Prometheus（三）监控服务器：node_exporter

发表于 | 分类于

若要监控服务器CPU、内存、磁盘、I/O等信息，需要node_exporter，作用是收集机器系统数据。

  1. 安装 node_exporter

    1
    2
    $ git clone https://github.com/prometheus/node_exporter.git
    $ mv node_exporter-0.18.1.linux-amd64 /usr/local/prometheus/node_exporter

  2. 创建Systemd服务

    1
    2
    3
    4
    5
    6
    7
    8
    9
    10
    11
    12
    $ vim /etc/systemd/system/node_exporter.service

    [Unit]
    Description=node_exporter
    After=network.target
    [Service]
    Type=simple
    User=prometheus
    ExecStart=/usr/local/prometheus/node_exporter/node_exporter
    Restart=on-failure
    [Install]
    WantedBy=multi-user.target

  3. 启动Node exporter

    1
    $ systemctl start node_exporter

  4. 验证Node exporter是否启动成功

    1
    2
    3
    4
    5
    6
    7
    8
    9
    10
    11
    $ systemctl status node_exporter

    ● node_exporter.service - node_exporter
       Loaded: loaded (/etc/systemd/system/node_exporter.service; disabled; vendor preset: enabled)
       Active: active (running) since Wed 2019-07-17 15:10:26 CST; 10s ago
     Main PID: 13085 (node_exporter)
        Tasks: 3
       Memory: 1.1M
          CPU: 5ms
       CGroup: /system.slice/node_exporter.service
               └─13085 /usr/local/prometheus/node_exporter/node_exporter

  5. 修改prometheus.yml，加入下面的监控目标：
    Node Exporter默认的抓取地址为http://IP:9100/metrics

    1
    2
    3
    4
    5
    6
    7
    $ vim  /usr/local/prometheus/prometheus.yml

      - job_name: 'linux'
        static_configs:
          - targets: ['localhost:9100']
            labels:
              instance: node1

prometheus.yml中一共定义了两个监控：一个是监控prometheus自身服务，另一个是监控Linux服务器。这里给个完整的示例：

1
2
3
4
5
6
7
8
9
10
11
- job_name: 'prometheus'
    # metrics_path defaults to '/metrics'
    # scheme defaults to 'http'.
    static_configs:
      - targets: ['localhost:9090']

  - job_name: 'linux'
    static_configs:
      - targets: ['localhost:9100']
        labels:
          instance: node1

  1. 重启Prometheus

    1
    $ systemctl restart prometheus

  2. 在Prometheus Web查看监控的目标

  • 访问Prometheus Web，在Status->Targets页面下，我们可以看到我们配置的两个Target，它们的State为UP。
  1. 使用Prometheus Web来验证Node Exporter的数据已经被正确的采集。
  • 查看当前主机的CPU负载情况

常用指令

  • systemctl daemon-reload
  • systemctl start node_exporter
  • systemctl stop node_exporter