On Wed, Jun 1, 2016 at 3:39 PM, John Nagle <[email protected]> wrote:

>
>>     What I want to do is to get people thinking about this as a
> solveable problem. There's a general impression that MITM attacks are
> fundamentally not detectable.


That might be a general impression among the community, but there are
people working on it. There is a "channel binding" standard that generates
a keypair that is bound to a particular TLS session [1], and this made it
into U2F:

https://fidoalliance.org/fido-technotes-channel-binding-and-fido/

Which means that the U2F token can include its "view" (the fingerprint of
the TLS session key seen by the client) of the TLS session in the package
it signs and sends up to the server. If the server's TLS session has a
different fingerprint, then it follows that there's an intermediary session
in the middle.

The "token binding" working group referenced there is where effort is being
directed now:
https://datatracker.ietf.org/wg/tokbind/documents/

Not saying it fully meets your goals, just pointing out that there is
active work on this issue, and that versions of this work have been
deployed in production via U2F.



>     I'm posting this because CA problems such as the Symantec/Blue Coat
> cert are becoming more common.


As discussed on another thread on this mailing list[2], the Symantec / Blue
Coat cert wasn't a CA problem. It was disclosed by Symantec, included in
their audits, and Symantec's possession of the key didn't give Blue Coat
powers to generate certificates for domains not under their control.

-- Eric

[1] https://tools.ietf.org/html/rfc5929
[2]
https://groups.google.com/d/msg/mozilla.dev.security.policy/akOzSAMLf_k/Y1D4-RXoAwAJ


>
>                                 John Nagle
>
> _______________________________________________
> dev-security-policy mailing list
> [email protected]
> https://lists.mozilla.org/listinfo/dev-security-policy
>



-- 
konklone.com | @konklone <https://twitter.com/konklone>
_______________________________________________
dev-security-policy mailing list
[email protected]
https://lists.mozilla.org/listinfo/dev-security-policy

Reply via email to