The MPLS WG Archive

["Matthew R. Meyer" <mmeyer@gblx.net>]

George, 

I see little value and lots of pain in changing the default
preemption methods of the overwhelming majority of the 
installed code base of MPLS boxes.  I see it unnecessarily 
stalling soft preemption, and I prefer that be avoided.

So I would fall in the 'happy where things stand' group.
(But all that should have been evident anyway).

Matthew

Thus spake George Swallow (swallow@cisco.com):

 |This thread has been quite for some time.  I'd like to see where the
 |workgroup is.  
 |
 |Also, as the primary author of RFC3209 I like to state that I believe
 |the RFC defines hard prememption, since it says one should send a RESV
 |Err message and doesn't say anything about setting the InPlace bit
 |(which I consider to be an exception that would need to be called out).
 |Further there is no discussion of 
 |
 |   a) lowering the forwarding priority
 |   b) how long to allow the situation to persist if the Source of the TE
 |      takes no corrective action.
 |
 |However, I will take a mea culpa on the draft being less than completely
 |clear.
 |
 |> But, if the WG is happy with the existing situation and drafts that is
 |> fine (although someone should write a BCP to ease the interpretation 
 |> of RFC3209).
 |
 |I don't think a BCP is needed, but any new RFC we produce will need to
 |include an applicability statement which includes the consensus
 |interpretation of RFC3209.
 |
 |So the question is, as Adrian put it, is "the WG is happy with the
 |existing situation and drafts"?
 |
 |...George
 |
 |========================================================================
 |George Swallow             Cisco Systems                  (978) 936-1398
 |                           1414 Massachusetts Avenue
 |                           Boxborough, MA 01719