OpenShift 之 Quarkus（3）用独立的Prometheus监控Quarkus应用

dawnsky.liu

已于 2023-08-09 08:55:51 修改

阅读量1k

点赞数

分类专栏： OpenShift 4 prometheus Quarkus 文章标签： prometheus grafana quarkus

于 2020-03-11 15:52:09 首次发布

本文链接：https://blog.csdn.net/weixin_43902588/article/details/104785542

版权

OpenShift 4 同时被 3 个专栏收录

271 篇文章 79 订阅

订阅专栏

Quarkus

7 篇文章 0 订阅

订阅专栏

prometheus

5 篇文章 1 订阅

订阅专栏

《OpenShift 4.x HOL教程汇总》

请先完成《 Quarkus（1）创建第一个Quarkus应用》

创建Quarkus测试应用

在第一个窗口创建带有“metrics”扩展的Quarkus测试应用，并运行。注意：为了能监控应用级别metrics，我们使用了Quarkus的“metrics”扩展。

$ mvn io.quarkus:quarkus-maven-plugin:$QUARKUS_VERSION.Final:create \
    -DprojectGroupId=com.example \
    -DprojectArtifactId=hello \
    -Dextensions="resteasy-reactive,micrometer-registry-prometheus"
 
$ cd hello 
$ mvn quarkus:dev

在第二个窗口执行以下命令测试应用，并查看应用级别的缺省metrics。

$ curl http://localhost:8080/hello
Hello from RESTEasy Reactive
$ curl http://localhost:8080/q/metrics
。。。
# TYPE jvm_memory_usage_after_gc_percent gauge
# HELP jvm_memory_usage_after_gc_percent The percentage of long-lived heap pool used after the last GC event, in the range [0..1]
jvm_memory_usage_after_gc_percent{area="heap",pool="long-lived"} 0.006293699835116974
# TYPE process_cpu_usage gauge
# HELP process_cpu_usage The \"recent cpu usage\" for the Java Virtual Machine process
process_cpu_usage 3.184822048068064E-4
# EOF

在第二个窗口编辑“helllo/src/main/java/org/acme/quickstart/GreetingResource.java”文件，增加标有“Add line”的代码行。

package org.acme.quickstart;
 
import javax.ws.rs.GET;
import javax.ws.rs.Path;
import javax.ws.rs.Produces;
import javax.ws.rs.core.MediaType;
import org.eclipse.microprofile.metrics.annotation.Counted;  //Add line.
 
@Path("/hello")
public class GreetingResource {
 
    @GET
    @Counted(name = "greetings", description = "How many greetings we've given.")    //Add line.
    @Produces(MediaType.TEXT_PLAIN)
    public String hello() {
        return "hello";
    }
}

在第二个窗口执行命令，确认已经有应用级别metrics：“application_org_acme_quickstart_GreetingResource_greetings_total”。

$ curl http://localhost:8080/q/metrics/application
# HELP application_org_acme_quickstart_GreetingResource_greetings_total How many greetings we've given.
# TYPE application_org_acme_quickstart_GreetingResource_greetings_total counter
application_org_acme_quickstart_GreetingResource_greetings_total 0.0

在第二个窗口创建“getting-started/src/main/java/org/acme/quickstart/PrimeNumberChecker.java”文件，内容如下。其中定义了3个定制应用级metrics：“performedChecks”、“checksTimer”、“highestPrimeNumberSoFar”。

package org.acme.quickstart;
 
import org.eclipse.microprofile.metrics.MetricUnits;
import org.eclipse.microprofile.metrics.annotation.Counted;
import org.eclipse.microprofile.metrics.annotation.Gauge;
import org.eclipse.microprofile.metrics.annotation.Timed;
 
import javax.ws.rs.GET;
import javax.ws.rs.Path;
import javax.ws.rs.PathParam;
import javax.ws.rs.Produces;
 
@Path("/prime")
public class PrimeNumberChecker {
 
    private long highestPrimeNumberSoFar = 2;
 
    @GET
    @Path("/{number}")
    @Produces("text/plain")
    @Counted(name = "performedChecks", description = "How many primality checks have been performed.")
    @Timed(name = "checksTimer", description = "A measure of how long it takes to perform the primality test.", unit = MetricUnits.MILLISECONDS)
    public String checkIfPrime(@PathParam("number") long number) {
        if (number < 1) {
            return "Only natural numbers can be prime numbers.";
        }
        if (number == 1) {
            return "1 is not prime.";
        }
        if (number == 2) {
            return "2 is prime.";
        }
        if (number % 2 == 0) {
            return number + " is not prime, it is divisible by 2.";
        }
        for (int i = 3; i < Math.floor(Math.sqrt(number)) + 1; i = i + 2) {
            if (number % i == 0) {
                return number + " is not prime, is divisible by " + i + ".";
            }
        }
        if (number > highestPrimeNumberSoFar) {
            highestPrimeNumberSoFar = number;
        }
        return number + " is prime.";
    }
 
    @Gauge(name = "highestPrimeNumberSoFar", unit = MetricUnits.NONE, description = "Highest prime number so far.")
    public Long highestPrimeNumberSoFar() {
        return highestPrimeNumberSoFar;
    }
}

在第二个窗口执行命令，测试本地Quarkus应用。

$ curl http://localhost:8080/prime/5
5 is prime.
$ curl http://localhost:8080/prime/350
351 is not prime, is divisible by 3.
$ curl http://localhost:8080/prime/359
359 is prime.

在第二个窗口执行命令查看包含“performedChecks”、“highestPrimeNumberSoFar”、“checksTimer”应用metrics监控值。

$ curl http://localhost:8080/metrics/application | grep performedChecks
# HELP application_org_acme_quickstart_PrimeNumberChecker_performedChecks_total How many primality checks have been performed.
# TYPE application_org_acme_quickstart_PrimeNumberChecker_performedChecks_total counter
application_org_acme_quickstart_PrimeNumberChecker_performedChecks_total 3.0 
 
$ curl http://localhost:8080/metrics/application | grep highestPrimeNumberSoFar
# HELP application_org_acme_quickstart_PrimeNumberChecker_highestPrimeNumberSoFar Highest prime number so far.
# TYPE application_org_acme_quickstart_PrimeNumberChecker_highestPrimeNumberSoFar gauge
application_org_acme_quickstart_PrimeNumberChecker_highestPrimeNumberSoFar 359.0 
 
$ curl http://localhost:8080/metrics/application | grep checksTimer
# TYPE application_org_acme_quickstart_PrimeNumberChecker_checksTimer_rate_per_second gauge
application_org_acme_quickstart_PrimeNumberChecker_checksTimer_rate_per_second 0.009651120871431985
# TYPE application_org_acme_quickstart_PrimeNumberChecker_checksTimer_one_min_rate_per_second gauge
application_org_acme_quickstart_PrimeNumberChecker_checksTimer_one_min_rate_per_second 0.0015203946095816126
# TYPE application_org_acme_quickstart_PrimeNumberChecker_checksTimer_five_min_rate_per_second gauge
application_org_acme_quickstart_PrimeNumberChecker_checksTimer_five_min_rate_per_second 0.07491811314937041
# TYPE application_org_acme_quickstart_PrimeNumberChecker_checksTimer_fifteen_min_rate_per_second gauge
application_org_acme_quickstart_PrimeNumberChecker_checksTimer_fifteen_min_rate_per_second 0.14412698663875181
# TYPE application_org_acme_quickstart_PrimeNumberChecker_checksTimer_min_seconds gauge
application_org_acme_quickstart_PrimeNumberChecker_checksTimer_min_seconds 1.07824E-4
# TYPE application_org_acme_quickstart_PrimeNumberChecker_checksTimer_max_seconds gauge
application_org_acme_quickstart_PrimeNumberChecker_checksTimer_max_seconds 8.55019E-4
# TYPE application_org_acme_quickstart_PrimeNumberChecker_checksTimer_mean_seconds gauge
application_org_acme_quickstart_PrimeNumberChecker_checksTimer_mean_seconds 4.0862461940924336E-4
# TYPE application_org_acme_quickstart_PrimeNumberChecker_checksTimer_stddev_seconds gauge
application_org_acme_quickstart_PrimeNumberChecker_checksTimer_stddev_seconds 3.0210634938071025E-4
# HELP application_org_acme_quickstart_PrimeNumberChecker_checksTimer_seconds A measure of how long it takes to perform the primality test.
# TYPE application_org_acme_quickstart_PrimeNumberChecker_checksTimer_seconds summary
application_org_acme_quickstart_PrimeNumberChecker_checksTimer_seconds_count 3.0
application_org_acme_quickstart_PrimeNumberChecker_checksTimer_seconds{quantile="0.5"} 4.36168E-4
application_org_acme_quickstart_PrimeNumberChecker_checksTimer_seconds{quantile="0.75"} 8.55019E-4
application_org_acme_quickstart_PrimeNumberChecker_checksTimer_seconds{quantile="0.95"} 8.55019E-4
application_org_acme_quickstart_PrimeNumberChecker_checksTimer_seconds{quantile="0.98"} 8.55019E-4
application_org_acme_quickstart_PrimeNumberChecker_checksTimer_seconds{quantile="0.99"} 8.55019E-4
application_org_acme_quickstart_PrimeNumberChecker_checksTimer_seconds{quantile="0.999"} 8.55019E-4

中断第一个窗口运行的应用，然后执行命令打包应用。

$ mvn clean package -DskipTests

以binary的方式向OpenShift部署Quarkus应用。

$ oc new-build registry.access.redhat.com/redhat-openjdk-18/openjdk18-openshift:1.5 --binary --name=primes -l app=primes
$ rm -rf target/binary && mkdir -p target/binary && cp -r target/*-runner.jar target/lib target/binary
$ oc new-app primes
$ oc expose service primes
$ oc rollout status -w dc/primes

完成部署后测试OpenShift上运行的Quarkus应用。

$ curl http://$(oc get route primes |awk 'NR==2 {print $2}')/prime/5
5 is prime.
$ curl http://$(oc get route primes |awk 'NR==2 {print $2}')/prime/351
351 is not prime, is divisible by 3.
$ curl http://$(oc get route primes |awk 'NR==2 {print $2}')/prime/359
359 is prime.

安装项目级Prometheus运行环境

执行命令，在quarkus-prometheus项目中部署Prometheus运行环境。

$ oc new-project quarkus-prometheus --display-name="Prometheus Monitoring Quarkus App"

创建以下内容的prometheus.yml文件，它是Prometheus的运行配置文件，定义了两个监控job，其中“localhost:9090”是Prometheus自身metrics地址，“primes:8080”是我们要运行的Quarkus应用的metrics地址。

global:
  scrape_interval:     15s
  evaluation_interval: 15s
alerting:
  alertmanagers:
  - static_configs:
    - targets:
scrape_configs:
  - job_name: 'prometheus'
    static_configs:
    - targets: ['localhost:9090']
  - job_name: 'hello-app'
    static_configs:
    - targets: ['primes:8080']

根据prometheus.yml创建Prometheus的运行配置ConfigMap对象。

$ oc create configmap prom --from-file=prometheus.yml=prometheus.yml
configmap/prom created

创建并部署Prometheus及其相关资源，包括Service、PVC等。

$ oc new-app prom/prometheus
$ oc expose svc/prometheus
$ oc set volume dc/prometheus --add -t configmap --configmap-name=prom -m /etc/prometheus/prometheus.yml --sub-path=prometheus.yml
info: Generated volume name: volume-2vbzq
deploymentconfig.apps.openshift.io/prometheus volume updated
 
$ oc rollout status -w dc/prometheus
Waiting for rollout to finish: 0 out of 1 new replicas have been updated...
Waiting for rollout to finish: 0 out of 1 new replicas have been updated...
Waiting for rollout to finish: 0 of 1 updated replicas are available...
Waiting for latest deployment config spec to be observed by the controller loop...
replication controller "prometheus-2" successfully rolled out

用Prometheus监控Quarkus应用

持续访问Quarkus应用，产生监控数据。

PRIME_URL=http://$(oc get route primes |awk 'NR==2 {print $2}')
while [ true ] ; do
        BITS=$(( ( RANDOM % 60 ) + 1 ))
        NUM=$(openssl prime -generate -bits $BITS)
        curl ${PRIME_URL}/prime/${NUM}  && echo
        sleep 2
done

通过名为prometheus的Route访问控制台。将“application_org_acme_quickstart_PrimeNumberChecker_performedChecks_total”填入查询框，然后点击Execute，此时会显示监控到的指标值。

$ oc get route prometheus

在这里插入图片描述
3. 点击Graph，然后将时间调整到5m（分钟）。此时将会以图形化方式显示监控结果。

4. 将监控指标改为“base_memory_usedHeap_bytes”，然后再Execute，可以查看监控到的内存指标情况。

用Grafana监控Quarkus应用

执行命令部署Grafana。

$ oc new-app grafana/grafana && oc expose svc/grafana
$ oc rollout status -w dc/grafana

通过名为grafana的Route访问控制台，用admin/admin登录，然后跳过修改密码进入Grafana控制台。

$ oc get route grafana

在这里插入图片描述
3. 在控制台初始化页面中点击“Add data source”。

4. 设置HTTP的URL为本地Prometheus的metrics地址“http://prometheus:9090”，然后点击“Save & Test”。

5. 在Grafana控制台中点击“+”，然后进入Dashboard菜单。在New dashborad界面中进入Add Query。
在这里插入图片描述
6. 在设置Query步骤中将Metrics设为“application_org_acme_quickstart_PrimeNumberChecker_performedChecks_total”。

7. 在设置General步骤中将Title设为“Prime Checks”。

8. 此时界面已经显示监控的指标图。点击右上方的Save Dashboard图标，并指定名称为“Quarkus Primes”，然后Save。
在这里插入图片描述
9. 按照以上的操作方式再新建以下的Dashboard。
指标名1：application_org_acme_quickstart_PrimeNumberChecker_checksTimer_seconds
保存名1：Primes Performance
指标名2：application_org_acme_quickstart_PrimeNumberChecker_highestPrimeNumberSoFar
图样式2：Singlestat
保存名2：Highest So Far
指标名3：base_memory_usedHeap_bytes
图样式3：Gauge
显示单位：Field Units to bytes
保存名3：Memory
10. 最后可以在Grafana的Dashboard显示如下图的监控控制台。
在这里插入图片描述