kubernetes grafana集成prometheus监控金丝雀部署

综合

发布日期: 2019-06-15

文章字数: 952

阅读时长: 4 分

阅读次数:

搭建springboot项目

demo地址：https://github.com/ciweigg2/test-walle.git

springboot2.x集成prometheus

        <dependency>
            <groupId>org.springframework.boot</groupId>
            <artifactId>spring-boot-starter-actuator</artifactId>
        </dependency>

        <dependency>
            <groupId>io.micrometer</groupId>
            <artifactId>micrometer-registry-prometheus</artifactId>
            <version>1.1.4</version>
        </dependency>

        <dependency>
            <groupId>org.springframework.boot</groupId>
            <artifactId>spring-boot-starter-aop</artifactId>
        </dependency>

application.properties

spring.application.name=springboot_prometheus
management.endpoints.web.exposure.include=*
management.metrics.tags.application=${spring.application.name}

核心aop监控请求代码也可以认为是埋点呀

/**
 * Created by Administrator on 2019/6/15.
 *
 * 集成prometheus监控接口流量 实现金丝雀部署
 */
@Component
@Aspect
public class PrometheusAop {

    @Pointcut("execution(public * com.example.testwalle.controller.*.*(..))")
    public void pointCut(){}

    //统计请求的处理时间
    ThreadLocal<Long> startTime = new ThreadLocal<>();

    @Autowired
    MeterRegistry registry;

    private Counter counter;

    @PostConstruct
    private void init(){
        counter = registry.counter("requests_total","status","success");
    }

    @Before("pointCut()")
    public void doBefore(JoinPoint joinPoint) throws Throwable{
        startTime.set(System.currentTimeMillis());

        counter.increment(); //记录系统总请求数
    }

    @AfterReturning(returning = "returnVal" , pointcut = "pointCut()")
    public void doAfterReturning(Object returnVal){
        //处理完请求后
        System.out.println("方法执行时间:"+ (System.currentTimeMillis() - startTime.get()));
    }

}

下载demo访问 http://ip:8080/walle

kubernetes搭建grafana prometheus

参考我的博客：kubernetes上用helm部署loki prometheus grafana

部署金丝雀

首先部署一个1.11.1版本的程序

prometheus.io/scrape: “true” 如果需要prometheus监控必须开启

prometheus.io/path: /actuator/prometheus springboot程序开启prometheus的地址

prometheus.io/port: “8080” 程序的端口

vi k8s-demo.yaml

apiVersion: v1
kind: Service
metadata:
  name: k8s-service
  namespace: default
  labels:
    app: k8s-service
spec:
  type: NodePort
  ports:
  - port: 8080
    nodePort: 30002
    targetPort: 8080
  selector:
    app: k8s-service

---

apiVersion: apps/v1
kind: Deployment
metadata:
  name: k8s-service
  labels:
    app: k8s-service
spec:
  replicas: 4
  selector:
    matchLabels:
      app: k8s-service
      release: 1.11.1
  template:
    metadata:
      labels:
        app: k8s-service
        release: 1.11.1
      annotations:
        prometheus.io/scrape: "true"
        prometheus.io/path: /actuator/prometheus
        prometheus.io/port: "8080"
    spec:
      containers:
      - name: k8s-service
        image: demo:1.0
        imagePullPolicy: IfNotPresent
        ports:
        - containerPort: 8080
        livenessProbe:
          tcpSocket:
            port: 8080
          periodSeconds: 15
          initialDelaySeconds: 30

kubectl create -f k8s-demo.yaml

再部署一个1.11.2版本的程序

不需要Service了一个Service可以对应多个Deployment

需要修改的地方

image: demo:1.0 改成 demo:2.0 (新版本，我测试就没打2个镜像呀)
release: 1.11.1 改成 release: 1.11.2
添加env(可以省略的)
name: k8s-service-v1.11.1 改成 name: k8s-service-v1.11.2

vi k8s-demo2.yaml

apiVersion: apps/v1
kind: Deployment
metadata:
  name: k8s-service-v1.11.2
  labels:
    app: k8s-service
spec:
  replicas: 1
  selector:
    matchLabels:
      app: k8s-service
      release: 1.11.2
  template:
    metadata:
      labels:
        app: k8s-service
        release: 1.11.2
      annotations:
        prometheus.io/scrape: "true"
        prometheus.io/path: /actuator/prometheus
        prometheus.io/port: "8080"
    spec:
      containers:
      - name: k8s-service
        image: demo:1.0
        imagePullPolicy: IfNotPresent
        ports:
        - containerPort: 8080
        env:
        - name: VERSION
          value: v1.11.2
        livenessProbe:
          tcpSocket:
            port: 8080
          periodSeconds: 15
          initialDelaySeconds: 30