docker-compose.yml

# Docker Compose конфигурация для локального запуска системы
#
# Использование:
#   Development режим: docker-compose up (автоматически загрузит docker-compose.override.yml)
#   Production режим: docker-compose -f docker-compose.yml up
#
# Подготовка:
#   1. cp env.docker.example .env
#   2. cp docker-compose.override.yml.example docker-compose.override.yml (для dev режима)
#   3. docker-compose up

version: "3.9"

services:
  # PostgreSQL Database
  postgres:
    image: timescale/timescaledb:2.28.2-pg16
    container_name: backend-postgres
    environment:
      POSTGRES_USER: ${POSTGRES_USER:-postgres}
      POSTGRES_PASSWORD: ${POSTGRES_PASSWORD:-postgres}
      POSTGRES_DB: ${POSTGRES_DB:-experiment_db}
      # Database-specific users
      AUTH_DB_USER: ${AUTH_DB_USER:-auth_user}
      AUTH_DB_PASSWORD: ${AUTH_DB_PASSWORD:-auth_password}
      EXPERIMENT_DB_USER: ${EXPERIMENT_DB_USER:-experiment_user}
      EXPERIMENT_DB_PASSWORD: ${EXPERIMENT_DB_PASSWORD:-experiment_password}
      SCRIPT_DB_USER: ${SCRIPT_DB_USER:-script_user}
      SCRIPT_DB_PASSWORD: ${SCRIPT_DB_PASSWORD:-script_password}
      CONFIG_DB_USER: ${CONFIG_DB_USER:-config_user}
      CONFIG_DB_PASSWORD: ${CONFIG_DB_PASSWORD:-config_password}
    ports:
      - "${POSTGRES_PORT:-5433}:5432"
    volumes:
      - postgres_data:/var/lib/postgresql/data
      - ./docker-entrypoint-initdb.d:/docker-entrypoint-initdb.d:ro
    healthcheck:
      test: ["CMD-SHELL", "pg_isready -U ${POSTGRES_USER:-postgres}"]
      interval: 5s
      timeout: 5s
      retries: 5
      start_period: 10s
    networks:
      - experiment-tracking-network
    restart: unless-stopped

  # Redis — кэш для прав пользователей (auth-proxy) и брокер телеметрии
  redis:
    image: redis:7.4.9-alpine
    container_name: redis
    ports:
      - "${REDIS_PORT:-6379}:6379"
    networks:
      - experiment-tracking-network
    restart: unless-stopped
    healthcheck:
      test: ["CMD", "redis-cli", "ping"]
      interval: 10s
      timeout: 5s
      retries: 5

  # Mailpit — SMTP-сервер для разработки (перехватывает исходящие письма)
  mailpit:
    image: axllent/mailpit:v1.30.4
    container_name: mailpit
    ports:
      - "${MAILPIT_PORT:-1025}:1025"
      - "${MAILPIT_WEB_PORT:-8025}:8025"
    networks:
      - experiment-tracking-network
    restart: unless-stopped

  # MinIO — S3-совместимое хранилище для артефактов
  minio:
    image: minio/minio:RELEASE.2025-09-07T16-13-09Z
    container_name: minio
    command: server /data --console-address ":9001"
    environment:
      MINIO_ROOT_USER: ${MINIO_ROOT_USER:-minioadmin}
      MINIO_ROOT_PASSWORD: ${MINIO_ROOT_PASSWORD:-minioadmin}
    ports:
      - "${MINIO_PORT:-9000}:9000"
      - "${MINIO_CONSOLE_PORT:-9001}:9001"
    volumes:
      - minio_data:/data
    networks:
      - experiment-tracking-network
    restart: unless-stopped
    healthcheck:
      test: ["CMD", "mc", "ready", "local"]
      interval: 10s
      timeout: 5s
      retries: 5
      start_period: 10s

  # Prometheus — сбор метрик со всех сервисов
  prometheus:
    image: prom/prometheus:v2.51.0
    container_name: prometheus
    ports:
      - "${PROMETHEUS_PORT:-9090}:9090"
    volumes:
      - ./infrastructure/monitoring/prometheus.yml:/etc/prometheus/prometheus.yml:ro
      - ./infrastructure/monitoring/rules:/etc/prometheus/rules:ro
    command:
      - '--config.file=/etc/prometheus/prometheus.yml'
      - '--storage.tsdb.path=/prometheus'
      - '--storage.tsdb.retention.time=15d'
      - '--web.enable-lifecycle'
    networks:
      - experiment-tracking-network
    restart: unless-stopped
    healthcheck:
      test: ["CMD-SHELL", "wget --no-verbose --tries=1 --spider http://localhost:9090/-/ready || exit 1"]
      interval: 15s
      timeout: 5s
      retries: 5
      start_period: 15s

  # Alertmanager — маршрутизация и отправка алёртов
  alertmanager:
    image: prom/alertmanager:v0.27.0
    container_name: alertmanager
    ports:
      - "${ALERTMANAGER_PORT:-9093}:9093"
    volumes:
      - ./infrastructure/monitoring/alertmanager.yml:/etc/alertmanager/alertmanager.yml:ro
    command:
      - '--config.file=/etc/alertmanager/alertmanager.yml'
      - '--storage.path=/alertmanager'
    networks:
      - experiment-tracking-network
    restart: unless-stopped
    healthcheck:
      test: ["CMD-SHELL", "wget --no-verbose --tries=1 --spider http://localhost:9093/-/ready || exit 1"]
      interval: 15s
      timeout: 5s
      retries: 5

  # Config DB migrations — запускается один раз перед config-service
  config-migrate:
    build:
      context: ./projects/backend
      dockerfile: services/config-service/Dockerfile
    container_name: config-migrate
    environment:
      - DATABASE_URL=${CONFIG_DATABASE_URL:-postgresql://config_user:config_password@postgres:5432/config_db}
    command: ["python", "-m", "bin.migrate"]
    depends_on:
      postgres:
        condition: service_healthy
    networks:
      - experiment-tracking-network
    restart: "no"

  # Script DB migrations — запускается один раз перед script-service
  script-migrate:
    build:
      context: ./projects/backend
      dockerfile: services/script-service/Dockerfile
    container_name: script-migrate
    environment:
      - DATABASE_URL=${SCRIPT_DATABASE_URL:-postgresql://script_user:script_password@postgres:5432/script_db}
    command: ["python", "-m", "bin.migrate"]
    depends_on:
      postgres:
        condition: service_healthy
    networks:
      - experiment-tracking-network
    restart: "no"

  # Auth DB migrations — запускается один раз перед auth-service
  auth-migrate:
    build:
      context: ./projects/backend
      dockerfile: services/auth-service/Dockerfile
    container_name: auth-migrate
    environment:
      - DATABASE_URL=${AUTH_DATABASE_URL:-postgresql://auth_user:auth_password@postgres:5432/auth_db}
    command: ["python", "-m", "bin.migrate"]
    depends_on:
      postgres:
        condition: service_healthy
    networks:
      - experiment-tracking-network
    restart: "no"

  # Experiment DB migrations — запускается один раз перед experiment-service
  experiment-migrate:
    build:
      context: ./projects/backend
      dockerfile: services/experiment-service/Dockerfile
    container_name: experiment-migrate
    environment:
      - DATABASE_URL=${EXPERIMENT_DATABASE_URL:-postgresql://experiment_user:experiment_password@postgres:5432/experiment_db}
    command: ["python", "-m", "bin.migrate"]
    depends_on:
      postgres:
        condition: service_healthy
    networks:
      - experiment-tracking-network
    restart: "no"

  # Telemetry-ingest DB migrations — мигрирует ту же experiment_db, что и
  # experiment-service, и пишет в общую таблицу schema_migrations. Запускается
  # строго после experiment-migrate, чтобы они не гонялись за CREATE TABLE IF NOT EXISTS.
  telemetry-ingest-migrate:
    build:
      context: ./projects/backend
      dockerfile: services/telemetry-ingest-service/Dockerfile
    container_name: telemetry-ingest-migrate
    environment:
      - DATABASE_URL=${EXPERIMENT_DATABASE_URL:-postgresql://experiment_user:experiment_password@postgres:5432/experiment_db}
    command: ["python", "-m", "bin.migrate"]
    depends_on:
      experiment-migrate:
        condition: service_completed_successfully
    networks:
      - experiment-tracking-network
    restart: "no"

  # Auth Service - сервис аутентификации
  auth-service:
    build:
      context: ./projects/backend
      dockerfile: services/auth-service/Dockerfile
    container_name: auth-service
    env_file:
      - .env
    environment:
      # Использует AUTH_DATABASE_URL из .env, если задан, иначе дефолт для auth_db с отдельным пользователем
      - DATABASE_URL=${AUTH_DATABASE_URL:-postgresql://auth_user:auth_password@postgres:5432/auth_db}
      - ENV=${AUTH_SERVICE_ENV:-development}
      - HOST=${AUTH_SERVICE_HOST:-0.0.0.0}
      - PORT=${AUTH_SERVICE_PORT:-8001}
      - DB_POOL_SIZE=${AUTH_SERVICE_DB_POOL_SIZE:-20}
      - JWT_SECRET=${JWT_SECRET:-dev-secret-key-change-in-production}
      - JWT_ALGORITHM=${JWT_ALGORITHM:-HS256}
      - ACCESS_TOKEN_TTL_SEC=${ACCESS_TTL_SEC:-900}
      - REFRESH_TOKEN_TTL_SEC=${REFRESH_TTL_SEC:-1209600}
      - BCRYPT_ROUNDS=${BCRYPT_ROUNDS:-12}
      - CORS_ALLOWED_ORIGINS=${CORS_ALLOWED_ORIGINS:-http://localhost:3000,http://localhost:8080}
      - ADMIN_BOOTSTRAP_SECRET=${ADMIN_BOOTSTRAP_SECRET:-dev-bootstrap-secret}
      - SMTP_HOST=${SMTP_HOST:-mailpit}
      - SMTP_PORT=${SMTP_PORT:-1025}
      - SMTP_USER=${SMTP_USER:-}
      - SMTP_PASSWORD=${SMTP_PASSWORD:-}
      - SMTP_FROM=${SMTP_FROM:-noreply@experiment-platform.local}
      - SMTP_FROM_NAME=${SMTP_FROM_NAME:-Experiment Platform}
      - SMTP_ENABLED=${SMTP_ENABLED:-true}
      - APP_URL=${APP_URL:-http://localhost:3000}
    ports:
      - "${AUTH_SERVICE_PORT:-8001}:8001"
    depends_on:
      auth-migrate:
        condition: service_completed_successfully
    networks:
      - experiment-tracking-network
    restart: unless-stopped
    healthcheck:
      test: ["CMD-SHELL", "python -c \"import urllib.request; urllib.request.urlopen('http://localhost:8001/health').read()\" || exit 1"]
      interval: 10s
      timeout: 5s
      retries: 3
      start_period: 20s

  # Experiment Service - Backend API
  experiment-service:
    build:
      context: ./projects/backend
      dockerfile: services/experiment-service/Dockerfile
    container_name: experiment-service
    env_file:
      - .env
    environment:
      # Использует EXPERIMENT_DATABASE_URL из .env, если задан, иначе дефолт для experiment_db с отдельным пользователем
      - DATABASE_URL=${EXPERIMENT_DATABASE_URL:-postgresql://experiment_user:experiment_password@postgres:5432/experiment_db}
      - ENV=${EXPERIMENT_SERVICE_ENV:-development}
      - HOST=${EXPERIMENT_SERVICE_HOST:-0.0.0.0}
      - PORT=${EXPERIMENT_SERVICE_PORT:-8002}
      - DB_POOL_SIZE=${EXPERIMENT_SERVICE_DB_POOL_SIZE:-20}
      - AUTH_SERVICE_URL=${AUTH_SERVICE_URL:-http://auth-service:8001}
      - RABBITMQ_URL=${RABBITMQ_URL:-amqp://guest:guest@rabbitmq:5672/}
      - TELEMETRY_BROKER_URL=${TELEMETRY_BROKER_URL:-redis://redis:6379/0}
      - S3_ENDPOINT_URL=${S3_ENDPOINT_URL:-http://minio:9000}
      - S3_ACCESS_KEY=${S3_ACCESS_KEY:-minioadmin}
      - S3_SECRET_KEY=${S3_SECRET_KEY:-minioadmin}
      - S3_BUCKET=${S3_BUCKET:-artifacts}
      - S3_PRESIGN_EXPIRE_SECONDS=${S3_PRESIGN_EXPIRE_SECONDS:-3600}
      - S3_PUBLIC_ENDPOINT_URL=${S3_PUBLIC_ENDPOINT_URL:-http://localhost:9000}
    command: ["python", "-m", "experiment_service.main"]
    ports:
      - "${EXPERIMENT_SERVICE_PORT:-8002}:8002"
    depends_on:
      experiment-migrate:
        condition: service_completed_successfully
      redis:
        condition: service_healthy
    networks:
      - experiment-tracking-network
    restart: unless-stopped
    healthcheck:
      test: ["CMD-SHELL", "python -c \"import urllib.request; urllib.request.urlopen('http://localhost:8002/health').read()\" || exit 1"]
      interval: 30s
      timeout: 10s
      retries: 3
      start_period: 40s

  # Telemetry Ingest Service - public ingest API for devices
  telemetry-ingest-service:
    build:
      context: ./projects/backend
      dockerfile: services/telemetry-ingest-service/Dockerfile
    container_name: telemetry-ingest-service
    env_file:
      - .env
    environment:
      - DATABASE_URL=${EXPERIMENT_DATABASE_URL:-postgresql://experiment_user:experiment_password@postgres:5432/experiment_db}
      - ENV=${TELEMETRY_INGEST_ENV:-development}
      - HOST=${TELEMETRY_INGEST_HOST:-0.0.0.0}
      - PORT=${TELEMETRY_INGEST_PORT:-8003}
      - DB_POOL_SIZE=${TELEMETRY_INGEST_DB_POOL_SIZE:-20}
      - CORS_ALLOWED_ORIGINS=${CORS_ALLOWED_ORIGINS:-http://localhost:3000,http://localhost:8080}
      - AUTH_SERVICE_URL=${AUTH_SERVICE_URL:-http://auth-service:8001}
      - CONFIG_CLIENT_ENABLED=${CONFIG_CLIENT_ENABLED:-true}
      - CONFIG_CLIENT_URL=${CONFIG_CLIENT_URL:-http://config-service:8005}
      - CONFIG_CLIENT_POLL_INTERVAL_SECONDS=${CONFIG_CLIENT_POLL_INTERVAL_SECONDS:-5.0}
      # Shared per-sensor rate-limit buckets; ingest falls back to in-memory if Redis is down.
      - RATE_LIMIT_REDIS_URL=${TELEMETRY_INGEST_RATE_LIMIT_REDIS_URL:-redis://redis:6379/0}
    command: ["python", "-m", "telemetry_ingest_service.main"]
    ports:
      - "${TELEMETRY_INGEST_PORT:-8003}:8003"
    depends_on:
      telemetry-ingest-migrate:
        condition: service_completed_successfully
      redis:
        condition: service_started
    networks:
      - experiment-tracking-network
    restart: unless-stopped
    healthcheck:
      test: ["CMD-SHELL", "python -c \"import urllib.request; urllib.request.urlopen('http://localhost:8003/health').read()\" || exit 1"]
      interval: 30s
      timeout: 10s
      retries: 3
      start_period: 20s

  # Script Service - управление и выполнение скриптов
  script-service:
    build:
      context: ./projects/backend
      dockerfile: services/script-service/Dockerfile
    container_name: script-service
    env_file:
      - .env
    environment:
      - DATABASE_URL=${SCRIPT_DATABASE_URL:-postgresql://script_user:script_password@postgres:5432/script_db}
      - ENV=${SCRIPT_SERVICE_ENV:-development}
      - HOST=${SCRIPT_SERVICE_HOST:-0.0.0.0}
      - PORT=${SCRIPT_SERVICE_PORT:-8004}
      - DB_POOL_SIZE=${SCRIPT_SERVICE_DB_POOL_SIZE:-10}
      - AUTH_SERVICE_URL=${AUTH_SERVICE_URL:-http://auth-service:8001}
      - RABBITMQ_URL=${RABBITMQ_URL:-amqp://guest:guest@rabbitmq:5672/}
    command: ["python", "-m", "script_service.main"]
    ports:
      - "${SCRIPT_SERVICE_PORT:-8004}:8004"
    depends_on:
      script-migrate:
        condition: service_completed_successfully
    networks:
      - experiment-tracking-network
    restart: unless-stopped
    healthcheck:
      test: ["CMD-SHELL", "python -c \"import urllib.request; urllib.request.urlopen('http://localhost:8004/health').read()\" || exit 1"]
      interval: 30s
      timeout: 10s
      retries: 3
      start_period: 40s

  # Config Service - runtime configuration (feature flags, QoS, kill-switches)
  config-service:
    build:
      context: ./projects/backend
      dockerfile: services/config-service/Dockerfile
    container_name: config-service
    env_file:
      - .env
    environment:
      - DATABASE_URL=${CONFIG_DATABASE_URL:-postgresql://config_user:config_password@postgres:5432/config_db}
      - ENV=${CONFIG_SERVICE_ENV:-development}
      - HOST=${CONFIG_SERVICE_HOST:-0.0.0.0}
      - PORT=${CONFIG_SERVICE_PORT:-8005}
      - DB_POOL_SIZE=${CONFIG_SERVICE_DB_POOL_SIZE:-10}
      - AUTH_SERVICE_URL=${AUTH_SERVICE_URL:-http://auth-service:8001}
      - CORS_ALLOWED_ORIGINS=${CORS_ALLOWED_ORIGINS:-http://localhost:3000,http://localhost:8080}
    command: ["python", "-m", "config_service.main"]
    ports:
      - "127.0.0.1:${CONFIG_SERVICE_PORT:-8005}:8005"
    depends_on:
      config-migrate:
        condition: service_completed_successfully
      postgres:
        condition: service_healthy
    networks:
      - experiment-tracking-network
    restart: unless-stopped
    healthcheck:
      test: ["CMD-SHELL", "python -c \"import urllib.request; urllib.request.urlopen('http://localhost:8005/health').read()\" || exit 1"]
      interval: 30s
      timeout: 10s
      retries: 3
      start_period: 20s

  # Auth Proxy - BFF для фронтенда
  auth-proxy:
    build:
      context: .
      dockerfile: projects/frontend/apps/auth-proxy/Dockerfile
      target: production
    container_name: auth-proxy
    env_file:
      - .env
    environment:
      - PORT=${AUTH_PROXY_PORT:-8080}
      - TARGET_EXPERIMENT_URL=${TARGET_EXPERIMENT_URL:-http://experiment-service:8002}
      - TARGET_TELEMETRY_URL=${TARGET_TELEMETRY_URL:-http://telemetry-ingest-service:8003}
      - TARGET_SCRIPT_URL=${TARGET_SCRIPT_URL:-http://script-service:8004}
      - TARGET_CONFIG_SERVICE_URL=${TARGET_CONFIG_SERVICE_URL:-http://config-service:8005}
      - AUTH_URL=${AUTH_URL:-http://auth-service:8001}
      - COOKIE_DOMAIN=${COOKIE_DOMAIN:-localhost}
      - COOKIE_SECURE=${COOKIE_SECURE:-false}
      - COOKIE_SAMESITE=${COOKIE_SAMESITE:-lax}
      - ACCESS_COOKIE_NAME=${ACCESS_COOKIE_NAME:-access_token}
      - REFRESH_COOKIE_NAME=${REFRESH_COOKIE_NAME:-refresh_token}
      - ACCESS_TTL_SEC=${ACCESS_TTL_SEC:-900}
      - REFRESH_TTL_SEC=${REFRESH_TTL_SEC:-1209600}
      - CORS_ORIGINS=${CORS_ORIGINS:-http://localhost:3000}
      - RATE_LIMIT_WINDOW_MS=${RATE_LIMIT_WINDOW_MS:-60000}
      - RATE_LIMIT_MAX=${RATE_LIMIT_MAX:-60}
      - LOG_LEVEL=${AUTH_PROXY_LOG_LEVEL:-info}
      - NODE_ENV=production
      - REDIS_URL=${REDIS_URL:-redis://redis:6379/0}
    ports:
      - "${AUTH_PROXY_PORT:-8080}:8080"
    depends_on:
      experiment-service:
        condition: service_healthy
      auth-service:
        condition: service_healthy
      redis:
        condition: service_healthy
      script-service:
        condition: service_healthy
      config-service:
        condition: service_healthy
    networks:
      - experiment-tracking-network
    restart: unless-stopped
    healthcheck:
      test: ["CMD-SHELL", "node -e \"require('http').get('http://localhost:8080/health', (r) => {process.exit(r.statusCode === 200 ? 0 : 1)}).on('error', () => process.exit(1))\" || exit 1"]
      interval: 30s
      timeout: 10s
      retries: 3
      start_period: 20s

  # Experiment Portal - Frontend SPA
  experiment-portal:
    build:
      context: .
      dockerfile: projects/frontend/apps/experiment-portal/Dockerfile
      target: production
    container_name: experiment-portal
    environment:
      - NODE_ENV=production
    ports:
      - "80:80"
    depends_on:
      - auth-proxy
    networks:
      - experiment-tracking-network
    restart: unless-stopped
    healthcheck:
      test: ["CMD-SHELL", "wget --no-verbose --tries=1 --spider http://localhost/ || exit 1"]
      interval: 30s
      timeout: 10s
      retries: 3
      start_period: 10s

  # Sensor Simulator - separate frontend SPA for telemetry simulation
  sensor-simulator:
    build:
      context: .
      dockerfile: projects/frontend/apps/sensor-simulator/Dockerfile
      target: production
    container_name: sensor-simulator
    environment:
      - NODE_ENV=production
    ports:
      - "8082:80"
    depends_on:
      telemetry-ingest-service:
        condition: service_healthy
    networks:
      - experiment-tracking-network
    restart: unless-stopped
    healthcheck:
      test: ["CMD-SHELL", "wget --no-verbose --tries=1 --spider http://127.0.0.1/ || exit 1"]
      interval: 30s
      timeout: 10s
      retries: 3
      start_period: 10s

  # Loki - система хранения и агрегации логов
  loki:
    image: grafana/loki:3.5.8
    container_name: loki
    ports:
      - "${LOKI_PORT:-3100}:3100"
    volumes:
      - ./infrastructure/logging/loki-config.yml:/etc/loki/local-config.yaml:ro
      - loki_data:/loki
    command: -config.file=/etc/loki/local-config.yaml
    networks:
      - experiment-tracking-network
    restart: unless-stopped
    healthcheck:
      test: ["CMD-SHELL", "wget --no-verbose --tries=1 --spider http://localhost:3100/ready || exit 1"]
      interval: 10s
      timeout: 5s
      retries: 5
      start_period: 10s

  # Grafana Alloy - сборщик логов из Docker контейнеров
  alloy:
    image: grafana/alloy:v1.17.1
    container_name: alloy
    volumes:
      - /var/lib/docker/containers:/var/lib/docker/containers:ro
      - /var/run/docker.sock:/var/run/docker.sock:ro
      - ./infrastructure/logging/alloy.river:/etc/alloy/config.river:ro
    command: run --server.http.listen-addr=0.0.0.0:12345 --storage.path=/var/lib/alloy/data /etc/alloy/config.river
    depends_on:
      loki:
        condition: service_healthy
    networks:
      - experiment-tracking-network
    restart: unless-stopped
    ports:
      - "12345:12345"  # HTTP сервер для метрик и UI
    healthcheck:
      test: ["CMD", "alloy", "fmt", "/etc/alloy/config.river"]
      interval: 15s
      timeout: 5s
      retries: 3
      start_period: 10s

  # Grafana - веб-интерфейс для визуализации логов
  grafana:
    image: grafana/grafana:12.4.5
    container_name: grafana
    ports:
      - "${GRAFANA_PORT:-3001}:3000"
    environment:
      - GF_SECURITY_ADMIN_USER=${GRAFANA_ADMIN_USER:-admin}
      - GF_SECURITY_ADMIN_PASSWORD=${GRAFANA_ADMIN_PASSWORD:-admin}
      - GF_AUTH_ANONYMOUS_ENABLED=${GRAFANA_ANONYMOUS_ENABLED:-false}
      - GF_AUTH_ANONYMOUS_ORG_ROLE=Admin
      - GF_SERVER_ROOT_URL=${GRAFANA_ROOT_URL:-http://localhost:3001}
      - GF_INSTALL_PLUGINS=
    volumes:
      - grafana_data:/var/lib/grafana
      - ./infrastructure/logging/grafana/provisioning:/etc/grafana/provisioning:ro
      - ./infrastructure/logging/grafana/dashboards:/var/lib/grafana/dashboards:ro
    depends_on:
      loki:
        condition: service_healthy
      prometheus:
        condition: service_healthy
    networks:
      - experiment-tracking-network
    restart: unless-stopped
    healthcheck:
      test: ["CMD-SHELL", "wget --no-verbose --tries=1 --spider http://localhost:3000/api/health || exit 1"]
      interval: 30s
      timeout: 10s
      retries: 3
      start_period: 40s

# Named volumes для персистентности данных
volumes:
  postgres_data:
    name: ${POSTGRES_DATA_VOLUME:-backend-postgres-data}
  loki_data:
    name: ${LOKI_DATA_VOLUME:-experiment-loki-data}
  grafana_data:
    name: ${GRAFANA_DATA_VOLUME:-experiment-grafana-data}
  minio_data:
    name: ${MINIO_DATA_VOLUME:-experiment-minio-data}

# Docker network для взаимодействия сервисов
# Ключ сети должен совпадать с name (или с меткой существующей сети), иначе compose не найдёт сеть
networks:
  experiment-tracking-network:
    name: ${DOCKER_NETWORK_NAME:-experiment-tracking-network}
    driver: bridge