mac 上學習k8s系列(27)Deployment Job CronJob

       在學習完上一講mac 上學習k8s系列（26）ReplicationController、ReplicaSet、StatefulSet中的ReplicationSet後，學習Depolyment就非常容易了。它只是在RS中增加了如下功能：1，可以查看Deployment的升級詳細進度和狀態2，可以使用回滾操作回滾到之前的任一版本3，每一次對Deployment的操作，都能夠保存下來4，對於每一次升級都能夠隨時暫停和啟動。整體來說就是運維更方便了。Job和CronJob對應的是任務和周期性任務。還是基於前面創建的apple:5678鏡像來學習，由於該鏡像沒有裝bash，所以學習Job的時候使用的是apline：3.13

1，Deployment 

apiVersion: apps/v1kind: Deployment metadata:  name: apple-deployment spec:  replicas: 3   selector: #標籤選擇器    matchLabels:      app: apple   template:    metadata:      labels:        app: apple     spec:      containers:         - name: apple-app         image: apple:5678         ports:        - containerPort: 5678
中間遇到的問題
% kubectl apply -f deployment/Deployment.yamlerror: error validating "deployment/Deployment.yaml": error validating data: ValidationError(Deployment.spec): missing required field "selector" in io.k8s.api.apps.v1.DeploymentSpec; if you choose to ignore these errors, turn validation off with --validate=false
原因：我們創建deployment的時候必須指定selector
 % kubectl apply -f deployment/Deployment.yamldeployment.apps/apple-deployment created
查看下我們的deployment和pod
% kubectl get deployment                     NAME                       READY   UP-TO-DATE   AVAILABLE   AGEapple-deployment           3/3     3            3           87s
 % kubectl get pods                           NAME                                        READY   STATUS      RESTARTS   AGEapple-app                                   1/1     Running     18         102dapple-deployment-fb8cfb965-bnjhr            1/1     Running     0          106sapple-deployment-fb8cfb965-fftd8            1/1     Running     0          106sapple-deployment-fb8cfb965-zznjr            1/1     Running     0          106s
當然，我們也可以更新鏡像和回滾鏡像
kubectl set image deployment/apple-deployment apple-app=gorse_server:latestdeployment.apps/apple-deployment image updated
% kubectl rollout history deployment/apple-deployment deployment.apps/apple-deployment REVISION  CHANGE-CAUSE1         <none>2         <none>
更新鏡像後我們發現有兩個版本的deployment，可以回滾掉剛才的更新
%  kubectl rollout undo deployment/apple-deploymentdeployment.apps/apple-deployment rolled back
當然我們也可以縮放我們的pod數量
% kubectl scale deployment apple-deployment --replicas 1deployment.apps/apple-deployment scaled
2，Job 
apiVersion: batch/v1kind: Jobmetadata:  name: alpine-jobspec:  template:    metadata:      name: alpine-job    spec:      restartPolicy: Never      containers:      - name: counter        image: alpine:3.13        command:        - "bin/sh"        - "-c"        - "for i in 9 8 7 6 5 4 3 2 1; do echo $i; done"
這裡需要注意的是Job的RestartPolicy僅支持Never和OnFailure兩種，不支持Always。
% kubectl apply -f deployment/job.yamljob.batch/alpine-job created
3，CronJob
apiVersion: batch/v1kind: CronJobmetadata:  name: alpine-cronjobspec:  schedule: "*/1 * * * *"  jobTemplate:    spec:      template:        spec:          restartPolicy: OnFailure          containers:          - name: alpine-cronjob            image: alpine:3.13            args:            - "bin/sh"            - "-c"            - "for i in 9 8 7 6 5 4 3 2 1; do echo $i; done"
CronJob裡面多了個schedule 它的配置和crontab配置一樣。可以用來執行周期性的任務。