2

我们对 Eclipse 微配置文件指标 @Timed 注释有问题。它似乎没有测量时间(只计算调用次数)。

设置是完整的 Payara 5。

通过http://localhost:8080/metrics/application的典型输出:

    # TYPE application:total_coffees_retrieved counter
application:total_coffees_retrieved 4
# TYPE application:total_coffees_consumed counter
application:total_coffees_consumed 4
# TYPE application:com_sebastian_daschner_hello_prometheus_coffees_retrieve_coffee_rate_per_second gauge
application:com_sebastian_daschner_hello_prometheus_coffees_retrieve_coffee_rate_per_second 1.4044150172871745
# TYPE application:com_sebastian_daschner_hello_prometheus_coffees_retrieve_coffee_one_min_rate_per_second gauge
application:com_sebastian_daschner_hello_prometheus_coffees_retrieve_coffee_one_min_rate_per_second 0.0
# TYPE application:com_sebastian_daschner_hello_prometheus_coffees_retrieve_coffee_five_min_rate_per_second gauge
application:com_sebastian_daschner_hello_prometheus_coffees_retrieve_coffee_five_min_rate_per_second 0.0
# TYPE application:com_sebastian_daschner_hello_prometheus_coffees_retrieve_coffee_fifteen_min_rate_per_second gauge
application:com_sebastian_daschner_hello_prometheus_coffees_retrieve_coffee_fifteen_min_rate_per_second 0.0
# TYPE application:com_sebastian_daschner_hello_prometheus_coffees_retrieve_coffee_mean_seconds gauge
application:com_sebastian_daschner_hello_prometheus_coffees_retrieve_coffee_mean_seconds 0.0
# TYPE application:com_sebastian_daschner_hello_prometheus_coffees_retrieve_coffee_max_seconds gauge
application:com_sebastian_daschner_hello_prometheus_coffees_retrieve_coffee_max_seconds 0.0
# TYPE application:com_sebastian_daschner_hello_prometheus_coffees_retrieve_coffee_min_seconds gauge
application:com_sebastian_daschner_hello_prometheus_coffees_retrieve_coffee_min_seconds 0.0
# TYPE application:com_sebastian_daschner_hello_prometheus_coffees_retrieve_coffee_stddev_seconds gauge
application:com_sebastian_daschner_hello_prometheus_coffees_retrieve_coffee_stddev_seconds 0.0
# TYPE application:com_sebastian_daschner_hello_prometheus_coffees_retrieve_coffee_seconds summary
application:com_sebastian_daschner_hello_prometheus_coffees_retrieve_coffee_seconds_count 4
application:com_sebastian_daschner_hello_prometheus_coffees_retrieve_coffee_seconds{quantile="0.5"} 0.0
application:com_sebastian_daschner_hello_prometheus_coffees_retrieve_coffee_seconds{quantile="0.75"} 0.0
application:com_sebastian_daschner_hello_prometheus_coffees_retrieve_coffee_seconds{quantile="0.95"} 0.0
application:com_sebastian_daschner_hello_prometheus_coffees_retrieve_coffee_seconds{quantile="0.98"} 0.0
application:com_sebastian_daschner_hello_prometheus_coffees_retrieve_coffee_seconds{quantile="0.99"} 0.0
application:com_sebastian_daschner_hello_prometheus_coffees_retrieve_coffee_seconds{quantile="0.999"} 0.0

平均值,最大值,分位数都是“0.0”,而计数效果很好(= 4)。

我们已经尝试了 Payara/JEE/Microprofile 版本的多种组合,每种都具有相同的结果。

以下是一些代码片段(基础取自https://github.com/sdaschner/hello-prometheus/tree/microprofile,仅添加了@Timed)

咖啡豆:

@ApplicationScoped
public class Coffees {


    @Counted(name = "total_coffees_retrieved", absolute = true, monotonic = true)
    @Timed   
    public String retrieveCoffee() {
        try {
            Thread.sleep((int) (1000 * Math.random()));
        } catch (InterruptedException ex) {
            Logger.getLogger(Coffees.class.getName()).log(Level.SEVERE, null, ex);
        }

        return "Coffee!";
    }

}

JAX-RS REST 资源:

import javax.inject.Inject;
import javax.ws.rs.GET;
import javax.ws.rs.Path;

@Path("coffee")
public class CoffeesResource {

    @Inject
    Coffees coffees;

    @GET
    public String getCoffee() {
        return coffees.retrieveCoffee();
    }

}

pom.xml

<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 http://maven.apache.org/xsd/maven-4.0.0.xsd">
    <modelVersion>4.0.0</modelVersion>
    <groupId>com.sebastian-daschner</groupId>
    <artifactId>hello-prometheus</artifactId>
    <version>1.0-SNAPSHOT</version>
    <packaging>war</packaging>

    <dependencies>
        <dependency>
            <groupId>javax</groupId>
            <artifactId>javaee-api</artifactId>
            <version>7.0</version>
            <scope>provided</scope>
        </dependency>
        <dependency>
            <groupId>org.eclipse.microprofile.metrics</groupId>
            <artifactId>microprofile-metrics-api</artifactId>
            <version>1.1.1</version> <!-- also tried 1.0 and 1.1 -->
            <scope>provided</scope>
        </dependency>
    </dependencies>

    <build>
        <finalName>hello-prometheus</finalName>
    </build>

    <properties>
        <maven.compiler.source>1.8</maven.compiler.source>
        <maven.compiler.target>1.8</maven.compiler.target>
        <failOnMissingWebXml>false</failOnMissingWebXml>
        <project.build.sourceEncoding>UTF-8</project.build.sourceEncoding>
    </properties>
</project>

任何想法如何解决这个问题?

编辑:使用 application/json 请求指标时不会发生

curl -H "Accept: application/json" http://localhost:8080/metrics/application

下一个 payara 版本的问题已经解决: https ://github.com/payara/Payara/issues/2970

4

1 回答 1

0

如果您想查看 0 以外的内容,您将需要一个需要更多时间执行的方法!简单地增加一个计数器并返回一个这样的字符串几乎不需要任何时间,绝对少于 0.1 秒,因此您永远不会在指标中看到任何值。

尝试添加Thread.sleep(1500)到该方法以查看计时器是否记录了 1.5 秒。

由于@Timed给了你很多统计数据,你也可以尝试Thread.sleep(Math.random() * 1000)给你一些执行时间的变化,并给你更多有趣的指标来检查。

编辑:正如下面评论中提到的,Payara Micro 5.182 中似乎存在一个错误,导致计时器不起作用。

于 2018-07-18T09:25:22.600 回答