The MPLS WG Archive

[Nabil Seddigh <nseddigh@tropicnetworks.com>]

Hi Kireeti,

> 
> Can you explain to me "specific to Generalized MPLS"?  If you do,
> I will write the applicability statement myself :-)

I replied to this in a previous post and I think others on this
thread essentially said the samething.  Box & stack vendors 
typically like to implement all of a draft. It is much easier and
more interoperable than picking and choosing sections from a 
variety of drafts. As for an applicability draft, the thoughts
are clear in my head even if it is not being expressed clearly 
via email. If there is interest, I could put something together.

There's a lot of new spec (objects & behavious) in the gmpls drafts 
(functional & signaling) that many plain router vendors or router 
stack vendors  would not implement - ie no desire to configure 
optics or tdm via signalling. Yet the unnumbered link & 
hitless restart features may be of interest to such vendors. 
It would be nice if they didn't have to wade through GMPLS docs
too extract this info.

> 
> > and I never understood the rational for rolling them into
> > the gmpls drafts. IMHO, it is better to remove the content in
> > 2 separate drafts.
> 
> Unnumbered *is* a separate draft!  (... as is LSP hierarchy, 
> before someone says that hierarchy is specific to GMPLS.)

True enough. However, part of section 9.1 of 
http://www.ietf.org/internet-drafts/draft-ietf-mpls-generalized-signaling-09.txt
contains content that's not in the unnum draft and which 
would be incorporated by someone implementing the unnum draft.

Best,
Nabil