I Created Linked Clone but no sign on volume IDyou can tell from the volid (if a volume is a linked clone, it will reference a base volume).
create linked clone of drive scsi0 (local-lvm:base-102-disk-0)
WARNING: You have not turned on protection against thin pools running out of space.
WARNING: Set activation/thin_pool_autoextend_threshold below 100 to trigger automatic extension of thin pools before they get full.
Logical volume "vm-113-disk-0" created.
WARNING: Sum of all thin volume sizes (1.61 TiB) exceeds the size of thin pool pve/data and the size of whole volume group (<893.72 GiB).
TASK OK
when try full clone get thisI Created Linked Clone but no sign on volume ID
So I'm missing something ?Code:create linked clone of drive scsi0 (local-lvm:base-102-disk-0) WARNING: You have not turned on protection against thin pools running out of space. WARNING: Set activation/thin_pool_autoextend_threshold below 100 to trigger automatic extension of thin pools before they get full. Logical volume "vm-113-disk-0" created. WARNING: Sum of all thin volume sizes (1.61 TiB) exceeds the size of thin pool pve/data and the size of whole volume group (<893.72 GiB). TASK OK
create full clone of drive ide2 (ssd4-out:vm-107-cloudinit)
Logical volume "vm-116-cloudinit" created.
create full clone of drive scsi0 (ssd4-out:vm-107-disk-0)
Logical volume "vm-116-disk-0" created.
Interesting! I would have expected something like that with ZFS (with its copy-on-write) instead but it appears not to be the case (as clone copies entire disks en linked won't allow removal).LVM-thin is the exception - linked clones in LVM thin are fully independent of their base image (each can be deleted without interfering with the other). so while the base image is renamed to "base-...", the linked images have a regular volume ID and continue to work even if you delete the base image/volume.
I think you need to seach all (other) VMs to see if the base-drive is referenced. Something likeI Created Linked Clone but no sign on volume ID
So I'm missing something ?
grep ':base-100-disk-[0-9]*/' /etc/pve/qemu-server/*.conf
for usages of the disks of VM template 100.pvesh get /nodes/<node name>/storage/<storage ID>/content
for the storage of the base image and see if it's mentioned in the parent
column somewhere.yes, LVM-thin is an exception here, because AFAIK the linked clones there are independent of the parent volume. Basically, every volume references its extents and as long as a single volume references an extent, the extent is not freed up. So you can remove or rename the parent without affecting the linked clone.Follow-up question (sorry to revive this thread, but it's not that old):
When I list the contents of the node (lvm-thin) like you explained, fiona, none of the LXCs I created as linked clones show up with a "parent" entry. Does that mean they aren't really linked clones but full clones? Or is the cloning handled by LVM somehow, but invisibly to the user?
If it's the latter, does that mean I can lvrename (and adapt the corresponding config file) or delete templates, even if I created linked clones from them, without destroying anything?
Or if I'm unhappy with a lvm-thin template because I forgot something, I could just clone it, add a missing feature, convert it to a new template and delete the old one? Even if I created linked clones from it?
I'm sorry I have to ask all these questions, I'm new to this and trying to grasp the concepts. So much to learn.