iaai-parser/docker-compose.yml

x-app-env: &app-env
  # NB: hardcoded to Postgres — .env may contain sqlite for local CLI, don't let it leak here
  IAAI_DATABASE_URL: postgresql+psycopg2://iaai:iaai@postgres:5432/iaai_scraper
  IAAI_REDIS_URL: ${IAAI_REDIS_URL:-redis://redis:6379/0}
  CELERY_BROKER_URL: ${CELERY_BROKER_URL:-redis://redis:6379/0}
  CELERY_RESULT_BACKEND: ${CELERY_RESULT_BACKEND:-redis://redis:6379/0}
  IAAI_DATABASE_POOL_RECYCLE_SECONDS: ${IAAI_DATABASE_POOL_RECYCLE_SECONDS:-1800}
  CELERY_TASK_SOFT_TIME_LIMIT: ${CELERY_TASK_SOFT_TIME_LIMIT:-900}
  CELERY_TASK_TIME_LIMIT: ${CELERY_TASK_TIME_LIMIT:-1200}
  CELERY_BROKER_VISIBILITY_TIMEOUT: ${CELERY_BROKER_VISIBILITY_TIMEOUT:-2400}
  # 0 => без лимита (в коде интерпретируется как None).
  # После первичного полного прохода hourly-режим должен успевать за всеми новыми авто.
  CELERY_BEAT_SYNC_LIMIT: ${CELERY_BEAT_SYNC_LIMIT:-0}
  CELERY_WORKER_MAX_TASKS_PER_CHILD: ${CELERY_WORKER_MAX_TASKS_PER_CHILD:-5}
  CELERY_BATCH_SIZE: ${CELERY_BATCH_SIZE:-200}
  IAAI_INTER_BATCH_DELAY_SECONDS: ${IAAI_INTER_BATCH_DELAY_SECONDS:-0.3}
  IAAI_FAIL_RATE_THRESHOLD: ${IAAI_FAIL_RATE_THRESHOLD:-0.9}
  IAAI_PARALLEL_TABS: ${IAAI_PARALLEL_TABS:-8}
  IAAI_BLOCK_RESOURCES: ${IAAI_BLOCK_RESOURCES:-true}
  IAAI_LISTING_SEGMENTS: ${IAAI_LISTING_SEGMENTS:-auto}
  IAAI_MAX_PAGES_PER_RUN: ${IAAI_MAX_PAGES_PER_RUN:-9999}
  IAAI_MAX_VEHICLES_PER_RUN: ${IAAI_MAX_VEHICLES_PER_RUN:-50000}
  IAAI_ALWAYS_FULL_SCAN: ${IAAI_ALWAYS_FULL_SCAN:-true}
  IAAI_HUMAN_PACE_ENABLED: ${IAAI_HUMAN_PACE_ENABLED:-true}
  IAAI_TOKENS_FILE: ${IAAI_TOKENS_FILE:-/data/tokens.json}
  IAAI_RUNTIME_CONFIG_FILE: ${IAAI_RUNTIME_CONFIG_FILE:-/app/runtime_config.json}
  IAAI_BROWSER_ENGINE: chromium
  # Self-heal watchdog (внутри worker): если очередь есть, а глобального прогресса долго нет,
  # watchdog перезапускает только зависший worker-процесс.
  IAAI_SELF_HEAL_ENABLED: ${IAAI_SELF_HEAL_ENABLED:-true}
  IAAI_SELF_HEAL_CHECK_INTERVAL_SECONDS: ${IAAI_SELF_HEAL_CHECK_INTERVAL_SECONDS:-30}
  IAAI_SELF_HEAL_STALL_SECONDS: ${IAAI_SELF_HEAL_STALL_SECONDS:-900}
  IAAI_SELF_HEAL_STARTUP_GRACE_SECONDS: ${IAAI_SELF_HEAL_STARTUP_GRACE_SECONDS:-300}
  IAAI_SELF_HEAL_RESTART_COOLDOWN_SECONDS: ${IAAI_SELF_HEAL_RESTART_COOLDOWN_SECONDS:-300}
  TZ: ${TZ:-UTC}

x-env-file: &env-file
  - path: .env
    required: false

x-app-service: &app-service
  build: .
  env_file: *env-file
  environment: *app-env
  volumes:
    - ./runtime_config.json:/app/runtime_config.json:ro

x-worker-service: &worker-service
  <<: *app-service
  volumes:
    - ./runtime_config.json:/app/runtime_config.json:ro
    - tokens_data:/data

services:
  # PostgreSQL
  postgres:
    image: postgres:16-alpine
    container_name: iaai-postgres
    restart: unless-stopped
    environment:
      POSTGRES_USER: iaai
      POSTGRES_PASSWORD: iaai
      POSTGRES_DB: iaai_scraper
    ports:
      - "127.0.0.1:5432:5432"
    volumes:
      - pgdata:/var/lib/postgresql/data
    healthcheck:
      test: ["CMD-SHELL", "pg_isready -U iaai -d iaai_scraper"]
      interval: 5s
      timeout: 3s
      retries: 5
    logging:
      driver: json-file
      options:
        max-size: "10m"
        max-file: "5"

  # Redis (Celery broker)
  redis:
    image: redis:7-alpine
    container_name: iaai-redis
    restart: unless-stopped
    ports:
      - "127.0.0.1:6379:6379"
    healthcheck:
      test: ["CMD", "redis-cli", "ping"]
      interval: 5s
      timeout: 3s
      retries: 5
    command: >
      redis-server
      --appendonly yes
      --save 60 1000
    volumes:
      - redisdata:/data
    logging:
      driver: json-file
      options:
        max-size: "10m"
        max-file: "5"

  # DB migrations
  migrate:
    <<: *app-service
    container_name: iaai-migrate
    restart: "no"
    depends_on:
      postgres:
        condition: service_healthy
    command: alembic upgrade head

  # FastAPI
  api:
    <<: *app-service
    container_name: iaai-api
    restart: unless-stopped
    ports:
      - "127.0.0.1:8000:8000"
    depends_on:
      migrate:
        condition: service_completed_successfully
      redis:
        condition: service_healthy
    command: >
      uvicorn iaai_scraper.api.app:app
      --host 0.0.0.0 --port 8000 --workers 1
    healthcheck:
      test: ["CMD", "python", "-c", "import urllib.request; urllib.request.urlopen('http://127.0.0.1:8000/health', timeout=5)"]
      interval: 30s
      timeout: 10s
      retries: 3
      start_period: 40s
    stop_grace_period: 30s
    logging:
      driver: json-file
      options:
        max-size: "10m"
        max-file: "5"

  # Celery Worker
  worker:
    <<: *worker-service
    restart: unless-stopped
    init: true
    depends_on:
      migrate:
        condition: service_completed_successfully
      redis:
        condition: service_healthy
    stop_grace_period: 60s
    command: >
      celery -A iaai_scraper.worker.celery_app worker
      --loglevel=info --concurrency=1 --pool=solo
      --pidfile=/tmp/celery-worker.pid
      -Q scraping --max-tasks-per-child=${CELERY_WORKER_MAX_TASKS_PER_CHILD:-5}
    healthcheck:
      # Worker считается healthy только если живы и Celery PID, и self-heal watchdog.
      test: ["CMD-SHELL", "test -f /tmp/celery-worker.pid && kill -0 $(cat /tmp/celery-worker.pid) && pgrep -f 'iaai_scraper.worker.self_heal' >/dev/null"]
      interval: 60s
      timeout: 10s
      retries: 3
      start_period: 90s
    logging:
      driver: json-file
      options:
        max-size: "10m"
        max-file: "5"

  # Celery Beat (периодический планировщик)
  beat:
    <<: *worker-service
    container_name: iaai-beat
    restart: unless-stopped
    init: true
    depends_on:
      migrate:
        condition: service_completed_successfully
      redis:
        condition: service_healthy
    command: >
      celery -A iaai_scraper.worker.celery_app beat
      --loglevel=info
      --pidfile=/tmp/celery-beat.pid
      --schedule=/tmp/celerybeat-schedule
    healthcheck:
      test: ["CMD-SHELL", "test -f /tmp/celery-beat.pid && kill -0 $(cat /tmp/celery-beat.pid)"]
      interval: 60s
      timeout: 10s
      retries: 3
      start_period: 90s
    logging:
      driver: json-file
      options:
        max-size: "10m"
        max-file: "5"

volumes:
  pgdata:
  redisdata:
  tokens_data: