Re: requirements draft revision

[RJ Atkinson <rja@inet.org>]

At 11:41 27/06/01, Joe Abley wrote:
>> I think we have to be more precise. Firstly it isn't the protocol that
>> survives, it's the session. Secondly, does this include UDP? Suggestion:
>> 
>>    Multihoming solutions must provide re-homing transparency for
>>    transport-layer sessions, including protocols such as
>>    TCP and SCTP, and purely stateless protocols built on UDP.
>
>That certainly makes things clearer.
>
>> [actually what about applications built on raw IP??]
>
>If we put a re-homing transparency requirement in for those, aren't we
>mandating that end-point IP addresses must survive through a re-homing?
>Doesn't that bring us back to exactly where we are today?

        Not necessarily.  One possible approach for transport-layer
protocols, one that seems reasonable to some folks, is that the 
Protocol Control Block for a given session would not contain an 
IP address, though it might contain some other identifier(s).  Also note 
that while UDP is characterised as stateless, all implementations 
that I'm familiar with have normal Protocol Control Blocks 
even for UDP sessions -- so as implemented and deployed there 
is some non-zero UDP state already.  For protocols built on raw IP, 
they could also use some other form of identifier in lieu of an 
IP address, if such an identifier were found and agreed upon.

        [Aside: it is in this respect that it would be nice if the
folks in the IRTF NSRG could focus on producing drafts on the
matter of alternate/additional namespaces.  Right now there's
some discussion, but few if any drafts over there, which seems
like a bug.]

Ran