Follow @Openwall on Twitter for new release announcements and other news
[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Date: Mon, 11 Mar 2013 20:03:26 +0000
From: Jeremy Stanley <fungi@...goth.org>
To: oss-security@...ts.openwall.com
Subject: Re: CVE Request: MD5 used for Download verification

On 2013-03-11 15:32:52 -0400 (-0400), Donald Stufft wrote:
[...]
> Setuptools (and it's fork distribute) utilize MD5 in order to
> verify that a download has not been tampered with.
[...]

While I'll be the first to agree that migrating to a more widely
accepted modern hashing scheme is a noble goal, I'm unconvinced you
present a security vulnerability in these tools' use of MD5.

1. Do the authors indicate in their documentation that this is
intended to protect against malicious actors altering data in
transit (a la MitM), and not just to identify corrupted downloads?

2. These tools are retrieving the checksums and files being
checksummed from the same location in many (most?) cases, right?

3. Can you come up with a reasonable case in which a collision
attack on MD5 would actually allow for maliciousness in this case
(note most common scenarios would require a preimage or second
preimage attack on MD5 instead, which still has yet to be
demonstrated)?
-- 
{ PGP( 48F9961143495829 ); FINGER( fungi@...ulhu.yuggoth.org );
WWW( http://fungi.yuggoth.org/ ); IRC( fungi@....yuggoth.org#ccl );
WHOIS( STANL3-ARIN ); MUD( kinrui@...arsis.mudpy.org:6669 ); }

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.