The benefits and tradeoffs are different for IPv6, which discards header checksums, making the pseudo-header check more important for reliability reasons.
Not convinced at all. If IP source or destination address bits get corrupted, a link-layer CRC will find it before the destination will ever get the packet.
Besides, shouldn't NAT be difficult?
With enough thrust, anything can fly. Doesn't mean it should. ;-) Dino
<http://www.ee.surrey.ac.uk/Personal/L.Wood/><[email protected]> -----Original Message----- From: [email protected] on behalf of Dino Farinacci Sent: Wed 2009-04-15 21:22 To: RJ Atkinson Cc: IRTF Routing RG Subject: Re: [rrg] Eliot's note > Frankly, the claim is just silly. There are *zero* security > benefits from having the transport-layer pseudo-header checksum > include the IP address. Zip. None. Zero. I would add there are very little *other* benefits as well. Definitely not at the cost of unnecessary complexity implementors have to go through to checksum UDP messages. Dino
_______________________________________________ rrg mailing list [email protected] http://www.irtf.org/mailman/listinfo/rrg
