SecurityContext problem about cloud-director-named-disk-csi-driver HOT 7 CLOSED

Hello @autokilla47 @arunmk

After reconfiguring the csidriver with

fsGroupPolicy: ReadWriteOnceWithFSType

VMware support provides the following fix

kubectl apply -f https://gist.githubusercontent.com/arunmk/f0c06b261e1181baf2d17ade1bbcfa4f/raw/051297d21e32eae4748a129d56e183105caa4d84/gistfile1.txt

from cloud-director-named-disk-csi-driver.

Yes This solves the problem but it creates another one. when two pods need to access the same PVC and they are not scheduled on the same node, one of the nodes fails to mount the volume

Warning FailedMount 13s kubelet MountVolume.MountDevice failed for volume "pvc-db604a6a-5241-494c-a934-94f7d9077010" : rpc error: code = Internal desc = unable to format and mount device [/dev/sdc] at path [/var/lib/kubelet/plugins/[kubernetes.io/csi/pv/pvc-db604a6a-5241-494c-a934-94f7d9077010/globalmount](https://fra01.safelinks.protection.outlook.com/?url=http%3A%2F%2Fkubernetes.io%2Fcsi%2Fpv%2Fpvc-db604a6a-5241-494c-a934-94f7d9077010%2Fglobalmount&data=05%7C01%7C%7C9f8b8e1f8bf044fb64db08da7b6d9ab3%7C06ba9ba160b04c909fe49f3621194d60%7C0%7C0%7C637958007964405507%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000%7C%7C%7C&sdata=nYpjlXFeXg6YFH3dDjfpcJHXmFHQxtDgBN99UMP1%2Fes%3D&reserved=0)] with fs [[rw]] and flags [mount failed: exit status 32 mounting arguments: -t ext4 -o rw,defaults /dev/sdc /var/lib/kubelet/plugins/[kubernetes.io/csi/pv/pvc-db604a6a-5241-494c-a934-94f7d9077010/globalmount](https://fra01.safelinks.protection.outlook.com/?url=http%3A%2F%2Fkubernetes.io%2Fcsi%2Fpv%2Fpvc-db604a6a-5241-494c-a934-94f7d9077010%2Fglobalmount&data=05%7C01%7C%7C9f8b8e1f8bf044fb64db08da7b6d9ab3%7C06ba9ba160b04c909fe49f3621194d60%7C0%7C0%7C637958007964405507%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000%7C%7C%7C&sdata=nYpjlXFeXg6YFH3dDjfpcJHXmFHQxtDgBN99UMP1%2Fes%3D&reserved=0) output: mount: /var/lib/kubelet/plugins/[kubernetes.io/csi/pv/pvc-db604a6a-5241-494c-a934-94f7d9077010/globalmount](https://fra01.safelinks.protection.outlook.com/?url=http%3A%2F%2Fkubernetes.io%2Fcsi%2Fpv%2Fpvc-db604a6a-5241-494c-a934-94f7d9077010%2Fglobalmount&data=05%7C01%7C%7C9f8b8e1f8bf044fb64db08da7b6d9ab3%7C06ba9ba160b04c909fe49f3621194d60%7C0%7C0%7C637958007964405507%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000%7C%7C%7C&sdata=nYpjlXFeXg6YFH3dDjfpcJHXmFHQxtDgBN99UMP1%2Fes%3D&reserved=0): mount(2) system call failed: Structure needs cleaning. ]: [%!!(MISSING)!(MISSING)v(MISSING)]

from cloud-director-named-disk-csi-driver.

The example of the DigitalOcean worked is below.

from cloud-director-named-disk-csi-driver.

I am solved this problem. Added CSIDriver spec fsGroupPolicy : File

apiVersion: storage.k8s.io/v1
kind: CSIDriver
metadata:
  name: named-disk.csi.cloud-director.vmware.com
spec:
  attachRequired: true
  podInfoOnMount: false
  fsGroupPolicy: File

from cloud-director-named-disk-csi-driver.

@Lorislombardi resolved?

from cloud-director-named-disk-csi-driver.

@Lorislombardi have you used ReadWriteMany as the access mode? We create a shared disk and allow mounts to multiple VMs only in the ReadWriteMany case.

from cloud-director-named-disk-csi-driver.

@erdembas would dc8f906 have solved your issue?

from cloud-director-named-disk-csi-driver.