다양한 OpenSource 모니터링 구성

Opensource Metrics 종류

툴 이름	수집기	데이터 핸들러	데이터 보관	시각화	비고
NFLUXDATA	Telegraf	Telegraf	Influxdb	Grafana	자체 UI있으나 Grafana 보다 기능이 떨어짐
Elastic Stack	Metricbeat	Logstash	Elasticsearch	Kibana
PROMETHUS	node_exporter	Promethus	Promethus	Grafana	자체 UI있으나 Grafana 보다 기능이 떨어짐

InfluxData 테스트

• 지원 OS : Docker, Ubuntu& debian, Redhat&CentOS, Windows(64Bit)
• telegraf 설치

  # wget https://dl.influxdata.com/telegraf/releases/telegraf-1.14.0_linux_amd64.tar.gz
  # tar xf telegraf-1.14.0_linux_amd64.tar.gz
  

• telegraf Metrics 설정(모니터링 을 위한 필요 Metrics 정보 취합만 설정)

    - In the Telegraf environment variables file (/etc/default/telegraf)
    Example)
    USER="alice"
    INFLUX_URL="http://localhost:8086"
    INFLUX_SKIP_DATABASE_CREATION="true"
    INFLUX_PASSWORD="monkey123"
  
    # vi telegraf/etc/telegraf/telegraf.conf
  
    [global_tags]
      service = "elastic"
    [agent]
      interval = "10s" # 데이터 수집 간격
      round_interval = true # 수집 간격을 반올림 여부(:00, :10, :20)
      metric_batch_size = 1000 # 최대 output size
      metric_buffer_limit = 10000 #output 에대한 최대 캐시(버퍼) 사이즈 최소 "metric_batch_size " 이상 배수
  
    [[inputs.cpu]] #https://github.com/influxdata/telegraf/blob/release-1.14/plugins/inputs/cpu/README.md
      percpu = false
      totalcpu = true
      collect_cpu_time = false
      report_active = false
  
    [[inputs.mem]] #https://github.com/influxdata/telegraf/blob/release-1.14/plugins/inputs/mem/README.md
  
    [[inputs.swap]]
  
    [[inputs.system]] #https://github.com/influxdata/telegraf/blob/release-1.14/plugins/inputs/system/README.md
      fielddrop = ["n_users", "n_cpus", "uptime", "uptime_format"]
  
    [[inputs.processes]] # https://github.com/influxdata/telegraf/blob/release-1.14/plugins/inputs/processes/README.md
  
    [[inputs.diskio]] # https://github.com/influxdata/telegraf/blob/release-1.14/plugins/inputs/diskio/README.md
      name_templates = ["$ID_FS_LABEL","$DM_VG_NAME/$DM_LV_NAME"]
  
    [[inputs.netstat]] # https://github.com/influxdata/telegraf/blob/master/plugins/inputs/net/NETSTAT_README.md
  
    [[inputs.net]]
  
    [[outputs.influxdb]]
      urls = ["http://localhost:8086"]
      database = "server"
  

• telegraf 실행

    # usr/bin/telegraf --config etc/telegraf/telegraf.conf > /dev/null &
  

• Influxdb 설치 (with docker)

    # docker pull influxdb
  

• Influxdb 실행

    # docker run -p 8086:8086 \
      -v $PWD:/var/lib/influxdb \
      influxdb
  
    # docker exec -it influxdb influx
  
    >>show databases
      name: databases
      name
      ----
      server
      _internal
  

• Grafana 설치(with Docker)

    # docker pull grafana/grafana
  

• Grafana 실행

    # docker run -d --name=grafana -p 3000:3000 grafana/grafana
  

• Grafana Dashboard

  

ELK (With Meticbeat) 테스트

• 지원 OS : Ubuntu& debian, Redhat&CentOS, Windows(32Bit/64Bit)
• metricbeat 설치

    # wget https://artifacts.elastic.co/downloads/beats/metricbeat/metricbeat-7.6.2-linux-x86_64.tar.gz
    # tar xvf metricbeat-7.6.2-linux-x86_64.tar.gz
  

• meticbeat config 설정(모니터링 을 위한 필요 Metrics 정보 취합만 설정)

    # vi modules.d/system.yml # https://www.elastic.co/guide/en/beats/metricbeat/7.6/metricbeat-module-system.html
      - module: system
        period: 10s
        metricsets:
          - cpu
          - load
          - memory
          - network
          - process_summary
          - socket_summary
          - diskio
  
    # vi metricbeat.yml
  
      metricbeat.config.modules:
        path: ${path.config}/modules.d/*.yml
        reload.enabled: false
        
      setup.template.settings:
        index.number_of_shards: 1
        index.codec: best_compression
        
      output.logstash:
        hosts: ["localhost:5556"]
  

• metricbeat 실행

    # ./metricbeat &
  

• Kibana Dashboard

promethus 테스트

• 지원 OS : Linux(64Bit), drawin
• node_exporter 설치

    # wget https://github.com/prometheus/node_exporter/releases/download/v1.0.0-rc.0/node_exporter-1.0.0-rc.0.linux-amd64.tar.gz
    # tar vf node_exporter-1.0.0-rc.0.linux-amd64.tar.gz
  

• node_exporter 실행

    # ./node_exporter &
  

• node_exporter 정상 확인

    > http://<server_ip>:9100/metrics
  

• promethus 설치

    # docker run \
        -p 9090:9090 \
        -v /tmp/prometheus.yml:/etc/prometheus/prometheus.yml \
        prom/prometheus
  

• promethus config

    # my global config
    global:
      scrape_interval:     10s # Set the scrape interval to every 15 seconds. Default is every 1 minute.
      evaluation_interval: 10s # Evaluate rules every 15 seconds. The default is every 1 minute.
      # scrape_timeout is set to the global default (10s).
  
    # Alertmanager configuration
    alerting:
      alertmanagers:
      - static_configs:
        - targets:
          # - alertmanager:9093
  
    # Load rules once and periodically evaluate them according to the global 'evaluation_interval'.
    rule_files:
      # - "first_rules.yml"
      # - "second_rules.yml"
  
    # A scrape configuration containing exactly one endpoint to scrape:
    # Here it's Prometheus itself.
    scrape_configs:
      # The job name is added as a label `job=<job_name>` to any timeseries scraped from this config.
      - job_name: 'metric'
  
        # metrics_path defaults to '/metrics'
        # scheme defaults to 'http'.
  
        static_configs:
        - targets: ['<server_ip>:9100']
  

• Grafana Dashboard