So I think we're already doing what's needed here. Please correct me if I'm wrong.
I'd support a feature request against maas/curtin for rich NVME support (creating namespaces, using the nvme command to modify/tweak features, functions, extract IDs etc); MAAS would need to use the nvme tool to extract this info, curtin would need a new storage type (type:nvme) and config to control etc).
"nvme0c33n1 is hidden is not shown in /dev/ (and lsblk does not report it) because GENHD_FL_HIDDEN is applied to those block device entries:"
Is this related at all to the request?
Further, curtin will emit dname rules that use ID_WWN, which from your paste before appears to include the NGUID, no?
ID_WWN= nvme.8086- 6e766d656330- 51454d55204e564 d65204374726c- 00000001
So I think we're already doing what's needed here. Please correct me if I'm wrong.
I'd support a feature request against maas/curtin for rich NVME support (creating namespaces, using the nvme command to modify/tweak features, functions, extract IDs etc); MAAS would need to use the nvme tool to extract this info, curtin would need a new storage type (type:nvme) and config to control etc).