저는 Debian 기반 배포판(LMDE)에 Minikube를 배포하고 nfs-kernel-server.
/srv/nfs-volume1 *(rw,sync,no_subtree_check)
그런 다음 아래 지침에 따라 다음 명령을 사용하여 PertantVolume(PV) 및 PertantVolumeClaim(PVC)을 Minikube에 추가했습니다 csi-driver-nfs.
https://discuss.kubernetes.io/t/use-nfs-for-pertant-volumes/19035
user@laptop1:/srv$ ls -lrt
total 4
drwxrwxrwx 4 nobody nogroup 4096 Aug 31 20:47 nfs-volume1
user@laptop1:/srv$
SC(StorageClass) 및 PVC는 다음과 같이 구성됩니다.
apiVersion: storage.k8s.io/v1
kind: StorageClass
metadata:
name: nfs-vol1
provisioner: nfs.csi.k8s.io
parameters:
server: 192.168.0.150
share: /srv/nfs-volume1
reclaimPolicy: Delete
volumeBindingMode: Immediate
mountOptions:
- hard
- nfsvers=4.2
apiVersion: v1
kind: PersistentVolumeClaim
metadata:
name: nfs-pvc
spec:
storageClassName: nfs-vol1
accessModes: [ReadWriteOnce]
resources:
requests:
storage: 5Gi
그런 다음 Kubeflow/JupyterLab 기반 노트북용 PVC를 사용합니다(Kubeflow UI에서 가져옴).
apiVersion: kubeflow.org/v1beta1
kind: Notebook
metadata:
annotations:
notebooks.kubeflow.org/server-type: jupyter
creationTimestamp: '2023-08-31T18:50:02Z'
generation: 1
labels:
app: volume-book
managedFields:
- apiVersion: kubeflow.org/v1beta1
fieldsType: FieldsV1
fieldsV1:
f:metadata:
f:annotations:
.: {}
f:notebooks.kubeflow.org/server-type: {}
f:labels:
.: {}
f:app: {}
f:spec:
.: {}
f:template:
.: {}
f:spec:
.: {}
f:containers: {}
f:serviceAccountName: {}
f:tolerations: {}
f:volumes: {}
manager: OpenAPI-Generator
operation: Update
time: '2023-08-31T18:50:02Z'
- apiVersion: kubeflow.org/v1beta1
fieldsType: FieldsV1
fieldsV1:
f:status:
.: {}
f:conditions: {}
f:containerState:
.: {}
f:running:
.: {}
f:startedAt: {}
f:readyReplicas: {}
manager: manager
operation: Update
subresource: status
time: '2023-08-31T18:50:06Z'
name: volume-book
namespace: kubeflow-user-example-com
resourceVersion: '110625'
uid: c92ba16b-4b0b-4399-b651-77d7a2eb730f
spec:
template:
spec:
containers:
- env: []
image: kubeflownotebookswg/jupyter-pytorch-cuda-full:v1.7.0-rc.0
imagePullPolicy: IfNotPresent
name: volume-book
resources:
limits:
cpu: '0.6'
memory: 1.2Gi
nvidia.com/gpu: '1'
requests:
cpu: '0.5'
memory: 1Gi
volumeMounts:
- mountPath: /dev/shm
name: dshm
- mountPath: /home/jovyan/kubeflow-volume
name: kubeflow-volume
- mountPath: /home/jovyan
name: volume-book-volume
serviceAccountName: default-editor
tolerations: []
volumes:
- emptyDir:
medium: Memory
name: dshm
- name: kubeflow-volume
persistentVolumeClaim:
claimName: kubeflow-volume
readOnly: false
- name: volume-book-volume
persistentVolumeClaim:
claimName: volume-book-volume
status:
conditions:
- lastProbeTime: '2023-08-31T18:50:06Z'
lastTransitionTime: '2023-08-31T18:50:05Z'
status: 'True'
type: Initialized
- lastProbeTime: '2023-08-31T18:50:06Z'
lastTransitionTime: '2023-08-31T18:50:06Z'
status: 'True'
type: Ready
- lastProbeTime: '2023-08-31T18:50:06Z'
lastTransitionTime: '2023-08-31T18:50:06Z'
status: 'True'
type: ContainersReady
- lastProbeTime: '2023-08-31T18:50:06Z'
lastTransitionTime: '2023-08-31T18:50:03Z'
status: 'True'
type: PodScheduled
containerState:
running:
startedAt: '2023-08-31T18:50:05Z'
readyReplicas: 1
그러나 볼륨에 쓰려고 하면 다음 오류가 발생합니다.
[I 2023-08-31 20:44:03.073 ServerApp] Creating new file in /kubeflow-volume
[W 2023-08-31 20:44:03.074 ServerApp] 403 POST /notebook/kubeflow-user-example-com/volume-book/api/contents/kubeflow-volume?1693514643048 (127.0.0.6): Permission denied: kubeflow-volume/untitled.txt
[W 2023-08-31 20:44:03.075 ServerApp] wrote error: 'Permission denied: kubeflow-volume/untitled.txt'
파일 시스템의 볼륨은 다음과 같습니다.
user@laptop1:/srv/nfs-volume1$ ll -d pvc-4a07985a-0207-41cc-8d2f-dec8ffbad3d7
drwxrwsr-x 2 nobody nogroup 4096 Aug 31 20:47 pvc-4a07985a-0207-41cc-8d2f-dec8ffbad3d7
소유권은 nfs 내보내기 디렉터리에서 가져오지만 권한은 가져오지 않습니다.
여러 노트북/포드에서 NFS 스토리지 클래스를 사용할 수 있도록 권한을 설정하려면 어떻게 해야 합니까?