Prometheus&Grafana

monitor

Prometheus

node_exporter

  • package

    1
    wget https://github.com/prometheus/node_exporter/releases/download/v1.8.2/node_exporter-1.8.2.linux-amd64.tar.gz

  • prepare user and directory

    1
    2
    3
    4
    5
    6
    7
    sudo groupadd -f dingofs
    sudo useradd -g dingofs dingofs
    sudo mkdir /etc/node_exporter
    sudo chown dingofs:dingofs /etc/node_exporter

    sudo cp /path/to/node_exporter-files/node_exporter /usr/bin/
    sudo chown dingofs:dingofs /usr/bin/node_exporter

  • config system service

    1
    2
    sudo vi /usr/lib/systemd/system/node_exporter.service
    sudo chmod 664 /usr/lib/systemd/system/node_exporter.service
    1
    2
    3
    4
    5
    6
    7
    8
    9
    10
    11
    12
    13
    14
    15
    16
    [Unit]
    Description=Node Exporter
    Documentation=https://prometheus.io/docs/guides/node-exporter/
    Wants=network-online.target
    After=network-online.target

    [Service]
    User=dingofs
    Group=dingofs
    Type=simple
    Restart=on-failure
    ExecStart=/usr/bin/node_exporter \
      --web.listen-address=:19100

    [Install]
    WantedBy=multi-user.target

  • start system service

    1
    2
    3
    4
    sudo systemctl daemon-reload
    sudo systemctl start node_exporter
    sudo systemctl status node_exporter
    sudo systemctl enable node_exporter.service # Configure node_exporter to start at boot

prometheus server

  • prometheus.yml

    1
    2
    3
    4
    5
    6
    7
    8
    9
    10
    11
    12
    13
    14
    15
    16
    17
    18
    19
    20
    21
    22
    23
    24
    25
    26
    27
    28
    29
    global:
      scrape_interval:     15s # By default, scrape targets every 15 seconds.
      evaluation_interval: 15s # Evaluate rules every 15 seconds.

      # Attach these extra labels to all timeseries collected by this Prometheus instance.
      external_labels:
        monitor: 'codelab-monitor'

    #rule_files:
    #- 'prometheus.rules.yml'

    scrape_configs:
      - job_name: 'prometheus'

        # Override the global default and scrape targets from this job every 5 seconds.
        scrape_interval: 5s

        static_configs:
          - targets: ['172.20.7.232:19090']

      - job_name: 'node'

        # Override the global default and scrape targets from this job every 5 seconds.
        scrape_interval: 5s

        static_configs:
          - targets: ['172.20.7.232:19100','172.20.7.233:19100','172.20.7.234:19100']
            labels:
              group: 'metric1'

  • docker container

1
2
3
4
5
6
7
8
# Create persistent volume for your data
docker volume create prometheus-data
# Start Prometheus container
docker run -d \
    -p 9090:9090 \
    -v /path/to/prometheus.yml:/etc/prometheus/prometheus.yml \
    -v prometheus-data:/prometheus \
    prom/prometheus

Grafana

  • deploy container

    1
    2
    3
    4
    5
    6
    7
    8
    9
    10
    11
    12
    13
    # create a persistent volume for your data
    docker volume create grafana-storage
    # verify that the volume was created correctly
    # you should see some JSON output
    docker volume inspect grafana-storage

    # start grafana
    docker run -d -p 13000:3000 --name=grafana \
      --volume grafana-storage:/var/lib/grafana \
      grafana/grafana

    # web login
    admin/admin

dashboard

https://grafana.com/grafana/dashboards/

just import from existed dashboard json

enjoy it