Binding a PersistentVolumeClaim to a Pod resource

Bind a PersistentVolumeClaim to a Pod resource as a Ceph File Systems (CephFS).

Procedure

  1. Bind the Pod to the PVC file created in Creating a PersistentVolumeClaim (PVC).
    • Bind the Pod with FUSE mounting, similar to the following cephfs-pod-fuse.yaml file. 
      kind: Pod
metadata:
name: cephfs-pod-fuse
labels:
app: cephfs-tester-fuse
spec:
containers:
- name:  cephfs-container-fuse
image: busybox
command: ["/bin/sh"]
args: ["-c", "while true; do echo 'Writing to CephFS'; dd if=/dev/zero of=/mnt/testfile bs=1M count=1024; sleep 10; done"]
volumeMounts:
- name: cephfs-volume-fuse
- mountPath: /mnt/
volumes:
- name: cephfs-volume-fuse
persistentVolumeClaim:
 claimName: cephfs-pvc-fuse
restartPolicy: Always
imagePullSecrets:
- name: dockerhub-secret
    • Bind the Pod with kernel mounting.
      1. If needed, first create a docker secret. 
        kubectl create secret docker-registry dockerhub-secret
 --docker-username=USERNAME
 --docker-password=PASSWORD
 --docker-email=EMAIL_ADDRESS
        For example, 
        kubectl create secret docker-registry dockerhub-secret
  --docker-username=user01
  --docker-password=mypassword
  --docker-email=your-email@example.com
      2. Create a YAML file, similar to the following cephfs-pod-kernel.yaml file. 
        kind: Pod
metadata:
name: cephfs-pod-kernel
labels:
  app: cephfs-tester-kernel
spec:
containers:
- name:  cephfs-container-kernel
  image: busybox
  command: ["/bin/sh"]
  args: ["-c", "while true; do echo 'Writing to CephFS'; dd if=/dev/zero of=/mnt/testfile bs=1M count=1024; sleep 10; done"]
  volumeMounts:
  - name: cephfs-volume-kernel
  - mountPath: /mnt/
volumes:
- name: cephfs-volume-kernel
  persistentVolumeClaim:
    claimName: cephfs-pvc-kernel
restartPolicy: Always
imagePullSecrets:
  - name: dockerhub-secret
  2. Apply the Pod, by using the kubectl apply command 
    kubectl apply -f FILENAME.YAML
    The pod/POD_NAME created message is emitted.

What to do next

Verify that the pods are bound and data is being written to the CephFS volume.
  1. Verify that the pods are bound, by using the kubectl get pods command.
    For example, 
    $ kubectl get pods
NAME                READY   STATUS    RESTARTS   AGE
cephfs-pod-fuse     1/1     Running   0          5h29m
cephfs-pod-kernel   1/1     Running   0          4s
  2. Verify that the data is writing to the CephFS volume. 
    ceph fs subvolume info
    For example, 
    [root@host01 ceph]# ceph fs subvolume info
cephfs csi-vol-ab1237a9-1fbd-43c1-8b41-0dfd9dc8dee9 csi
{
   "atime": "2025-01-10 14:05:36",
   "bytes_pcent": "9.49",
   "bytes_quota": 10737418240,
   "bytes_used": 1019084800,
   "created_at": "2025-01-10 14:05:36",
   "ctime": "2025-01-10 14:14:54",
   "data_pool": "cephfs.cephfs.data",
   "earmark": "",
   "features": [
       "snapshot-clone",
       "snapshot-autoprotect",
       "snapshot-retention"
   ],
   "flavor": 2,
   "gid": 0,
   "mode": 16877,
   "mon_addrs": [
       "10.0.67.29:6789",
       "10.0.65.240:6789",
       "10.0.65.252:6789"
   ],
   "mtime": "2025-01-10 14:14:54",
   "path": "/volumes/csi/csi-vol-ab1237a9-1fbd-43c1-8b41-0dfd9dc8dee9/36070c5a-208e-4bc5-94f9-f24fe0afc71a",
   "pool_namespace": "",
   "state": "complete",
   "type": "subvolume",
   "uid": 0
}