Discussion:
FYI: Dan's talk on 27c3 in Berlin
Erwin Hoffmann
2011-02-12 18:07:09 UTC
Permalink
Hi together,

for those, who are interested:

DJB gave a talk on 27c3 'Hacker congress' (at December 28th, 2010) in Berlin:

"High-speed high-security cryptography encrypting and authenticating the whole internet"

In essence, Dan

- critices DNSSec from first principles ('CIA') and explaining possible amplification attacks, and addressing the problem of static signing keys,

- introduces briefly DNSSec with ECC and NYM deployed Public Keys,

- outlines CurveCP, a new protocol, using UDP services while encrypting the payload (asymmetrically) by means of ECC. This could be used for general HTTP traffic (instead using standard TCP).

--

What is interesting, challenging, and extraordinary is the approach - unlike TLS - to directly encrypt data with ECC and not to first negotiate a shared secret for (later) symmetrical en/de-cryption. Dan tries to convince the public that asymmetric cryptography by ECC is not heavy burdon on today's CPUs.

Sources:

His talk: http://cr.yp.to/talks/2010.12.28/slides.pdf

His life presentation: http://vimeo.com/18279777

--

Interesting enough, apart from Dan's approach, Google also tries to tie down the latency introduced by TLS (for instant HTTP traffic):

http://tools.ietf.org/html/draft-agl-tls-snapstart-00

--

Thus, given the current hardware capabilities, not the CPU load is problematic for encryption, but rather the (slow) current approach, to at first set up a security context/session and negotiate on a cipher.


Enjoy!

regards.
--eh.

PS: Sorry for potentially receive this mail twice. It is worth it!
--
Dr. Erwin Hoffmann | FEHCom | http://www.fehcom.de
David Nicol
2011-02-13 18:46:10 UTC
Permalink
Post by Erwin Hoffmann
- outlines CurveCP, a new protocol, using UDP services while encrypting the payload (asymmetrically) by means of ECC. This could be used for general HTTP traffic (instead using standard TCP).
Even after reading the presentation, i don't understand why CurveCP is
envisioned as riding UDP instead of as an additional protocol, like
SCTP, except as an initial phase to transition out of.

Is it political? If DnsCurve gets UDP:53, the same process can be the
communications endpoint, in effect tunneling everything through UDP53
for maximum opacity?
Kevin Chadwick
2011-02-14 00:17:55 UTC
Permalink
Forwarded it to unbound and got this in response.

Begin forwarded message:

Date: Sun, 13 Feb 2011 13:03:32 -0500 (EST)
From: Paul Wouters
To: Kevin Chadwick
Cc: unbound-***@unbound.net
Subject: Re: [Unbound-users] unbound 1.4.6 released
That was not a talk. That was a rant devoid of facts and filled with
unsubstantiated and by now disproven claims. Both me and Kaminsky
already spend too much time debunking his shit. Let's not reitterate
that nonsense here.

http://dankaminsky.com/2011/01/05/djb-ccc/

Paul
Markus Stumpf
2011-02-14 15:26:37 UTC
Permalink
Post by Kevin Chadwick
Forwarded it to unbound and got this in response.
[ ... ]
Post by Kevin Chadwick
Post by Kevin Chadwick
That was not a talk. That was a rant devoid of facts and filled with
unsubstantiated and by now disproven claims. Both me and Kaminsky
already spend too much time debunking his shit. Let's not reitterate
that nonsense here.
http://dankaminsky.com/2011/01/05/djb-ccc/
*> But these are not limitations to DNSSEC as a protocol.
*> They’re implementation artifacts, no more inherent to DNSSEC
*> than publicfile‘s inability to support PHP. (Web servers
*> were not originally designed to support dynamic content, the occasional
*> cgi-bin notwithstanding. So, we wrote better web servers!

*LMAO* Sorry, but totally *LMAO*
Please note that he didn't write "more secure" but "better".

While IMHO some of the arguments of both sides are "valid" nobody should
think that "securing" DNS is the next best thing to sliced bread.
There will be tons of servers hacked or admins tricked to give away
passwords [1] which will (in the case of large hosters or large sites (think
twitter [2])) still have the same effect as it has now.
And it will not secure https/ssl/x.509 as this is f'up by design.

[1] HBGary Federal Hacked by Anonymous
http://krebsonsecurity.com/2011/02/hbgary-federal-hacked-by-anonymous/
[2] DNS attack hijacks Twitter
http://www.theregister.co.uk/2009/12/18/dns_twitter_hijack/

\Maex
Erwin Hoffmann
2011-02-14 20:20:18 UTC
Permalink
Hi,

On Mon, 14 Feb 2011 00:17:55 +0000
Post by Kevin Chadwick
Forwarded it to unbound and got this in response.
Date: Sun, 13 Feb 2011 13:03:32 -0500 (EST)
From: Paul Wouters
To: Kevin Chadwick
Subject: Re: [Unbound-users] unbound 1.4.6 released
Paul Wouters:

" That was not a talk. That was a rant devoid of facts and filled with
unsubstantiated and by now disproven claims. Both me and Kaminsky
already spend too much time debunking his shit. Let's not reitterate
that nonsense here."

Just took a look at Paul's presentation at the Black Hat DC 2009 (!) [to be curious about his arguments]:

http://secdocs.lonerunners.net/documents/details/415-defending-your-dns-in-a-post-kaminsky-world

"Why DNSCURVE sucks ...
...
I still need to punch him in the face for qmail"

Palpable arguments!

regards.
--eh.
Post by Kevin Chadwick
http://dankaminsky.com/2011/01/05/djb-ccc/
Paul
--
Dr. Erwin Hoffmann | FEHCom | http://www.fehcom.de
Loading...