VM disks show incorrect size in Proxmox GUI after manual migration

[aifiedler]

Dear Proxmox Team,

Our Proxmox Version: 8.1.4

We are running multiple clusters. As soon as we need to update one of these clusters to a newer OS/PVE version, we migrate all vms
manually to a different cluster.

We are using ceph, so we copy the biggest part of the VM disk, stop the VM on the source host, copy the rbd diff of the image and start the vm
on the destination host again. This does work pretty well, but we have a few VMs that report an incorrect disk size in the Proxmox GUI:

bootdisk: virtio0
cores: 2
cpu: host
ide2: none,media=cdrom
memory: 4096
name: redacted
net0: virtio=XX:XX:XX:XX:XX:XX,bridge=vmbr111,rate=125
net1: virtio=XX:XX:XX:XX:XX:XX,bridge=vmbr116
numa: 0
ostype: l26
parent: autodaily240610011501
scsihw: virtio-scsi-pci
smbios1: uuid=b091df0a-656e-43de-9b53-c46181845b45
sockets: 1
virtio0: cluster4-rbd:vm-100-disk-0,discard=on,size=2G

The user in the Proxmox GUI would assume that the VM's disk is only 2GB in size, however, running lsblk on the VM shows a different picture:

NAME        MAJ:MIN RM  SIZE RO TYPE MOUNTPOINT
sr0          11:0    1 1024M  0 rom
vda         254:0    0   32G  0 disk
├─vda1      254:1    0  512M  0 part /boot
├─vda2      254:2    0    2G  0 part [SWAP]
├─vda3      254:3    0    9G  0 part /
└─vda4      254:4    0 20.5G  0 part
  └─vg-data 253:0    0   20G  0 lvm  /data

Do you have any idea what may cause this incorrect reporting?

 

[leesteken]

Do you have any idea what may cause this incorrect reporting?

No idea, but it's only cosmetic. Try running a qm disk rescan --vmid 100 to have Proxmox determine the current size and update the VM configuration.

 

[aifiedler]

No idea, but it's only cosmetic. Try running a qm disk rescan --vmid 100 to have Proxmox determine the current size and update the VM configuration.

Thanks for the hint, the command works.