Driver fail to list datasets under certain circumstances

[angarod]

What steps did you take and what happened: Create a persistent volume claim and a persistent volume on a server that use zfs as filesystem. ZFS topology:

$ zfs list
nvme                                242K  9.20G       24K  none
nvme/openebs                        120K  9.20G       24K  none
nvme/openebs/dataset                24K   9.20G       24K  none

What did you expect to happen: PVC and PV are successfully created without error.

Anything else you would like to add: The driver cannot list datasets since depth level is hard coded to 1, which prevents the driver from reading my topology. Seems to be related to https://github.com/openebs/zfs-localpv/commit/ba0e1749ec1841bb8f9d704225e6df04787dc608. openebs-zfs-plugin E1229 09:29:32.162493 1 zfs_util.go:931] zfs: could not list zpools cmd [list -d 1 -s name -o name,guid,available -H -p]: exit status 127

Environment:

• ZFS-LocalPV version: zfs-localpv-2.3.0
• Kubernetes version (use kubectl version): v1.28.2
• Kubernetes installer & version: v1.28.2
• Cloud provider or hardware configuration: bare-metal
• OS (e.g. from /etc/os-release): Ubuntu 20.04.6 LTS

[Val]

@lowjoel what do you propose to meet the need for topology on several levels?

[avishnu]

Yes, this could be the issue. Any takers for fixing? Possible approaches could be determine the depth based on the input dataset or checking for the existence of the input dataset..