All crypto fails in the presence of a compromised PRNG. You can't accuse DSA of 
that without also admitting it for everything else.

And since you lifted the "10-foot pole" comment from schneier.com, I'll lift 
Bruce's response to that very comment:

"In general, I don't think there is a difference. Cryptanalytic advances 
against one transfer to the other."

I don't really want to get argumentative or off-topic though. Alex's response 
is probably the most thorough piece in a single place on the subject, so I 
think we can safely put this tangent to rest.

-- 
Brian Bennett
Systems Engineer, Cloud Operations
Joyent, Inc. | www.joyent.com

> On Aug 31, 2015, at 12:59 PM, Steve <[email protected]> wrote:
> 
> On 08/31/2015 02:33 AM, Brian Bennett wrote:
>>> On Aug 28, 2015, at 8:41 PM, Steve <[email protected]> wrote:
>>> 
>>> In my mind, it ought to be disabled by default so that you have to
>>> know you are lowering, *almost* to the point of entirely loosing
>>> your security, when you activate it.
>> 
>> Not to get too far off topic, but is there actually any evidence to
>> back up that statement? I've been searching for a number of years for
>> someone who can speak intelligently on the topic. As I understand it,
>> statements like this are parroted simply due to key sizes. While
>> ssh-keygen can only create DSA keys of 1024 bits, openssl can
>> generate arbitrarily large DSA keys that can be used with OpenSSH.
>> 
>> Do you know of any specific weaknesses of DSA? If DSA is inherently
>> weak, wouldn't that also render ECDSA similarly weak?
> 
> DSA fails horribly if you ever use a key on a system with a broken PRNG.
> Since PRNGs are obviously a prime target for subversion, my gut feeling
> would be not to touch DSA with a 10-foot pole.
> 
> 
> http://www.debian.org/security/2008/dsa-1571 "Furthermore, all DSA keys
> ever used on affected Debian systems for signing or authentication
> purposes should be considered compromised;"
> 
> Good general view:
> https://www.schneier.com/blog/archives/2013/09/the_nsa_is_brea.html
> 
> Good read how PRNGs are compromised:
> https://www.schneier.com/paper-yarrow.pdf
> 
> Detail on DSA:
> https://www.schneier.com/paper-prngs.pdf
> 
> excerpts:
> 
> The DSA PRNG
> 
> Since this generator appears to come with an NSA stamp of approval, it
> has been used and proposed for applications quite different than those
> for which it was originally designed.
> 
> Summary The DSA standard’s PRNG appears to be quite secure when used in
> the application for which it was designed: DSA signature parameter
> generation. However, it doesn’t perform well as a general-purpose
> cryptographic PRNG because it handles its inputs poorly, and because it
> recovers more slowly from state compromise than it should.
> 
> 
> 
>> Bonus points if you work for the NSA and have something to disclose.
> 
> Haha! No can't say I do or would for that matter. But you may have 2nd
> thoughts about using anything they approve of.
> 
> --
> Steve
> 

Attachment: smime.p7s
Description: S/MIME cryptographic signature




-------------------------------------------
smartos-discuss
Archives: https://www.listbox.com/member/archive/184463/=now
RSS Feed: https://www.listbox.com/member/archive/rss/184463/25769125-55cfbc00
Modify Your Subscription: 
https://www.listbox.com/member/?member_id=25769125&id_secret=25769125-7688e9fb
Powered by Listbox: http://www.listbox.com

Reply via email to