docker-compose.yml

services:
  postgres:
    image: postgres:16-alpine
    container_name: events-postgres
    environment:
      POSTGRES_DB: events_db
      POSTGRES_USER: events_user
      POSTGRES_PASSWORD: events_pass
      POSTGRES_SHARED_BUFFERS: 4GB
      POSTGRES_EFFECTIVE_CACHE_SIZE: 12GB
      POSTGRES_WORK_MEM: 64MB
      POSTGRES_MAINTENANCE_WORK_MEM: 1GB
    ports:
      - "5432:5432"
    volumes:
      - postgres_data:/var/lib/postgresql/data
    command: |
      postgres
      -c max_connections=1000
      -c shared_buffers=4GB
      -c effective_cache_size=12GB
      -c work_mem=64MB
      -c maintenance_work_mem=1GB
      -c wal_buffers=128MB
      -c synchronous_commit=off
      -c commit_delay=5000
      -c commit_siblings=10
      -c max_wal_size=16GB
      -c min_wal_size=4GB
      -c checkpoint_timeout=20min
      -c checkpoint_completion_target=0.9
      -c effective_io_concurrency=300
      -c random_page_cost=1.1
      -c seq_page_cost=1.0
      -c autovacuum_max_workers=6
      -c autovacuum_naptime=5s
      -c autovacuum_vacuum_scale_factor=0.02
      -c autovacuum_analyze_scale_factor=0.01
      -c bgwriter_delay=100ms
      -c bgwriter_lru_maxpages=200
      -c bgwriter_lru_multiplier=4.0
      -c max_wal_senders=0
      -c wal_level=minimal
    healthcheck:
      test: ["CMD-SHELL", "pg_isready -U events_user -d events_db"]
      interval: 10s
      timeout: 5s
      retries: 5
    networks:
      - events-network
    deploy:
      resources:
        limits:
          cpus: '4'
          memory: 16G
        reservations:
          cpus: '2'
          memory: 8G

  # NATS JetStream - optimized for 50k events/sec
  nats:
    image: nats:2.10-alpine
    container_name: events-nats
    ports:
      - "4222:4222"
      - "8222:8222"
    command: ["-c", "/config/nats.conf"]
    volumes:
      - nats_data:/data
      - ./nats.conf:/config/nats.conf:ro
    healthcheck:
      test: ["CMD", "wget", "--spider", "-q", "http://localhost:8222/healthz"]
      interval: 10s
      timeout: 5s
      retries: 5
    networks:
      - events-network

  # NATS Prometheus Exporter
  nats-exporter:
    image: natsio/prometheus-nats-exporter:latest
    container_name: nats-exporter
    command:
      - -varz
      - -connz
      - -subz
      - -jsz=all
      - http://nats:8222
    ports:
      - "7777:7777"
    depends_on:
      nats:
        condition: service_healthy
    networks:
      - events-network
    restart: unless-stopped

  ingestion-service:
    build:
      context: ./services/ingestion-service
      dockerfile: Dockerfile
    environment:
      NODE_ENV: production
      PORT: 3001
      NATS_URL: nats://nats:4222
      NATS_STREAM: EVENTS
      NATS_SUBJECT_PREFIX: events
      NATS_CONSUMER_NAME: events_processor
    ports:
      - "3001"
    depends_on:
      nats:
        condition: service_healthy
    networks:
      - events-network
    restart: unless-stopped
    deploy:
      resources:
        limits:
          cpus: '2'
          memory: 2G
        reservations:
          cpus: '1'
          memory: 1G

  # Worker Service - Replicas 1-12 (batch processing, optimized for 50k+ events/sec)
  worker-service:
    build:
      context: ./services/worker-service
      dockerfile: Dockerfile
    environment:
      NODE_ENV: production
      PORT: 3000
      DATABASE_HOST: postgres
      DATABASE_PORT: 5432
      DATABASE_NAME: events_db
      DATABASE_USER: events_user
      DATABASE_PASSWORD: events_pass
      DB_LOGGING: false
      DB_POOL_MAX: 75
      DB_POOL_MIN: 10
      NATS_URL: nats://nats:4222
      NATS_STREAM: EVENTS
      NATS_SUBJECT_PREFIX: events
      NATS_CONSUMER_NAME: events_processor
      # Removed NATS_QUEUE_GROUP - each worker has its own durable consumer
      NATS_BATCH_SIZE: 5000
      NATS_CONCURRENCY: 300
    # Expose port for Prometheus scraping
    ports:
      - "3000"
    depends_on:
      postgres:
        condition: service_healthy
      nats:
        condition: service_healthy
    networks:
      - events-network
    restart: unless-stopped
    deploy:
      replicas: 4
      resources:
        limits:
          cpus: '3'
          memory: 4G
        reservations:
          cpus: '2'
          memory: 3G

  # Analytics Service - read-only API
  analytics-service:
    build:
      context: ./services/analytics-service
      dockerfile: Dockerfile
    container_name: analytics-service
    environment:
      NODE_ENV: production
      PORT: 3002
      DATABASE_HOST: postgres
      DATABASE_PORT: 5432
      DATABASE_NAME: events_db
      DATABASE_USER: events_user
      DATABASE_PASSWORD: events_pass
      DB_LOGGING: false
      DB_POOL_MAX: 50
      DB_POOL_MIN: 10
    ports:
      - "3002:3002"
    depends_on:
      postgres:
        condition: service_healthy
    networks:
      - events-network
    restart: unless-stopped
    deploy:
      resources:
        limits:
          cpus: '1'
          memory: 1G
        reservations:
          cpus: '0.5'
          memory: 512M

  # Prometheus - metrics collection
  prometheus:
    image: prom/prometheus:latest
    container_name: prometheus
    ports:
      - "9090:9090"
    volumes:
      - ./prometheus.yml:/etc/prometheus/prometheus.yml
      - prometheus_data:/prometheus
    command:
      - '--config.file=/etc/prometheus/prometheus.yml'
      - '--storage.tsdb.path=/prometheus'
      - '--storage.tsdb.retention.time=30d'
    networks:
      - events-network
    restart: unless-stopped

  # Grafana - dashboards
  grafana:
    image: grafana/grafana:latest
    container_name: grafana
    ports:
      - "3003:3000"
    volumes:
      - grafana_data:/var/lib/grafana
      - ./grafana/provisioning:/etc/grafana/provisioning
      - ./grafana/provisioning/dashboards:/var/lib/grafana/dashboards
    environment:
      GF_SECURITY_ADMIN_PASSWORD: admin
      GF_USERS_ALLOW_SIGN_UP: false
    networks:
      - events-network
    restart: unless-stopped
    depends_on:
      - prometheus
  publisher:
    image: andriiuni/events
    container_name: events-publisher
    environment:
      EVENT_ENDPOINT: http://ingestion-service:3001/webhook
      EVENTS_PER_SECOND: 5000
    depends_on:
      - ingestion-service
    networks:
      - events-network

networks:
  events-network:
    driver: bridge

volumes:
  postgres_data:
  nats_data:
  prometheus_data:
  grafana_data: