News for hackers. Stuff that matters.
Re:Crypto breakthrough shows Flame was designed by world-cla... by 0x0 on Thursday June 07, 2012 @02:56PM (#4080686)
What exactly was the requirement for a "0day" MD5 collision attack here; I thought the Terminal Server Licensing thing at Microsoft spat out good-for-code-signing certificates by itself?
Re:Crypto breakthrough shows Flame was designed by world-cla... by emily37 on Thursday June 07, 2012 @03:01PM (#4080724)
They did, but it seems that those certificates had some extensions that made the code-signing attack difficult to carry out on some versions of Windows, so the collisions were used to generate certs without those extensions. (I read this at http://blog.cryptographyengineering.com/2012/06/flame-certif...)
Re:Crypto breakthrough shows Flame was designed by world-cla... by 0x0 on Thursday June 07, 2012 @03:30PM (#4080899)
I see; on Vista and above, certain fields in the certificate prevent it from being accepted for code signature. So they did a collision attack to create a slightly twisted certificate where all those fields are tucked away in a useless segment and ignored.

Are there any details on what's new with this particular attack, compared to the known previously published ones? Why wouldn't earlier public research (such as that ps3 fake SSL CA stunt) suffice?

Re:Crypto breakthrough shows Flame was designed by world-cla... by marshray on Thursday June 07, 2012 @04:56PM (#4081294)
I believe the phrase you're searching for is "awesome applied crypto research." :-) http://www.win.tue.nl/hashclash/rogue-ca/

We don't yet have details on the differences. I've looked at the evil colliding cert and, AFAICT, the "MD5 Considered Harmful" technique would probably have been sufficient to pull this off.

Re:Crypto breakthrough shows Flame was designed by world-cla... by 0x0 on Thursday June 07, 2012 @05:12PM (#4081363)
Thanks for the clarification.

The rogue-ca thing is awesome indeed! :)