ceph, no thin provisioning with vm clone

[komar07]

Hi I run proxmox 3 node cluster with ceph.
Everything it's ok when we create vm from scratch, we make vm with 100GB disks in raw format, thin provisioning works ok, if vm have 8gb data on disk - 8gb is on ceph but when we start to clone that vm, ceph/proxmox? allocate whole 100GB for new cloned vm.
Where is the problem? Is it on ceph? is it on proxmox clone options?

proxmox-ve: 4.1-25 (running kernel: 4.2.6-1-pve)
pve-manager: 4.0-64 (running version: 4.0-64/fc76ac6c)
pve-kernel-4.2.6-1-pve: 4.2.6-25
pve-kernel-4.2.2-1-pve: 4.2.2-16
lvm2: 2.02.116-pve2
corosync-pve: 2.3.5-2
libqb0: 0.17.2-1
pve-cluster: 4.0-29
qemu-server: 4.0-39
pve-firmware: 1.1-7
libpve-common-perl: 4.0-41
libpve-access-control: 4.0-10
libpve-storage-perl: 4.0-37
pve-libspice-server1: 0.12.5-2
vncterm: 1.2-1
pve-qemu-kvm: 2.4-16
pve-container: 1.0-32
pve-firewall: 2.0-14
pve-ha-manager: 1.0-14
ksm-control-daemon: 1.2-1
glusterfs-client: 3.5.2-2+deb8u1
lxc-pve: 1.1.5-5
lxcfs: 0.13-pve1
cgmanager: 0.39-pve1
criu: 1.6.0-1
zfsutils: 0.6.5-pve6~jessie

 

[Q-wulf]

that sounds like a bug.

 

[Q-wulf]

Just FYI:

apt-get update
apt-get upgrade

your not running the latest versions.
Also keep check if this bug might affect you:
https://forum.proxmox.com/threads/25094-Do-NOT-use-SUSE-virtio-drivers-from-Windows-Update

 

[komar07]

all latest version, cluster was installed and upgraded 3 days ago

 

[Q-wulf]

not according to your pveversion -v


compare:

pveversion -v
proxmox-ve: 4.1-26 (running kernel: 4.2.6-1-pve)
pve-manager: 4.0-66 (running version: 4.0-66/8ba0405d)
pve-kernel-4.2.6-1-pve: 4.2.6-26
pve-kernel-4.2.2-1-pve: 4.2.2-16
pve-kernel-4.2.3-2-pve: 4.2.3-22
lvm2: 2.02.116-pve2
corosync-pve: 2.3.5-2
libqb0: 0.17.2-1
pve-cluster: 4.0-29
qemu-server: 4.0-41
pve-firmware: 1.1-7
libpve-common-perl: 4.0-41
libpve-access-control: 4.0-10
libpve-storage-perl: 4.0-38
pve-libspice-server1: 0.12.5-2
vncterm: 1.2-1
pve-qemu-kvm: 2.4-17
pve-container: 1.0-32
pve-firewall: 2.0-14
pve-ha-manager: 1.0-14
ksm-control-daemon: 1.2-1
glusterfs-client: 3.5.2-2+deb8u1
lxc-pve: 1.1.5-5
lxcfs: 0.13-pve1
cgmanager: 0.39-pve1
criu: 1.6.0-1
zfsutils: 0.6.5-pve6~jessie
openvswitch-switch: 2.3.2-2

 

[komar07]

now upgradeing, at the morning i will check that bug at the morning.

 

[dietmar]

You create a 'linked' clone, or a 'full' clone?

 

[komar07]

full clone
now I'm up to date

proxmox-ve: 4.1-26 (running kernel: 4.2.6-1-pve)
pve-manager: 4.0-66 (running version: 4.0-66/8ba0405d)
pve-kernel-4.2.6-1-pve: 4.2.6-26
pve-kernel-4.2.2-1-pve: 4.2.2-16
lvm2: 2.02.116-pve2
corosync-pve: 2.3.5-2
libqb0: 0.17.2-1
pve-cluster: 4.0-29
qemu-server: 4.0-41
pve-firmware: 1.1-7
libpve-common-perl: 4.0-41
libpve-access-control: 4.0-10
libpve-storage-perl: 4.0-38
pve-libspice-server1: 0.12.5-2
vncterm: 1.2-1
pve-qemu-kvm: 2.4-17
pve-container: 1.0-32
pve-firewall: 2.0-14
pve-ha-manager: 1.0-14
ksm-control-daemon: 1.2-1
glusterfs-client: 3.5.2-2+deb8u1
lxc-pve: 1.1.5-5
lxcfs: 0.13-pve1
cgmanager: 0.39-pve1
criu: 1.6.0-1
zfsutils: 0.6.5-pve6~jessie

still same problem

my ceph config use cachetering (cache pool) on fasd ssd as a front of sata backend. Maybe this is a problem?
On the other hand, when restore backup or make new vm thin provisioning works ok.

 

[dietmar]

full clone

If you create a template firsts, you can use linked clones. Those are more efficient and use less space.

 

[komar07]

Yes I know, but I need full clone.

 

[spirit]

It's a known bug in rbd driver and block mirror.

One workaround : use virtio-scsi + discard, and do a trim

 

[komar07]

One workaround : use virtio-scsi + discard, and do a trim

virtio-scsi + discard ok
do a trim ???

 

[spirit]

virtio-scsi + discard ok
do a trim ???

if you use linux, you can use "fstrim" command in your guest.

fstrim /mountpoint (for each mount point)

 

[rumors]

it seems this bug still persist .., just tested with 5.1.46 and found this thread

 

[aderumier]

yes, it still missing support in librbd

https://tracker.ceph.com/issues/20070