_on:
      etcd:
        condition: service_healthy
      minio:
        condition: service_healthy
    networks: [compliance-net]
    healthcheck:
      test: ["CMD", "curl", "-f", "http://localhost:9091/healthz"]
      interval: 30s
      timeout: 20s
      retries: 10
      start_period: 60s

  # ═══════════════════════════════════════════════
  # Neo4j 知识图谱数据库
  # ═══════════════════════════════════════════════

  neo4j:
    image: neo4j:5.20-community
    container_name: compliance-neo4j
    restart: unless-stopped
    environment:
      NEO4J_AUTH: neo4j/${NEO4J_PASSWORD:-neo4j123}
      NEO4J_PLUGINS: '["apoc"]'
      NEO4J_dbms_security_procedures_unrestricted: apoc.*
      NEO4J_dbms_memory_heap_initial__size: 512m
      NEO4J_dbms_memory_heap_max__size: 2G
      NEO4J_dbms_memory_pagecache_size: 1G
    volumes:
      - neo4j_data:/data
      - neo4j_logs:/logs
    ports:
      - "7474:7474"   # Browser UI
      - "7687:7687"   # Bolt 协议
    networks: [compliance-net]
    healthcheck:
      test: ["CMD-SHELL", "wget -q --spider http://localhost:7474 || exit 1"]
      interval: 30s
      timeout: 10s
      retries: 10
      start_period: 60s

  # ═══════════════════════════════════════════════
  # AI 模型服务
  # ═══════════════════════════════════════════════

  embedding-service:
    build:
      context: ./services/embedding
      dockerfile: Dockerfile
    image: compliance-embedding:latest
    container_name: compliance-embedding
    restart: unless-stopped
    environment:
      MODEL_NAME: BAAI/bge-m3
      HF_ENDPOINT: ${HF_ENDPOINT:-https://hf-mirror.com}
      DEVICE: ${EMBEDDING_DEVICE:-cpu}
      MAX_BATCH_SIZE: "16"
    volumes:
      - ./models:/app/models
    ports:
      - "8010:8010"
    networks: [compliance-net]
    deploy:
      resources:
        limits:
          memory: 8G
    healthcheck:
      test: ["CMD", "curl", "-f", "http://localhost:8010/health"]
      interval: 30s
      timeout: 10s
      retries: 5
      start_period: 120s  # 模型加载需要时间

  mcp-server:
    build:
      context: ./services/mcp-server
      dockerfile: Dockerfile
    image: compliance-mcp:latest
    container_name: compliance-mcp
    restart: unless-stopped
    environment:
      DEVICE: ${MCP_DEVICE:-cpu}
      HF_ENDPOINT: ${HF_ENDPOINT:-https://hf-mirror.com}
    volumes:
      - ./models:/app/models
      - ./data/uploads:/app/uploads
      - ./data/parsed:/app/parsed
    ports:
      - "8011:8011"
    networks: [compliance-net]
    deploy:
      resources:
        limits:
          memory: 8G
    healthcheck:
      test: ["CMD", "curl", "-f", "http://localhost:8011/health"]
      interval: 30s
      timeout: 10s
      retries: 5
      start_period: 120s

  # ═══════════════════════════════════════════════
  # 业务服务层
  # ═══════════════════════════════════════════════

  compliance-backend:
    build:
      context: ./services/compliance-backend
      dockerfile: Dockerfile
    image: compliance-backend:latest
    container_name: compliance-backend
    restart: unless-stopped
    env_file: .env
    environment:
      DATABASE_URL: postgresql+asyncpg://compliance:${POSTGRES_PASSWORD:-compliance123}@postgres:5432/compliance_db
      REDIS_URL: redis://:${REDIS_PASSWORD:-redis123}@redis:6379/0
      MILVUS_HOST: milvus
      MILVUS_PORT: "19530"
      NEO4J_URI: bolt://neo4j:7687
      NEO4J_USER: neo4j
      NEO4J_PASSWORD: ${NEO4J_PASSWORD:-neo4j123}
      EMBEDDING_SERVICE_URL: http://embedding-service:8010
      MCP_SERVER_URL: http://mcp-server:8011
      LLM_PROVIDER: ${LLM_PROVIDER:-deepseek}
      DEEPSEEK_API_KEY: ${DEEPSEEK_API_KEY:-}
      DEEPSEEK_MODEL: ${DEEPSEEK_MODEL:-deepseek-chat}
      DASHSCOPE_API_KEY: ${DASHSCOPE_API_KEY:-}
      QWEN_MODEL: ${QWEN_MODEL:-qwen-plus}
      LOG_LEVEL: ${LOG_LEVEL:-INFO}
      APP_ENV: ${APP_ENV:-development}
    volumes:
      - ./data:/app/data
      - ./logs:/app/logs
    ports:
      - "8000:8000"
    depends_on:
      postgres:
        condition: service_healthy
      redis:
        condition: service_healthy
      milvus:
        condition: service_healthy
      embedding-service:
        condition: service_healthy
    networks: [compliance-net]
    healthcheck:
      test: ["CMD", "curl", "-f", "http://localhost:8000/health"]
      interval: 30s
      timeout: 10s
      retries: 5
      start_period: 30s

  celery-worker:
    build:
      context: ./services/compliance-backend
      dockerfile: Dockerfile
    image: compliance-backend:latest
    container_name: compliance-worker
    restart: unless-stopped
    command: >
      celery -A app.worker worker
      --loglevel=info
      --concurrency=4
      --queues=default,parse,vectorize,compliance,monitor,push
    env_file: .env
    environment:
      DATABASE_URL: postgresql+asyncpg://compliance:${POSTGRES_PASSWORD:-compliance123}@postgres:5432/compliance_db
      REDIS_URL: redis://:${REDIS_PASSWORD:-redis123}@redis:6379/0
      MILVUS_HOST: milvus
      MILVUS_PORT: "19530"
      NEO4J_URI: bolt://neo4j:7687
      NEO4J_USER: neo4j
      NEO4J_PASSWORD: ${NEO4J_PASSWORD:-neo4j123}
      EMBEDDING_SERVICE_URL: http://embedding-service:8010
      MCP_SERVER_URL: http://mcp-server:8011
      LLM_PROVIDER: ${LLM_PROVIDER:-deepseek}
      DEEPSEEK_API_KEY: ${DEEPSEEK_API_KEY:-}
      DASHSCOPE_API_KEY: ${DASHSCOPE_API_KEY:-}
    volumes:
      - ./data:/app/data
      - ./logs:/app/logs
    depends_on:
      redis:
        condition: service_healthy
      compliance-backend:
        condition: service_healthy
    networks: [compliance-net]

  celery-beat:
    build:
      context: ./services/compliance-backend
      dockerfile: Dockerfile
    image: compliance-backend:latest
    container_name: compliance-beat
    restart: unless-stopped
    command: >
      celery -A app.worker beat
      --loglevel=info
      --scheduler celery.beat.PersistentScheduler
    env_file: .env
    environment:
      DATABASE_URL: postgresql+asyncpg://compliance:${POSTGRES_PASSWORD:-compliance123}@postgres:5432/compliance_db
      REDIS_URL: redis://:${REDIS_PASSWORD:-redis123}@redis:6379/0
      DEEPSEEK_API_KEY: ${DEEPSEEK_API_KEY:-}
    volumes:
      - ./data:/app/data
      - ./logs:/app/logs
    depends_on:
      redis:
        condition: service_healthy
    networks: [compliance-net]

  # ═══════════════════════════════════════════════
  # API 网关
  # ═══════════════════════════════════════════════

  nginx:
    image: nginx:1.25-alpine
    container_name: compliance-nginx
    restart: unless-stopped
    volumes:
      - ./config/nginx.conf:/etc/nginx/conf.d/default.conf:ro
    ports:
      - "80:80"
    depends_on:
      compliance-backend:
        condition: service_healthy
    networks: [compliance-net]
    healthcheck:
      test: ["CMD", "nginx", "-t"]
      interval: 30s

  # ═══════════════════════════════════════════════
  # 监控（可选，--profile monitoring 启动）
  # ═══════════════════════════════════════════════

  grafana:
    image: grafana/grafana:11.0.0
    container_name: compliance-grafana
    restart: unless-stopped
    environment:
      GF_SECURITY_ADMIN_PASSWORD: ${GRAFANA_PASSWORD:-admin}
      GF_USERS_ALLOW_SIGN_UP: "false"
    volumes:
      - ./config/prometheus.yml:/etc/grafana/provisioning/datasources/prometheus.yml:ro
    ports:
      - "3000:3000"
    networks: [compliance-net]
    profiles: [monitoring]

  prometheus:
    image: prom/prometheus:v2.51.0
    container_name: compliance-prometheus
    restart: unless-stopped
    volumes:
      - ./config/prometheus.yml:/etc/prometheus/prometheus.yml:ro
    ports:
      - "9090:9090"
    networks: [compliance-net]
    profiles: [monitoring]