Spring Boot 2.x监控数据可视化(Actuator + Prometheus + Grafana手把手)

TIPS

本文基于Spring Boot 2.1.4，理论支持Spring Boot 2.x所有版本

众所周知，Spring Boot有个子项目Spring Boot Actuator，它为应用提供了强大的监控能力。从Spring Boot 2.0开始，Actuator将底层改为Micrometer，提供了更强、更灵活的监控能力。Micrometer是一个监控门面，可以类比成监控界的 Slf4j 。

借助Micrometer，应用能够对接各种监控系统，例如：

• AppOptics
• Atlas
• Datadog
• Dynatrace
• Elastic
• Ganglia
• Graphite
• Humio
• Influx
• JMX
• KairosDB
• New Relic
• Prometheus
• SignalFx
• Simple (in-memory)
• StatsD
• Wavefront

下面演示如何对接 Prometheus ，并使用 Grafana 实现数据的可视化。

TIPS

童鞋们对Prometheus或Grafana不熟悉也没关系，本文是手把手文章，按步骤操作即可。

编码

编写代码

• 加依赖

    org.springframework.boot  spring-boot-starter-actuator  io.micrometer  micrometer-registry-prometheus

  这里，我们为应用引入了 micrometer-registry-prometheus ，事实上，你想对接上文列表中的哪款监控系统，就写啥。例如想对接 Influx ，则需添加依赖 micrometer-registry-influx 。

• 写配置

  server:  port: 8080spring:  application:    name: prometheus-testmanagement:  endpoints:    web:      exposure:        include: 'prometheus'  metrics:    tags:      application: ${spring.application.name}

  如配置所示，指定应用名为 prometheus-test ，并将 Actuator 的 /actuator/prometheus 端点暴露出来； management.metrics.tags.application=prometheus-test 作用是为指标设置一个名为application="prometheus-test" 的Tag，Tag是Prometheus提供的一种能力，从而实现更加灵活的筛选。

测试

• 启动应用

• 访问 http://localhost:8080/actuator/prometheus 可获得类似如下的结果：

  # HELP jvm_memory_used_bytes The amount of used memory# TYPE jvm_memory_used_bytes gaugejvm_memory_used_bytes{application="prometheus-test",area="heap",id="PS Old Gen",} 2.1193976E7jvm_memory_used_bytes{application="prometheus-test",area="nonheap",id="Metaspace",} 3.8791688E7jvm_memory_used_bytes{application="prometheus-test",area="heap",id="PS Survivor Space",} 0.0jvm_memory_used_bytes{application="prometheus-test",area="nonheap",id="Compressed Class Space",} 5303976.0jvm_memory_used_bytes{application="prometheus-test",area="heap",id="PS Eden Space",} 8.2574816E7jvm_memory_used_bytes{application="prometheus-test",area="nonheap",id="Code Cache",} 8693824.0# HELP tomcat_global_received_bytes_total  # TYPE tomcat_global_received_bytes_total countertomcat_global_received_bytes_total{application="prometheus-test",name="http-nio-8080",} 0.0# HELP jvm_threads_daemon_threads The current number of live daemon threads# TYPE jvm_threads_daemon_threads gaugejvm_threads_daemon_threads{application="prometheus-test",} 20.0# HELP tomcat_sessions_alive_max_seconds  # TYPE tomcat_sessions_alive_max_seconds gaugetomcat_sessions_alive_max_seconds{application="prometheus-test",} 0.0# HELP jvm_buffer_memory_used_bytes An estimate of the memory that the Java virtual machine is using for this buffer pool# TYPE jvm_buffer_memory_used_bytes gaugejvm_buffer_memory_used_bytes{application="prometheus-test",id="mapped",} 0.0jvm_buffer_memory_used_bytes{application="prometheus-test",id="direct",} 90112.0# HELP jvm_threads_states_threads The current number of threads having NEW state# TYPE jvm_threads_states_threads gaugejvm_threads_states_threads{application="prometheus-test",state="runnable",} 9.0jvm_threads_states_threads{application="prometheus-test",state="new",} 0.0jvm_threads_states_threads{application="prometheus-test",state="terminated",} 0.0jvm_threads_states_threads{application="prometheus-test",state="blocked",} 0.0jvm_threads_states_threads{application="prometheus-test",state="waiting",} 12.0jvm_threads_states_threads{application="prometheus-test",state="timed-waiting",} 3.0# HELP process_cpu_usage The "recent cpu usage" for the Java Virtual Machine process# TYPE process_cpu_usage gaugeprocess_cpu_usage{application="prometheus-test",} 0.0030590633504868434# HELP logback_events_total Number of error level events that made it to the logs# TYPE logback_events_total counterlogback_events_total{application="prometheus-test",level="info",} 7.0logback_events_total{application="prometheus-test",level="warn",} 0.0logback_events_total{application="prometheus-test",level="trace",} 0.0logback_events_total{application="prometheus-test",level="debug",} 0.0logback_events_total{application="prometheus-test",level="error",} 0.0# HELP tomcat_global_sent_bytes_total  # TYPE tomcat_global_sent_bytes_total countertomcat_global_sent_bytes_total{application="prometheus-test",name="http-nio-8080",} 195356.0# HELP process_files_max_files The maximum file descriptor count# TYPE process_files_max_files gaugeprocess_files_max_files{application="prometheus-test",} 10240.0# HELP tomcat_threads_busy_threads  # TYPE tomcat_threads_busy_threads gaugetomcat_threads_busy_threads{application="prometheus-test",name="http-nio-8080",} 1.0# HELP process_files_open_files The open file descriptor count# TYPE process_files_open_files gaugeprocess_files_open_files{application="prometheus-test",} 101.0# HELP tomcat_sessions_active_current_sessions  # TYPE tomcat_sessions_active_current_sessions gaugetomcat_sessions_active_current_sessions{application="prometheus-test",} 0.0# HELP jvm_classes_unloaded_classes_total The total number of classes unloaded since the Java virtual machine has started execution# TYPE jvm_classes_unloaded_classes_total counterjvm_classes_unloaded_classes_total{application="prometheus-test",} 2.0# HELP jvm_memory_committed_bytes The amount of memory in bytes that is committed for the Java virtual machine to use# TYPE jvm_memory_committed_bytes gaugejvm_memory_committed_bytes{application="prometheus-test",area="heap",id="PS Old Gen",} 1.5466496E8jvm_memory_committed_bytes{application="prometheus-test",area="nonheap",id="Metaspace",} 4.1418752E7jvm_memory_committed_bytes{application="prometheus-test",area="heap",id="PS Survivor Space",} 1.6252928E7jvm_memory_committed_bytes{application="prometheus-test",area="nonheap",id="Compressed Class Space",} 5767168.0jvm_memory_committed_bytes{application="prometheus-test",area="heap",id="PS Eden Space",} 1.73539328E8jvm_memory_committed_bytes{application="prometheus-test",area="nonheap",id="Code Cache",} 8716288.0# HELP http_server_requests_seconds  # TYPE http_server_requests_seconds summaryhttp_server_requests_seconds_count{application="prometheus-test",exception="None",method="GET",outcome="SUCCESS",status="200",uri="/actuator/prometheus",} 16.0http_server_requests_seconds_sum{application="prometheus-test",exception="None",method="GET",outcome="SUCCESS",status="200",uri="/actuator/prometheus",} 0.326299973http_server_requests_seconds_count{application="prometheus-test",exception="None",method="GET",outcome="SUCCESS",status="200",uri="/**/favicon.ico",} 3.0http_server_requests_seconds_sum{application="prometheus-test",exception="None",method="GET",outcome="SUCCESS",status="200",uri="/**/favicon.ico",} 0.028434427# HELP http_server_requests_seconds_max  # TYPE http_server_requests_seconds_max gaugehttp_server_requests_seconds_max{application="prometheus-test",exception="None",method="GET",outcome="SUCCESS",status="200",uri="/actuator/prometheus",} 0.120627904http_server_requests_seconds_max{application="prometheus-test",exception="None",method="GET",outcome="SUCCESS",status="200",uri="/**/favicon.ico",} 0.015596512# HELP tomcat_global_error_total  # TYPE tomcat_global_error_total countertomcat_global_error_total{application="prometheus-test",name="http-nio-8080",} 0.0# HELP jvm_gc_max_data_size_bytes Max size of old generation memory pool# TYPE jvm_gc_max_data_size_bytes gaugejvm_gc_max_data_size_bytes{application="prometheus-test",} 2.863661056E9# HELP jvm_gc_live_data_size_bytes Size of old generation memory pool after a full GC# TYPE jvm_gc_live_data_size_bytes gaugejvm_gc_live_data_size_bytes{application="prometheus-test",} 2.1193976E7# HELP system_load_average_1m The sum of the number of runnable entities queued to available processors and the number of runnable entities running on the available processors averaged over a period of time# TYPE system_load_average_1m gaugesystem_load_average_1m{application="prometheus-test",} 3.9423828125# HELP process_uptime_seconds The uptime of the Java virtual machine# TYPE process_uptime_seconds gaugeprocess_uptime_seconds{application="prometheus-test",} 173.424# HELP tomcat_sessions_expired_sessions_total  # TYPE tomcat_sessions_expired_sessions_total countertomcat_sessions_expired_sessions_total{application="prometheus-test",} 0.0# HELP jvm_gc_pause_seconds Time spent in GC pause# TYPE jvm_gc_pause_seconds summaryjvm_gc_pause_seconds_count{action="end of minor GC",application="prometheus-test",cause="Metadata GC Threshold",} 1.0jvm_gc_pause_seconds_sum{action="end of minor GC",application="prometheus-test",cause="Metadata GC Threshold",} 0.012jvm_gc_pause_seconds_count{action="end of major GC",application="prometheus-test",cause="Metadata GC Threshold",} 1.0jvm_gc_pause_seconds_sum{action="end of major GC",application="prometheus-test",cause="Metadata GC Threshold",} 0.045# HELP jvm_gc_pause_seconds_max Time spent in GC pause# TYPE jvm_gc_pause_seconds_max gaugejvm_gc_pause_seconds_max{action="end of minor GC",application="prometheus-test",cause="Metadata GC Threshold",} 0.012jvm_gc_pause_seconds_max{action="end of major GC",application="prometheus-test",cause="Metadata GC Threshold",} 0.045# HELP jvm_gc_memory_promoted_bytes_total Count of positive increases in the size of the old generation memory pool before GC to after GC# TYPE jvm_gc_memory_promoted_bytes_total counterjvm_gc_memory_promoted_bytes_total{application="prometheus-test",} 1.06878E7# HELP jvm_gc_memory_allocated_bytes_total Incremented for an increase in the size of the young generation memory pool after one GC to before the next# TYPE jvm_gc_memory_allocated_bytes_total counterjvm_gc_memory_allocated_bytes_total{application="prometheus-test",} 7.8713648E7# HELP tomcat_global_request_seconds  # TYPE tomcat_global_request_seconds summarytomcat_global_request_seconds_count{application="prometheus-test",name="http-nio-8080",} 19.0tomcat_global_request_seconds_sum{application="prometheus-test",name="http-nio-8080",} 0.432# HELP jvm_threads_live_threads The current number of live threads including both daemon and non-daemon threads# TYPE jvm_threads_live_threads gaugejvm_threads_live_threads{application="prometheus-test",} 24.0# HELP jvm_buffer_count_buffers An estimate of the number of buffers in the pool# TYPE jvm_buffer_count_buffers gaugejvm_buffer_count_buffers{application="prometheus-test",id="mapped",} 0.0jvm_buffer_count_buffers{application="prometheus-test",id="direct",} 11.0# HELP process_start_time_seconds Start time of the process since unix epoch.# TYPE process_start_time_seconds gaugeprocess_start_time_seconds{application="prometheus-test",} 1.556017153576E9# HELP tomcat_threads_config_max_threads  # TYPE tomcat_threads_config_max_threads gaugetomcat_threads_config_max_threads{application="prometheus-test",name="http-nio-8080",} 200.0# HELP system_cpu_count The number of processors available to the Java virtual machine# TYPE system_cpu_count gaugesystem_cpu_count{application="prometheus-test",} 8.0# HELP tomcat_threads_current_threads  # TYPE tomcat_threads_current_threads gaugetomcat_threads_current_threads{application="prometheus-test",name="http-nio-8080",} 10.0# HELP jvm_memory_max_bytes The maximum amount of memory in bytes that can be used for memory management# TYPE jvm_memory_max_bytes gaugejvm_memory_max_bytes{application="prometheus-test",area="heap",id="PS Old Gen",} 2.863661056E9jvm_memory_max_bytes{application="prometheus-test",area="nonheap",id="Metaspace",} -1.0jvm_memory_max_bytes{application="prometheus-test",area="heap",id="PS Survivor Space",} 1.6252928E7jvm_memory_max_bytes{application="prometheus-test",area="nonheap",id="Compressed Class Space",} 1.073741824E9jvm_memory_max_bytes{application="prometheus-test",area="heap",id="PS Eden Space",} 1.395654656E9jvm_memory_max_bytes{application="prometheus-test",area="nonheap",id="Code Cache",} 2.5165824E8# HELP jvm_threads_peak_threads The peak live thread count since the Java virtual machine started or peak was reset# TYPE jvm_threads_peak_threads gaugejvm_threads_peak_threads{application="prometheus-test",} 29.0# HELP jvm_classes_loaded_classes The number of classes that are currently loaded in the Java virtual machine# TYPE jvm_classes_loaded_classes gaugejvm_classes_loaded_classes{application="prometheus-test",} 7936.0# HELP tomcat_sessions_active_max_sessions  # TYPE tomcat_sessions_active_max_sessions gaugetomcat_sessions_active_max_sessions{application="prometheus-test",} 0.0# HELP tomcat_global_request_max_seconds  # TYPE tomcat_global_request_max_seconds gaugetomcat_global_request_max_seconds{application="prometheus-test",name="http-nio-8080",} 0.161# HELP tomcat_sessions_created_sessions_total  # TYPE tomcat_sessions_created_sessions_total countertomcat_sessions_created_sessions_total{application="prometheus-test",} 0.0# HELP tomcat_sessions_rejected_sessions_total  # TYPE tomcat_sessions_rejected_sessions_total countertomcat_sessions_rejected_sessions_total{application="prometheus-test",} 0.0# HELP jvm_buffer_total_capacity_bytes An estimate of the total capacity of the buffers in this pool# TYPE jvm_buffer_total_capacity_bytes gaugejvm_buffer_total_capacity_bytes{application="prometheus-test",id="mapped",} 0.0jvm_buffer_total_capacity_bytes{application="prometheus-test",id="direct",} 90112.0# HELP system_cpu_usage The "recent cpu usage" for the whole system# TYPE system_cpu_usage gaugesystem_cpu_usage{application="prometheus-test",} 0.2523364485981308

  该端点返回的数据结构是Prometheus需要使用的。对照注释，相信大家能看懂每一项的含义。例如：

  # HELP jvm_memory_used_bytes The amount of used memory# TYPE jvm_memory_used_bytes gaugejvm_memory_used_bytes{application="prometheus-test",area="heap",id="PS Old Gen",} 2.1193976E7

  表示：prometheus-test 应用堆内存中的 PS Old Gen 区域占用的空间是 2.1193976E7 字节。

对接Prometheus

Prometheus是一款开源的监控 + 时序数据库 + 报警软件，由SoundCloud公司开发的。目前在CNCF基金会托管，并已成功孵化。

官方网站

• Prometheus官方网站
• Prometheus GitHub

安装Prometheus

TIPS

简单期间，本文使用Docker方式安装，其他方式安装请详见 Prometheus安装说明 。

• 准备配置文件，命名为prometheus.yml

  scrape_configs:# 任意写，建议英文，不要包含特殊字符- job_name: 'spring'  # 多久采集一次数据  scrape_interval: 15s  # 采集时的超时时间  scrape_timeout: 10s  # 采集的路径是啥  metrics_path: '/actuator/prometheus'  # 采集服务的地址，设置成上面Spring Boot应用所在服务器的具体地址。  static_configs:  - targets: ['30.1.229.89:8080']

  这样配置后，Prometheus服务器就会自动每隔15秒请求 http://30.1.229.89:8080/actuator/prometheus 。更多配置项详见 Prometheus Configuration官方文档

• 在prometheus.yml 文件所在目录执行如下命令，即可启动Prometheus

  docker run -d -p 9090:9090 \    -v $(pwd)/prometheus.yml:/etc/prometheus/prometheus.yml \    prom/prometheus --config.file=/etc/prometheus/prometheus.yml

测试

• 访问http://localhost:9090 ，可看到类似如下的界面

  

• 点击 Insert metric at cursor ，即可选择监控指标；点击 Graph ，即可让指标以图表方式展示；点击Execute 按钮，即可看到类似下图的结果：

Grafana可视化

至此，已经用Prometheus实现了监控数据的可视化，然而使用体验并不好。下面来用Grafana实现更友好、更贴近生产的监控可视化。

Grafana是一个开源的跨平台度量分析和可视化 + 告警工具。

官方网站

• Grafana官方网站
• Grafana GitHub

安装Grafana

TIPS

简单期间，本文使用Docker方式安装，其他方式安装请详见 Grafana安装说明 。

• 启动

  docker run -d -p 3000:3000 grafana/grafana

配置

• 登录：访问 http://localhost:3000/login ，初始账号/密码为：admin/admin ，登录后，即可看到类似如下的界面

  

• 点击Add data source ，即可看到类似如下的界面：

  

• 点击Prometheus，即可看到类似如下界面：

  

• 填写Prometheus的首页地址，并点击 Save & Test

创建监控Dashboard

• 点击导航栏上的 + 按钮，并点击Dashboard，将会看到类似如下的界面：

• 点击 Add Query ，即可看到类似如下的界面：

  

  在白框标记的位置输入指标，指标的取值详见Spring Boot应用的 /actuator/prometheus 端点，例如jvm_memory_used_bytes 、jvm_threads_states_threads 、jvm_threads_live_threads 等，Grafana会给你较好的提示，并且支持较为复杂的计算，例如聚合、求和、平均等。如果想要绘制多个线条，可点击Add Query 按钮，如上图所示，笔者为图表绘制了两条线，分别代表daemon以及peak线程。

• 点击下图的按钮，并填入Title，即可设置图表标题。

  

• 按下图步骤操作，即可为Dashboard添加新的图表。

  

Dashboard市场

至此，已可用Grafana实现较为丰富的图表展示——可将关心的监控指标放置到Dashboard上，并且非常灵活！然而，这个配置的操作虽然不难，但还是挺费时间的……

那么是否有配置好的又强大、又通用、拿来即用的Dashboard呢？答案是肯定的！前往 Grafana Lab - Dashboards ，输入关键词即可搜索指定Dashboard。

如上图所示，可以找到若干款以 Prometheus 作为数据源，支持Micrometer的Dashboard！下面，来演示如何使用 JVM(Micrometer) 这个Dashboard。

其他Dashboard使用方式大同小异，建议使用前先点击了解下详情，一般Dashboard中都会有说明。

• 点击 JVM(Micrometer) 进入Dashboard详情介绍页，如下图所示：

  

  如图已详细描述了该Dashboard的特性、配置。其中的management.metrics.tags.application ，前面已经配置过了；本例也不打算演示 Process memory 特性，所以 micrometer-jvm-extras 怎么用也不打算引入了(这是该Dashboard作者为Micrometer提供的增强包，有兴趣的童鞋可自行点击过去研究)。

  细心的童鞋会发现，该页的右上角有个 4701 这是一个非常重要的数字！

• 访问http://localhost:3000 ，按下图步骤操作：

  

• 输入后即可看到类似如下的界面，选择数据源，并点击Import。

  

• 此时，即可看到类似如下的界面，如图所示，我们常关心的指标该Dashboard均已支持！

  

比较好用的Dashboard

• JVM (Micrometer)
• JVM (Actuator)
• Spring Boot Statistics ，注：该Dashboard可能有Bug，笔者未能直接测试成功，做了一些改动后才能正常显示。

以上三款是笔者亲测较为全面、好用的Dashboard，仅供参考。

告警

Grafana支持的告警渠道非常丰富，例如邮件、钉钉、Slack、Webhook等，非常实用！比较简单，各位看官自行研究下吧。

配套代码

• GitHub
• Gitee

其中 prometheus+grafana 目录已提供Docker一键部署Prometheus + Grafana的Shell。