跳到主要内容

Grafana Prometheus

阅读需 2 分钟

windows 进程 监控​

  • windows wmi-exporter 进程监控
    • 对应的看板 dashboard Id 10467
    • 没有进程数据
      • 服务器管理器找到 WMI exporter
        • 添加参数 --collectors.enabled "[defaults],process,container"
Grafana Status Map Panel not working, shows "Panel plugin not found: flant-statusmap-panel" #197
  • 解决方法
    • Open your Grafana instance and go to the Plugins section in the left-hand -nu.
    • Search for "statusmap" in the search bar.
    • Click on the "Install" button for the "statusmap" plugin.
    • Wait for the installation to complete.\

Grafana + prometheus + node-exporter docker-compose.yml​

vim prometheus/docker-compose.yml

version: "3.7"
services:
  node_exporter:
    image: quay.io/prometheus/node-exporter:latest
    container_name: node_exporter
    command:
      - '--path.rootfs=/host'
    user: "1000"
    network_mode: host
    # pid: host
    restart: unless-stopped
    volumes:
      - '/:/host:ro,rslave'
  prometheus:
    image: prom/prometheus:latest
    container_name: "prometheus"
    restart: always
    ports:
      - "9111:9090"
    volumes:
      - "./prometheus.yml:/etc/prometheus/prometheus.yml"
      - "./prometheus_data:/prometheus"
  grafana:
    image: grafana/grafana
    container_name: "grafana"
    ports:
      - "3000:3000"
    restart: always
    volumes:
      - "./grafana_data:/var/lib/grafana"
  process-exporter:
    image: ncabatoff/process-exporter
    command: --procfs /host/proc -config.path /config/process-exporter-config.yml
    privileged: true
    volumes:
      - /proc:/host/proc
      - ./process-exporter-config.yml:/config/process-exporter-config.yml
    ports:
      - "9256:9256"
    restart: always

vim prometheus/prometheus.yml - 需要根据自己的 ip 进行修改器

global:
  scrape_interval: 15s # 默认抓取周期
  external_labels:
    monitor: "codelab-monitor"
scrape_configs:
  - job_name: "node-exporter" #服务的名称
    scrape_interval: 5s
    metrics_path: /metrics #获取指标的 url
    static_configs:
      - targets: ["192.168.31.2:9100"] # 这个为监听指定服务服务的 ip 和 port，需要修改为自己的 ip，貌似云服务必须用公网 ip
  - job_name: "process-exporter"
    scrape_interval: 5s
    metrics_path: /metrics
    static_configs:
      - targets: ["192.168.31.2:9256"]
  - job_name: "windows-laptop"
    scrape_interval: 5s
    metrics_path: /metrics
    static_configs:
      - targets: ["192.168.31.3:9182"]
  - job_name: "other-node-192.168.31.5-node-exporter"
    scrape_interval: 5s
    metrics_path: /metrics
    static_configs:
      - targets: ["192.168.31.5"]

node-exporter linux 服务器状态监控​

prometheus 相关​

看板 Grafana​

  • http://ip:9111/3000

  • 配合 node - 的 Dashborad 编号是 9276

    • Grafana 直接搜索 id 就能找到别人分享的看板。
    • 需要单独配置下 网络 io 图表。
    • 都是可视化的 先再 node-exporter 里找到相关指标，然后可视化点点点就能出现数据。
Loading Comments...
0%