public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Kent Borg <kentborg@borg.org>
To: Andi Kleen <freitag@alancoxonachip.com>
Cc: linux-kernel@vger.kernel.org, ehaase@inf.fu-berlin.de
Subject: Re: ext2 not NULLing deleted files?
Date: Fri, 17 Aug 2001 10:20:24 -0400	[thread overview]
Message-ID: <20010817102024.A19505@borg.org> (raw)
In-Reply-To: <01081709381000.08800@haneman.suse.lists.linux.kernel> <oupitfnw1st.fsf@pigdrop.muc.suse.de>
In-Reply-To: <oupitfnw1st.fsf@pigdrop.muc.suse.de>; from freitag@alancoxonachip.com on Fri, Aug 17, 2001 at 10:03:46AM +0200

Andi Kleen <freitag@alancoxonachip.com> writes:
>Just NULLing alone is quite useless anyways; just 0ed data can be
>easily recovered in a special laboratory by using old traces of
>magnetism on the surfaces.  If you care about real data deletion you
>should probably use an utility like wipe which does about 20-30
>passes with random data.

The services of such a laboratory are quite expensive, and invasive
(they need the disk).  An unerase utility is quite cheap (free) and
can be quietly run from the other side of the planet.

It seems to me there be room for something simple that raises the cost
of recovering deleted files to a price significantly above the current
sale price of "free".  Simple NULLing would do that and it could be
done cheaply by a low priority daemon that goes around sweeping up
deleted bits when nothing much else is happening.

Yes, there would still be a window when files will not have been
NULLed, and some machines are too busy to allow such a daemon to run
(are those machines also too busy to do encryption?), but it would be
much better than the case now where nearly all of us have tons of
deleted stuff just sitting there.  (Do you?)

In the physical world documents are sometimes shredded.  Yes, there is
a window between when a document is designated to be shredded and when
it can be shredded, and, yes, most shredders leave big enough pieces
to reassemble the original.  But shredders significantly lower one's
exposure and they significantly raise the cost of recovering that
data.  Just because they are not perfect doesn't mean they are "quite
useless".  The same is true of NULLing deleted files.

Also, I note that such a userland daemon is not a kernel issue.


-kb, the Kent who doesn't consider seatbelts "quite useless" just
because there are accidents for which they will not save his life.


P.S.  We still don't know what was in the 18-1/2 minute gap.  Maybe,
after all these years, we will soon find out, but most of the folks
involved are now dead.  (How many on this list were not yet born then?
Do they even know what I am talking about?  That's pretty good
security for daily use.)  Simple erasure is not perfect security, but
it is pretty damn good, and all it took was Rosemary Woods stretching
to reach that "record" button.

  reply	other threads:[~2001-08-17 14:20 UTC|newest]

Thread overview: 13+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
     [not found] <01081709381000.08800@haneman.suse.lists.linux.kernel>
2001-08-17  8:03 ` ext2 not NULLing deleted files? Andi Kleen
2001-08-17 14:20   ` Kent Borg [this message]
2001-08-17 22:05 Jesse Pollard
  -- strict thread matches above, loose matches on Subject: below --
2001-08-17  7:38 Enver Haase
2001-08-17  7:47 ` Robert Love
2001-08-17 17:40   ` Mark H. Wood
2001-08-17  7:56 ` Thomas Pornin
2001-08-17  8:02 ` Andreas Dilger
2001-08-17 17:55   ` Mark H. Wood
2001-08-17 20:09     ` Andreas Dilger
2001-08-21 14:19   ` Andreas Bombe
2001-08-17 16:32 ` Marc SCHAEFER
2001-08-17 17:25   ` Andreas Dilger

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20010817102024.A19505@borg.org \
    --to=kentborg@borg.org \
    --cc=ehaase@inf.fu-berlin.de \
    --cc=freitag@alancoxonachip.com \
    --cc=linux-kernel@vger.kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox