The MPLS WG Archive

Cell Relay Retreat>MPLS WG Archive>month:2003-Mar> msg00327



[Date Prev][Date Next][Thread Prev][Thread Next]  
  [Date Index][Thread Index][Author Index][Subject Index]

[PWE3] MPLS PID

  • From: Curtis Villamizar <curtis@fictitious.org>
  • Date: Wed, 26 Mar 2003 15:09:28 -0500
  • cc: "'David Allan'" <dallan@nortelnetworks.com>, "'Yaakov Stein'" <yaakov_s@rad.com>, "W. Mark Townsley" <townsley@cisco.com>, "Andrew G. Malis" <Andy.Malis@vivacenetworks.com>, pwe3@ietf.org, mpls@UU.NET


In message <4B6D09F3B826D411A67300D0B706EFDE0115C830@nt-exch-yow.pmc-sierra.bc.
ca>, Shahram Davari writes:
> This message is in MIME format. Since your mail reader does not understand
> this format, some or all of this message may not be legible.
> 
> ------_=_NextPart_001_01C2F3CB.763A5440
> Content-Type: text/plain;
> 	charset="iso-8859-1"
> 
> I also agree with Dave/Andy/Ron/Alia. If PID is needed, do it properly by def
> ining a standard protocol Mux header such as LLC/SNAP. Those interested in id
> entifying the protocol in the interior routers would then require to carry th
> is header in all packets.
>  
> -Shahram


LLC/SNAP is overkill.  Something along the lines of the 4 byte L3PID
would probably be better.  It might be worth adding a value in L3PID
or the GMPLS GPID to indicate that the LSP payload has the PID in the
top 4 bytes and use the same set of values rather than inventing new
number spaces for new PIDs (whether PID is payload ID or packet ID).

Curtis

  • References: