Removing erroneously created disks?

[surfrock66]

I recently set up ZFS over Iscsi on my truenas using a new 10G fiber network. While setting it up I had a few failures (binding the iscsi to the right nics on the TrueNAS, etc). The result was I tried to move a disk, but the move failed with an error. it's all since been resolved.

In doing those tests, objects were created on the iscsi target, but now I cannot delete them. The source disk is fine and the VM only sees 1 disk, but the 3 failed disks from attempted storage moves are still present on the iscsi storage and I can't remove them in the UI. What's the safe way to purge them, just using the terminal on the nas and purging the files? Is there some command to unhook and remove them from the UI?

 

[bbgeek17]

Perform a "qm disk rescan" from PVE shell, then remove the "unused" disks from VM hardware panel

Good luck

---

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

 

[surfrock66]

Curious, when doing that i get this:

root@sr66-prox-02:~# qm disk rescan
rescan volumes...
Could not find lu_name for zvol vm-117-disk-0 at /usr/share/perl5/PVE/Storage/ZFSPlugin.pm line 118.

I still can't remove it from the LUN's GUI, and the extra disks don't show in the hardware UI for the VM.

 

[bbgeek17]

I still can't remove it from the LUN's GUI, and the extra disks don't show in the hardware UI for the VM.

Sounds like plugin is unable to handle orphaned disks.

I think you have two choices:
- Find reproducible sequence of steps to get into this situation and report it here: https://bugzilla.proxmox.com
- Cleanup the orphaned disks directly on the backend storage

Good luck

---

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

 

[surfrock66]

That did it, from truenas' web UI I just deleted the datasets for the VM and nothing bad happened. When it was clear I migrated the storage over which is how I wanted it, and it all went fine.