global:

  scrape_interval: 15s           ##默认拉取频率

  evaluation_interval: 15s      ##执行规则频率

  scrape_timeout: 5s         ##拉取超时时间

alerting:

  alertmanagers:

    - static_configs:

        - targets: ["192.168.48.128:9093"]           ##alertmanager服务地址端口指定

      path_prefix: /root/alertmanager-0.28.0.linux-amd64

rule_files:

  - "alert.yml"                         ##告警规则文件地址文件名指定

scrape_configs:                       ##具体的拉取数据的相关配置

  - job_name: "prometheus"

    static_configs:

      - targets: ["localhost:9090"]

  - job_name: "node-exporter"

    static_configs:

      - targets: ["192.168.48.129:9100"]   ##被监控主机的ip：port

        labels:

          group: 'linux'                          ##自定义标签

      - targets: ['192.168.48.131:9182']

        labels:

          group: 'win10'

groups:
- name: example
  rules:
  
  # 实例不可达警报
  - alert: InstanceDown
    expr: up == 0
    for: 5s
    labels:
      severity: critical
    annotations:
      summary: "Instance {{ $labels.instance }} down"
      description: "实例down"
  # 内存使用率过高警报
  - alert: HighMemoryUsage
    expr: 100 - (node_memory_MemAvailable_bytes / node_memory_MemTotal_bytes * 100) > 80
    for: 10s
    labels:
      severity: warning
    annotations:
      summary: "High Memory Usage on {{ $labels.instance }}"
      description: "Memory使用率超过80%"

[Unit]

Description=prometheus

After=network.target

[Service]

WorkingDirectory=/root/prometheus-3.2.1.linux-amd64                  ##工作路径

ExecStart=/root/prometheus-3.2.1.linux-amd64/prometheus \

 --config.file="prometheus.yml" \                                                        ##配置文件指定

 --web.listen-address=:9090 \                                                     ##运行端口指定

 --storage.tsdb.path /root/prometheus-3.2.1.linux-amd64/data            ##数据存储路径指定

StandardOutput=append:/var/log/prometheus.log

StandardError=append:/var/log/prometheus_error.log

[Install]

WantedBy=default.target

[Unit]

Description=Prometheus Node Exporter

After=network.target

[Service]

Type=simple

User=root

Group=root

ExecStart=/root/node_exporter-1.9.0.linux-amd64/node_exporter \

        --collector.systemd \

        --collector.textfile.directory=/root/node_exporter-1.9.0.linux-amd64/textfiles \

        --web.listen-address=:9100

StandardOutput=append:/var/log/node_exporter.log

StandardError=append:/var/log/node_exporter_error.log

SyslogIdentifier=node_exporter

Restart=always

[Install]

WantedBy=default.target

#!/bin/bash

# 设置你要监控的进程名称

PROCESSNAME="sshd"

# 使用pgrep找到进程ID，注意这里可能返回多个PID，取决于是否有多个实例运行

PIDS=$(pgrep -f "$PROCESSNAME")

# 一个进程名称对应多个进程时，对进程进行编号区分

PROCESSNUM=0

# 清空历史数据

cat /dev/null > /root/node_exporter-1.9.0.linux-amd64/textfiles/process_pid.prom

# 为每个PID创建一个临时文件，格式符合Prometheus文本格式

for PID in $PIDS; do

         ((PROCESSNUM++))

         echo "process_"$PROCESSNAME"_$PROCESSNUM{process=\"$PROCESSNAME\", instance=\"$HOSTNAME\"} $PID" >> /root/node_exporter-1.9.0.linux-amd64/textfiles/process_pid.prom

done

echo “collectors:

  enabled: cpu,cpu_info,exchange,iis,logical_disk,logon,memory,net,os,performancecounter,process,remote_fx,service,system,tcp,time,terminal_services,textfile

  textfile_collector:

    directory: C:\Program Files\windows_exporter\textfile_inputs

log:

  level: warn

  format: logfmt

  file: C:\Program Files\windows_exporter\log\windows_exporter.log” > config.yaml

sc.exe create windows_exporter binPath=""E:\windows_exporter\windows_exporter-windows_exporter-0.30.4-amd64.exe" --collectors.enabled="cpu,cpu_info,exchange,iis,logical_disk,logon,memory,net,os,performancecounter,process,remote_fx,service,system,tcp,time,terminal_services,textfile" --web.listen-address :9182 --collector.textfile.directories="E:\windows_exporter\textfile_inputs" --log.file "E:\windows_exporter\log\windows_exporter.log" --log.format logfmt --log.level debug" start=auto obj="NT AUTHORITY\LocalService"

注：

--web.listen-address :9182                                                                  #监听地址和端口

--collector.textfile.directories="E:\\windows_exporter\\textfile"           #textfile指标文件目录

--log.file "E:\windows_exporter\log\windows_exporter.log"                   #日志输出目标

--log.format logfmt                                                                           #日志格式（logfmt、json、text）

--log.level info"                                                                                 #日志级别（info、debug、warn）

@echo off

setlocal enabledelayedexpansion

:: ----------------------------------------

:: 配置区域（根据实际环境修改）

:: ----------------------------------------

set TEXTFILE_DIR="E:\windows_exporter\textfile_inputs"

set PROCESS_NAME=PowerShell

set LOG_FILE= "E:\windows_exporter\log\textfile.log"

set process_status=2

:: ----------------------------------------

:: 检测PowerShell进程状态（0=未运行，1=运行）

:: ----------------------------------------

tasklist /FI "IMAGENAME eq %PROCESS_NAME%*" 2>nul | find /i "%PROCESS_NAME%" >nul

if %errorlevel% equ 0 (

    set process_status=1

) else (

    set process_status=0

)

:: 确保目录存在并赋权

if not exist "%TEXTFILE_DIR%" (

    mkdir "%TEXTFILE_DIR%"

)

:: 测试环境简化权限

icacls "%TEXTFILE_DIR%" /grant "Everyone:(OI)(CI)W"

:: ----------------------------------------

:: 生成Prometheus指标文件

:: ----------------------------------------

(

    echo # HELP powershell_process_status powershell status 0=running,1=stoped

    echo # TYPE powershell_process_status gauge

    echo powershell_process_status !process_status!

) > "%TEXTFILE_DIR%\process_status.prom"

:: 原子操作：临时文件替换确保采集器读取完整性

:: move /Y %TEXTFILE_DIR%\process_status.prom.tmp %TEXTFILE_DIR%\process_status.prom >nul

:: =============================================

:: 日志记录（带时间戳和状态记录）

:: =============================================

echo [%date% %time%] ^

process_status=!process_status!  >> %LOG_FILE%

endlocal

exit /b 0

global:

  # 处理警报的SMTP服务器的配置

  smtp_smarthost: 'smtp.qq.com:587'

  smtp_from: '6742*****@qq.com'

  smtp_auth_username: '6742*****@qq.com'

  smtp_auth_password: '**************'                        #登录qq邮箱，申请开启smtp服务，会生成此密码

  resolve_timeout: 5m

# 定义路由树，用于决定警报应该发送到哪里

route:

  group_by: ['alertname']

  group_wait: 5s

  group_interval: 10s

  repeat_interval: 12h

  receiver: 'team-X-mails'

  # 可以定义子路由，例如根据严重性或团队进行细分

  #routes:

  #- match:

  #    severity: 'critical'

  #  receiver: 'team-X-pager'

  #  group_wait: 10s

templates:

- '/root/alertmanager-0.28.0.linux-amd64/templates/email_template.tmpl' #模板文件路径

receivers:

- name: 'team-X-mails'

  email_configs:

  - to: '6742*****@qq.com'

    send_resolved: true

    html: '{{ template "email.default.html" . }}'  # 引用模板

    headers:

      Subject: '{{ template "email.default.subject" . }}'  # 邮件主题

  webhook_configs:

  - url: 'http://192.168.48.128:8080/prometheusalert?type=alydx&tpl=prometheus-alydx&phone=153********'

    send_resolved: true

[Unit]

Description=Alertmanager for Prometheus

After=network.target

[Service]

ExecStart=/root/alertmanager-0.28.0.linux-amd64/alertmanager \

  --web.route-prefix=/root/alertmanager-0.28.0.linux-amd64 \

  --web.listen-address=:9093 \

  --config.file=/root/alertmanager-0.28.0.linux-amd64/alertmanager.yml \

  --storage.path=/root/alertmanager-0.28.0.linux-amd64/data

StandardOutput=append:/var/log/alertmanager.log

StandardError=append:/var/log/alertmanager_error.log

[Install]

WantedBy=default.target

#---------------------↓阿里云接口-----------------------

#是否开启阿里云短信告警通道,可同时开始多个通道0为关闭,1为开启

open-alydx=1

#阿里云短信主账号AccessKey的ID

ALY_DX_AccessKeyId=LTAI*********************

#阿里云短信接口密钥

ALY_DX_AccessSecret=****************************************

#阿里云短信签名名称

ALY_DX_SignName=**************

#阿里云短信模板ID

ALY_DX_Template=SMS_***********

[Unit]

Description=PrometheusAlert Service

After=network.target

[Service]

Type=simple

User=root  # 建议使用非root用户，需提前创建

Group=root

WorkingDirectory=/root/PrometheusAlert  # 替换为你的安装路径

ExecStart=/root/PrometheusAlert/PrometheusAlert  # 指向可执行文件路径

Restart=always

StandardOutput=/var/log/prometheusalert.log

StandardError=/var/log/prometheusalert_error.log

SyslogIdentifier=prometheusalert

[Install]

WantedBy=default.target

[Unit]

Description=Pushgetway

After=network.target

[Service]

Type=simple

Group=prometheus

ExecStart=/opt/prometheus/pushgateway/pushgateway

Restart=on-failure

[Install]

WantedBy=default.target

  - job_name: "pushgateway"

    scrape_interval: 30s

    honor_labels: true  //加上此配置，exporter节点上传数据中的一些标签将不会被pushgateway节点的相同标签覆盖

    static_configs:

    - targets: ["localhost:9091"]

      labels:

        instance: pushgateway