K8s部署发布Golang应用程序的实现方法

2021-07-16 10:06:38 作者：luoguo

本文主要介绍了K8s部署发布Golang应用程序的实现方法，文中通过示例代码介绍的非常详细，需要的朋友们下面随着小编来一起学习学习吧

alertGo程序可以参考上篇文章，主要用于alertmanager实现钉钉报警

创建dockerfile

FROM golang:1.14-alpine
ENV GOPROXY=https://goproxy.cn
WORKDIR /build
COPY . .
EXPOSE 8088
RUN mkdir /app
RUN  go mod tidy
RUN go build -o /app/alertGo alertGo.go
WORKDIR /app
CMD ["/app/alertGo"]

打包并且推送

docker build -t 10.206.16.4/k8s-go/alert.sentsss.com:v2 .
docker push 10.206.16.4/k8s-go/alert.sentsss.com:v2

创建namespace

apiVersion: v1
kind: Namespace
metadata:
  name: k8s-go

创建deployment

apiVersion: apps/v1
kind: Deployment
metadata:
  name: alertgo
  namespace: k8s-go
spec:
  selector:
    matchLabels:
      app: alertgo
  replicas: 2
  template:
    metadata:
      labels:
        app: alertgo
    spec:
      imagePullSecrets:
      - name: registry-pull-secret
      containers:
        - name: alertgo
          image: 10.206.16.4/k8s-go/alert.sentsss.com:v2
          ports:
            - containerPort: 8088
          livenessProbe:
            httpGet:
              path: /
              port: 8088
            initialDelaySeconds: 30
            periodSeconds: 10
            successThreshold: 1
            failureThreshold: 3
            timeoutSeconds: 1
          readinessProbe:
            httpGet:
              path: /
              port: 8088
            initialDelaySeconds: 30
            periodSeconds: 10
            successThreshold: 1
            failureThreshold: 3
            timeoutSeconds: 1
          lifecycle:
            preStop:
              exec:
                command: ["/bin/bash","-c","sleep 20"]
          resources:
            limits:
              cpu: 20m
              memory: 20Mi
            requests:
       cpu: 10m
              memory: 10Mi

创建service

apiVersion: v1
kind: Service
metadata:
  name: alertgo
  namespace: k8s-go
spec:
  selector:
    app: alertgo
  ports:
    - port: 80
      targetPort: 8088

创建ingress

kind: Ingress # 对象类型
apiVersion: networking.k8s.io/v1beta1
metadata:
  name: alertgo
  namespace: k8s-go
spec:
  rules:
    - host: alertgo.xxx.com
      http:
        paths:
        - path: /
          backend:
            serviceName: alertgo 
            servicePort: 80

创建hpa

kind: HorizontalPodAutoscaler # 对象类型，简称 hpa，水平自动伸缩
apiVersion: autoscaling/v2beta2 # autoscaling/v2beta2 与 autoscaling/v1 的 API 有很大的不同，注意识别两者的差异
metadata:
  name: alertgo
  namespace: fronted
spec:
  scaleTargetRef: # 伸缩的目标对象
    apiVersion: apps/v1 # 对象版本
    kind: Deployment # 目标对象的类型
    name: alertgo # 目标对象的名称
  minReplicas: 3 # 最小副本数
  maxReplicas: 6 # 最大副本数
  metrics: # 指标
    - type: Resource # 类型：资源
      resource:
        name: memory # 内存
        target:
          type: Utilization
          averageUtilization: 70 # 1% 这个值是为了实验，具体值请参考业务方实际情况而定

    - type: Resource
      resource:
        name: cpu
        target:
          type: Utilization
          averageUtilization: 70

结果查看

[root@k8s-master-01 alertGo]# kubectl get pods,svc,ingress,hpa -n k8s-go
NAME                           READY   STATUS              RESTARTS   AGE
pod/alertgo-5bc79ccd65-8thmw   1/1     Running             0          37m
pod/alertgo-5bc79ccd65-dm8ll   1/1     Running             0          38m
pod/alertgo-5bc79ccd65-m9cd4   0/1     ContainerCreating   0          0s

NAME              TYPE        CLUSTER-IP     EXTERNAL-IP   PORT(S)   AGE
service/alertgo   ClusterIP   10.1.140.126   <none>        80/TCP    65m

NAME                         HOSTS                 ADDRESS   PORTS   AGE
ingress.extensions/alertgo   alertgo.sentsss.com             80      34m

NAME                                          REFERENCE            TARGETS            MINPODS   MAXPODS   REPLICAS   AGE
horizontalpodautoscaler.autoscaling/alertgo   Deployment/alertgo   79%/70%, 10%/70%   2         6         2          15s

到此这篇关于K8s部署发布Golang应用程序的实现方法的文章就介绍到这了,更多相关K8s部署发布Golang内容请搜索脚本之家以前的文章或继续浏览下面的相关文章希望大家以后多多支持脚本之家！