prometheus 监控系统

一. 安装docker环境

#01 安装docker
# step 1: 安装必要的一些系统工具
sudo yum install -y yum-utils device-mapper-persistent-data lvm2
# Step 2: 添加软件源信息
sudo yum-config-manager --add-repo https://mirrors.aliyun.com/docker-ce/linux/centos/docker-ce.repo
# Step 3
sudo sed -i 's+download.docker.com+mirrors.aliyun.com/docker-ce+' /etc/yum.repos.d/docker-ce.repo
# Step 4: 更新并安装Docker-CE
sudo yum makecache fast
sudo yum -y install docker-ce
# Step 4: 开启Docker服务
sudo service docker start
 
#02 设置镜像加速
sudo mkdir -p /etc/docker
sudo tee /etc/docker/daemon.json <<-'EOF'
{
  "registry-mirrors": ["https://9akof579.mirror.aliyuncs.com"]
}
EOF
sudo systemctl daemon-reload
sudo systemctl restart docker
 
#03 安装 docker-compose
sudo curl -L "https://github.com/docker/compose/releases/download/1.25.5/docker-compose-$(uname -s)-$(uname -m)" -o /usr/local/bin/docker-compose
sudo chmod +x /usr/local/bin/docker-compose

 

二. 安装prometheus

2.1 编辑配置文件

#01 编辑配置文件 邮件相关
1) alertmanager.yaml
[root@iZ2ze9nv7e19wvoseb2sl1Z prometheus]# cat alertmanager.yaml
global:
  smtp_smarthost: 'z15119911990@163.com:25'　　#163服务器
  smtp_from: '522308001@qq.com'　　　　　　　　#发邮件的邮箱
  smtp_auth_username: 'z15119911990'　　#发邮件的邮箱用户名，也就是你的邮箱
  smtp_auth_password: 'RVDDRXWLPQHVYNOB'#发邮件的邮箱密码
  smtp_require_tls: false　　　　　　　　#不进行tls验证
 
route:
  group_by: ['alertname']
  group_wait: 10s
  group_interval: 10s
  repeat_interval: 10m
  receiver: live-monitoring
 
receivers:
- name: 'live-monitoring'
  email_configs:
  - to: 'z15119911990@163.com'
 
2) node_down.yml 自定义的告警规则文件
[root@iZ2ze9nv7e19wvoseb2sl1Z prometheus]# cat node_down.yml
groups:
- name: node_down
  rules:
  - alert: InstanceDown
    expr: up == 0
    for: 1m
    labels:
      user: test
    annotations:
      summary: "Instance {{ $labels.instance }} down"
      description: "{{ $labels.instance }} of job {{ $labels.job }} has been down for more than 1 minutes."
 
 
3）主配置文件
[root@iZ2ze9nv7e19wvoseb2sl1Z prometheus]# grep -Ev '^#|^$' prometheus.yml
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).
alerting:
  alertmanagers:
  - static_configs:
    - targets: ['172.17.24.204:9093']
      # - alertmanager:9093
rule_files:
  - "node_down.yml"
  # - "first_rules.yml"
  # - "second_rules.yml"
scrape_configs:
  # The job name is added as a label `job=<job_name>` to any timeseries scraped from this config.
  - job_name: 'prometheus'
    static_configs:
    - targets: ['172.17.24.204:9090']
  - job_name: 'cadvisor'
    static_configs:
    - targets: ['172.17.24.204:8080']
  - job_name: 'node'
    scrape_interval: 8s
    static_configs:
      - targets: ['172.17.24.204:9100','172.17.24.203:9100']    #抓取被监控主机IP

 

2.2 编辑docker-compose

#01 书写docker-compose
[root@iZ2ze9nv7e19wvoseb2sl1Z prometheus]# cat docker-compose.yaml
version: '2'
 
networks:
    mynet:
        driver: bridge
 
services:
    prometheus:
        image: prom/prometheus
        container_name: prometheus
        hostname: prometheus
        restart: always
        volumes:
            - ./prometheus.yml:/etc/prometheus/prometheus.yml
            - ./node_down.yml:/etc/prometheus/node_down.yml
        ports:
            - "9090:9090"
        networks:
            - mynet
 
    alertmanager:
        image: prom/alertmanager
        container_name: alertmanager
        hostname: alertmanager
        restart: always
        volumes:
            - ./alertmanager.yaml:/etc/alertmanager/alertmanager.yaml
        ports:
            - "9093:9093"
        networks:
            - mynet
 
    grafana:
        image: grafana/grafana
        container_name: grafana
        hostname: grafana
        restart: always
        ports:
            - "3000:3000"
        networks:
            - mynet
 
    node-exporter:
        image: prom/node-exporter
        #image: quay.io/prometheus/node-exporter
        container_name: node-exporter
        hostname: node-exporter
        restart: always
        ports:
            - "9100:9100"
        networks:
            - mynet
    cadvisor:
        image: google/cadvisor:latest
        container_name: cadvisor
        hostname: cadvisor
        restart: always
        volumes:
            - /:/rootfs:ro
            - /var/run:/var/run:rw
            - /sys:/sys:ro
            - /var/lib/docker/:/var/lib/docker:ro
        ports:
            - "8080:8080"
        networks:
            - mynet
 
#02 启动 docker-compose up -d
 
#03 web验证
 
http://172.17.24.204:9090/

 

三. grafana 展示

#01 登入地址
172.17.24.204:3000
user: admin
passwd: admin
 
#02 使用node 模版

 

 

四 添加监控节点

  • 选择对应的 exporter
#01 安装exporter
docker pull prom/node-exporter
docker run -d -p 9100:9100 --restart=always  prom/node-exporter
 
 
#02 访问验证
http://localhost:9100/
 
#03 修改服务端 prometheus配置文件
  - job_name: 'node'
    scrape_interval: 8s
    static_configs:
      - targets: ['172.17.24.204:9100','172.17.24.203:9100']    #抓取被监控主机IP

 

五. 监控 java进程

 

#01 下载
wget https://repo1.maven.org/maven2/io/prometheus/jmx/jmx_prometheus_javaagent/0.17.0/jmx_prometheus_javaagent-0.17.0.jar
 
#02 书写配置文件
[root@iZ2ze9nv7e19wvoseb2skzZ prometheus]# cat simple-config.yml
lowercaseOutputLabelNames: true
lowercaseOutputName: true
whitelistObjectNames: ["java.lang:type=OperatingSystem"]
blacklistObjectNames: []
rules:
  - pattern: 'java.lang<type=OperatingSystem><>(committed_virtual_memory|free_physical_memory|free_swap_space|total_physical_memory|total_swap_space)_size:'
    name: os_$1_bytes
    type: GAUGE
    attrNameSnakeCase: true
  - pattern: 'java.lang<type=OperatingSystem><>((?!process_cpu_time)w+):'
    name: os_$1
    type: GAUGE
    attrNameSnakeCase: true
 
#03 启动服务
java -javaagent:./jmx_prometheus_javaagent-0.17.0.jar=9400:simple-config.yaml -jar <你的jar>
nohup java -javaagent:./jmx_prometheus_javaagent-0.17.0.jar=9400:simple-config.yml -jar /data/project/pigx-upms-biz/pigx-upms-biz.jar -java.tmp.dir=/data/upload_tmp &
 
 
#04 书写启动脚本
/opt/prometheus/jmx_prometheus_javaagent-0.17.0.jar
/opt/prometheus/simple-config.yml
-javaagent:/opt/prometheus/jmx_prometheus_javaagent-0.17.0.jar=9400:/opt/prometheus/simple-config.yml
 
[root@iZ2ze9nv7e19wvoseb2skzZ prometheus]# cat  /root/sh/pigx-upms-biz.sh
#!/bin/bash
source /etc/profile
PORT=4000
kill -9 $(lsof -i:$PORT | grep LISTEN | awk '{print $2}')
sleep 5s
echo "$name kill"
name_tmp=${0%\.*}
name=${name_tmp##*/}
LOG_PATH=${name%%.jar}
Promtheus="-javaagent:/opt/prometheus/jmx_prometheus_javaagent-0.17.0.jar=9400:/opt/prometheus/simple-config.yml "
echo $name
#nohup java -jar -Dfile.encoding=UTF-8 -Xms128m -Xmx1g -Dlog.path=/pigx/$LOG_PATH /data/project/$name/$name.jar -java.tmp.dir=/data/upload_tmp >> /data/project/$name/$name.log 2>&1 &
nohup   java ${Promtheus} -jar -Dfile.encoding=UTF-8 -Xms128m -Xmx1g -Dlog.path=/pigx/$LOG_PATH /data/project/$name/$name.jar -java.tmp.dir=/data/upload_tmp >> /data/project/$name/$name.log 2>&1 &

 

 

六. 自动化监控服务

#依赖项目
         <dependency>
               <groupId>io.micrometer</groupId>
               <artifactId>micrometer-registry-prometheus</artifactId>
           </dependency>
           <dependency>
               <groupId>org.springframework.boot</groupId>
               <artifactId>spring-boot-starter-actuator</artifactId>
               <version>2.2.0.RELEASE</version>
           </dependency>
  • 启动类
@SpringBootApplication
public class Springboot2PrometheusApplication {
  
    public static void main(String[] args) {
        SpringApplication.run(Springboot2PrometheusApplication.class, args);
    }
    @Bean
    MeterRegistryCustomizer<MeterRegistry> configurer(
            @Value("${spring.application.name}") String applicationName) {
        return (registry) -> registry.config().commonTags("application", applicationName);
    }
}

 

  • 配置

management.endpoints.web.exposure.include=*
management.metrics.tags.application=${spring.application.name}
micrometer-registry-prometheus 指定版本可能会导致不兼容

 

management.endpoints.web.base-path=/admin
management.server.port=8090

 

  • grafana id = 4701
posted @   宁采臣open  阅读(62)  评论(0)  编辑  收藏  举报
相关博文：
· Zabbix MQQT协议监控 loT设备
· kubernetes 1.20版本 二进制部署
· prometheus监控系统搭建
· 监控工具prometheus配置-docker版
· 1-3 docker 安装 prometheus
阅读排行：
· 全程不用写代码，我用AI程序员写了一个飞机大战
· DeepSeek 开源周回顾「GitHub 热点速览」
· 记一次.NET内存居高不下排查解决与启示
· MongoDB 8.0这个新功能碉堡了，比商业数据库还牛
· .NET10 - 预览版1新功能体验（一）
点击右上角即可分享
微信分享提示
AI FOR CODE 大赛