RE: pppcn

Kehn, Doug writes:
> > Obviously, such a trick is ineffective if the authentication mechanism
> > is more obtuse than either PAP or CHAP.
> > 
> It also seems that the trick would be ineffective in the presence of
> L2TP.  Especially, when the LAC uses one authentication protocol to
> determine which LNS to setup a tunnel with and the LNS end of the tunnel
> renegotiates LCP to a different authentication protocol.

Not really.  In either case, both ends must be able to predict the
exact packet sequence that leads to success.  It doesn't matter how
that sequence is formed.

I'll grant that if the LAC doesn't bother to buffer some small number
of packets until the tunnel is established, then it'll most likely
fall back to traditional PPP negotiation.  Again, that possible
buffering issue seems like an internal quality-of-implementation
detail to me.

-- 
James Carlson, IP Systems Group                <[email protected]>
Sun Microsystems / 1 Network Drive         71.234W   Vox +1 781 442 2084
MS UBUR02-212 / Burlington MA 01803-2757   42.497N   Fax +1 781 442 1677