一、普通部署

1.1 Deployment

---
apiVersion: apps/v1
kind: Deployment
metadata:
  namespace: a-env
  name: docker-hub
spec:
  selector:
    matchLabels:
      app: docker-hub
  replicas: 1
  template:
    metadata:
      labels:
        app: docker-hub
    spec:
      # 限定为master节点
      # nodeName: master
      containers:
        - name: docker-hub
          image: registry
          imagePullPolicy: IfNotPresent
          ports:
            - name: hubssl
              containerPort: 5000
              protocol: TCP
          env:
            - name: REGISTRY_HTTP_ADDR
              value: "0.0.0.0:5000"
            - name: REGISTRY_HTTP_TLS_CERTIFICATE
              value: "/certs/hub.crt"
            - name: REGISTRY_HTTP_TLS_KEY
              value: "/certs/hub.key"
          volumeMounts:
            - name: data
              mountPath: /var/lib/registry
            - name: certs
              mountPath: /certs
      restartPolicy: Always
      volumes:
        - name: data
          hostPath: 
            path: /a_soft/docker-hub/data
            type: DirectoryOrCreate
        - name: conf
          hostPath: 
            path: /a_soft/docker-hub/conf
            type: DirectoryOrCreate
        - name: certs
          hostPath: 
            path: /a_soft/ca/docker-hub
            type: DirectoryOrCreate
 

1.2 Service

---
apiVersion: v1
kind: Service
metadata:
  namespace: a-env
  name: docker-hub
spec:
  selector:
    app: docker-hub
  ports:
    - name: hubssl
      targetPort: hubssl
      protocol: TCP
      port: 5000
      nodePort: 5000
  type: NodePort

 

 

二、定时任务

2.1 CronJob

---
apiVersion: batch/v1beta1
kind: CronJob
metadata:
  namespace: a-job
  name: t3
spec: 
  schedule: "*/1 7 * * *"          # 分 时 日 月 星期 
  concurrencyPolicy: Replace       # 并发策略：（1）Allow（默认，并发）。（2）Forbid（保留旧的，跳过新的）。（3）Replace（跳过旧的，保留新的）。-- 附：错过的job会在启动时跟新的一起开
  startingDeadlineSeconds: 7200    # 超时时间，默认永久
  successfulJobsHistoryLimit: 1    # 历史成功任务显示个数
  failedJobsHistoryLimit: 1        # 历史失败任务显示个数
  
  jobTemplate: 
    spec:
      backoffLimit: 5    # 重试最大次数
      template: 
        # metadata:   # 可以不写
          # labels:
            # app: test 
        spec:
          restartPolicy: Never  # 重启策略：（1）Never（2）OnFailure
          containers:
            - name: test 
              image: busybox
              imagePullPolicy: IfNotPresent
              command:
                - /bin/sh
                - -c
                - date; echo Hello from the Kubernetes cluster

 

 

三、外部存储

3.1 PersistentVolume

---
apiVersion: v1
kind: PersistentVolume
metadata:
  name: a-soft-harbor-data
  labels:
    name: a-soft-harbor-data
spec:
  # storageClassName: 
  accessModes:
    - ReadWriteOnce
    - ReadWriteMany
  capacity:
    storage: 20Gi
  persistentVolumeReclaimPolicy: Retain   #其他：Recycle、Delete
  nfs:
    server: 192.168.15.135
    path: /a_soft/harbor/data

3.2 PersistentVolumeClaim

---
apiVersion: v1
kind: PersistentVolumeClaim
metadata:
  namespace: a-env
  name: a-soft-harbor-data
spec:
  # storageClassName: 
  accessModes:
    - ReadWriteMany
  resources:
    requests:
      storage: 20Gi
  selector:
    matchLabels:
      name: a-soft-harbor-data

 

 

 

目录：https://blog.csdn.net/u013595395/article/details/114527658