camel系列-Tracer

发表于 2022-04-08 更新于 2025-04-18

概念

Camel 的Tracer(跟踪器)用于在路由期间记录消息详细信息，您可以在其中看到每条消息发生时的路由路径。还会记录消息的详细信息，例如消息正文。

开启Tracer

public static void main(String[] args) throws Exception {
    Main main = new Main();
    main.configure().withRoutesIncludePattern("routes/*.xml");
    main.configure().withTracing(true);
    main.run(args);
}

运行日志

日志会打印路由相关的 Exchange 信息

15:54:08.704 [DefaultQuartzScheduler-MyXmlCamel_Worker-2] INFO  org.apache.camel.Tracing - *--> [foo         ] [from[quartz:foo?cron={{myCron}}] ] Exchange[Id: 89D3473A20DFC41-0000000000000000, BodyType: null, Body: [Body is null]]
15:54:08.704 [DefaultQuartzScheduler-MyXmlCamel_Worker-1] INFO  org.apache.camel.Tracing - *--> [foo         ] [from[quartz:foo?cron={{myCron}}] ] Exchange[Id: 89D3473A20DFC41-0000000000000001, BodyType: null, Body: [Body is null]]
15:54:08.708 [DefaultQuartzScheduler-MyXmlCamel_Worker-2] INFO  org.apache.camel.Tracing -      [foo         ] [bean[ref:myBean method:hello]    ] Exchange[Id: 89D3473A20DFC41-0000000000000000, BodyType: null, Body: [Body is null]]
15:54:08.708 [DefaultQuartzScheduler-MyXmlCamel_Worker-1] INFO  org.apache.camel.Tracing -      [foo         ] [bean[ref:myBean method:hello]    ] Exchange[Id: 89D3473A20DFC41-0000000000000001, BodyType: null, Body: [Body is null]]
15:54:08.721 [DefaultQuartzScheduler-MyXmlCamel_Worker-1] INFO  org.apache.camel.Tracing -      [foo         ] [log                              ] Exchange[Id: 89D3473A20DFC41-0000000000000001, BodyType: String, Body: Hello how are you?]
15:54:08.721 [DefaultQuartzScheduler-MyXmlCamel_Worker-2] INFO  org.apache.camel.Tracing -      [foo         ] [log                              ] Exchange[Id: 89D3473A20DFC41-0000000000000000, BodyType: String, Body: Hello how are you?]
15:54:08.721 [DefaultQuartzScheduler-MyXmlCamel_Worker-1] INFO  foo - Hello how are you?
15:54:08.721 [DefaultQuartzScheduler-MyXmlCamel_Worker-2] INFO  foo - Hello how are you?
15:54:08.724 [DefaultQuartzScheduler-MyXmlCamel_Worker-1] INFO  org.apache.camel.Tracing -      [foo         ] [bean[ref:myBean method:bye]      ] Exchange[Id: 89D3473A20DFC41-0000000000000001, BodyType: String, Body: Hello how are you?]
15:54:08.724 [DefaultQuartzScheduler-MyXmlCamel_Worker-2] INFO  org.apache.camel.Tracing -      [foo         ] [bean[ref:myBean method:bye]      ] Exchange[Id: 89D3473A20DFC41-0000000000000000, BodyType: String, Body: Hello how are you?]
15:54:08.725 [DefaultQuartzScheduler-MyXmlCamel_Worker-2] INFO  org.apache.camel.Tracing -      [foo         ] [log                              ] Exchange[Id: 89D3473A20DFC41-0000000000000000, BodyType: String, Body: Bye World]
15:54:08.725 [DefaultQuartzScheduler-MyXmlCamel_Worker-1] INFO  org.apache.camel.Tracing -      [foo         ] [log                              ] Exchange[Id: 89D3473A20DFC41-0000000000000001, BodyType: String, Body: Bye World]
15:54:08.725 [DefaultQuartzScheduler-MyXmlCamel_Worker-2] INFO  foo - Bye World
15:54:08.725 [DefaultQuartzScheduler-MyXmlCamel_Worker-1] INFO  foo - Bye World

ExchangeFormatter

我们可以通过 ExchangeFormatter 来实现自定义的 Exchange 内容格式输出，系统默认格式实现为DefaultExchangeFormatter

自定义格式示例

实现 ExchangeFormatter 接口的 format 方法

public class CustomExchangeFormatter implements ExchangeFormatter {
    @Override
    public String format(Exchange exchange) {
        return "customFormat:"+exchange.getIn().getBody();
    }
}

使用 Tracer 的 setExchangeFormatter 方法设置 ExchangeFormatter

public static void main(String[] args) throws Exception {
	// use Camels Main class
	Main main = new Main();
	main.configure().withTracing(true);
	main.addMainListener(new MainListenerSupport() {
		@Override
		public void afterConfigure(BaseMainSupport main) {
			main.getCamelContext().getTracer().
				setExchangeFormatter(new CustomExchangeFormatter());
		}
	});
	main.run(args);
}

测试运行日志

1 2	INFO org.apache.camel.Tracing - [foo ] [bean[ref:myBean method:bye] ] customFormat:Hello how are you? INFO org.apache.camel.Tracing - [foo ] [log ] customFormat:Bye World

Backlog Tracer

Backlog Tracer 将跟踪消息存储在积压队列中，需要手动调用BacklogTracer 的相关 dump 方法来消费消息

开启 Backlog Tracer

public static void main(String[] args) throws Exception {

        Main main = new Main();
        main.configure().withBacklogTracing(true);
        main.addMainListener(new MainListenerSupport() {
            @Override
            public void afterConfigure(BaseMainSupport main) {
                BacklogTracer tracer=main.getCamelContext().getExtension(BacklogTracer.class);
                tracer = BacklogTracer.createTracer(main.getCamelContext());
                main.getCamelContext().setExtension(BacklogTracer.class, tracer);
                tracer.setEnabled(true);
            }
        });
        main.run(args);
    }

运行效果

内存队列数量

调用 dumpAllTracedMessagesAsXml 后效果

<backlogTracerEventMessages>
  <backlogTracerEventMessage>
    <uid>58</uid>
    <timestamp>2022-04-08T16:39:57.704+0800</timestamp>
    <routeId>foo</routeId>
    <toNode>log1</toNode>
    <exchangeId>800F9FA329C6554-000000000000000B</exchangeId>
    <message exchangeId="800F9FA329C6554-000000000000000B">
      <headers>
        <header key="CamelMessageTimestamp" type="java.lang.Long">1649407156007</header>
        <header key="calendar"></header>
        <header key="fireTime" type="java.util.Date">Fri Apr 08 16:39:16 CST 2022</header>
        <header key="jobDetail" type="org.quartz.impl.JobDetailImpl">JobDetail 'Camel_MyXmlCamel.foo':  jobClass: 'org.apache.camel.component.quartz.CamelJob concurrentExectionDisallowed: false persistJobDataAfterExecution: false isDurable: false requestsRecovers: false</header>
        <header key="jobInstance" type="org.apache.camel.component.quartz.CamelJob">org.apache.camel.component.quartz.CamelJob@329ec7ad</header>
        <header key="jobRunTime" type="java.lang.Long">-1</header>
        <header key="mergedJobDataMap" type="org.quartz.JobDataMap">org.quartz.JobDataMap@23f33ac1</header>
        <header key="nextFireTime" type="java.util.Date">Fri Apr 08 16:39:18 CST 2022</header>
        <header key="previousFireTime" type="java.util.Date">Fri Apr 08 16:39:14 CST 2022</header>
        <header key="refireCount" type="java.lang.Integer">0</header>
        <header key="result"></header>
        <header key="scheduledFireTime" type="java.util.Date">Fri Apr 08 16:39:16 CST 2022</header>
        <header key="scheduler" type="org.quartz.impl.StdScheduler">org.quartz.impl.StdScheduler@281258a7</header>
        <header key="trigger" type="org.quartz.impl.triggers.CronTriggerImpl">Trigger 'Camel_MyXmlCamel.foo':  triggerClass: 'org.quartz.impl.triggers.CronTriggerImpl calendar: 'null' misfireInstruction: 1 nextFireTime: Fri Apr 08 16:39:18 CST 2022</header>
        <header key="triggerGroup" type="java.lang.String">Camel_MyXmlCamel</header>
        <header key="triggerName" type="java.lang.String">foo</header>
      </headers>
      <body type="java.lang.String">Hello how are you?</body>
    </message>
  </backlogTracerEventMessage>
</backlogTracerEventMessages>

和 TRACER 的区别

Backlog Tracer将消息的捕获存储在内部积压队列中。

Tracer基于事件并在消息发生时记录消息（或路由到另一个 Camel 目的地）。

当您只需要在跟踪消息发生时记录它们时，请使用Tracer 。

积压跟踪器允许您按需从积压队列中提取消息。积压跟踪器与支持 JMX 的工具一起工作得更好，因为它更简单，允许以 POJO 或 XML 格式批量转储所有跟踪的消息。

概念