Can't migrate VM in Cluster

[bigAboo]

HI, I have cluster with three nodes - pve19lvm)-pve2(lvm)-pve3(zfs). pve 3 installed on ZFS RAID10, I add rpool as zfs-data storage in Datacenter menu. Pve1 is a main cluster node. Both pve 1 and pve2 configured with LVM. When I try to migrate to pve3(zfs) from pve2(lvm), I see error

2024-03-06 11:58:02 starting migration of VM 103 to node 'pve3' (172.16.17.130)
2024-03-06 11:58:03 found generated disk 'local-lvm:vm-103-disk-0' (in current VM config)
2024-03-06 11:58:03 found local disk 'local-lvm:vm-103-disk-1' (attached)
2024-03-06 11:58:03 found local disk 'local-lvm:vm-103-disk-2' (attached)
2024-03-06 11:58:03 copying local disk images
2024-03-06 11:58:03 ERROR: storage migration for 'local-lvm:vm-103-disk-0' to storage 'zfs-data' failed - cannot migrate from storage type 'lvmthin' to 'zfspool'
2024-03-06 11:58:03 aborting phase 1 - cleanup resources
2024-03-06 11:58:03 ERROR: migration aborted (duration 00:00:01): storage migration for 'local-lvm:vm-103-disk-0' to storage 'zfs-data' failed - cannot migrate from storage type 'lvmthin' to 'zfspool'
TASK ERROR: migration aborted

storage.cfg

root@pve3:~# cat /etc/pve/storage.cfg
dir: local
        path /var/lib/vz
        content iso,vztmpl,backup
        shared 1

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

nfs: NFS_STORAGE
        export /volume1/pve_data
        path /mnt/pve/NFS_STORAGE
        server 172.16.17.67
        content vztmpl,images,snippets,iso,rootdir,backup
        prune-backups keep-all=1

lvm: hdd500gb
        vgname hdd500gb
        content rootdir,images
        nodes pve1
        shared 1

lvm: pve2-lvm-pool
        vgname pve2-lvm-pool
        content rootdir,images
        nodes pve2
        shared 1

zfspool: zfs-data
        pool rpool/data
        content rootdir,images
        nodes pve3
        sparse 1

 

[fiona]

Hi,
offline migration is currently not possible between all pairs of storage types. Try using online migration instead (you can pause the VM right after start to avoid copying lots of RAM).

 

[bigAboo]

Hi,
offline migration is currently not possible between all pairs of storage types. Try using online migration instead (you can pause the VM right after start to avoid copying lots of RAM).

Actually I started online migration, but when I paused vm, still receved same message

()
2024-03-06 13:27:39 starting migration of VM 103 to node 'pve3' (172.16.17.130)
2024-03-06 13:27:40 found generated disk 'local-lvm:vm-103-disk-0' (in current VM config)
2024-03-06 13:27:40 found local disk 'local-lvm:vm-103-disk-1' (attached)
2024-03-06 13:27:40 found local disk 'local-lvm:vm-103-disk-2' (attached)
2024-03-06 13:27:40 copying local disk images
2024-03-06 13:27:40 ERROR: storage migration for 'local-lvm:vm-103-disk-0' to storage 'zfs-data' failed - cannot migrate from storage type 'lvmthin' to 'zfspool'
2024-03-06 13:27:40 aborting phase 1 - cleanup resources
2024-03-06 13:27:40 ERROR: migration aborted (duration 00:00:01): storage migration for 'local-lvm:vm-103-disk-0' to storage 'zfs-data' failed - cannot migrate from storage type 'lvmthin' to 'zfspool'
TASK ERROR: migration aborted

 

[fiona]

Actually I started online migration, but when I paused vm, still receved same message

()
2024-03-06 13:27:39 starting migration of VM 103 to node 'pve3' (172.16.17.130)
2024-03-06 13:27:40 found generated disk 'local-lvm:vm-103-disk-0' (in current VM config)
2024-03-06 13:27:40 found local disk 'local-lvm:vm-103-disk-1' (attached)
2024-03-06 13:27:40 found local disk 'local-lvm:vm-103-disk-2' (attached)
2024-03-06 13:27:40 copying local disk images
2024-03-06 13:27:40 ERROR: storage migration for 'local-lvm:vm-103-disk-0' to storage 'zfs-data' failed - cannot migrate from storage type 'lvmthin' to 'zfspool'
2024-03-06 13:27:40 aborting phase 1 - cleanup resources
2024-03-06 13:27:40 ERROR: migration aborted (duration 00:00:01): storage migration for 'local-lvm:vm-103-disk-0' to storage 'zfs-data' failed - cannot migrate from storage type 'lvmthin' to 'zfspool'
TASK ERROR: migration aborted

Oh, that's because of the generated disk 'local-lvm:vm-103-disk-0' (either cloud-init or TPM) which is migrated separately, even if the VM is running.

Since you have a shared storage NFS_STORAGE, you can try putting the problematic image on there and migrate afterwards. For TPM state, you can do it via Disk Action > Move storage. For cloud-init, you need to delete it and create a new one on the new storage.

 

[bigAboo]

Ok thank you