[cinder-csi-plugin] Resize raw block device

[sergelogvinov]

/kind bug

What happened:

During the volume expansion process, there is a risk of accidentally resizing an unmanaged (raw) block device if it already has a file system on it. Since raw block devices are not tied to any file system by Kubernetes, operations like resizing can potentially corrupt the existing file system, as it is unaware of the underlying changes.

cloud-provider-openstack/pkg/csi/cinder/nodeserver.go

Lines 365 to 402 in 04bd8e1

	klog.V(4).Infof("NodeExpandVolume: called with args %+v", protosanitizer.StripSecrets(*req))
	volumeID := req.GetVolumeId()
	if len(volumeID) == 0 {
	return nil, status.Error(codes.InvalidArgument, "Volume ID not provided")
	}
	volumePath := req.GetVolumePath()
	if len(volumePath) == 0 {
	return nil, status.Error(codes.InvalidArgument, "Volume path not provided")
	}
	_, err := blockdevice.IsBlockDevice(volumePath)
	if err != nil {
	return nil, status.Errorf(codes.NotFound, "Failed to determine device path for volumePath %s: %v", volumePath, err)
	}
	output, err := ns.Mount.GetMountFs(volumePath)
	if err != nil {
	return nil, status.Errorf(codes.Internal, "Failed to find mount file system %s: %v", volumePath, err)
	}
	devicePath := strings.TrimSpace(string(output))
	if devicePath == "" {
	return nil, status.Error(codes.Internal, "Unable to find Device path for volume")
	}
	if ns.Opts.RescanOnResize {
	// comparing current volume size with the expected one
	newSize := req.GetCapacityRange().GetRequiredBytes()
	if err := blockdevice.RescanBlockDeviceGeometry(devicePath, volumePath, newSize); err != nil {
	return nil, status.Errorf(codes.Internal, "Could not verify %q volume size: %v", volumeID, err)
	}
	}
	r := mountutil.NewResizeFs(ns.Mount.Mounter().Exec)
	if _, err := r.Resize(devicePath, volumePath); err != nil {
	return nil, status.Errorf(codes.Internal, "Could not resize volume %q: %v", volumeID, err)
	}

What you expected to happen:

How to reproduce it:

Anything else we need to know?:

Environment:

• csi plugin version: master branch
• OpenStack version:
• Others:

[k8s-triage-robot]

The Kubernetes project currently lacks enough contributors to adequately respond to all issues.

This bot triages un-triaged issues according to the following rules:

• After 90d of inactivity, lifecycle/stale is applied
• After 30d of inactivity since lifecycle/stale was applied, lifecycle/rotten is applied
• After 30d of inactivity since lifecycle/rotten was applied, the issue is closed

You can:

• Mark this issue as fresh with /remove-lifecycle stale
• Close this issue with /close
• Offer to help out with Issue Triage

Please send feedback to sig-contributor-experience at kubernetes/community.

/lifecycle stale

[sergelogvinov]

/remove-lifecycle stale