Spring Boot自定义监控指标的详细过程
作者:BUG弄潮儿
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,根据实际场景自由调整即可。
3. 测试
定义两个controller分别使用NativeMetricsMontior
和PrometheusMetricsMonitor
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
改变amount多次方式http://127.0.0.1:8080/order?amount=100
和http://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
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
到此这篇关于Spring Boot如何自定义监控指标的文章就介绍到这了,更多相关Spring Boot自定义监控指标内容请搜索脚本之家以前的文章或继续浏览下面的相关文章希望大家以后多多支持脚本之家!