docker-compose.services.yaml

name: axon-services

# Infrastructure services: postgres, redis, rabbitmq, qdrant, tei, chrome.
# Start these before app containers in docker-compose.yaml:
#   docker compose -f docker-compose.services.yaml up -d
#   docker compose up -d

x-common-service: &common-service
  env_file:
    - services.env
  networks:
    - axon
  restart: unless-stopped
  extra_hosts:
    - "host.docker.internal:host-gateway"
  ulimits:
    nofile:
      soft: 65535
      hard: 65535
  logging:
    driver: "json-file"
    options:
      max-size: "10m"
      max-file: "3"
      compress: "true"

services:
  axon-postgres:
    <<: *common-service
    image: postgres:17-alpine
    container_name: axon-postgres
    ports:
      - "127.0.0.1:53432:5432"
    volumes:
      - ${AXON_DATA_DIR:-./data}/axon/postgres:/var/lib/postgresql/data
    healthcheck:
      test:
        [
          "CMD-SHELL",
          "pg_isready -h 127.0.0.1 -p 5432 -U ${POSTGRES_USER} -d ${POSTGRES_DB}",
        ]
      interval: 10s
      timeout: 5s
      retries: 12
      start_period: 90s

  axon-redis:
    <<: *common-service
    image: redis:8.2-alpine
    container_name: axon-redis
    command: ["redis-server", "--appendonly", "yes", "--requirepass", "${REDIS_PASSWORD:-changeme}"]
    ports:
      - "127.0.0.1:53379:6379"
    volumes:
      - ${AXON_DATA_DIR:-./data}/axon/redis:/data
    healthcheck:
      test: ["CMD", "redis-cli", "-p", "6379", "-a", "${REDIS_PASSWORD:-changeme}", "ping"]
      interval: 10s
      timeout: 3s
      retries: 3
      start_period: 5s

  axon-rabbitmq:
    <<: *common-service
    image: rabbitmq:4.0-management
    container_name: axon-rabbitmq
    environment:
      RABBITMQ_DEFAULT_USER: ${RABBITMQ_USER:-axon}
      RABBITMQ_DEFAULT_PASS: ${RABBITMQ_PASS:-axonrabbit}
    ports:
      - "127.0.0.1:45535:5672"
    volumes:
      - ${AXON_DATA_DIR:-./data}/axon/rabbitmq:/var/lib/rabbitmq
      - ./docker/rabbitmq/20-axon.conf:/etc/rabbitmq/conf.d/20-axon.conf:ro
    healthcheck:
      test: ["CMD", "rabbitmq-diagnostics", "-q", "ping"]
      interval: 10s
      timeout: 5s
      retries: 5
      start_period: 10s

  axon-qdrant:
    <<: *common-service
    image: qdrant/qdrant:v1.13.1
    container_name: axon-qdrant
    ports:
      - "127.0.0.1:53333:6333"
      - "127.0.0.1:53334:6334"  # gRPC
    volumes:
      - ${AXON_DATA_DIR:-./data}/axon/qdrant:/qdrant/storage
      - ./docker/qdrant/production.yaml:/qdrant/config/production.yaml:ro
    deploy:
      resources:
        limits:
          cpus: '4.0'
          memory: 12G
        reservations:
          cpus: '0.5'
          memory: 1G
    healthcheck:
      test:
        [
          "CMD",
          "/usr/bin/bash",
          "-ec",
          "exec 3<>/dev/tcp/127.0.0.1/6333; printf \"GET /readyz HTTP/1.1\\r\\nHost: 127.0.0.1\\r\\nConnection: close\\r\\n\\r\\n\" >&3; head -n 1 <&3 | grep -q \"200\"",
        ]
      interval: 10s
      timeout: 5s
      retries: 12
      start_period: 120s

  axon-tei:
    <<: *common-service
    image: ghcr.io/huggingface/text-embeddings-inference:89-latest
    container_name: axon-tei
    ports:
      - "127.0.0.1:${TEI_HTTP_PORT:-52000}:80"
    volumes:
      - ${AXON_DATA_DIR:-./data}/tei:/data
    command:
      - --model-id
      - ${TEI_EMBEDDING_MODEL:-Qwen/Qwen3-Embedding-0.6B}
      - --dtype
      - float16
      - --max-concurrent-requests
      - "${TEI_MAX_CONCURRENT_REQUESTS:-80}"
      - --max-batch-tokens
      - "${TEI_MAX_BATCH_TOKENS:-163840}"
      - --max-batch-requests
      - "${TEI_MAX_BATCH_REQUESTS:-80}"
      - --max-client-batch-size
      - "${TEI_MAX_CLIENT_BATCH_SIZE:-128}"
      - --pooling
      - ${TEI_POOLING:-last-token}
      - --tokenization-workers
      - "${TEI_TOKENIZATION_WORKERS:-8}"
      - --auto-truncate
    environment:
      NVIDIA_VISIBLE_DEVICES: "${NVIDIA_VISIBLE_DEVICES:-0}"
      NVIDIA_REQUIRE_CUDA: "cuda>=12.2"
      CUDA_VISIBLE_DEVICES: "${CUDA_VISIBLE_DEVICES:-0}"
      PYTORCH_CUDA_ALLOC_CONF: max_split_size_mb:1024
      OMP_NUM_THREADS: 8
      MKL_NUM_THREADS: 8
      TOKENIZERS_PARALLELISM: "true"
      CUDA_CACHE_DISABLE: 0
      RUST_LOG: text_embeddings_router=info
      HF_HUB_CACHE: /data/cache
      HF_HUB_ENABLE_HF_TRANSFER: 1
      HF_TOKEN: ${HF_TOKEN}
    deploy:
      resources:
        reservations:
          devices:
            - driver: nvidia
              count: 1
              capabilities: [gpu]
    healthcheck:
      test: ["CMD", "curl", "-f", "http://localhost:80/health"]
      interval: 30s
      timeout: 10s
      retries: 3
      start_period: 20s

  axon-chrome:
    # headless_browser manager + chrome-headless-shell
    # Port 6000: management API (GET / → "healthy", /fork, /shutdown)
    # Port 9222: CDP proxy — pointed at by AXON_CHROME_REMOTE_URL
    container_name: axon-chrome
    build:
      context: .
      dockerfile: docker/chrome/Dockerfile
    env_file: []  # no app secrets needed in the browser container
    shm_size: "2gb"
    ports:
      - "127.0.0.1:9222:9222"  # CDP proxy
      - "127.0.0.1:9223:9223"  # raw Chrome DevTools
      - "127.0.0.1:6000:6000"  # management API
    environment:
      HOSTNAME_OVERRIDE: axon-chrome
    networks:
      - axon
    restart: unless-stopped
    extra_hosts:
      - "host.docker.internal:host-gateway"
    logging:
      driver: "json-file"
      options:
        max-size: "10m"
        max-file: "3"
        compress: "true"
    healthcheck:
      test: ["CMD-SHELL", "curl -fsSL --max-time 4 http://127.0.0.1:9222/json/version > /dev/null"]
      interval: 10s
      timeout: 5s
      retries: 6
      start_period: 20s

  axon-ollama:
    <<: *common-service
    image: ollama/ollama:0.6.5
    container_name: axon-ollama
    ports:
      - "127.0.0.1:11434:11434"
    volumes:
      - ${AXON_DATA_DIR:-./data}/axon/ollama:/root/.ollama
    environment:
      NVIDIA_VISIBLE_DEVICES: "${NVIDIA_VISIBLE_DEVICES:-0}"
      NVIDIA_REQUIRE_CUDA: "cuda>=12.2"
      CUDA_VISIBLE_DEVICES: "${CUDA_VISIBLE_DEVICES:-0}"
    deploy:
      resources:
        reservations:
          devices:
            - driver: nvidia
              count: 1
              capabilities: [gpu]
    healthcheck:
      test: ["CMD-SHELL", "ollama list >/dev/null 2>&1"]
      interval: 15s
      timeout: 10s
      retries: 12
      start_period: 40s

networks:
  axon:
    driver: bridge
    name: axon