pinpoint 설치하기

오픈소스 APM 중 하나인 pinpoint를 docker compose를 통해 설치하는 과정을 정리.

 

GitHub - pinpoint-apm/pinpoint-docker: Official Dockerized components of the Pinpoint

 

GCP VM 두개을 활용하여 Agent 서버와 Collector 서버로 구분

1. Pinpoint Collector, Pinpoint Web etc..
2. Web Application(Java Application)과 Pinpoint Agent

 

1. Collector Server

docker-compose.yml 파일 내 pinpoint-agent와 pinpoint-quickstart 두 가지를 제거하고 docker compose 실행

git clone https://github.com/pinpoint-apm/pinpoint-docker.git
cd pinpoint-docker

//docker-compose.yml에서 pinpoint-agent, pinpoint-quickstart 제거

docker compose up -d

docker compose 실행 시 service 목록

 

docker-compose.yml에서 삭제할 부분

# docker-compose.yml

pinpoint-agent:
    build:
      context: ./pinpoint-agent/
      dockerfile: Dockerfile
      args:
        - PINPOINT_VERSION=${PINPOINT_VERSION}

    container_name: "${PINPOINT_AGENT_NAME}"
    image: "pinpointdocker/pinpoint-agent:${PINPOINT_VERSION}"
    restart: unless-stopped
    networks:
      - pinpoint
    volumes:
      - data-volume:/pinpoint-agent
    environment:
      - SPRING_PROFILES=${SPRING_PROFILES}
       ....
    depends_on:
      - pinpoint-collector
   
   ...
   
   pinpoint-quickstart:
    build:
      context: ./pinpoint-quickstart/
      dockerfile: Dockerfile

    container_name: "pinpoint-quickstart"
    image: "pinpointdocker/pinpoint-quickstart"
    ports:
      - "${APP_PORT:-8085}:8080"
    volumes:
      - data-volume:/pinpoint-agent
    environment:
      JAVA_OPTS: "-javaagent:/pinpoint-agent/pinpoint-bootstrap-${PINPOINT_VERSION}.jar -Dpinpoint.agentId=${AGENT_ID} -Dpinpoint.applicationName=${APP_NAME} -Dpinpoint.profiler.profiles.active=${SPRING_PROFILES}"
    networks:
      - pinpoint
    depends_on:
      - pinpoint-agent

 

2. Agent Server 

docker compose에서 환경변수를 정의하는 파일인 .env를 다음과 같이 수정

환경변수 명	설명
SPRING_PROFILES	Pinpoint 환경 구분, Debug Level이나 Sampling 등을 개별적으로 관리할 수 있다.
COLLECTOR_IP	Pinpoint Collector Server IP
AGENT_ID	Web Application의 고유한 이름
APP_NAME	Web Application이 Pinpoint Web에서 보여지는 이름

 

.env

# .env

PINPOINT_VERSION=2.5.2
SPRING_PROFILES=release #pinpoint profile local|release 

### Pinpoint-Agent

PINPOINT_AGENT_NAME=pinpoint-agent

COLLECTOR_IP=34.64.XXX.XX # Collector Server IP
PROFILER_TRANSPORT_AGENT_COLLECTOR_PORT=9991
PROFILER_TRANSPORT_METADATA_COLLECTOR_PORT=9991
PROFILER_TRANSPORT_STAT_COLLECTOR_PORT=9992
PROFILER_TRANSPORT_SPAN_COLLECTOR_PORT=9993

COLLECTOR_TCP_PORT=9994
COLLECTOR_STAT_PORT=9995
COLLECTOR_SPAN_PORT=9996

# Sampling Configurations
PROFILER_SAMPLING_TYPE=COUNTING
PROFILER_SAMPLING_COUNTING_SAMPLING_RATE=1
PROFILER_SAMPLING_PERCENT_SAMPLING_RATE=100
PROFILER_SAMPLING_NEW_THROUGHPUT=0
PROFILER_SAMPLING_CONTINUE_THROUGHPUT=0

AGENT_ID=board # Application Unique ID
APP_NAME=board # Application Name For View

AGENT_DEBUG_LEVEL=INFO

### Board
APP_PORT=8081

 

docker-compose.yml 에서는 Web Application과 Pinpoint-Agent를 service로 선언

docker volume을 Application과 Agent가 공유하는 구조

Application Container 내 디렉토리 구조

 

docker-compose.yml

# docker-compose.yml

services:
  board:
    container_name: "board"
    image: "ghcr.io/sukkyun2/sbb"

    restart: always
    ports:
      - "${APP_PORT:-8081}:8080"
    volumes:
      - data-volume:/pinpoint-agent
    environment:
      JAVA_OPTS: "-javaagent:/pinpoint-agent/pinpoint-bootstrap-${PINPOINT_VERSION}.jar -Dpinpoint.agentId=${AGENT_ID} -Dpinpoint.applicationName=${APP_NAME} -Dpinpoint.profiler.profiles.active=${SPRING_PROFILES}"
    networks:
      - pinpoint
    depends_on:
      - pinpoint-agent

  pinpoint-agent:
    container_name: "${PINPOINT_AGENT_NAME}"
    image: "pinpointdocker/pinpoint-agent:${PINPOINT_VERSION}"

    restart: unless-stopped

    networks:
      - pinpoint
    volumes:
      - data-volume:/pinpoint-agent
    environment:
      - COLLECTOR_IP=${COLLECTOR_IP}
      - COLLECTOR_TCP_PORT=${COLLECTOR_TCP_PORT}
      - COLLECTOR_STAT_PORT=${COLLECTOR_STAT_PORT}
      - COLLECTOR_SPAN_PORT=${COLLECTOR_SPAN_PORT}
      - PROFILER_SAMPLING_TYPE=${PROFILER_SAMPLING_TYPE} 
      - PROFILER_SAMPLING_COUNTING_SAMPLING_RATE=${PROFILER_SAMPLING_COUNTING_SAMPLING_RATE}
      - PROFILER_SAMPLING_PERCENT_SAMPLING_RATE=${PROFILER_SAMPLING_PERCENT_SAMPLING_RATE}
      - PROFILER_SAMPLING_NEW_THROUGHPUT=${PROFILER_SAMPLING_NEW_THROUGHPUT}
      - PROFILER_SAMPLING_CONTINUE_THROUGHPUT=${PROFILER_SAMPLING_CONTINUE_THROUGHPUT}
      - DEBUG_LEVEL=${AGENT_DEBUG_LEVEL}

volumes:
  data-volume:

networks:
  pinpoint:
    driver: bridge

 

# 방화벽 설정

• Agent Servcer에서 Collector Server로 로그 전송
• 외부에서 Pinpoint Web을 접속

두 가지 측면에서 방화벽을 설정해줘야 한다.

 

Collector Server의 Inbound Port 설정을 8080, 9991 ~ 9996까지 열어주면 설정 완료!

 

pinpoint port