Snapshot and quota problem in proxmox 4.1.

dirkschmid

New Member
Mar 28, 2016
4
0
1
42
Hi,

I'm just trying out the newest version of proxmox (4.1-22) and having some problems.

1. Problem: When I try to start an lxc container with enabled quota on an lvm-thin storage the container starts to fail. When I disable quota it just works fine and it's also possible to take snapshots of the VMs.

2. Problem: When I run an VM from an ext4 storage it works with quota, but I no longer have the option to use snapshot backups.

Is it correct, that with the recent version I have to decide between snapshots or quota? I really need both features. Is there any kind of workaround?

Thanks,
Dirk
 
When I understand it correctly snapshot support is coming in version 1.054 of pve-container. Will it than work also on lvm-thin volumes? When will be this version released?

storage types: with the next updates all ext4 based storage types except for passed-through /dev nodes (a patch for that is already waiting) should™ work (If I'm not mistaken then that should be all except for bind mounts, zfs subvolumes or when using `size=0`).
as for migrating storage: not yet but soon, that part's overdue by now, sorry...
about the last error: Right, the updates to the autodev hook didn't make it into the last package. Will be fixed with pve-container>=1.0-54 where LVM devices will appear the usual way with /dev/dm-* as node and the usual named symlinks in /dev/mapper/.
 
1.0-54 will work with lvm-thin volumes.
The forum's not dead. It's just that sometimes problems get reported in multiple places (bugzilla, mailing list, forum threads) and it can happen that we lose track of the places we need to reply to (and the forum forum doesn't send out new notification emails until you explicitly visit or mark a thread as read, so there's that...)

As for when the package will be available, well. It should hit the pvetest repo within the next couple of days.
 
HI wbumiller, thanks for the reply. I will look forward for the new version and will test it out as soon as it get's available. Thanks, Dirk