Re: I-D Action:draft-ietf-nsis-ntlp-sctp-08.txt

Roland Bless <[email protected]> Wed, 20 Jan 2010 20:52:38 +0100
Newsgroups gmane.ietf.nsis
Organization Institute of Telematics, Karlsruhe Institute of Technology (KIT)
Message-ID <[email protected]>
Hi Jukka and all,

Jukka Manner wrote:
> The SCTP transport for GIST has been on our agenda. However, there has
> been little discussion about it in the last few years. The chairs and
> our AD want to see active discussion and clear support for the work to

We support this draft. We also implemented it, i.e., our implementation
supports SCTP, too. The DTLS part is new, however, though we have to
check it...

> warrant us spending more cycles on it, similar to the tunneling
> document. We already have TCP and UDP available, SCTP (and DCCP) has
> well-known deployment challenges, and lack of clear use case and
> "client" for the work. If there is no strong support, we will drop the

I think that SCTP can be beneficially used within or between network
domains for longer term MAs, i.e. when NAT traversal isn't an issue.
Since SCTP was primarily designed for transport of signaling data, e.g.,
one can advantageously multiplex different NSLP sessions onto different
streams.

Regards,
 Roland