Possible crypto backdoor in RFC-2631 Diffie-Hellman Key Agreement Method

[Georgi Guninski]

On Thu, Sep 03, 2015 at 11:59:11AM +0000, Peter Gutmann wrote:
> One saving grace about RFC 2631 was that it was pretty much universally
> ignored for the reason that it was, well, a pretty stupid way to do things, so
> the number of affected implementations would be approximately zero.
> 

Anyway, I would appreciate if someone checks if current implementations
accept composite $q$.


> (I only know of one, rather minor, vendor who implemented it.  Microsoft
> implemented it in receive-only mode solely so that they couldn't be accused of
> being non-standards-compliant, but I'd be very surprised if there was anything
> still around that supported it.  For starters you'd need to be able to find a
> CA that could issue you a DH certificate...).
> 

What do you mean by DH certificate?

Can DH sign?

> Peter.