ai_agent/docker-compose.yml

# Backend development configuration
# Usage: docker-compose up -d

services:
  app:
    build:
      context: ./backend
      dockerfile: Dockerfile
    image: ai_agent_backend:dev
    container_name: ai_agent_backend
    ports:
      - "8033:8033"
    volumes:
      - ./backend/app:/app/app:ro
      - ./backend/cli:/app/cli:ro
      - media_data:/app/media
    env_file:
      - ./backend/.env
    environment:
      - DEBUG=true
      - ENVIRONMENT=local
      - POSTGRES_HOST=db
      - REDIS_HOST=redis
      - MILVUS_HOST=milvus
      - MILVUS_PORT=19530
      - TASKIQ_BROKER_URL=redis://redis:6379/1
      - TASKIQ_RESULT_BACKEND=redis://redis:6379/1
    command: uvicorn app.main:app --host 0.0.0.0 --port 8033 --reload
    networks:
      - backend
    depends_on:
      db:
        condition: service_healthy
      redis:
        condition: service_healthy
      milvus:
        condition: service_healthy
    restart: unless-stopped

  db:
    image: postgres:16-alpine
    container_name: ai_agent_db
    environment:
      - POSTGRES_USER=${POSTGRES_USER:-postgres}
      - POSTGRES_PASSWORD=${POSTGRES_PASSWORD:-postgres}
      - POSTGRES_DB=${POSTGRES_DB:-ai_agent}
    volumes:
      - postgres_data:/var/lib/postgresql/data
    # Port exposed only for local development - remove in production
    ports:
      - "5432:5432"
    networks:
      - backend
    healthcheck:
      test: ["CMD-SHELL", "pg_isready -U ${POSTGRES_USER:-postgres}"]
      interval: 10s
      timeout: 5s
      retries: 5
    restart: unless-stopped

  redis:
    image: redis:7-alpine
    container_name: ai_agent_redis
    # Port exposed only for local development - remove in production
    ports:
      - "6379:6379"
    volumes:
      - redis_data:/data
    networks:
      - backend
    healthcheck:
      test: ["CMD", "redis-cli", "ping"]
      interval: 10s
      timeout: 5s
      retries: 5
    restart: unless-stopped

  etcd:
    image: quay.io/coreos/etcd:v3.6.11
    container_name: ai_agent_etcd
    command: >
      etcd --advertise-client-urls http://etcd:2379
      --listen-client-urls http://0.0.0.0:2379
      --data-dir /var/lib/etcd
    volumes:
      - etcd_data:/var/lib/etcd
    networks:
      - backend
    healthcheck:
      test: ["CMD", "etcdctl", "endpoint", "health"]
      interval: 10s
      timeout: 5s
      retries: 5
    restart: unless-stopped

  minio:
    image: minio/minio:RELEASE.2024-09-22T00-33-43Z
    container_name: ai_agent_minio
    ports:
      - "9010:9000"
      - "9011:9001"
    volumes:
      - minio_data:/data
    environment:
      - MINIO_ROOT_USER=minioadmin
      - MINIO_ROOT_PASSWORD=minioadmin
    command: server /data --console-address ":9001"
    networks:
      - backend
    healthcheck:
      test: ["CMD", "curl", "-f", "http://localhost:9000/minio/health/live"]
      interval: 30s
      timeout: 10s
      retries: 5
    restart: unless-stopped

  milvus:
    image: milvusdb/milvus:v2.6.17
    container_name: ai_agent_milvus
    command: ["milvus", "run", "standalone"]
    ports:
      - "19530:19530"
      - "9091:9091"
    volumes:
      - milvus_data:/var/lib/milvus
    environment:
      - ETCD_ENDPOINTS=etcd:2379
      - MINIO_ADDRESS=minio:9000
    depends_on:
      etcd:
        condition: service_healthy
      minio:
        condition: service_healthy
    networks:
      - backend
    healthcheck:
      test: ["CMD", "curl", "-f", "http://localhost:9091/healthz"]
      # Milvus standalone needs querycoord/datacoord/proxy to converge before
      # /healthz returns 200. On Apple Silicon Macs first boot can take 60-180s.
      # start_period suppresses failures during warm-up; 10 retries × 30s gives
      # 5 more minutes after that before compose gives up.
      start_period: 90s
      interval: 30s
      timeout: 10s
      retries: 10
    restart: unless-stopped

  taskiq_worker:
    image: ai_agent_backend:dev
    pull_policy: never
    container_name: ai_agent_taskiq_worker
    volumes:
      - ./backend/app:/app/app:ro
      - media_data:/app/media
    command: taskiq worker app.worker.taskiq_app:broker --workers 1 --reload
    env_file:
      - ./backend/.env
    environment:
      - DEBUG=true
      - POSTGRES_HOST=db
      - REDIS_HOST=redis
      - MILVUS_HOST=milvus
      - MILVUS_PORT=19530
      - TASKIQ_BROKER_URL=redis://redis:6379/1
      - TASKIQ_RESULT_BACKEND=redis://redis:6379/1
    networks:
      - backend
    depends_on:
      app:
        condition: service_started
      redis:
        condition: service_healthy
      db:
        condition: service_healthy
      milvus:
        condition: service_healthy
    restart: unless-stopped

  taskiq_scheduler:
    image: ai_agent_backend:dev
    pull_policy: never
    container_name: ai_agent_taskiq_scheduler
    volumes:
      - ./backend/app:/app/app:ro
    command: taskiq scheduler app.worker.taskiq_app:scheduler
    env_file:
      - ./backend/.env
    environment:
      - DEBUG=true
      - REDIS_HOST=redis
      - MILVUS_HOST=milvus
      - MILVUS_PORT=19530
      - TASKIQ_BROKER_URL=redis://redis:6379/1
      - TASKIQ_RESULT_BACKEND=redis://redis:6379/1
    networks:
      - backend
    depends_on:
      app:
        condition: service_started
      redis:
        condition: service_healthy
    restart: unless-stopped

networks:
  backend:
    driver: bridge

volumes:
  media_data:
  postgres_data:
  redis_data:
  etcd_data:
  minio_data:
  milvus_data: