docker-compose.prod.yml

name: catchtable-clone

services:
  app:
    image: ghcr.io/catchtable-clone/backend:${IMAGE_TAG}
    container_name: catchtable-app
    env_file: .env
    expose:
      - "8080"
    # t3.small 메모리 한도 고려한 app 한도 (OTel Agent 80MB 포함 600M)
    environment:
      # 분산 트레이싱(Jaeger + OTel Collector) 제거 — OTel javaagent SDK 전체 비활성화.
      # 메트릭은 Micrometer/Prometheus, 로그는 Promtail/Loki가 담당하므로 트레이싱만 끔.
      # 트레이싱 복원 시: 이 줄 제거 후 export/sampler 설정 복구 + otel-collector/jaeger 서비스 재추가.
      OTEL_SDK_DISABLED: "true"
    deploy:
      resources:
        limits:
          memory: 1000M
        reservations:
          memory: 400M
    healthcheck:
      test: ["CMD", "wget", "-qO-", "http://localhost:8080/actuator/health"]
      interval: 30s
      timeout: 3s
      retries: 3
      start_period: 120s
    restart: unless-stopped
    networks:
      - catchtable-net
    logging:
      driver: "json-file"
      options:
        max-size: "10m"
        max-file: "3"

  nginx:
    image: nginx:1.26.3-alpine
    container_name: catchtable-nginx
    ports:
      - "80:80"
      - "443:443"
    volumes:
      - ./nginx/nginx.conf:/etc/nginx/nginx.conf:ro
      - ./certbot/conf:/etc/letsencrypt:ro
      - ./certbot/www:/var/www/certbot:ro
    depends_on:
      app:
        condition: service_healthy
    deploy:
      resources:
        limits:
          memory: 50M
    restart: unless-stopped
    networks:
      - catchtable-net

  certbot:
    image: certbot/certbot:v3.2.0
    container_name: catchtable-certbot
    volumes:
      - ./certbot/conf:/etc/letsencrypt
      - ./certbot/www:/var/www/certbot
    entrypoint: "/bin/sh -c 'trap exit TERM; while :; do certbot renew --quiet; sleep 12h & wait $${!}; done;'"
    deploy:
      resources:
        limits:
          memory: 30M
    restart: unless-stopped

  prometheus:
    image: prom/prometheus:v2.55.0
    container_name: catchtable-prometheus
    volumes:
      - ./monitoring/prometheus.yml:/etc/prometheus/prometheus.yml:ro
      - prometheus-data:/prometheus
    command:
      - '--config.file=/etc/prometheus/prometheus.yml'
      - '--storage.tsdb.retention.time=7d'
      - '--storage.tsdb.retention.size=1GB'
      - '--web.enable-lifecycle'
      - '--web.enable-remote-write-receiver'
    ports:
      - "9090:9090"
    deploy:
      resources:
        limits:
          memory: 800M
    restart: unless-stopped
    networks:
      - catchtable-net

  loki:
    image: grafana/loki:3.2.0
    container_name: catchtable-loki
    volumes:
      - ./monitoring/loki-config.yml:/etc/loki/local-config.yaml:ro
      - loki-data:/loki
    command: -config.file=/etc/loki/local-config.yaml
    expose:
      - "3100"
    deploy:
      resources:
        limits:
          memory: 300M
    restart: unless-stopped
    networks:
      - catchtable-net

  promtail:
    image: grafana/promtail:3.2.0
    container_name: catchtable-promtail
    volumes:
      - ./monitoring/promtail-config.yml:/etc/promtail/config.yml:ro
      - /var/run/docker.sock:/var/run/docker.sock:ro
      - /var/lib/docker/containers:/var/lib/docker/containers:ro
      - promtail-positions:/tmp
    command: -config.file=/etc/promtail/config.yml
    depends_on:
      - loki
    deploy:
      resources:
        limits:
          memory: 200M
    restart: unless-stopped
    networks:
      - catchtable-net

  grafana:
    image: grafana/grafana:11.3.0
    container_name: catchtable-grafana
    environment:
      GF_SECURITY_ADMIN_USER: ${GRAFANA_USER:-admin}
      GF_SECURITY_ADMIN_PASSWORD: ${GRAFANA_PASSWORD:-admin}
      GF_USERS_ALLOW_SIGN_UP: "false"
    volumes:
      - ./grafana/provisioning:/etc/grafana/provisioning:ro
      - grafana-data:/var/lib/grafana
    ports:
      - "3001:3000"  # Grafana UI (외부 접근, 3000은 프론트 dev와 충돌 회피)
    depends_on:
      - prometheus
      - loki
    deploy:
      resources:
        limits:
          memory: 300M
    restart: unless-stopped
    networks:
      - catchtable-net

  # ============================================================
  # 고도화 인프라 (Kafka는 해당 작업 시작 시 주석 해제)
  # ============================================================

  redis:
    image: redis:7-alpine
    container_name: catchtable-redis
    command:
      - redis-server
      - --maxmemory
      - 80mb
      - --maxmemory-policy
      - allkeys-lru
      - --appendonly
      - "yes"
    expose:
      - "6379"
    volumes:
      - redis-data:/data
    deploy:
      resources:
        limits:
          memory: 200M
    restart: unless-stopped
    networks:
      - catchtable-net

  kafka:
    image: apache/kafka:3.7.0
    container_name: catchtable-kafka-prod
    user: "0"  # 볼륨 마운트 폴더에 로그를 쓰고 지울 수 있도록 최고 관리자 권한 부여

    # 외부(내 PC 등)로 포트를 무방비하게 노출하지 않고, 오직 'catchtable-net' 내부에서만 통신하도록 격리
    expose:
      - "9092"
      - "9093"

    environment:
      # KRaft 모드 핵심 식별자 설정
      KAFKA_CLUSTER_ID: "MkU3OEV5NURaZW52OTYyM2RjY1JqcA" # 운영 환경 고유 클러스터 ID
      KAFKA_NODE_ID: 1
      KAFKA_PROCESS_ROLES: 'broker,controller'
      KAFKA_CONTROLLER_LISTENER_NAMES: 'CONTROLLER'

      # 모든 인터페이스(0.0.0.0)로 귀를 열어두고, 운영 서버 안의 다른 컨테이너(Spring Boot)에는 서비스 명인 'kafka:9092'로 광고합니다.
      KAFKA_LISTENERS: 'PLAINTEXT://0.0.0.0:9092,CONTROLLER://0.0.0.0:9093'
      KAFKA_ADVERTISED_LISTENERS: 'PLAINTEXT://kafka:9092'
      KAFKA_LISTENER_SECURITY_PROTOCOL_MAP: 'CONTROLLER:PLAINTEXT,PLAINTEXT:PLAINTEXT'
      KAFKA_INTER_BROKER_LISTENER_NAME: 'PLAINTEXT'
      KAFKA_CONTROLLER_QUORUM_VOTERS: '1@127.0.0.1:9093' # 루프백 주소로 바인딩 충돌 방지

      # 메모리 제한 설정 (서버 스펙에 맞게 조정 가능, 최소 256m~512m 추천)
      KAFKA_HEAP_OPTS: "-Xms256m -Xmx256m"

      # 단일 broker 운영용 internal topic 설정.
      # default(3)면 __consumer_offsets/__transaction_state 가 영원히 생성 실패 → consumer group join 불가.
      KAFKA_OFFSETS_TOPIC_REPLICATION_FACTOR: 1
      KAFKA_TRANSACTION_STATE_LOG_REPLICATION_FACTOR: 1
      KAFKA_TRANSACTION_STATE_LOG_MIN_ISR: 1
    # 임시 폴더인 /tmp 대신 컨테이너가 내려가도 데이터와 오프셋 장부가 보존되도록 볼륨 볼륨 마운트 경로 고정
    volumes:
      - kafka-prod-data:/var/lib/kafka/data

    # 리소스 상한선 설정 (메모리 릭으로 인한 전체 서버 다운 방지)
    deploy:
      resources:
        limits:
          memory: 600M

    restart: unless-stopped
    networks:
      - catchtable-net

networks:
  catchtable-net:
    driver: bridge

volumes:
  prometheus-data:
  loki-data:
  grafana-data:
  promtail-positions:
  kafka-prod-data:
  redis-data: