Re: (ASCEND) Ascend RADIUS - what now?

[Matt Holdrege <matt@ascend.com>]

At 11:55 PM 3/4/99 -0500, Jason Nealis wrote:
>Ascend has always stated they are always standard compliant, I'm sure they
>will move up their bogus attributes, In my mind they don't have a choice if
>they want to continually make the statement about standard compliant.
>
>
>IMO, The right thing is to stay compliant with the IETF spec's.

Normally, but in the case of Radius, the IETF failed. Some of the purists
in the IETF really hated the direction that Radius went and all the work
that Ascend and other vendors wanted to do with Radius was stalled. It's
been years since any significant Radius changes have made it into RFC's. In
fact we still only have the two RFC's for Radius and Radius Accounting from
April 1997. Check out http://www.ietf.org/html.charters/radius-charter.html

Ascend and many other vendors have had lots of ideas for improving Radius.
We've wanted to standardize all the special Radius attributes that we added
as result of our customers requests. But the Radius working group chair and
the IETF area directors were unresponsive. I believe that much of this is
detailed in the Radius mailing list archives.

Unfortunately the IETF has not fixed the Radius working group nor have they
allowed any other AAA protocol to begin. They are just now starting a
generic AAA working group, but it has a very wide scope which (IMHO) will
prevent any real replacement for Radius for a very long time.

So in the meantime we added all those so called bogus attributes in
response to customer demand. Also due to customer demand we've now added
VSA support. It's there both in the 7.0 TAOS code and in our commercial
Radius server. The free unsupported Radius server that we offer is just
that. Free and unsupported. Most of our customers that use it end up
customizing it for themselves and I believe it would be pretty easy to add
on VSA support for yourself if you know C.

Part of my job is to make sure we comply with standards and to help steer
the standards to suit the developing network architectures of our
customers. I just hosted on behalf of Ascend the most recent IETF meeting
in Minneapolis and I can tell you that we are heavily involved in all of
the working groups related to our customers networks. But in the case of
Radius, we didn't fail. The IETF did.

Going forward, we will continue to work with each of our customers to make
sure that their networks work with whatever AAA configuration their
business requires.

++ Ascend Users Mailing List ++
To unsubscribe:	send unsubscribe to ascend-users-request@bungi.com
To get FAQ'd:	<http://www.nealis.net/ascend/faq>