Hosting Provided By

[Hipsec] Pseudo Header on IPv4

From: Derek Fawcus <dfawcus(at)cisco.com>
Date: Tue Mar 18 2003 - 17:31:58 EST

On Mon, Mar 17, 2003 at 09:22:57PM -0800, Henderson, Thomas R wrote:
> Sec 3.2. Local Scope Identity

Actually, I'd like to suggest another idea...

How about always using the IPv6 style of pseudo header calculation, even when sending IPv4 packets. i.e. use the HITs in the checksum calc.

The reason for this is that it would seem to give us the equiv. of NAT-PT for free - i.e. we can connect from a IPv6 host to an IPv4 host and still have correct end2end checksum. We'd still need something like SIIT for the ICMP transformation...

Given that I'd expect the pseudo header to be precalculated when the HIP session is started, I don't see any perf issue. Just a simple matter of coding :-)

Thoughts?

Hipsec mailing list
Hipsec@lists.freeswan.org
http://lists.freeswan.org/mailman/listinfo/hipsec Received on Tue Mar 18 18:34:51 2003

Do you need help?

This message: [ Message body ]
Next message: Henderson, Thomas R: "[Hipsec] Re: The exact method used to generate a HIT from a HI"
Previous message: Derek Fawcus: "[Hipsec] One HIP session for multiple flows, or one per flow?"
In reply to Henderson, Thomas R: "[Hipsec] comments on new HIP draft"
Next in thread: Pekka Nikander: "Re: [Hipsec] Pseudo Header on IPv4"
Reply: Pekka Nikander: "Re: [Hipsec] Pseudo Header on IPv4"

Contemporary messages sorted: [ By Date ] [ By Thread ] [ By Subject ] [ By Author ] [ By messages with attachments ]

This archive was generated by hypermail 2.1.8 : Wed Aug 23 2006 - 12:59:58 EDT