[lustre-discuss] undelete

[E.S. Rosenberg]

Dragging the old discussion back up....
First of thanks for all the replies last time!

Last time in the end we didn't need to recover but now another user made a
bigger mistake and we do need to recover data.
I have shut down our Lustre filesystem and am going to do some simulations
on a test system trying various undelete tools.

autopsy (sleuthkit) on the metadata shows that at least the structure is
still there and hopefully we'll be able to recover more.

Has anyone ever done true recovery of Lustre or is it all just theoretical
knowledge at the moment?

What are the consequences of say undeleting data on OSTs that is then not
referenced on the MDS? Could I cause corruption of the whole filesystem by
doing stuff like that?
(As far as the files themselves go they are most likely all single striped
since that is our default and we are pre PFL so that should be easier I
think).
Thanks,
Eli


On Thu, May 4, 2017 at 2:21 AM, Dilger, Andreas <andreas.dilger at intel.com>
wrote:

> On Apr 27, 2017, at 05:43, E.S. Rosenberg <esr+lustre at mail.hebrew.edu>
> wrote:
> >
> > A user just rm'd a big archive of theirs on lustre, any way to recover
> it before it gets destroyed by other writes?
>
> Just noticed this email.
>
> In some cases, an immediate power-off followed by some ext4 recovery tools
> (e.g. ext3grep) might get you some data back, but that is very uncertain.
>
> With ZFS MDT/OST filesystems (or to a lesser extent LVM) it is possible to
> create periodic snapshots of the filesystems for recovery purposes.  ZFS
> handles this fairly well performance wise, LVM much less so.  With Lustre
> 2.10 there are new tools to manage the ZFS snapshots and allow mounting
> them as a separate Lustre filesystem.
>
> Cheers, Andreas
> --
> Andreas Dilger
> Lustre Principal Architect
> Intel Corporation
>
>
>
>
>
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.lustre.org/pipermail/lustre-discuss-lustre.org/attachments/20180123/d425c182/attachment.html>