eap Mailing List: Re: [Issue 248] Comments on EAP state machine v4

[Nick Petroni (npetroni]

> >Comment #14 - Technical
> >  Request text to describe the possible DoS issues and possible
> >  mitigation techniques. Specific changes to the SM necessary to
> >  achieve such mitigations would be great.
> >
> I am reluctant to provide text and modifications to the SM for this one
> (*although I already did in some of my previous mails*) because my
> understanding was that the group had not reached consensus on whether
> this issue has to be handled and how this has to be done...
Sorry, I should have been more clear on this. I think we are agreed that
no changes should take place to the SM itself. However, some of the
possible protocol changes pointed out by yourself and John Vollbrecht, and
discussed on this list, could be described at least in terms of the
vulnerabilities themselves. Jari mentioned that such a description might
be fitting:
http://mail.frascone.com/pipermail/eap/2004-June/002580.html

"As a result, I would recommend documenting the specific
vulnerabilities to accepting NAKs and Failures. I think
RFC 3748 already has some general text about this, but it
would be OK for the state machine document to talk about
specific issues related to specific state transitions.
I am a bit uneasy about changing the actual diagram or
behaviour, however. "

This seems the most prudent to me and it was the text I was requesting.

> If the group is (apart from silent) against this issue (and the stupid
> DoS ones in general) which seems to be the case IINM, then I might want
> to save my text/modifications for some other purposes... ;-)
I agree modifications would not be the consensus of the group, but I think
everyone recognizes these attacks are feasible.

Thanks,
nick