math2-platform/docker-compose.prod.yml

# ============================================================
# DOCKER COMPOSE PRODUCTION - ENTERPRISE GRADE
# Math Platform with SSL/TLS, Health Checks, and Monitoring
# ============================================================

version: '3.9'

services:
  # ========================================
  # PostgreSQL Database (Production Optimized)
  # ========================================
  postgres:
    image: postgres:15.4-alpine
    container_name: math-postgres-prod
    environment:
      POSTGRES_USER: ${DB_USER:-mathuser}
      POSTGRES_PASSWORD: ${DB_PASSWORD}
      POSTGRES_DB: ${DB_NAME:-mathdb}
      PGDATA: /var/lib/postgresql/data/pgdata
    volumes:
      - postgres_data:/var/lib/postgresql/data
      - ./docker/init-scripts:/docker-entrypoint-initdb.d:ro
      - ./backups:/backups
    ports:
      - "127.0.0.1:5432:5432"
    command:
      - "postgres"
      - "-c"
      - "max_connections=200"
      - "-c"
      - "shared_buffers=2GB"
      - "-c"
      - "effective_cache_size=6GB"
      - "-c"
      - "maintenance_work_mem=512MB"
      - "-c"
      - "checkpoint_completion_target=0.9"
      - "-c"
      - "wal_buffers=16MB"
      - "-c"
      - "default_statistics_target=100"
      - "-c"
      - "random_page_cost=1.1"
      - "-c"
      - "effective_io_concurrency=200"
      - "-c"
      - "work_mem=5242kB"
      - "-c"
      - "min_wal_size=1GB"
      - "-c"
      - "max_wal_size=4GB"
      - "-c"
      - "max_worker_processes=4"
      - "-c"
      - "max_parallel_workers_per_gather=2"
      - "-c"
      - "max_parallel_workers=4"
      - "-c"
      - "max_parallel_maintenance_workers=2"
    healthcheck:
      test: ["CMD-SHELL", "pg_isready -U ${DB_USER:-mathuser} -d ${DB_NAME:-mathdb}"]
      interval: 10s
      timeout: 5s
      retries: 5
      start_period: 30s
    deploy:
      resources:
        limits:
          cpus: '2'
          memory: 4G
        reservations:
          cpus: '0.5'
          memory: 1G
    networks:
      - backend
    restart: unless-stopped
    logging:
      driver: "json-file"
      options:
        max-size: "50m"
        max-file: "5"

  # ========================================
  # Redis Cache & Queue (Production Optimized)
  # ========================================
  redis:
    image: redis:7.2.3-alpine
    container_name: math-redis-prod
    command: >
      redis-server
      --requirepass ${REDIS_PASSWORD}
      --appendonly yes
      --appendfsync everysec
      --maxmemory 512mb
      --maxmemory-policy allkeys-lru
      --tcp-backlog 511
      --timeout 0
      --tcp-keepalive 300
      --save 900 1
      --save 300 10
      --save 60 10000
    volumes:
      - redis_data:/data
    ports:
      - "127.0.0.1:6379:6379"
    healthcheck:
      test: ["CMD", "redis-cli", "-a", "${REDIS_PASSWORD}", "ping"]
      interval: 10s
      timeout: 3s
      retries: 5
      start_period: 10s
    deploy:
      resources:
        limits:
          cpus: '0.5'
          memory: 512M
    networks:
      - backend
    restart: unless-stopped
    logging:
      driver: "json-file"
      options:
        max-size: "50m"
        max-file: "5"

  # ========================================
  # Backend API (Production - Multi-Instance)
  # ========================================
  backend:
    build:
      context: .
      dockerfile: docker/Dockerfile.backend
      target: production
    image: math-backend:${VERSION:-1.0.0}
    # NOTA: container_name omitido para permitir múltiples réplicas con Docker Swarm
    environment:
      NODE_ENV: production
      DATABASE_URL: postgresql://${DB_USER:-mathuser}:${DB_PASSWORD}@postgres:5432/${DB_NAME:-mathdb}
      REDIS_URL: redis://:${REDIS_PASSWORD}@redis:6379
      REDIS_HOST: redis
      REDIS_PORT: 6379
      REDIS_PASSWORD: ${REDIS_PASSWORD}
      AI_API_BASE_URL: ${AI_API_BASE_URL}
      AI_API_KEY: ${AI_API_KEY}
      AI_MODEL: ${AI_MODEL}
      TELEGRAM_BOT_TOKEN: ${TELEGRAM_BOT_TOKEN}
      TELEGRAM_ADMIN_CHAT_ID: ${TELEGRAM_ADMIN_CHAT_ID}
      JWT_SECRET: ${JWT_SECRET}
      JWT_EXPIRES_IN: ${JWT_EXPIRES_IN:-15m}
      BACKEND_PORT: 3001
      LOG_LEVEL: ${LOG_LEVEL:-info}
      CORS_ORIGIN: ${CORS_ORIGIN:-https://localhost}
      AUTH_RATE_LIMIT_WINDOW_MS: ${AUTH_RATE_LIMIT_WINDOW_MS:-900000}
      AUTH_RATE_LIMIT_MAX: ${AUTH_RATE_LIMIT_MAX:-20}
    volumes:
      - backend_logs:/app/logs
      - ./pdfs:/app/pdfs:ro
    expose:
      - "3001"
    depends_on:
      postgres:
        condition: service_healthy
      redis:
        condition: service_healthy
    healthcheck:
      test: ["CMD", "wget", "--quiet", "--tries=1", "--spider", "http://localhost:3001/health"]
      interval: 30s
      timeout: 10s
      retries: 3
      start_period: 40s
    deploy:
      replicas: 2
      update_config:
        parallelism: 1
        delay: 10s
        order: start-first
      restart_policy:
        condition: on-failure
        delay: 5s
        max_attempts: 3
      resources:
        limits:
          cpus: '1'
          memory: 1G
        reservations:
          cpus: '0.25'
          memory: 256M
    networks:
      - backend
      - frontend
    logging:
      driver: "json-file"
      options:
        max-size: "50m"
        max-file: "5"

  # ========================================
  # Frontend (Next.js - Production)
  # ========================================
  frontend:
    build:
      context: .
      dockerfile: docker/Dockerfile.frontend
    image: math-frontend:${VERSION:-1.0.0}
    # NOTA: container_name omitido para permitir múltiples réplicas con Docker Swarm
    environment:
      NODE_ENV: production
      NEXT_PUBLIC_API_URL: ${NEXT_PUBLIC_API_URL:-/api}
      NEXT_PUBLIC_APP_NAME: ${NEXT_PUBLIC_APP_NAME:-Plataforma de Álgebra Lineal}
    expose:
      - "3000"
    depends_on:
      backend:
        condition: service_healthy
    healthcheck:
      test: ["CMD", "wget", "--quiet", "--tries=1", "--spider", "http://localhost:3000"]
      interval: 30s
      timeout: 10s
      retries: 3
      start_period: 40s
    deploy:
      replicas: 2
      resources:
        limits:
          cpus: '0.5'
          memory: 512M
    networks:
      - frontend
    logging:
      driver: "json-file"
      options:
        max-size: "50m"
        max-file: "5"

  # ========================================
  # Nginx Reverse Proxy with SSL
  # ========================================
  nginx:
    image: nginx:1.25-alpine
    container_name: math-nginx-prod
    volumes:
      - ./docker/nginx/nginx.prod.conf:/etc/nginx/nginx.conf:ro
      - ./docker/ssl:/etc/nginx/ssl:ro
      - certbot-data:/etc/letsencrypt
      - certbot-www:/var/www/certbot
      - ./docker/logs:/var/log/nginx
    ports:
      - "80:80"
      - "443:443"
    depends_on:
      frontend:
        condition: service_healthy
      backend:
        condition: service_healthy
    healthcheck:
      test: ["CMD", "wget", "--quiet", "--tries=1", "--spider", "http://localhost/health"]
      interval: 30s
      timeout: 10s
      retries: 3
      start_period: 10s
    command: "/bin/sh -c 'while :; do sleep 6h & wait $${!}; nginx -s reload; done & nginx -g \"daemon off;\"'"
    deploy:
      resources:
        limits:
          cpus: '0.25'
          memory: 128M
    networks:
      - frontend
      - backend
    restart: unless-stopped
    logging:
      driver: "json-file"
      options:
        max-size: "50m"
        max-file: "5"

  # ========================================
  # Certbot for SSL Certificates
  # ========================================
  certbot:
    image: certbot/certbot:latest
    container_name: math-certbot
    volumes:
      - certbot-data:/etc/letsencrypt
      - certbot-www:/var/www/certbot
    entrypoint: "/bin/sh -c 'trap exit TERM; while :; do certbot renew; sleep 12h & wait $${!}; done;'"
    networks:
      - frontend
    restart: unless-stopped

  # ========================================
  # PDF Processing Worker
  # ========================================
  pdf-worker:
    build:
      context: .
      dockerfile: docker/Dockerfile.worker
      target: pdf-worker
    image: math-worker:${VERSION:-1.0.0}
    container_name: math-pdf-worker-prod
    environment:
      NODE_ENV: production
      DATABASE_URL: postgresql://${DB_USER:-mathuser}:${DB_PASSWORD}@postgres:5432/${DB_NAME:-mathdb}
      REDIS_HOST: redis
      REDIS_PORT: 6379
      REDIS_PASSWORD: ${REDIS_PASSWORD}
      WORKER_TYPE: pdf
      LOG_LEVEL: ${LOG_LEVEL:-info}
      HEALTH_PORT: 3002
    volumes:
      - worker_logs:/app/logs
      - ./pdfs:/app/pdfs:ro
      - ./pdfs/processed:/app/pdfs/processed
    depends_on:
      postgres:
        condition: service_healthy
      redis:
        condition: service_healthy
    healthcheck:
      test: ["CMD", "wget", "--quiet", "--tries=1", "--spider", "http://localhost:3002/health"]
      interval: 30s
      timeout: 10s
      retries: 3
      start_period: 30s
    deploy:
      resources:
        limits:
          cpus: '0.5'
          memory: 512M
    networks:
      - backend
    restart: unless-stopped
    logging:
      driver: "json-file"
      options:
        max-size: "50m"
        max-file: "5"

  # ========================================
  # Exercise Generation Worker (AI)
  # ========================================
  exercise-worker:
    build:
      context: .
      dockerfile: docker/Dockerfile.worker
      target: exercise-worker
    image: math-worker:${VERSION:-1.0.0}
    container_name: math-exercise-worker-prod
    environment:
      NODE_ENV: production
      DATABASE_URL: postgresql://${DB_USER:-mathuser}:${DB_PASSWORD}@postgres:5432/${DB_NAME:-mathdb}
      REDIS_HOST: redis
      REDIS_PORT: 6379
      REDIS_PASSWORD: ${REDIS_PASSWORD}
      AI_API_BASE_URL: ${AI_API_BASE_URL}
      AI_API_KEY: ${AI_API_KEY}
      AI_MODEL: ${AI_MODEL}
      WORKER_TYPE: exercise
      LOG_LEVEL: ${LOG_LEVEL:-info}
      HEALTH_PORT: 3003
    volumes:
      - worker_logs:/app/logs
    depends_on:
      postgres:
        condition: service_healthy
      redis:
        condition: service_healthy
    healthcheck:
      test: ["CMD", "wget", "--quiet", "--tries=1", "--spider", "http://localhost:3003/health"]
      interval: 30s
      timeout: 10s
      retries: 3
      start_period: 30s
    deploy:
      resources:
        limits:
          cpus: '0.5'
          memory: 512M
    networks:
      - backend
    restart: unless-stopped
    logging:
      driver: "json-file"
      options:
        max-size: "50m"
        max-file: "5"

  # ========================================
  # Notification Worker (Telegram)
  # ========================================
  notification-worker:
    build:
      context: .
      dockerfile: docker/Dockerfile.worker
      target: notification-worker
    image: math-worker:${VERSION:-1.0.0}
    container_name: math-notification-worker-prod
    environment:
      NODE_ENV: production
      DATABASE_URL: postgresql://${DB_USER:-mathuser}:${DB_PASSWORD}@postgres:5432/${DB_NAME:-mathdb}
      REDIS_HOST: redis
      REDIS_PORT: 6379
      REDIS_PASSWORD: ${REDIS_PASSWORD}
      TELEGRAM_BOT_TOKEN: ${TELEGRAM_BOT_TOKEN}
      TELEGRAM_ADMIN_CHAT_ID: ${TELEGRAM_ADMIN_CHAT_ID}
      WORKER_TYPE: notification
      LOG_LEVEL: ${LOG_LEVEL:-info}
      HEALTH_PORT: 3004
    volumes:
      - worker_logs:/app/logs
    depends_on:
      postgres:
        condition: service_healthy
      redis:
        condition: service_healthy
    healthcheck:
      test: ["CMD", "wget", "--quiet", "--tries=1", "--spider", "http://localhost:3004/health"]
      interval: 30s
      timeout: 10s
      retries: 3
      start_period: 30s
    deploy:
      resources:
        limits:
          cpus: '0.25'
          memory: 256M
    networks:
      - backend
    restart: unless-stopped
    logging:
      driver: "json-file"
      options:
        max-size: "50m"
        max-file: "5"

# ========================================
# Named Volumes
# ========================================
volumes:
  postgres_data:
    driver: local
  redis_data:
    driver: local
  backend_logs:
    driver: local
  worker_logs:
    driver: local
  certbot-data:
    driver: local
  certbot-www:
    driver: local

# ========================================
# Networks
# ========================================
networks:
  backend:
    driver: bridge
    internal: true
    ipam:
      config:
        - subnet: 172.20.0.0/16
  frontend:
    driver: bridge
    ipam:
      config:
        - subnet: 172.21.0.0/16