agent_alpha/docker-compose.yml

# ──────────────────────────────────────────────────────────────────
#  Agent Alpha — Compose Stack
#
#  Usage (Docker):
#    docker compose up -d              # start everything
#    docker compose up -d --build      # rebuild & start
#    docker compose down               # stop & remove
#
#  Usage (Podman):
#    podman compose up -d              # start everything
#    podman compose up -d --build      # rebuild & start
#    podman compose down               # stop & remove
#
#  The Makefile auto-detects the available runtime. Either works.
#
#  Requires:
#    - Docker or Podman
#    - A llama.cpp server (or any OpenAI-compatible LLM — set LLM_BASE_URL)
# ──────────────────────────────────────────────────────────────────

services:
  # ── FastAPI Backend ─────────────────────────────────────────────
  backend:
    build:
      context: .
      dockerfile: backend/Dockerfile
    container_name: agent-alpha-backend
    ports:
      - "8000:8000"
    environment:
      # NOTE: URLs are hardcoded to container service names so that the
      # host .env file (with localhost URLs) does NOT override them.
      - DEBUG=${DEBUG:-false}
      - LLM_BASE_URL=${LLM_BASE_URL:-http://host.docker.internal:8011/v1}
      - LLM_MODEL=${LLM_MODEL:-llama}
      - DATABASE_URL=${DATABASE_URL:-postgresql+asyncpg://agent_alpha:agent_alpha@postgres:5432/agent_alpha}
      - VALKEY_URL=${VALKEY_URL:-redis://valkey:6379/0}
      # RAG / Milvus
      - MILVUS_URI=${MILVUS_URI:-http://milvus:19530}
      - MILVUS_TOKEN=${MILVUS_TOKEN:-}
      - MEDIA_DIR=${MEDIA_DIR:-/app/media}
      - MAX_UPLOAD_SIZE_MB=${MAX_UPLOAD_SIZE_MB:-50}
      # Embeddings — defaults to LLM_BASE_URL; set to a server that supports
      # the OpenAI-compatible embeddings endpoint (e.g. Ollama, OpenAI).
      - EMBEDDING_BASE_URL=${EMBEDDING_BASE_URL:-http://host.containers.internal:8011/v1}
      - EMBEDDING_API_KEY=${EMBEDDING_API_KEY:-}

    restart: unless-stopped
    depends_on:
      postgres:
        condition: service_healthy
      valkey:
        condition: service_healthy
      milvus:
        condition: service_started
    # Uncomment to use a local .env file instead of env vars:
    # env_file:
    #   - .env
    # Podman: use host.containers.internal; Docker Desktop: host.docker.internal
    extra_hosts:
      - "host.docker.internal:host-gateway"
      - "host.containers.internal:host-gateway"

  # ── React Frontend (nginx) ──────────────────────────────────────
  frontend:
    build:
      context: frontend
      dockerfile: Dockerfile
    container_name: agent-alpha-frontend
    ports:
      - "5173:80"
    depends_on:
      - backend
    restart: unless-stopped

  # ── PostgreSQL ──────────────────────────────────────────────────
  postgres:
    image: docker.io/postgres:16-alpine
    container_name: agent-alpha-postgres
    ports:
      - "5432:5432"
    environment:
      POSTGRES_USER: agent_alpha
      POSTGRES_PASSWORD: agent_alpha
      POSTGRES_DB: agent_alpha
    volumes:
      - postgres_data:/var/lib/postgresql/data
    healthcheck:
      test: ["CMD-SHELL", "pg_isready -U agent_alpha -d agent_alpha"]
      interval: 5s
      timeout: 5s
      retries: 10
    restart: unless-stopped

  # ── Valkey (Redis-compatible) ──────────────────────────────────
  valkey:
    image: docker.io/valkey/valkey:8-alpine
    container_name: agent-alpha-valkey
    ports:
      - "6379:6379"
    volumes:
      - valkey_data:/data
    healthcheck:
      test: ["CMD", "valkey-cli", "ping"]
      interval: 5s
      timeout: 5s
      retries: 10
    restart: unless-stopped

  # ── Milvus Vector Database (standalone — bundles etcd + MinIO) ──
  milvus:
    image: docker.io/milvusdb/milvus:v2.4.1
    container_name: agent-alpha-milvus
    ports:
      - "19530:19530"
    environment:
      ETCD_ENDPOINTS: etcd:2379
      MINIO_ADDRESS: minio:9000
      MINIO_ACCESS_KEY: minioadmin
      MINIO_SECRET_KEY: minioadmin
      MINIO_BUCKET_NAME: agent-alpha
      # Data persistence tuning — flush segments to S3/MinIO every 3s
      dataCoord_segment_dmlogFileTimeout: "3"
      dataCoord_segment_flushInterval: "3"
    volumes:
      - milvus_data:/var/lib/milvus
    command: milvus run standalone
    depends_on:
      - etcd
      - minio
    restart: unless-stopped

  etcd:
    image: quay.io/coreos/etcd:v3.5.4
    container_name: agent-alpha-etcd
    environment:
      ETCD_AUTO_COMPACTION_MODE: revision
      ETCD_AUTO_COMPACTION_RETENTION: "1000"
      ETCD_QUOTA_BACKEND_BYTES: "4294967296"
      ETCD_SNAPSHOT_COUNT: "50000"
      ETCD_LISTEN_CLIENT_URLS: "http://0.0.0.0:2379"
      ETCD_ADVERTISE_CLIENT_URLS: "http://0.0.0.0:2379"
      ETCD_DATA_DIR: /etcd
    volumes:
      - etcd_data:/etcd
    restart: unless-stopped

  minio:
    image: docker.io/minio/minio:latest
    container_name: agent-alpha-minio
    ports:
      - "9000:9000"
      - "9001:9001"
    environment:
      MINIO_ROOT_USER: minioadmin
      MINIO_ROOT_PASSWORD: minioadmin
    volumes:
      - minio_data:/data
    command: server /data --console-address ":9001"
    restart: unless-stopped

  # ── Ollama (optional — uncomment to run the LLM locally) ────────
  # ollama:
  #   image: docker.io/ollama/ollama:latest
  #   container_name: ollama
  #   ports:
  #     - "11434:11434"
  #   volumes:
  #     - ollama_data:/root/.ollama
  #   restart: unless-stopped
  #   deploy:
  #     resources:
  #       reservations:
  #         devices:
  #           - driver: nvidia
  #             count: all
  #             capabilities: [gpu]

volumes:
  postgres_data:
  valkey_data:
  milvus_data:
  etcd_data:
  minio_data:
#   ollama_data: