[Cryptography] Long-term security (was Re: ratcheting DH strengths over time)

John-Mark Gurney jmg at funkthat.com
Tue Nov 17 02:32:59 EST 2015


Peter Gutmann wrote this message on Tue, Nov 17, 2015 at 04:22 +0000:
> John-Mark Gurney <jmg at funkthat.com> writes:
> 
> >Then I thought about it, putting any type of security on it gave the user of
> >the device a false sense of security.  There is no way that I could properly
> >secure the device from a remote attacker, so I decided that no security was
> >the best security...  This would hopefully require the user to use a bastion
> >host to access the device..
> 
> That's the "false sense of security" bugbear I mentioned in an earlier post.
> What about just moving comms to a nonstandard port, or (better) requiring port
> knocking, or at least doing something?  Relying on someone else to secure it
> means you're going to end up with your totally unprotected device sitting on
> the public Internet.  Even just using a nonstandard port will protect against
> port-scanning attacks, and then you can add further defences as required, with
> each one boxing the attack onto a smaller and smaller attack surface.  This
> isn't a "false sense of security", this is actual security that's reducing
> your attack surface, even if it's not theoretically perfect cryptographic
> security.

Only changes your attack surface...  Most of the ones mentioned will
easily fall to passive listening...  What happens when it's well known
that a device needs to be port knocked to get into?  Then some attacker
will just passively listen in to the knock sequence...  If you say make
it cryptographicly secure, w/ challenge, then we're back to implementing
crypto...  Easier than ssh, but still complicated, and even then, if
someone MitM's while you're connected, you've lost the ability to
authenticate the communications...

Either you implement full crypto to get security, or you're vulnerable
to any number of attacks...  Some of those attacks may fall outside
you're threat model, but won't for others..

-- 
  John-Mark Gurney				Voice: +1 415 225 5579

     "All that I will do, has been done, All that I have, has not."


More information about the cryptography mailing list