1. 头疼编写K8S部署文件？

  • K8S yaml 参数很多，须要边写边查？
  • 保留回滚版本数怎么设？
  • 如何探测启动胜利，如何探活？
  • 如何调配和限度资源？
  • 如何设置时区？否则打印日志是GMT规范工夫
  • 如何裸露服务供其它服务调用？
  • 如何依据CPU和内存使用率来配置程度伸缩？

首先，你须要晓得有这些知识点，其次要把这些知识点都搞明确也不容易，再次，每次编写仍然容易出错！

2. 创立服务镜像

前一篇文章 解说了如何疾速创立本人的服务镜像，不过为了演示，这篇文章咱们以 redis:6-alpine 镜像为例。

3. 残缺K8S部署文件编写过程

  • 首先装置 goctl 工具

    GO111MODULE=on GOPROXY=https://goproxy.cn/,direct go get -u github.com/tal-tech/go-zero/tools/goctl

  • 一键生成K8S部署文件

    goctl kube deploy -name redis -namespace adhoc -image redis:6-alpine -o redis.yaml -port 6379

    生成的 yaml 文件如下：

    apiVersion: apps/v1kind: Deploymentmetadata:  name: redis  namespace: adhoc  labels:    app: redisspec:  replicas: 3  revisionHistoryLimit: 5  selector:    matchLabels:      app: redis  template:    metadata:      labels:        app: redis    spec:      containers:      - name: redis        image: redis:6-alpine        lifecycle:          preStop:            exec:              command: ["sh","-c","sleep 5"]        ports:        - containerPort: 6379        readinessProbe:          tcpSocket:            port: 6379          initialDelaySeconds: 5          periodSeconds: 10        livenessProbe:          tcpSocket:            port: 6379          initialDelaySeconds: 15          periodSeconds: 20        resources:          requests:            cpu: 500m            memory: 512Mi          limits:            cpu: 1000m            memory: 1024Mi        volumeMounts:        - name: timezone          mountPath: /etc/localtime      volumes:        - name: timezone          hostPath:            path: /usr/share/zoneinfo/Asia/Shanghai---apiVersion: v1kind: Servicemetadata:  name: redis-svc  namespace: adhocspec:  ports:    - port: 6379  selector:    app: redis---apiVersion: autoscaling/v2beta1kind: HorizontalPodAutoscalermetadata:  name: redis-hpa-c  namespace: adhoc  labels:    app: redis-hpa-cspec:  scaleTargetRef:    apiVersion: apps/v1    kind: Deployment    name: redis  minReplicas: 3  maxReplicas: 10  metrics:  - type: Resource    resource:      name: cpu      targetAverageUtilization: 80---apiVersion: autoscaling/v2beta1kind: HorizontalPodAutoscalermetadata:  name: redis-hpa-m  namespace: adhoc  labels:    app: redis-hpa-mspec:  scaleTargetRef:    apiVersion: apps/v1    kind: Deployment    name: redis  minReplicas: 3  maxReplicas: 10  metrics:  - type: Resource    resource:      name: memory      targetAverageUtilization: 80

  • 部署服务，如果 adhoc namespace不存在的话，请先通过 kubectl create namespace adhoc 创立

    $ kubectl apply -f redis.yamldeployment.apps/redis createdservice/redis-svc createdhorizontalpodautoscaler.autoscaling/redis-hpa-c createdhorizontalpodautoscaler.autoscaling/redis-hpa-m created

  • 查看服务容许状态

    $ kubectl get all -n adhocNAME                         READY   STATUS    RESTARTS   AGEpod/redis-585bc66876-5ph26   1/1     Running   0          6m5spod/redis-585bc66876-bfqxz   1/1     Running   0          6m5spod/redis-585bc66876-vvfc9   1/1     Running   0          6m5sNAME                TYPE        CLUSTER-IP    EXTERNAL-IP   PORT(S)    AGEservice/redis-svc   ClusterIP   172.24.15.8   <none>        6379/TCP   6m5sNAME                    READY   UP-TO-DATE   AVAILABLE   AGEdeployment.apps/redis   3/3     3            3           6m6sNAME                               DESIRED   CURRENT   READY   AGEreplicaset.apps/redis-585bc66876   3         3         3       6m6sNAME                                              REFERENCE          TARGETS   MINPODS   MAXPODS   REPLICAS   AGEhorizontalpodautoscaler.autoscaling/redis-hpa-c   Deployment/redis   0%/80%    3         10        3          6m6shorizontalpodautoscaler.autoscaling/redis-hpa-m   Deployment/redis   0%/80%    3         10        3          6m6s

  • 测试服务

    $ kubectl run -i --tty --rm cli --image=redis:6-alpine -n adhoc -- sh/data # redis-cli -h redis-svcredis-svc:6379> set go-zero greatOKredis-svc:6379> get go-zero"great"

4. 总结

goctl 工具极大简化了 K8S yaml 文件的编写，提供了开箱即用的最佳实际，并且反对了模板自定义。

如果感觉文章有帮忙，欢送 star ????

5. 我的项目地址

https://github.com/tal-tech/go-zero

我的项目地址：
https://github.com/tal-tech/go-zero