Re: [radext] Review request for draft-ietf-softwire-map-radius

[Stefan Winter <[e-mail]>]

Hello,

and here is a quick question to begin with: the document states that the
attributes can occur in an Accounting-Request packet. However, nowhere
in the document is the corresponding semantics of that transmission defined.

I.e. why would a BNG or other RADIUS Client send this in an
Accounting-Request? What would it want to achieve? And how does the
RADIUS server react if it gets conflicting information there (say, IP
addresses in the sub-TLVs differ in the Accounting-Request from what was
sent in the preceding Access-Accept)?

Greetings,

Stefan Winter

Am 12.11.18 um 13:25 schrieb Stefan Winter:
> Hello radext,
>
> we have received a request for review of an I-D from the softwire
> working group, please see below.
>
> I would appreciate if you could take a few moments of your time to
> review this document and respond here to the working group, with the
> draft's mail address in cc.
>
> Greetings,
>
> Stefan Winter
>
> -------- Weitergeleitete Nachricht --------
>
> We have the following wg document, RADIUS Attributes for Address plus
> Port (A+P) based Software Mechanisms.
> https://datatracker.ietf.org/doc/draft-ietf-softwire-map-radius/
>
> There has been a lot of work on getting draft ready for publication from
> the opinion of softwire wg. However, because there are several new
> attributes and a lot of sub-TLVs, we think we need to get RADEXT to help
> review, before we advance this doc to IESG.
>