ubuntu 22.04 配置 Prometheus 和 Grafana 服务器监控

由于内存白菜价， 家里的虚拟机越来越多了。

1个是难以管理hostname 和 ip， 只靠/etc/hosts 去记住其实并不方便

试过宝塔linux面板， 可惜是单机的，不适用于管理集群linux服务器， 而且里面一股金钱气息。
至于他们另1个产品堡塔云， 估计是集群管理， 但是安装失败， 而且看了价格感觉像智商税。

公司用过Appdynamic， 功能强大可惜更不是我能想的。

只能还是去寻找开源方案。好在外能的google提供了一套成熟的开源方案

Prometheus - 用于管理机 - 集群服务器数据收集
node_exporter - 用于被管理节点 - 提供节点本身的服务器数据给Prometheus
Grafana - 数据展示工具

下面就会写出详细的安装步骤

1. 服务器准备

2. 在宿主机安装Prometheus

2.1 安装

个人觉得不用下载安装包这么复杂， 直接从ubuntu源安装

sudo apt-get install prometheus
1

检查prometheus 有没有启动

gateman@amdeuc-vm0:~$ systemctl status prometheus
● prometheus.service - Monitoring system and time series database
     Loaded: loaded (/lib/systemd/system/prometheus.service; enabled; vendor preset: enabled)
     Active: active (running) since Thu 2022-11-24 15:58:52 UTC; 4h 3min ago
       Docs: https://prometheus.io/docs/introduction/overview/
             man:prometheus(1)
   Main PID: 23080 (prometheus)
      Tasks: 7 (limit: 23310)
     Memory: 44.6M
        CPU: 30.744s
     CGroup: /system.slice/prometheus.service
             └─23080 /usr/bin/prometheus
1
2
3
4
5
6
7
8
9
10
11
12

2.2 检查配置文件

gateman@amdeuc-vm0:/etc/prometheus$ cat prometheus.yml 
# Sample config for Prometheus.

global:
  scrape_interval:     15s # Set the scrape interval to every 15 seconds. Default is every 1 minute.
  evaluation_interval: 15s # Evaluate rules every 15 seconds. The default is every 1 minute.
  # scrape_timeout is set to the global default (10s).

  # Attach these labels to any time series or alerts when communicating with
  # external systems (federation, remote storage, Alertmanager).
  external_labels:
      monitor: 'example'

# Alertmanager configuration
alerting:
  alertmanagers:
  - static_configs:
    - targets: ['localhost: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=` to any timeseries scraped from this config.
  - job_name: 'prometheus'

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

    # metrics_path defaults to '/metrics'
    # scheme defaults to 'http'.

    static_configs:
      - targets: ['localhost:9090']

  - job_name: node
    # If prometheus-node-exporter is installed, grab stats about the local
    # machine by default.
    static_configs:
      - targets: ['localhost:9100']
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
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45

可以见prometheus 的 控制端口是9090

在浏览器访问服务器ip和9090 端口， 如果 prometheus 页面能打开就没什么问题

3. 在被监控节点安装node exporter

也是直接从ubuntu源里安装

sudo apt-get install prometheus-node-exporter
1

检查node exporter 服务有没有启动

gateman@amdeuc-vm3:~$ systemctl status prometheus-node-exporter
● prometheus-node-exporter.service - Prometheus exporter for machine metrics
     Loaded: loaded (/lib/systemd/system/prometheus-node-exporter.service; enabled; vendor preset: enabled)
     Active: active (running) since Fri 2022-11-25 11:49:24 UTC; 5min ago
       Docs: https://github.com/prometheus/node_exporter
   Main PID: 1567 (prometheus-node)
      Tasks: 4 (limit: 4460)
     Memory: 2.2M
        CPU: 7ms
     CGroup: /system.slice/prometheus-node-exporter.service
             └─1567 /usr/bin/prometheus-node-exporter

Nov 25 11:49:24 amdeuc-vm3 prometheus-node-exporter[1567]: ts=2022-11-25T11:49:24.830Z caller=node_exporter.go:115 level=info collector=thermal_zone
Nov 25 11:49:24 amdeuc-vm3 prometheus-node-exporter[1567]: ts=2022-11-25T11:49:24.830Z caller=node_exporter.go:115 level=info collector=time
Nov 25 11:49:24 amdeuc-vm3 prometheus-node-exporter[1567]: ts=2022-11-25T11:49:24.830Z caller=node_exporter.go:115 level=info collector=timex
Nov 25 11:49:24 amdeuc-vm3 prometheus-node-exporter[1567]: ts=2022-11-25T11:49:24.830Z caller=node_exporter.go:115 level=info collector=udp_queues
Nov 25 11:49:24 amdeuc-vm3 prometheus-node-exporter[1567]: ts=2022-11-25T11:49:24.830Z caller=node_exporter.go:115 level=info collector=uname
Nov 25 11:49:24 amdeuc-vm3 prometheus-node-exporter[1567]: ts=2022-11-25T11:49:24.830Z caller=node_exporter.go:115 level=info collector=vmstat
Nov 25 11:49:24 amdeuc-vm3 prometheus-node-exporter[1567]: ts=2022-11-25T11:49:24.830Z caller=node_exporter.go:115 level=info collector=xfs
Nov 25 11:49:24 amdeuc-vm3 prometheus-node-exporter[1567]: ts=2022-11-25T11:49:24.830Z caller=node_exporter.go:115 level=info collector=zfs
Nov 25 11:49:24 amdeuc-vm3 prometheus-node-exporter[1567]: ts=2022-11-25T11:49:24.830Z caller=node_exporter.go:199 level=info msg="Listening on" address=:9100
Nov 25 11:49:24 amdeuc-vm3 prometheus-node-exporter[1567]: ts=2022-11-25T11:49:24.830Z caller=tls_config.go:195 level=info msg="TLS is disabled." http2=false
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22

可以简单监控端口是9100

4. 配置监控机的 prometheus.xml 加入被监控的节点

  - job_name: node
    # If prometheus-node-exporter is installed, grab stats about the local
    # machine by default.
    static_configs:
      - targets: ['localhost:9100', '10.0.1.156:9100', '10.0.1.157:9100', '10.0.1.158:9100']
1
2
3
4
5

然后重启服务

 sudo systemctl restart prometheus
1

再打开 prometheus 的监控列表， 发现 已经有4个节点被监控了， 包括监控机本身

5.安装Grafana

没想到grafana居然不在ubuntu的默认仓库…

sudo wget -q -O - https://packages.grafana.com/gpg.key | sudo apt-key add -
echo "deb https://packages.grafana.com/oss/deb stable main" | sudo tee -a /etc/apt/sources.list.d/grafana.list
sudo apt-get update
sudo apt install grafana
1
2
3
4

安装完后检查服务有没有启动

 systemctl status grafana-server
1

没有就手动起来

gateman@amdeuc-vm0:/etc/prometheus$ sudo systemctl start grafana-server
gateman@amdeuc-vm0:/etc/prometheus$ sudo systemctl enable grafana-server
1
2

grafana 的默认端口是3000
再浏览器打开ip:3000， 如果见到grafana页面就代表安装好了

6.配置grafana

默认账号和密码都是admin 先登录再说
登录后马上要求你改密码的

6.1 添加数据源

Configuration -> Data Sources ->add data source -> Prometheus

6.2 添加Dashboard

接下来我们先去下载一个模板
https://grafana.com/grafana/dashboards/?search=8919

把Json文件下载下来就好

然后在grafana
New Dashboard -> Import Json
就得到1个比较完善的linux主机详情模板了

看着dashboard 效果， 各台服务器的状态一目了然， 不比那宝塔面板吊多了？

如何你的服务器太多， 想分组看?
那么prometheus 的配置文件可以这样编写

  - job_name: node
    # If prometheus-node-exporter is installed, grab stats about the local
    # machine by default.
    static_configs:
      - targets: 
        - '10.0.1.156:9100'
        - '10.0.1.157:9100'
        - '10.0.1.158:9100'
        - '10.0.1.158:9100'
        - '10.0.1.154:9100'
        - '10.0.1.152:9100'
        labels:
          group: 'VM Group'
      - targets: 
        - '10.0.1.198:9100'
        - '10.0.1.122:9100'
        - '10.0.1.107:9100'
        - '10.0.1.223:9182'
        labels:
          group: 'PHY Group'
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20

在grafana中就可以分组查看了