RE: SCTP multihoming issues draft

[Coene Lode <Lode.Coene@siemens.atea.be>]

-----Original Message-----
From: Tim Chown [mailto:tjc@ecs.soton.ac.uk]
Sent: Wednesday, February 06, 2002 3:31 PM
To: Coene Lode
Cc: Transport Area wg (E-mail); SIGTRAN (E-mail); 'multi6@ops.ietf.org'
Subject: Re: SCTP multihoming issues draft


>On Wed, 6 Feb 2002, Coene Lode wrote:
>
>> If by then end of next week, no further issues have been identified, then
I
>> would ask for this draft to be moved to informational.
>
>Does the draft have to only consider two network interfaces for the
>multihoming issues?   In IPv6, as you point out, each interface can have
>more than one IPv6 address (and addresses of many scopes), so the simplest
>multihoming case for IPv6 is a host (endpoint) with one interface, 
>recieving two different /64 prefix RAs, and thus having two IPv6 addresses
>from two different providers.   The dual interface scenario may be useful 
>of course, e.g. when considering wired and air interfaces.

It considers any configuration with multiple interface cards or one
interface card with multiple addresses(in both cases possibly going to
infinity:-)
The examples are for multiple interface cards as that was the first use we
thougth of.

>
>When you say "it is recommended that IP addresses in a multihomed endpoint 
>be assigned IP endpoints from different TLA's to ensure against network 
>failure", this would be the default case for IPv6.  
>
Great, that is a plus for letting SCTP run on IPv6
>
>
>For IPv6, delete section 2.3 :-)

Unfortunaly, I have also to describe the present situation with IPv4 which
as you can derive from paragraph 2.3 is less than ideal. The last sentence
is a gentle push for people NOT to use IPv4 NAT boxes if people really want
to use multihoming. The result should be to use IPv6 for such cases.

>
>In general, more consideration could be given for IPv6, perhaps?  (you 
>did cc multi6 :-)   If I've misinterpreted the text, perhaps the title
>of the draft needs revision?

Well, this draft has to address all possibilities whether I(and my
comrades-in-arms) like them or not. And the issue of multihoming of SCTP in
IPv4 networks together with NAT's is one of them. That's why the TSVWG and
SIGTRAN are on also.

And if this issue list makes that more people would use IPv6 networks in
real situation, ,that would certainly make IPv6 proponents happy.(besides
the SCTP folks)

>
>Best wishes,
>Tim
>

yours sincerly,
Lode