root@castiana:~# zfs set quota=2EB encrypted/lxd/containers/zesty
root@castiana:~# echo $?
0

root@castiana:~# zfs get quota encrypted/lxd/containers/zesty
NAME                            PROPERTY  VALUE  SOURCE
encrypted/lxd/containers/zesty  quota     2E     local

So as can be seen, ZFS allows settings quotas that far exceed the pool
size and does so without returning an error. There is therefore no way
for LXD to return an error since the operation succeeded.

My guess is that ZFS will always shows your volume total size to be the
smallest of its quota or any of the quota in the tree, including the
tree root (the pool itself).

** Changed in: lxc (Ubuntu)
       Status: New => Invalid

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to lxc in Ubuntu.
https://bugs.launchpad.net/bugs/1636617

Title:
  lxc set device root size fails silently when out of space

Status in lxc package in Ubuntu:
  Invalid

Bug description:
  On 16.04.1, use lxd init with default responses (zfs, loopback, 10G).
  Create container with lxc launch. Attempt to resize container root to
  20G (larger than loopback file).

  lxc set device root size 20G reports no error, but the container still
  has a 1.1G root as shown by lxc exec <container> df -h

  It would be ideal to have lxc throw an error of some kind when it does
  not have enough space to complete the operation.

  For documentation purposes, the fix is to grow the zfs.img file with
  truncate, turn on ZFS autoexpand, then perform a zpool online resize.
  After adding enough loopback device storage, the set command behaves
  identically but achieves the intended result.

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/lxc/+bug/1636617/+subscriptions

-- 
Mailing list: https://launchpad.net/~touch-packages
Post to     : [email protected]
Unsubscribe : https://launchpad.net/~touch-packages
More help   : https://help.launchpad.net/ListHelp

Reply via email to