Prometheusbu deployment Use -1

Prometheus + grafana deployment uses

Host List:

192.168.161.130 ： Prometheus
192.168.161.128 : node-1
192.168.161.129 ： node-2

Binary manual deployment

Extracting prometheus-2.14.0.linux-amd64.tar.gz packet
decompression discharge to the / usr / local

Service startup script editor

vim /usr/lib/systemd/system/prometheus.service

[Unit]
Description=Prometheus server daemon

[Service]
ExecStart=/usr/local/prome/prometheus --config.file=/usr/local/prome/prometheus.yml
ExecReload=/bin/kill -HUP $MAINPID
KillMode=process
Restart=on-failure
RestartSec=42s

[Install]
WantedBy=multi-user.target

systemctl daemon-reload  # 加载后台服务
systemctl start | stop | restart | status prometheus

Server monitoring profile contents

global:    # 全局配置段
  scrape_interval:      15s
  evaluation_interval:  15s
  external_labels:
      monitor: 'codelab-monitor'
rule_files:     # 告警规则配置

scrape_configs:     # 监控配置
  - job_name: 'prometheus'   #监控任务名称


    static_configs:  # 静态监控配置
      - targets: ['192.168.235.130:9090']
        labels:
          node: prome_server

With Dynamic Discovery Add Host

global:    # 全局配置段
  scrape_interval:      15s
  evaluation_interval:  15s
  external_labels:
      monitor: 'codelab-monitor'
rule_files:     # 告警规则配置

scrape_configs:     # 监控配置
  - job_name: 'prometheus'   #监控任务名称


#    static_configs:  # 静态监控配置
#      - targets: ['192.168.235.130:9090']
#        labels:
#          node: prome_server
    file_sd_configs: # 基于配置文件动态发现添加主机
       - files: ['/usr/local/prome/sd_config/*.yml']  # 定义服务发现配置路径
         refresh_interval: 5s  # 5秒发现刷洗发现一次

Create a service discovery profile

mkdir /usr/local/prome/sd_config

Edit prometheus monitor configuration file

- targets:            # 直接指定动态发现下的 targets 目标主机
  - 192.168.235.130:9090
    labels:           # 给目标主机添加标签标识
      node: prome_server

Overload Services

systemctl reload prometheus

Monitored data acquisition tools mounted node node-exporter

Decompression

[root@localhost ~]# tar zxf node_exporter-0.18.1.linux-amd64.tar.gz 
[root@localhost ~]# mv node_exporter-0.18.1.linux-amd64 /usr/local/

Service management system will be made operational service

[Unit]
Description=Node_exporter server daemon

[Service]
ExecStart=/usr/local/node_exporter/node_exporter
Restart=on-failure

[Install]
WantedBy=multi-user.target

prometheus Add Remote Host Monitoring

Add the task in the original prometheus.yml, the monitoring remote host

# my global config
global:
  evaluation_interval: 15s
  scrape_interval:     15s

rule_files:
  # - "first_rules.yml"
  # - "second_rules.yml"

scrape_configs:
  - job_name: 'prometheus'
    static_configs:
    - targets: ['192.168.235.130:9090']
      labels:
        node: prome_server
#    file_sd_configs:
#      - files: ['/usr/local/prome/sd_config/*.yml']
#        refresh_interval: 5s
  - job_name: 'node-1'   # 添加主机node-1
    static_configs:
    - targets: ['192.168.235.128:9100']
      labels:
        node: node-1

  - job_name: 'node-2'  # 添加主机node-2
    static_configs:
    - targets: ['192.168.235.129:9100']
      labels:
        node: node-2

Add the node_exporter service monitoring

node_exporter can also service status monitoring system, the startup parameter is specified to enable service monitoring--collector.systemd --collector.systemd.unit-whitelist=服务名称

[Unit]
Description=Node_exporter server daemon

[Service]
ExecStart=/usr/local/node_exporter/node_exporter --collector.systemd  --collector.systemd.unit-whitelist=(sshd|docker|nginx).service
Restart=on-failure

[Install]
WantedBy=multi-user.target

--collector.systemd: Enable express service monitoring
--collector.systemd.unit-whitelist = (sshd | docker | nginx) .service: representation monitors multiple service

grafan quickly use monitoring templates to quickly monitor

grafana offers a number of monitoring templates for a variety of data sources and exporter produced a different monitoring templates can be used directly in the lead grafana

https://grafana.com/grafana/dashboards

Directly into the ID value monitoring templates

Select a monitoring template, find the Get this dashboard:value, there is a good template for the ID template node-exporter of: 8919, it can quickly realize monitor host resources.
Similarly, container monitoring, too. You can find the right template