K8s, CSI-NFS, PVC и Statefulsets: «селектор для *v1.PersistentVolumeClaim не реализован»

Question

Я использую https://github.com/kubernetes-csi/csi-driver-nfs , чтобы предоставить PVC NFS для моих наборов состояний. Но после нового автоматизированного развертывания (возможно, это происходит какое-то время, я не следил за кластером) журналы PVC сообщают следующее:

      $ kubectl get pod,pvc,storageclass

NAME                              READY   STATUS    RESTARTS   AGE
pod/kafka-0                       0/1     Pending   0          17m
pod/kafka-1                       0/1     Pending   0          17m
pod/kafka-2                       0/1     Pending   0          17m
pod/nfs-server-595df7c955-gtsv9   1/1     Running   0          22m
pod/zookeeper-0                   0/1     Pending   0          17m

NAME                                          STATUS    VOLUME   CAPACITY   ACCESS MODES   STORAGECLASS   AGE
persistentvolumeclaim/kafka-kafka-0           Pending                                      kafka          17m
persistentvolumeclaim/kafka-kafka-1           Pending                                      kafka          17m
persistentvolumeclaim/kafka-kafka-2           Pending                                      kafka          17m
persistentvolumeclaim/zookeeper-zookeeper-0   Pending                                      zookeeper      17m

NAME                                    PROVISIONER      RECLAIMPOLICY   VOLUMEBINDINGMODE   ALLOWVOLUMEEXPANSION   AGE
storageclass.storage.k8s.io/kafka       nfs.csi.k8s.io   Delete          Immediate           false                  22m
storageclass.storage.k8s.io/zookeeper   nfs.csi.k8s.io   Delete          Immediate           false                  22m

$ kubectl logs persistentvolumeclaim/zookeeper-zookeeper-0

error: cannot get the logs from *v1.PersistentVolumeClaim: selector for *v1.PersistentVolumeClaim not implemented

$ kubectl describe persistentvolumeclaim/zookeeper-zookeeper-0
Name:          zookeeper-zookeeper-0
Namespace:     meia
StorageClass:  
Status:        Pending
Volume:        
Labels:        app.kubernetes.io/component=zookeeper
               app.kubernetes.io/instance=kafka
               app.kubernetes.io/name=zookeeper
Annotations:   <none>
Finalizers:    [kubernetes.io/pvc-protection]
Capacity:      
Access Modes:  
VolumeMode:    Filesystem
Used By:       zookeeper-0
Events:
  Type    Reason         Age               From                         Message
  ----    ------         ----              ----                         -------
  Normal  FailedBinding  4s (x4 over 41s)  persistentvolume-controller  no persistent volumes available for this claim and no storage class is set

Любопытно, что эта же ошибка возникает и на локальном кластере Kind. Это происходит с несколькими версиями образов zookeeper/kafka и версиями 3.1 и 4.0 csi-nfs, это происходит либо в моем локальном кластере Kind, либо в моем чистом 5-узловом кластере kubernetes.

Это мой класс хранения:

      $ cat storageclass-zookeeper.yaml 
kind: StorageClass
apiVersion: storage.k8s.io/v1
metadata:
  name: zookeeper
  labels:
    name: zookeeper
provisioner: nfs.csi.k8s.io
parameters:
  server: nfs-server.coucou.svc.cluster.local
  share: /coucou-nfs/data/zookeeper
reclaimPolicy: Delete
volumeBindingMode: Immediate
mountOptions:
  - hard
  - nfsvers=4.1

И мой набор состояний:

      $ cat statefulset-zookeeper.yaml 
apiVersion: apps/v1
kind: StatefulSet
metadata:
  name: zookeeper
...
          volumeMounts:
            - name: scripts
              mountPath: /scripts/setup.sh
              subPath: setup.sh
            - name: zookeeper
              mountPath: /bitnami/zookeeper
      volumes:
        - name: scripts
          configMap:
            name: zookeeper-scripts
            defaultMode: 0755
  volumeClaimTemplates:
    - metadata:
        name: zookeeper
        annotations:
      spec:
        accessModes: [ "ReadWriteOnce" ]
        storageClassName: "zookeeper"
        resources:
          requests:
            storage: 8Gi

Я не могу найти конкретную ошибку в Google. Итак, в чем проблема? Что-то с CSI-NFS? (проверено несколько версий). Проблемы с CRI/Containerd?

На моем локальном типе:

      WARNING: This version information is deprecated and will be replaced with the output from kubectl version --short.  Use --output=yaml|json to get the full version.
Client Version: version.Info{Major:"1", Minor:"24", GitVersion:"v1.24.0", GitCommit:"4ce5a8954017644c5420bae81d72b09b735c21f0", GitTreeState:"clean", BuildDate:"2022-05-03T13:46:05Z", GoVersion:"go1.18.1", Compiler:"gc", Platform:"linux/amd64"}
Kustomize Version: v4.5.4
Server Version: version.Info{Major:"1", Minor:"23", GitVersion:"v1.23.4", GitCommit:"e6c093d87ea4cbb530a7b2ae91e54c0842d8308a", GitTreeState:"clean", BuildDate:"2022-03-06T21:32:53Z", GoVersion:"go1.17.7", Compiler:"gc", Platform:"linux/amd64"}

В моем кластере k8s:

      WARNING: This version information is deprecated and will be replaced with the output from kubectl version --short.  Use --output=yaml|json to get the full version.
Client Version: version.Info{Major:"1", Minor:"24", GitVersion:"v1.24.0", GitCommit:"4ce5a8954017644c5420bae81d72b09b735c21f0", GitTreeState:"clean", BuildDate:"2022-05-03T13:46:05Z", GoVersion:"go1.18.1", Compiler:"gc", Platform:"linux/amd64"}
Kustomize Version: v4.5.4
Server Version: version.Info{Major:"1", Minor:"24", GitVersion:"v1.24.0", GitCommit:"4ce5a8954017644c5420bae81d72b09b735c21f0", GitTreeState:"clean", BuildDate:"2022-05-03T13:38:19Z", GoVersion:"go1.18.1", Compiler:"gc", Platform:"linux/amd64"}