Re: [BEHAVE] Very late query on stateful NAT64

[marcelo bagnulo braun <[e-mail]>]

  El 22/10/10 17:55, Dan Wing escribió:
>
> This text is related to existing NAPT44 behavior, where a mapping
> is kept active so long as there is an inside->outside flow (which
> makes sense, and is usually 15-30 seconds) and is kept active longer
> if there is an outside->inside response (which makes sense, too,
> and is usually ~3 minutes).  The next characteristic is that the
> NAT shouldn't keep a mapping alive if there is solely and only
> an outside->inside flow.  This is discussed (but with a focus on
> RTP flows) in draft-ietf-avt-app-rtp-keepalive, and the best
> reference is probably REQ-6 of RFC4787.
>
Right, as several have stated already, this is the same as REQ 6 in 
RFC4787 and as a guideline we have in general tried to honour the behave 
requirements (while trnaslating them to the v6 v4 translation)


> Would it be helpful to add a citation to "Section 4.3 of [RFC4787]"
> during that document's AUTH48?
>
> (The wording should be tweaked, anyway, from "MAY choose not to
> extend" to "MAY choose to not extend".)
>
let me know what you want to do and i can drop a note to the RFC editor. 
I am fine with Dan's proposed changes.

Regards, marcelo


> -d
>
>
>