[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Breaks compatablity
Most of the changes to the ipsec drafts since the last bakeoff don't
appear to break compatablity with the implementations that were tested
there, except the following.
I just thought that this item should REALLY be EMPHASIZED. As
Tero Kivinen points out:
>Did you add any clarification of the calculation of authentication
>hash? Most of the vendors in interop used only IP address instead of
>full ID payload (without generic headers, but with protocol and port
>number). I had to add compat option for that in interop...
>
> HASH_I = prf(SKEYID, g^xi | g^xr | CKY-I | CKY-R | SAp | IDii)
> ^^^^
> HASH_R = prf(SKEYID, g^xr | g^xi | CKY-R | CKY-I | SAp | IDir)
^^^^
And as Daniel Harkins responds concerning the correction to the
resolution doc incorporated into version 5:
>Thank you! I forgot to mention that. It says:
> "The entire ID payload (including ID type, port, and protocol
> but excluding the generic header) is hashed into both HASH_I
> and HASH_R."
I know that everybody we tested with at the last bakeoff did NOT
include the type, port, protocol as part of the HASH, and since
the three postings to the mailing list that mention this change,
mention this item amidst many other items, many people may have
overlooked it AND WILL NEED to correct there implementations to
bring them inline with the latest resolution draft.
Does anyone know of any other changes to the drafts that outright
break compatability with what was tested at the last bakeoff?
-dave