The MPLS WG Archive

Cell Relay Retreat>MPLS WG Archive>month:2003-Mar> msg00343



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

[PWE3] MPLS PID

  • From: Alia Atlas <aatlas@avici.com>
  • Date: Wed, 26 Mar 2003 18:43:34 -0500
  • Cc: "'Thomas Nadeau'" <tnadeau@cisco.com>, "'George Swallow'" <swallow@cisco.com>, "W. Mark Townsley" <townsley@cisco.com>, "Andrew G. Malis" <Andy.Malis@vivacenetworks.com>, "'mpls@uu.net'" <mpls@UU.NET>

At 01:33 PM 3/26/2003 -0800, Shahram Davari wrote:
>Tweaking IETF standards to try to support all possible proprietary hashing
>algorithms is not what I call standards development. What if a vendor decides
>to distinguish IP protocol using IP header CRC. Will we now mandate that 
>all encapsulated
>protocols should not yield a valid CRC at that specific position in the 
>header?

There is equipment that does look at the IP header CRC as well as the 
version field and various others for sanity-checking...  but I really think 
that any equipment which is sophisticated enough to compute and verify the 
IP header checksum will also check the version field.

>Egress always knows what protocol is encapsulated after the bottom label.

Not necessarily whether it is IPv4 or IPv6 except by checking the IP header.

Alia

  • References: