backing up zfs dataset

Freddie Cash fjwcash at gmail.com
Fri Nov 25 00:15:50 UTC 2011


On Thu, Nov 24, 2011 at 4:12 PM, Johannes Totz <johannes at jo-t.de> wrote:

> On 25/11/2011 00:07, Freddie Cash wrote:
>
>> On Thu, Nov 24, 2011 at 3:58 PM, Andrew Reilly<areilly at bigpond.net.au>**
>> wrote:
>>
>>  I do think that backup is something of a weakness for ZFS at
>>> the moment.  Sure, live filesystems and snapshots are clearly
>>> cool, and the modern way and all, but there is an awful lot of
>>> flexibility and ease of undersanding in the model of a "backup
>>> file on a tape."  Doesn't have to be on a tape, but the moral
>>> equivalent to dump/restore would (in my book) be a wonderful
>>> addition to ZFS, if anyone felt inclined.  Just padding the
>>> send/receive serialisation format with enough checksum and
>>> restart information to allow detection and graceful recovery
>>> from read errors in the backup medium would do the job.
>>>
>>
>>
>> One could probably work around this by doing a zfs send to a file, then
>> running it through parchive [1] to generate all the redundancy data.
>>  Granted, I've never used par, so it may or may not be feasible.
>>
>> [1] http://parchive.sourceforge.**net <http://parchive.sourceforge.net>
>>
>
> yeah, was thinking of doing that at some stage. but then send/receive
> format is not guaranteed to be stable (has anybody done any tests on this?
> try a v1 send with a v28 receive?).
>

I haven't tried it, but the consensus I've seen on the zfs-discuss mailing
list is that so long as the receive side is of a higher (or equal) version
to the send side, it's supposed to work.  In theory.  ;)

-- 
Freddie Cash
fjwcash at gmail.com


More information about the freebsd-fs mailing list