背景

多个计算集群且计算机群使用外部ceph,但是计算集群的pod需要从一个集群迁移到另一个集群

实现方式

假设是从计算集群1迁移计算集群2,计算集群1中的pv内容如下

apiVersion: v1
kind: PersistentVolume
metadata:
  annotations:
    pv.kubernetes.io/provisioned-by: kubesphere-rook-ceph.rbd.csi.ceph.com
  creationTimestamp: "2023-05-08T11:41:15Z"
  finalizers:
  - kubernetes.io/pv-protection
  name: pvc-716a6f63-2b4e-4b42-bd90-a8231deab0fd
  resourceVersion: "239077529"
  uid: 853cb97b-6d5e-4ac2-b345-15592390e1df
spec:
  accessModes:
  - ReadWriteOnce
  capacity:
    storage: 1Gi
  claimRef:
    apiVersion: v1
    kind: PersistentVolumeClaim
    name: www-nginxdemo-0
    namespace: kube-system
    resourceVersion: "239077512"
    uid: 716a6f63-2b4e-4b42-bd90-a8231deab0fd
  csi:
    controllerExpandSecretRef:
      name: ceph-wwq
      namespace: kubesphere-rook-ceph
    driver: kubesphere-rook-ceph.rbd.csi.ceph.com
    fsType: ext4
    nodeStageSecretRef:
      name: ceph-wwq
      namespace: kubesphere-rook-ceph
    volumeAttributes:
      clusterID: 1319f9dd-2326-4297-a486-e57cc8c6df1b
      imageFeatures: layering
      imageFormat: "2"
      imageName: csi-vol-fb6c7473-ed94-11ed-a41e-027bc4f73bc2
      journalPool: rbd-wwq
      pool: rbd-wwq
      storage.kubernetes.io/csiProvisionerIdentity: 1672975589281-8081-kubesphere-rook-ceph.rbd.csi.ceph.com
    volumeHandle: 0001-0024-1319f9dd-2326-4297-a486-e57cc8c6df1b-000000000000001e-fb6c7473-ed94-11ed-a41e-027bc4f73bc2
  persistentVolumeReclaimPolicy: Delete
  storageClassName: cephblk-wwq
  volumeMode: Filesystem

复制计算集群1中的pv,在计算集群2中手动创建,只需要修改storage.kubernetes.io/csiProvisionerIdentity,修改后的内容是计算集群2中的实际值(随便创建一个pv就可以得到这个值)