Strange disc names on Linstor storage

[Osorkonus]

Hi all.
I have a question regarding proxmox+linstor. On all video manuals (example (https://www.youtube.com/watch?v=pP7nS_rmhmE)) resources in linstor are created as vm-100-disk-0, but in my linstor for some reason it is some pm-47b985ee. Though in lvmthin and in ceph everything is named normally. In this case everything works, but with such names it is difficult to debug problems, you can't understand what this disc is from and where to look for its vm. Actually the question: is it so and it should be and how it can (if) be done in a human way?
PVE - 7.4-17
linstor controller 1.27.1

 

[bbgeek17]

is it so and it should be and how it can (if) be done in a human way?
PVE - 7.4-17
linstor controller 1.27.1

Have you tried to ping linstor about it? Proxmox has no way to control or influence how external 3rd party storage manages the disks.

---

Blockbridge : Ultra low latency all-NVME shared storage for Proxmox - https://www.blockbridge.com/proxmox

 

[Osorkonus]

Yeah, I've been thinking about it. However, on the other hand linstore creates resources with the names the client tells it. And it is logical that it is proxmox that generates these names. After all, it is proxmox that writes such names into VM configs, linstore does not do it

 

[Osorkonus]

Does no one use linstor in proxmox or does everyone's drive names look normal?

 

[SiriusNZ]

I do, and see the same. Before I go diving in to find out where these pm_randomhexvaule_VMID names come from, did you have any luck tracking it down?

Seemed to start when I upgraded Proxmox VE from v6 to v7.

edit: The storage backend must have some say - I have noticed one difference between Linstor and Ceph is that Ceph's numbering would (use to) start at 0 (eg vm-100-disk-0) whereas with Linstor the numbering would start at 1.

Can't comment about Ceph now as I've changed jobs from a workplace that used Proxmox+Ceph while they were still on PVE v6, and where I am now we figured Proxmox+Linstor suits our smaller scale better (server cluster is only 3-node).

 

[SiriusNZ]

Update: It's Linstor. https://forums.linbit.com/t/linstor-proxmox-v8-0-0/58

Reassining disks:
In order to allow reassigning disks between VMs we had to change the
naming scheme. New disks (also clones of old ones) will have names like
“pm-12cf742a_101” on PVE level, and “pm-12cf742a” on LINSTOR/DRBD level.
This is a static prefix (“pm-”), 8 characters of a UUID, and on PVE
level the VMID separated by an underscore (“_101”). The new naming
obsoletes the old “vm-101-disk-1” like names. Old VMs with legacy names
still work with version 8 of the plugin. For the user this change should
be completely transparent and should not require any changes besides
getting accustomed to the new names.