[pve-devel] [PATCH manager] report: add WWN column to lsblk output

[Michael Köppl]

On 3/31/25 11:41, Friedrich Weber wrote:
> WWNs (WWIDs) are useful information for debugging setups where
> multipath is not (correctly) set up, as the WWN reveals which block
> devices point to the same LUN. The report currently does not provide
> WWNs in such scenarios. There is a symlink for the WWN under
> /dev/disk/by-id/wwn-*, but if there are multiple block devices with
> the same WWN, the symlink just points to one of them.
> 
> Hence, add the WWN column to the lsblk output.
> 
> Signed-off-by: Friedrich Weber <f.weber at proxmox.com>
> ---
>   PVE/Report.pm | 2 +-
>   1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/PVE/Report.pm b/PVE/Report.pm
> index 1ed91c8e..8f53823e 100644
> --- a/PVE/Report.pm
> +++ b/PVE/Report.pm
> @@ -117,7 +117,7 @@ my $init_report_cmds = sub {
>   	'block devices' => {
>   	    order => 80,
>   	    cmds => [
> -		'lsblk --ascii -M -o +HOTPLUG,ROTA,PHY-SEC,FSTYPE,MODEL,TRAN',
> +		'lsblk --ascii -M -o +HOTPLUG,ROTA,PHY-SEC,FSTYPE,MODEL,TRAN,WWN',
>   		'ls -l /dev/disk/by-*/',
>   		'iscsiadm -m node',
>   		'iscsiadm -m session',

Tested this by adding SCSI, IDE, and SATA hard disks to a VM and 
manually setting a WWN via qm set. I noticed that the output of lsblk 
does only show a WWN for the SCSI disk, but not for the others. smartctl 
-i does return a WWN for the SATA disk and scsi_id also seems to 
recognize the manually set value. I think it could make sense to add a 
note to the documentation regarding the limitations of the lsblk output 
in this regard?