Re: [v4v6interim] [BEHAVE] [46translation] Proposal for new BEHAVE charter

[Mark Townsley <[e-mail]>]

Pekka Savola wrote:
> On Thu, 23 Oct 2008, Mark Townsley wrote:
>>>  If #2 does not remind you of how we got into so much trouble nat44, 
>>> well
>>>  it should. Because of the impact on NAT66 on applications, if IETF 
>>> does
>>>  decide to do NAT66 specifications, I think it is very important 
>>> that the
>>>  specification is developed not only in the context of v6ops people 
>>> but is
>>>  also developed with input of folks from applications that need to 
>>> use it.
>>>  Today that would roughly mean behave.
>>
>> I think that something that operates only at the IP layer could stay 
>> in the int-area.
>
> But it doesn't; a 1:1 NAT66 would break every app (FTP, SIP, etc.) 
> that's broken today with NAT4 (with or without port translation) 
> unless the NAT has ALGs.  To implement such ALGs, NAT66 would need to 
> operate beyond the IP layer.
Of course, anything that IP does affects things above and below. By this 
argument we could collapse whole portions of int area into transport or 
vice-versa (which may be a fait accomplis anyhow, but I digress).

All of these translator things cross layer boundaries. My argument is, 
simply, that if any of it could land on the int area side, a strictly 
1:1 NAT might be the most likely candidate. Remember also, this was in 
response also to Magnus raising his arms overhead and looking for a 
liferaft amidst the new work behave already has. Honestly, if he is 
willing to take it on in behave, that's perfectly fine with me as long 
as it receives enough attention from int-area folks. I'm far less 
concerned about that last issue right now than I was, say, 6 months or 
so ago when we first started talking about bringing IPv6 translation 
into behave.

This discussion has already gotten ahead of itself though. Let's get the 
-00 out, discuss it where best we can find the time and cycles, and then 
decide where (and if) it should land as chartered work. Margaret asked 
for a slot in int-area, perhaps there will be time in behave as well, so 
for the moment some of you might see this discussed twice - as is the 
case often with work that is yet unchartered.

- Mark

_______________________________________________
v4v6interim mailing list
[e-mail]
https://www.ietf.org/mailman/listinfo/v4v6interim