springboot2.x暴露健康状况通过prometheus监控
加入依赖
org.springframework.boot spring-boot-starter-actuator io.micrometer micrometer-registry-prometheus
application.yml加入相关配置
management: security: enabled: false #prometheus+grafana+springboot2监控集成配置 metrics: export: prometheus: enabled: true jmx: enabled: true endpoints: web: exposure: include: '*' base-path: /metrics #prometheus+grafana+springboot2监控集成配置
主启动类加入配置
package com.drore.saas;
import com.drore.saas.services.service.StorageService;
import io.micrometer.core.instrument.MeterRegistry;
import org.springframework.boot.CommandLineRunner;
import org.springframework.boot.SpringApplication;
import org.springframework.boot.actuate.autoconfigure.metrics.MeterRegistryCustomizer;
import org.springframework.boot.autoconfigure.SpringBootApplication;
import org.springframework.boot.web.servlet.ServletComponentScan;
import org.springframework.context.annotation.Bean;
@SpringBootApplication
@ServletComponentScan
public class TenantappApplication {
public static void main(String[] args) {
SpringApplication.run(TenantappApplication.class, args);
}
@Bean
CommandLineRunner init(StorageService storageService) {
return (args) -> {
storageService.init();
};
}
#prometheus+grafana+springboot2监控集成配置
@Bean
MeterRegistryCustomizer meterRegistryCustomizer(MeterRegistry meterRegistry) {
return meterRegistry1 -> {
meterRegistry.config()
.commonTags("application", "Tenantapp");
};
}
#prometheus+grafana+springboot2监控集成配置
}
启动之后通过路径访问查看健康状况
http://xxxxxx/metrics/prometheus
查看到的数据如下：
[root@saas98 /]$ curl "http://10.98.94.220:80/ts/metrics/prometheus"
# HELP process_start_time_seconds Start time of the process since unix epoch.
# TYPE process_start_time_seconds gauge
process_start_time_seconds{application="Tenantapp",} 1.556068841226E9
# HELP tomcat_threads_busy_threads
# TYPE tomcat_threads_busy_threads gauge
tomcat_threads_busy_threads{application="Tenantapp",name="http-nio-9081",} 1.0
# HELP tomcat_sessions_expired_sessions_total
# TYPE tomcat_sessions_expired_sessions_total counter
tomcat_sessions_expired_sessions_total{application="Tenantapp",} 0.0
# HELP tomcat_sessions_active_current_sessions
# TYPE tomcat_sessions_active_current_sessions gauge
tomcat_sessions_active_current_sessions{application="Tenantapp",} 0.0
# 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 counter
jvm_gc_memory_promoted_bytes_total{application="Tenantapp",} 1.18894656E8
# HELP tomcat_global_request_max_seconds
# TYPE tomcat_global_request_max_seconds gauge
tomcat_global_request_max_seconds{application="Tenantapp",name="http-nio-9081",} 3.366
# HELP jvm_memory_used_bytes The amount of used memory
# TYPE jvm_memory_used_bytes gauge
jvm_memory_used_bytes{application="Tenantapp",area="heap",id="Survivor Space",} 653880.0
jvm_memory_used_bytes{application="Tenantapp",area="nonheap",id="Metaspace",} 1.36445248E8
jvm_memory_used_bytes{application="Tenantapp",area="heap",id="Eden Space",} 7511504.0
jvm_memory_used_bytes{application="Tenantapp",area="nonheap",id="Code Cache",} 3.8031424E7
jvm_memory_used_bytes{application="Tenantapp",area="heap",id="Tenured Gen",} 1.3880212E8
jvm_memory_used_bytes{application="Tenantapp",area="nonheap",id="Compressed Class Space",} 1.7220968E7
# HELP tomcat_sessions_created_sessions_total
# TYPE tomcat_sessions_created_sessions_total counter
tomcat_sessions_created_sessions_total{application="Tenantapp",} 0.0
# HELP system_cpu_count The number of processors available to the Java virtual machine
# TYPE system_cpu_count gauge
system_cpu_count{application="Tenantapp",} 1.0
# HELP tomcat_global_sent_bytes_total
# TYPE tomcat_global_sent_bytes_total counter
tomcat_global_sent_bytes_total{application="Tenantapp",name="http-nio-9081",} 8168269.0
# HELP jvm_threads_daemon_threads The current number of live daemon threads
# TYPE jvm_threads_daemon_threads gauge
jvm_threads_daemon_threads{application="Tenantapp",} 34.0
# HELP logback_events_total Number of error level events that made it to the logs
# TYPE logback_events_total counter
logback_events_total{application="Tenantapp",level="debug",} 0.0
logback_events_total{application="Tenantapp",level="error",} 965.0
logback_events_total{application="Tenantapp",level="warn",} 4.0
logback_events_total{application="Tenantapp",level="info",} 1047.0
logback_events_total{application="Tenantapp",level="trace",} 0.0
# HELP tomcat_cache_access_total
# TYPE tomcat_cache_access_total counter
tomcat_cache_access_total{application="Tenantapp",} 0.0
# HELP tomcat_servlet_error_total
# TYPE tomcat_servlet_error_total counter
tomcat_servlet_error_total{application="Tenantapp",name="dispatcherServlet",} 0.0
tomcat_servlet_error_total{application="Tenantapp",name="statViewServlet",} 0.0
tomcat_servlet_error_total{application="Tenantapp",name="default",} 0.0
# HELP jvm_buffer_total_capacity_bytes An estimate of the total capacity of the buffers in this pool
prometheus-operator监控java应用整合
目前的prometheus-operator目录下的相关文件
https://github.com/coreos/prometheus-operator.git
已经移植到
https://github.com/coreos/kube-prometheus.git
具体原因参照官网。
下面也有自己总结的文档的地址，大家可以参照下：
https://github.com/hkj123/kube-prometheus-manifests.git
具体的文件如下图展示，这里有很多文件，我们只介绍关注的几个文件，其他的可以参照官网了解：
构建springboot项目的时候需要注意的点：
--- kind: Service apiVersion: v1 metadata: labels: app: ms #prometheus配置监控需要通过app去监控，这个k8s-app和app都可以 namespace: drore-saas name: ms spec: ports: - name: http-ms #prometheus配置监控需要通过port name去识别，当然也可以直接配置端口，不过这样看起来更专业 nodePort: 31945 protocol: TCP port: 80 targetPort: 80 selector: app: ms type: NodePort ---
部署prometheus需要注意的点
https://github.com/coreos/kube-prometheus/manifests
提供了一套整合的例子：
需要修改的文件: prometheus-clusterRole.yaml
默认的文件内容：
apiVersion: rbac.authorization.k8s.io/v1 kind: ClusterRole metadata: name: prometheus-k8s rules: - apiGroups: - "" resources: - nodes/metrics verbs: - get - nonResourceURLs: - /metrics verbs: - get
可以发现这里的监控的resources：- nodes/metrics
我们需要监控自己构建的java应用需要扩大权限
apiVersion: rbac.authorization.k8s.io/v1 kind: ClusterRole metadata: name: prometheus-k8s rules: - apiGroups: - "" resources: - nodes - services - endpoints - pods - nodes/proxy verbs: - get - list - watch - apiGroups: - "" resources: - configmaps - nodes/metrics verbs: - get - nonResourceURLs: - /metrics verbs: - get
书写monitor监控文件，提供了一下参考的文件：prometheus-serviceMonitor.yaml
自己编写的 ：prometheus-serviceMonitorSelfServiceMs.yaml
apiVersion: monitoring.coreos.com/v1 kind: ServiceMonitor metadata: name: drore-saas-ms namespace: monitoring spec: endpoints: - interval: 15s port: http-ms #prometheus配置注意这个port的名称 path: /ms/metrics/prometheus #prometheus配置注意这个路径 jobLabel: app #prometheus配置注意我们用的是app在springboot构建的时候 namespaceSelector: # 表示去匹配某一命名空间中的service，如果想从所有的namespace中匹配用any: true matchNames: - drore-saas selector: matchLabels: app: ms #prometheus配置注意我们用的是app在springboot构建的时候
配置完毕之后使用
kubectl apply -f prometheus-clusterRole.yaml -f prometheus-serviceMonitorSelfServiceMs.yaml
就可以登录管理界面去看了
http://118.31.17.205:31144/targets
报警规则可以通过 prometheus-rules.yaml 去配置
参考
https://github.com/coreos/prometheus-operator
补充：spring boot2.x暴露监控endpoint并配置prometheus及grafana对多个targets进行监控
1. build.gradle中添加依赖
compile group: 'org.springframework.boot', name: 'spring-boot-starter-actuator'
compile group: 'org.springframework.boot', name: 'spring-boot-starter-security'
compile('io.micrometer:micrometer-registry-prometheus')
2. spring boot 2.x暴露信息
不同于之前的Actuator 1.x，Actuator 2.x 的大多数端点默认被禁掉。 Actuator 2.x 中的默认端点增加了/actuator前缀。
3. spring boot服务端改造
配置信息
server.port=8085 management.endpoints.web.exposure.include=prometheus management.endpoint.metrics.enabled=true spring.security.user.name=xxx spring.security.user.password=123456 spring.security.user.roles=ACTUATOR_ADMIN management.server.port=8090 management.endpoints.web.base-path=/admin
指定权限
@Configuration
public class ActuatorSecurityConfig extends WebSecurityConfigurerAdapter {
@Override
protected void configure(HttpSecurity http) throws Exception {
http.requestMatcher(EndpointRequest.toAnyEndpoint()).authorizeRequests().antMatchers("/actuator/*").hasRole(
"ACTUATOR_ADMIN").and().httpBasic();
}
}
这样我们可以看到：
这个spring boot服务启动了两个端口：
8085端口，进行正常的业务处理
8090端口，admin，暴露出监控信息，并通过spring security进行权限控制
打开浏览器访问http://192.168.211.2:8090/admin/prometheus
输入用户名 密码
4. Prometheus监控多个spring boot服务
下载prometheus-2.11.1.linux-amd64.tar.gz
编写配置文件prometheus.yml
global: scrape_interval: 10s scrape_timeout: 10s evaluation_interval: 10m scrape_configs: - job_name: app-gateway scrape_interval: 5s scrape_timeout: 5s metrics_path: /admin/prometheus scheme: http basic_auth: username: root password: 1230456 static_configs: - targets: - 192.168.211.2:8099 - job_name: search scrape_interval: 5s scrape_timeout: 5s metrics_path: /admin/prometheus scheme: http basic_auth: username: root password: 1230456 static_configs: - targets: - 192.168.211.2:8090
其中，我的spring boot服务部署在192.168.211.2上，而prometheus和grafana都部署在192.168.211.101上
配置了两个服务
执行以下命令启动 ./prometheus --config.file=prometheus.yml &
访问http://192.168.211.101:9090/targets
可以看到
5. grafana可视化监控多个prometheus targets
下载grafana-6.2.5.linux-amd64.tar.gz
启动grafana
./grafana-server &
http://192.168.211.101:3000
在dashboard中配置时，要采用这样的方式
jvm_gc_pause_seconds_count{action="end of minor GC",cause="Metadata GC Threshold",instance="192.168.211.2:8090",job="search"}
通过instance和job来进行区分prometheus的targets
可以看到dashboard中有两个prometheus的targets
以上为个人经验，希望能给大家一个参考，也希望大家多多支持脚本之家。如有错误或未考虑完全的地方，望不吝赐教。