Re: [mpls] comments on draft-kompella-mpls-special-purpose-labels-02

[Jeff Wheeler <[e-mail]>]

On Fri, Mar 22, 2013 at 11:03 PM, Curtis Villamizar <[e-mail]> wrote:
> Many LSR use a table for the ILM.  Since the receving end specifies
> which label to use in both LDP and RSVP-TE (and PW), an LSR with less
> than the full 20 bits of space can be assured that all labels it will
> ever see will fall within that table.

I don't think anyone has suggested this is a problem for PE -- except
carrier-of-carrier PE.  Ignoring that case for now, keep in mind that
a PE isn't going to signal willingness to accept entropy labels, or
any future feature, that might require the use of extended
special-purpose labels.

Aside from carrier-of-carrier issues, there is no sense in even
discussing PEs.  Either they'll support it, or they won't; but PEs are
in a position to choose.

> I'm liking Pablo's suggestion a little bit because of the small
> extension to the space, except for the problem of LSR that don't know
> the meaning of 15 and then misinterpret the next label (for example,
> as ELI).  So I like using 15 and then a small space such as 17-1023.

The 15 prefix is my gripe.  I do not believe hardware exists that
can't inter-operate in a P role if 16-1023 (or whatever) suddenly
became reserved labels.  If the community can't come up with a single
example of such hardware, then it shouldn't penalize everyone with
this ridiculous prefix label.

Plenty of routers exist that are unwilling to look at a label stack
deeper than 3..5 labels.  This draft just adds to label bloat, and
that costs data-plane processing time or makes features impossible to
implement without new hardware.


I also would like to understand why several consecutive 15 labels
should be processed.  That seems pointless.  Is there a reason for it?

--
Jeff S Wheeler <[e-mail]>
Sr Network Operator  /  Innovative Network Concepts