Re: CVE request for ecryptfs

[Jamie Strandboge <jamie () canonical com>]

On Wed, 29 Oct 2008, Tomas Hoger wrote:

> Hi Jamie!
>
> On Thu, 23 Oct 2008 16:16:28 -0500 Jamie Strandboge
> <jamie () canonical com> wrote:
>
> > While reviewing ecryptfs, I discovered an information disclosure
> > vulnerability in ecryptfs-setup-private and notified upstream. This
> > helper script was known as ecryptfs-setup-confidential in earlier
> > releases.
> >
> > The problem arises when ecryptfs-setup-private invokes
> > ecryptfs-wrap-passphrase and ecryptfs-add-passphrase with command line
> > arguments that include the user's existing login password as well as
> > the newly created mount password. As a result, these passwords can be
> > snooped in the process table.
>
> Well the question is whether this should be worded as
> ecryptfs-setup-{private,confidential} issue, or more generic issue
> affecting various ecryptfs-* command line utilities, that only accept
> passwords as command line arguments (i.e. no interactive prompt).  So
> even though there's not ecryptfs-setup-* script to fix in older
> versions, steps done by ecryptfs-setup-* are likely to be performed by
> the user manually, resulting in the same risk of leak as with helper
> script.  Or do I miss anything?

That's exactly right, which reminded me, documentation surrounding
ecryptfs also should be updated. I pinged upstream about it.

I was also notified of an additional commit that is desirable (a bugix
for the patch to ecryptfs-setup-private):
http://git.kernel.org/?p=linux/kernel/git/mhalcrow/ecryptfs-utils.git;a=commit;h=2c422e6d2549f90258cddeebf105b066b598bdbb

Jamie

-- 
Ubuntu Security Engineer     | http://www.ubuntu.com/
Canonical Ltd.               | http://www.canonical.com/

Attachment: signature.asc
Description: Digital signature