VM Migration from ZFS to LVM thin

[shofman]

Hello,
I'm quite new on Proxmox and I'm playing with it on a lab.
2 hosts with Proxmox 8.2.4

1 host has ZFS pool, and the other one has an LVM thin storage on RAID Hardware.
So I try to move a VM (Windows 11, without snapshot) from one host (pve1) to another (pvetest).

I have this error message :
2024-08-23 18:27:07 starting migration of VM 102 to node 'pvetest' (192.168.1.23)
2024-08-23 18:27:07 found local disk 'DATA:vm-102-disk-0' (attached)
2024-08-23 18:27:07 found local disk 'DATA:vm-102-disk-1' (attached)
2024-08-23 18:27:07 found generated disk 'DATA:vm-102-disk-2' (in current VM config)
2024-08-23 18:27:07 copying local disk images
2024-08-23 18:27:07 ERROR: storage migration for 'DATA:vm-102-disk-2' to storage 'RAID' failed - cannot migrate from storage type 'zfspool' to 'lvmthin'
2024-08-23 18:27:07 aborting phase 1 - cleanup resources
2024-08-23 18:27:07 ERROR: migration aborted (duration 00:00:00): storage migration for 'DATA:vm-102-disk-2' to storage 'RAID' failed - cannot migrate from storage type 'zfspool' to 'lvmthin'
TASK ERROR: migration aborted

root@pve1:~# cat /etc/pve/storage.cfg
dir: local
path /var/lib/vz
content iso,backup,vztmpl

zfspool: local-zfs
pool rpool/data
content images,rootdir
nodes pve1
sparse 1

dir: ISO
path /DATA/ISO
content iso,vztmpl
nodes pve1
prune-backups keep-all=1
shared 0

zfspool: DATA
pool DATA
blocksize 64k
content rootdir,images
mountpoint /DATA
nodes pve1
sparse 1

lvmthin: RAID
thinpool RAID
vgname RAID
content rootdir,images
nodes pvetest



Can someone explain what I'm doing wrong ?

 

[shofman]

It's very strange, because migration from LVM Thin to ZFS wotks...

Another problem : the migration is not thin.
The drive is 32Gb, and it use 32gb on ZFS :

~# zfs list
NAME USED AVAIL REFER MOUNTPOINT
DATA 65.3G 2.55T 186K /DATA
DATA/ISO 6.56G 2.55T 6.56G /DATA/ISO
DATA/base-101-disk-0 2.94G 2.55T 2.94G -
DATA/vm-100-disk-00 2.94G 2.55T 2.94G -
DATA/vm-102-disk-0 668K 2.55T 668K -
DATA/vm-102-disk-1 20.9G 2.55T 20.9G -
DATA/vm-102-disk-2 145K 2.55T 145K -
DATA/vm-103-disk-0 32.0G 2.55T 32.0G -
rpool 2.06G 227G 104K /rpool
rpool/ROOT 2.05G 227G 96K /rpool/ROOT
rpool/ROOT/pve-1 2.05G 227G 2.05G /
rpool/data 96K 227G 96K /rpool/data
rpool/var-lib-vz 96K 227G 96K /var/lib/vz

 

[shofman]

Another try : Ubuntu go back to pvetest.
And now it works... but with 100% space used

# lvs
LV VG Attr LSize Pool Origin Data% Meta% Move Log Cpy%Sync Convert
RAID5 RAID5 twi-aotz-- <1.07t 2.93 0.26
vm-103-disk-0 RAID5 Vwi-aotz-- 32.00g RAID5 100.00
data pve twi-aotz-- 337.86g 0.00 0.50
root pve -wi-ao---- 96.00g
swap pve -wi-ao---- 8.00g



I don't understand !

 

[gleira]

Hello,

I have the same problem, and I don't understand. Migration should be just an operation consisting in copying bytes. Why it can't migrate from zfspool to lvmthin?

root@pve1:~# qm migrate 107 pve2 --targetstorage local-lvm --with-local-disks
2025-04-11 14:00:34 starting migration of VM 107 to node 'pve2' (192.168.56.58)
2025-04-11 14:00:34 found local disk 'local-zfs:vm-107-disk-0' (attached)
2025-04-11 14:00:34 copying local disk images
2025-04-11 14:00:34 ERROR: storage migration for 'local-zfs:vm-107-disk-0' to storage 'local-lvm' failed - cannot migrate from storage type 'zfspool' to 'lvmthin'
2025-04-11 14:00:34 aborting phase 1 - cleanup resources
2025-04-11 14:00:34 ERROR: migration aborted (duration 00:00:00): storage migration for 'local-zfs:vm-107-disk-0' to storage 'local-lvm' failed - cannot migrate from storage type 'zfspool' to 'lvmthin'
migration aborted

I thought it might be a problem with with storage.cfg, but I think it is ok now:

dir: local
path /var/lib/vz
content backup,iso,vztmpl

zfspool: local-zfs
pool rpool/data
content rootdir,images
nodes pve1
sparse 1

iscsi: iscsi-01
portal 192.168.56.101
target iqn.1992-08.com.netapp:sn.96f11dad058711f0955b000c29ab0ade:vs.12
content images

lvm: lvm-lun0
vgname volgroup0
base iscsi-01:0.0.0.scsi-3600a09807770457a7a5d4f6f6f62375a
content images,rootdir
saferemove 0
shared 1

nfs: NFSAlmacen
export /Almacen
path /mnt/pve/NFSAlmacen
server 192.168.56.94
content images
prune-backups keep-all=1

lvmthin: local-lvm
thinpool data
vgname pve
content images,rootdir
nodes pve2,pve3

There are no snapshots, and version is 8.4.1. But it has happened since we installed 8.3. We expected to move it to production, but this is quite a roadblock.

Any suggestions?

Best regards.