Follow @Openwall on Twitter for new release announcements and other news
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date: Mon, 14 Mar 2011 08:56:45 -0400
From: Dan Rosenberg <>
Cc: Stephan Mueller <>, Vasiliy Kulikov <>
Subject: Re: Untrusted fs and invalid filenames

On Mon, Mar 14, 2011 at 4:12 AM, Stephan Mueller
<> wrote:
> Am Samstag, 12. März 2011, um 18:03:45 schrieb Vasiliy Kulikov:
> Therefore, if you consider a file system untrusted, a simple flag "untrusted"
> which disables some high-level logic (like symlinks across partitions or funky
> file names) may just be window-dressing until the entire parsing of the
> physical data structure layout is hardened.

I'd like to add that while this kind of hardening would be nice in
theory, there is little urgency in making these improvements since the
proposed attack vectors are extremely limited.  As I see it, there are
four scenarios where this might matter:

1. An attacker convinces a victim to download an evil filesystem image
and manually mount it.

2. An attacker with physical access leverages automounting features to
cause the mounting of evil filesystems residing on external media.

3. An attacker wishes to escalate from CAP_SYS_ADMIN to full root by
mounting a malformed an evil filesystem.

4. An attacker leverages a setuid mount helper to mount an evil
filesystem image.

The first case is clearly unlikely.  The second case can be addressed
by restricting automounting in circumstances where it is
inappropriate, such as when the screen is locked.  The third case is
silly, since being able to mount arbitrary filesystems can easily get
you root without having to trick someone into doing something
inappropriate with an evil filesystem (e.g. mount over /etc/pam.d).

The final case is something distros can do something about now.  It
should be solved by restricting the usage of setuid root mount
helpers.  There have been far too many vulnerabilities in these types
of utilities to be worth the risk - I think distros should strip the
setuid bits from these helpers when possible, and otherwise ship these
helpers with 4750 permissions and restrict their execution to trusted
groups.  I understand that FUSE must be an exception on some
distributions (such as Ubuntu), but other helpers (cifs, ncpfs, hgfs,
etc.) can probably be restricted a bit more.

So while improving this aspect of the kernel is on my longterm
wishlist, I think once the actual threat model is considered, these
kinds of attacks pose little security risk in real life that can't be
handled by fixing problems outside the kernel.


Powered by blists - more mailing lists

Please check out the Open Source Software Security Wiki, which is counterpart to this mailing list.

Confused about mailing lists and their use? Read about mailing lists on Wikipedia and check out these guidelines on proper formatting of your messages.