] I'd go a bit further. I messed around with a large bridged
] network for some years, and this included messing with ARP
] proxies and all the troubles they cause. Basically, making
] a level 2 device simulate level 3 functions is a kludge, and it
] gets even worse when attempting to "bridge" different LAN
] technologies.

Lately I find myself wondering if there's not room for a uniform layer 2.5
interface that is designed to work over a set of bridged 802-style layer 2 
networks, but presents a slightly different interface to the host. So for
instance hosts would be explicitly required to announce presence on a link (no
more passive waiting for ARP requests), there would be an explicit L2.5
multicast join/leave (no more sniffing for L3 multicast requests), and there
would be a uniform way for a host to authenticate to the link (no more PPPoE),
with hubs having a way to know whether a particular host/port/address was
authenticated and the ability to switch that host's traffic differently
depending on that bit.  The same authentication interface could be used as a
gateway to control access to a larger network.

The point is to explicitly design such an interface rather than resorting to
more and more tricks.

(has this already been worked on and I just don't know about it?)

Keith
--------------------------------------------------------------------
IETF IPng Working Group Mailing List
IPng Home Page:                      http://playground.sun.com/ipng
FTP archive:                      ftp://playground.sun.com/pub/ipng
Direct all administrative requests to [EMAIL PROTECTED]
--------------------------------------------------------------------

Reply via email to