[lxc-users] zfs snapshot point

[Tamas Papp]

On 09/24/2014 06:28 PM, Serge Hallyn wrote:
> Quoting Chris Burroughs (chris.burroughs at gmail.com):
>> On 07/08/2014 11:51 AM, Serge Hallyn wrote:
>>> Oh is that possible?  With rootfs being a 'sub-dataset' or something
>>> in /lxc/foo?  And both /lxc/foo and /lxc/foo/rootfs can be
>>> snapshotted?  That'd be neat.
>> Sorry for the long delay.  Yes ZFS datasets can be nested and this
>> is a pretty common thing to do (if a ZFS dataset is /, then I
>> suppose you can call all of the other datasets nested).  If they are
>> nested you can also recessively snapshot them as one atomic action,
>> which I think is what we would want to do for /lxc/foo and
>> /lxc/foo/rootfs
>>
>>
>> https://pthree.org/2012/12/17/zfs-administration-part-x-creating-filesystems/
>>
>> http://docs.oracle.com/cd/E19253-01/819-5461/gdfdt/index.html
> Ok, thanks, then we should do that.  I won't have time soon-ish, so
> if someone wants to come up with the patch that'd be great.  Else
> I should get to it eventually.

My opinion is still that having one dataset like '/lxc/foo/' is simpler 
and the simpler is usually the better.

Last time we discussed this on the list Serge wrote, that he would 
accept patches, just someone has to implement it... though it needs C 
coding capabilities:)


tamas