1.创建项目

pom.xml引入相关依赖

<project xmlns="http://maven.apache.org/POM/4.0.0" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://maven.apache.org/POM/4.0.0 https://maven.apache.org/xsd/maven-4.0.0.xsd">
	<modelVersion>4.0.0</modelVersion>
	<groupId>com.olive</groupId>
	<artifactId>prometheus-meter-demo</artifactId>
	<version>0.0.1-SNAPSHOT</version>
	<parent>
		<groupId>org.springframework.boot</groupId>
		<artifactId>spring-boot-starter-parent</artifactId>
		<version>2.3.7.RELEASE</version>
		<relativePath />
	</parent>
	<properties>
		<java.version>1.8</java.version>
		<project.build.sourceEncoding>UTF-8</project.build.sourceEncoding>
		<project.reporting.outputEncoding>UTF-8</project.reporting.outputEncoding>
		<spring-boot.version>2.3.7.RELEASE</spring-boot.version>
	</properties>
	<dependencies>
		<dependency>
			<groupId>org.springframework.boot</groupId>
			<artifactId>spring-boot-starter-aop</artifactId>
		</dependency>
		<dependency>
			<groupId>org.springframework.boot</groupId>
			<artifactId>spring-boot-starter-web</artifactId>
		</dependency>
		<dependency>
			<groupId>org.springframework.boot</groupId>
			<artifactId>spring-boot-starter-actuator</artifactId>
		</dependency>
		<!-- Micrometer Prometheus registry  -->
		<dependency>
			<groupId>io.micrometer</groupId>
			<artifactId>micrometer-registry-prometheus</artifactId>
		</dependency>
	</dependencies>
	<dependencyManagement>
		<dependencies>
			<dependency>
				<groupId>org.springframework.boot</groupId>
				<artifactId>spring-boot-dependencies</artifactId>
				<version>${spring-boot.version}</version>
				<type>pom</type>
				<scope>import</scope>
			</dependency>
		</dependencies>
	</dependencyManagement>
</project>

2.自定义指标

方式一

直接使用micrometer核心包的类进行指标定义和注册

package com.olive.monitor;
 
import javax.annotation.PostConstruct;
 
import org.springframework.beans.factory.annotation.Autowired;
import org.springframework.stereotype.Component;
 
import io.micrometer.core.instrument.Counter;
import io.micrometer.core.instrument.DistributionSummary;
import io.micrometer.core.instrument.MeterRegistry;
 
@Component
public class NativeMetricsMontior {
 
	/**
	 * 支付次数
	 */
	private Counter payCount;
 
	/**
	 * 支付金额统计
	 */
	private DistributionSummary payAmountSum;
 
	@Autowired
	private MeterRegistry registry;
 
	@PostConstruct
	private void init() {
		payCount = registry.counter("pay_request_count", "payCount", "pay-count");
		payAmountSum = registry.summary("pay_amount_sum", "payAmountSum", "pay-amount-sum");
	}
 
	public Counter getPayCount() {
		return payCount;
	}
 
	public DistributionSummary getPayAmountSum() {
		return payAmountSum;
	}
 
}

方式二

通过引入micrometer-registry-prometheus包,该包结合prometheus,对micrometer进行了封装

<dependency>
			<groupId>io.micrometer</groupId>
			<artifactId>micrometer-registry-prometheus</artifactId>
		</dependency>

同样定义两个metrics

package com.olive.monitor;
 
import javax.annotation.PostConstruct;
 
import org.springframework.beans.factory.annotation.Autowired;
import org.springframework.stereotype.Component;
 
import io.prometheus.client.CollectorRegistry;
import io.prometheus.client.Counter;
 
@Component
public class PrometheusMetricsMonitor {
 
	/**
	 * 订单发起次数
	 */
	private Counter orderCount;
 
	/**
	 * 金额统计
	 */
	private Counter orderAmountSum;
	
	@Autowired
	private CollectorRegistry registry;
	@PostConstruct
	private void init() {
		orderCount = Counter.build().name("order_request_count")
				.help("order request count.")
				.labelNames("orderCount")
				.register();
		orderAmountSum = Counter.build().name("order_amount_sum")
				.help("order amount sum.")
				.labelNames("orderAmountSum")
				.register();
		registry.register(orderCount);
		registry.register(orderAmountSum);
	}
 
	public Counter getOrderCount() {
		return orderCount;
	}
 
	public Counter getOrderAmountSum() {
		return orderAmountSum;
	}
 
}

prometheus 4种常用Metrics

Counter

连续增加不会减少的计数器,可以用于记录只增不减的类型,例如:网站访问人数,系统运行时间等。

对于Counter类型的指标,只包含一个inc()的方法,就是用于计数器+1.

一般而言,Counter类型的metric指标在冥冥中我们使用_total结束,如http_requests_total.

Gauge

可增可减的仪表盘,曲线图

对于这类可增可减的指标,用于反应应用的当前状态。

例如在监控主机时,主机当前空闲的内存大小,可用内存大小等等。

对于Gauge指标的对象则包含两个主要的方法inc()和dec(),用于增加和减少计数。

Histogram

主要用来统计数据的分布情况,这是一种特殊的metrics数据类型,代表的是一种近似的百分比估算数值,统计所有离散的指标数据在各个取值区段内的次数。例如:我们想统计一段时间内http请求响应小于0.005秒、小于0.01秒、小于0.025秒的数据分布情况。那么使用Histogram采集每一次http请求的时间,同时设置bucket。

Summary

Summary和Histogram非常相似,都可以统计事件发生的次数或者大小,以及其分布情况,他们都提供了对时间的计数_count以及值的汇总_sum,也都提供了可以计算统计样本分布情况的功能,不同之处在于Histogram可以通过histogram_quantile函数在服务器计算分位数。而Sumamry的分位数则是直接在客户端进行定义的。因此对于分位数的计算,Summary在通过PromQL进行查询的时候有更好的性能表现,而Histogram则会消耗更多的资源,但是相对于客户端而言Histogram消耗的资源就更少。用哪个都行,根据实际场景*调整即可。

3. 测试

定义两个controller分别使用NativeMetricsMontiorPrometheusMetricsMonitor

package com.olive.controller;
 
import java.util.Random;
 
import javax.annotation.Resource;
 
import org.springframework.web.bind.annotation.RequestMapping;
import org.springframework.web.bind.annotation.RequestParam;
import org.springframework.web.bind.annotation.RestController;
 
import com.olive.monitor.NativeMetricsMontior;
 
@RestController
public class PayController {
 
	@Resource
	private NativeMetricsMontior monitor;
 
	@RequestMapping("/pay")
	public String pay(@RequestParam("amount") Double amount) throws Exception {
		// 统计支付次数
		monitor.getPayCount().increment();
 
		Random random = new Random();
		//int amount = random.nextInt(100);
		if(amount==null) {
			amount = 0.0;
		}
		// 统计支付总金额
		monitor.getPayAmountSum().record(amount);
		return "支付成功, 支付金额: " + amount;
	}
 
}
package com.olive.controller;
 
import java.util.Random;
 
import javax.annotation.Resource;
 
import org.springframework.web.bind.annotation.RequestMapping;
import org.springframework.web.bind.annotation.RequestParam;
import org.springframework.web.bind.annotation.RestController;
 
import com.olive.monitor.PrometheusMetricsMonitor;
 
@RestController
public class OrderController {
 
	@Resource
	private PrometheusMetricsMonitor monitor;
 
	@RequestMapping("/order")
	public String order(@RequestParam("amount") Double amount) throws Exception {
		// 订单总数
		monitor.getOrderCount()
			.labels("orderCount")
			.inc();
 
		Random random = new Random();
		//int amount = random.nextInt(100);
		if(amount==null) {
			amount = 0.0;
		}
		// 统计订单总金额
		monitor.getOrderAmountSum()
			.labels("orderAmountSum")
			.inc(amount);
		return "下单成功, 订单金额: " + amount;
	}
 
}

启动服务

访问http://127.0.0.1:9595/actuator/prometheus;正常看到监测数据

Spring Boot自定义监控指标的详细过程

改变amount多次方式http://127.0.0.1:8080/order?amount=100http://127.0.0.1:8080/pay?amount=10后;再访问http://127.0.0.1:9595/actuator/prometheus。查看监控数据

Spring Boot自定义监控指标的详细过程

4.项目中的应用

项目中按照上面说的方式进行数据埋点监控不太现实;在spring项目中基本通过AOP进行埋点监测。比如写一个切面Aspect;这样的方式就非常友好。能在入口就做了数据埋点监测,无须在controller里进行代码编写。

package com.olive.aspect;
 
import java.time.LocalDate;
import java.util.concurrent.TimeUnit;
 
import javax.servlet.http.HttpServletRequest;
 
import org.aspectj.lang.ProceedingJoinPoint;
import org.aspectj.lang.annotation.Around;
import org.aspectj.lang.annotation.Aspect;
import org.aspectj.lang.annotation.Pointcut;
import org.springframework.stereotype.Component;
import org.springframework.util.StringUtils;
import org.springframework.web.context.request.RequestContextHolder;
import org.springframework.web.context.request.ServletRequestAttributes;
 
import io.micrometer.core.instrument.Metrics;
 
@Aspect
@Component
public class PrometheusMetricsAspect {
 
    // 切入所有controller包下的请求方法
    @Pointcut("execution(* com.olive.controller..*.*(..))")
    public void controllerPointcut() {
    }
 
    @Around("controllerPointcut()")
    public Object MetricsCollector(ProceedingJoinPoint joinPoint) throws Throwable {
 
        HttpServletRequest request = ((ServletRequestAttributes) RequestContextHolder.getRequestAttributes()).getRequest();
        String userId = StringUtils.hasText(request.getParameter("userId")) ? 
        		request.getParameter("userId") : "no userId";
        
        // 获取api url
        String api = request.getServletPath();
        // 获取请求方法
        String method = request.getMethod();
        long startTs = System.currentTimeMillis();
        LocalDate now = LocalDate.now();
        String[] tags = new String[10];
        tags[0] = "api";
        tags[1] = api;
        tags[2] = "method";
        tags[3] = method;
        tags[4] = "day";
        tags[5] = now.toString();
        tags[6] = "userId";
        tags[7] = userId;
        
        String amount = StringUtils.hasText(request.getParameter("amount")) ? 
        		request.getParameter("amount") : "0.0";
        
        tags[8] = "amount";
        tags[9] = amount;
        // 请求次数加1
        //自定义的指标名称:custom_http_request_all,指标包含数据
        Metrics.counter("custom_http_request_all", tags).increment();
        Object object = null;
        try {
            object = joinPoint.proceed();
        } catch (Exception e) {
            //请求失败次数加1
            Metrics.counter("custom_http_request_error", tags).increment();
            throw e;
        } finally {
            long endTs = System.currentTimeMillis() - startTs;
            //记录请求响应时间
           Metrics.timer("custom_http_request_time", tags).record(endTs, TimeUnit.MILLISECONDS);
        }
        return object;
    }
}

编写好切面后,重启服务;访问controller的接口,同样可以进行自定义监控指标埋点

Spring Boot自定义监控指标的详细过程

原文地址:https://www.cnblogs.com/happyhuangjinjin/p/17177241.html

相关文章: