The MPLS WG Archive

Cell Relay Retreat>MPLS WG Archive>month:2000-Dec> msg00262



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

MPLS and fragmentation...

  • From: Brandon Ross <bross@netrail.net>
  • Date: Thu, 14 Dec 2000 18:12:40 -0500 (EST)
  • cc: Ben Black <ben@layer8.net>, mpls@UU.NET

On Thu, 14 Dec 2000, Eric Gray wrote:

> If the major market leaders do not feel that it is necessary to support
> one mechanism to do MTU discovery, what is to make anyone feel that they
> will support a new one? 

Because if that's the case (and I'm not convinced it is), then it's
fundamentally and operationally broken.  PMTU will not work if the ingress
LSR does not either fragment packets to fit through the LSP or in the case
of DF packets, drop and send ICMP needs fragmentation messages.

I do know for a fact that Juniper's GRE functionality is broken for
similar reasons, or at least it was the last time I tried it.

Brandon Ross                                                 404-522-5400
EVP Engineering, NetRail                           http://www.netrail.net
AIM:  BrandonNR                                             ICQ:  2269442
Read RFC 2644!