On Fri, 18 Feb 2005, Jakob Hirsch wrote:
> From: Jakob Hirsch <jh@???>
> Resent-From: D.H.Davis@???
> To: 'Exim-users' <exim-users@???>
> Date: Fri, 18 Feb 2005 16:51:23 +0100
> Subject: [exim] CRAM-MD5 with Courier authdaemon (with one wishlist suggestion
> and a security question)
...
> Even though everybody now thinks SHA-1 is insecure...
Not quite. See:
http://www.financialcryptography.com/mt/archives/000355.html
which includes the comment:
"it seems that Schneier forgot to mention that the paper has a footnote
which says that the attack on full SHA-1 only works if some padding
(which SHA-1 requires) is not done."
It seems that the attack weakens full SHA-1 without the padding and
reduced round versions of SHA-1. The results look important and perhaps
can be improved. But for practical purposes I think I'll stick with SHA-1
rather than reverting to MD5. No panic yet, certainly not for the low-key
uses I need!
--
Dennis Davis, BUCS, University of Bath, Bath, BA2 7AY, UK
D.H.Davis@??? Phone: +44 1225 386101
