Thin Volume pve maps more than assigned

[Ivanoz]

Hi all,

I'm facing a problem with a proxmox I have, I moved several VMs from vmware to proxmox and I just realised that I get this warning after noticing that my thin volume is running out of space:

# lvs
WARNING: Thin volume pve/vm-104-disk-0 maps 274.99 GiB while the size is only 150.00 GiB.
WARNING: Thin volume pve/vm-105-disk-0 maps 196.69 GiB while the size is only 150.00 GiB.
LV VG Attr LSize Pool Origin Data% Meta% Move Log Cpy%Sync Convert
data pve twi-aotz-- 794.29g 97.68 4.71
root pve -wi-ao---- 96.00g
swap pve -wi-ao---- 8.00g
vm-100-disk-0 pve Vwi-aotz-- 100.00g data 95.11
vm-101-disk-0 pve Vwi-aotz-- 150.00g data 59.81
vm-101-disk-1 pve Vwi-aotz-- 40.00g data 63.64
vm-103-disk-1 pve Vwi-aotz-- 100.00g data 93.92
vm-104-disk-0 pve Vwi-aotz-- 150.00g data 100.00
vm-105-disk-0 pve Vwi-aotz-- 150.00g data 100.00

After logging into VM 104 and VM 105 I realized that the OS (Linux) is using the assigned 150 GB each. actually the disk is under that size, it is like 130.

how can I reduce that extra 100 GB that apparently the VMs are using?

Thanks a lot

 

[wolfgang]

Hi,

ensure you have enabled discard on the vdisk.
the you can use fstrim to release this space.

 

[Ivanoz]

Hi All,

the problem is not inside the VM, as I used systemRescueCD with Gparted and I shrunk it.

The problem was with proxmox itself, so I found this entry that sent me to this bug, at the end I did what was suggested in the thread and basically made a backup and restore (vzdump & qmrestore) and now the size is OK and my VMs are working as before:

# lvs
LV VG Attr LSize Pool Origin Data% Meta% Move Log Cpy%Sync Convert
data pve twi-aotz-- 794.29g 71.53 3.44
root pve -wi-ao---- 96.00g
swap pve -wi-ao---- 8.00g
vm-100-disk-0 pve Vwi-aotz-- 100.00g data 95.11
vm-101-disk-0 pve Vwi-aotz-- 150.00g data 72.36
vm-101-disk-1 pve Vwi-aotz-- 40.00g data 63.64
vm-103-disk-1 pve Vwi-aotz-- 100.00g data 93.92
vm-104-disk-0 pve Vwi-aotz-- 150.00g data 78.76
vm-105-disk-0 pve Vwi-aotz-- 150.00g data 84.67

as you can see I dont get the warnings nor my VMs are using more HDD space (that was dangerous).

If someone knows if the bug was fixed, will be great to know.

Thanks.