The MPLS WG Archive

[Giles Heron <giles@packetexchange.net>]

On Wed, 2002-07-24 at 18:22, Jothilingam, Vasudevan (Vasu) wrote:
[snip]
> I would like to say, yeah RSVP-TE is the way adn CR-LDP could be
> informational RFC.
> Being said that, there is lots of work going of L2 Transportation over MPLS,
> using LDP.

LDP != CR-LSP

Giles

> Well, even though I am not for that approach of using RSVP-TE and LDP ( two
> big protocols ??)
> for L2 transport, we need to see where that goes..
> 
> Thanks
> Vasu
> 
> 
> 
> 
> Vasu Jothilingam
> Lucent Technologies
> 1 Robbins Rd
> Westford, MA-01886
> PH:   (978)-952-7533
> FAX: (978)-952-7977
> 
> 
> -----Original Message-----
> From: Ong, Lyndon [mailto:LyOng@ciena.com]
> Sent: Wednesday, July 24, 2002 2:00 PM
> To: 'David Charlap'; IETF MPLS List
> Subject: RE: initiating a discussion on mpls singaling protocols
> 
> 
> Hi David,
> 
> Actually, SCTP is an alternative that supports message
> passing and boundary preservation, plus other features
> for reliability and detection of signaling path failure.
> It was intended to relieve the application from having to
> do continual verification of the signaling path and support
> of a backup path.  However, it was being defined at about
> the same time as MPLS, so it was never considered as an
> alternative for transport of MPLS signaling.  
> 
> A reliable UDP was one of the alternatives considered when doing
> SCTP, but it still leaves the application to do testing
> of the signaling path and message rerouting if that fails, 
> and does not have congestion control built in.
> 
> Cheers,
> 
> Lyndon
> 
> -----Original Message-----
> From: David Charlap [mailto:david.charlap@marconi.com]
> Sent: Wednesday, July 24, 2002 7:45 AM
> To: IETF MPLS List
> Subject: Re: initiating a discussion on mpls singaling protocols
> 
> As for the rest, TCP is not an appropriate transport for signaling 
> protocols in the first place.  RSVP-TE, LDP and CR-LDP all use a 
> message-passing mechanism.  They are not designed around streaming data. 
>   Use of a streaming protocol means that the protocol has to deal with 
> the issue of detecting message boundaries and half-received messages.
> 
> Unfortunately, there is no standard for anything resembling reliable UDP 
> - that would be ideal.
> 
-- 
=================================================================
Giles Heron    Principal Network Architect    PacketExchange Ltd.
ph: +44 7880 506185              "if you build it they will yawn"
=================================================================