fix: tmp path mount permission error on charts

[andyzhangx]

What type of PR is this?
/kind bug

What this PR does / why we need it:
fix: tmp path mount permission error

Which issue(s) this PR fixes:

Fixes #

Special notes for your reviewer:

Does this PR introduce a user-facing change?:

fix: tmp path mount permission error

[k8s-ci-robot]

[APPROVALNOTIFIER] This PR is APPROVED

This pull-request has been approved by: andyzhangx

The full list of commands accepted by this bot can be found here.

The pull request process is described here

Needs approval from an approver in each of these files:

• OWNERS [andyzhangx]

Approvers can indicate their approval by writing /approve in a comment
Approvers can cancel approval by writing /approve cancel in a comment

[coveralls]

Pull Request Test Coverage Report for Build 10060391750

Details

• 0 of 0 changed or added relevant lines in 0 files are covered.
• No unchanged relevant lines lost coverage.
• Overall coverage remained the same at 78.793%

---

Totals
Change from base Build 10059037242:	0.0%
Covered Lines:	888
Relevant Lines:	1127

---

💛 - Coveralls

[rhessus]

Hi,

Since this issue was delivered (since v4.8.0 of the chart), I had to apply a Kustomize patch to restore the tmp-dir volume, which was removed and caused the following error:

failed to provision volume with StorageClass "nfs-csi-xxx-sc": rpc error: code = Internal desc = failed to mount nfs server: rpc error: code = Internal desc = mkdir /tmp/pvc-6a531286-d87b-4e11-95cd-39cfa7a9d6c0: read-only file system

Here is the patch Kustomize i applied to work around the issue:

apiVersion: apps/v1
kind: Deployment
metadata:
  name: csi-nfs-controller
  namespace: kube-system
spec:
  template:
    spec:
      containers:
        - name: nfs
          volumeMounts:
          - mountPath: /tmp
            name: tmp-dir
      volumes:
        - emptyDir: {}
          name: tmp-dir

Thanks to this patch, I was finally able to upgrade the chart from version 4.7.0 to 4.11.0.

For context, I am running this within a MicroK8s cluster on three Raspberry Pi 5 nodes.

Could you confirm if the removal of tmp-dir was intentional? If so, what would be the recommended way to handle this situation?

[andyzhangx]

Hi,

Since this issue was delivered (since v4.8.0 of the chart), I had to apply a Kustomize patch to restore the tmp-dir volume, which was removed and caused the following error:

failed to provision volume with StorageClass "nfs-csi-xxx-sc": rpc error: code = Internal desc = failed to mount nfs server: rpc error: code = Internal desc = mkdir /tmp/pvc-6a531286-d87b-4e11-95cd-39cfa7a9d6c0: read-only file system

Here is the patch Kustomize i applied to work around the issue:

apiVersion: apps/v1
kind: Deployment
metadata:
  name: csi-nfs-controller
  namespace: kube-system
spec:
  template:
    spec:
      containers:
        - name: nfs
          volumeMounts:
          - mountPath: /tmp
            name: tmp-dir
      volumes:
        - emptyDir: {}
          name: tmp-dir

Thanks to this patch, I was finally able to upgrade the chart from version 4.7.0 to 4.11.0.

For context, I am running this within a MicroK8s cluster on three Raspberry Pi 5 nodes.

Could you confirm if the removal of tmp-dir was intentional? If so, what would be the recommended way to handle this situation?

@rhessus yes, removal of tmp-dir is intentional, default workingMountDir is /tmp on the node now, you could customize this setting to use other place.

csi-driver-nfs/charts/latest/csi-driver-nfs/values.yaml

Line 64 in 3e18952

workingMountDir: /tmp