> because with ipsec the content inside the encrypted payload (or even
> the content after the AH header) is not intended to be played with by
> anyone along the path. Detection of bit flippage in the packet is the
> point of ipsec.
> 
> BGP requires that folk along the path adjust metrics, communities,
> localpref (inside their ASNs) etc. It's not appropriate to 'secure the
> whole of the update' since that can not be done.

But you're missing the point. Sandy said, "all we're trying to do is
secure the existing protocol semantics." As Sandy herself said, the only
"semantics" in IP are things that are contained within the header. So
why not just secure the header.

So the "securing the semantics" argument won't fly. Tell me what you
want to secure, not how you intend to secure it. If the charter says,
"we want to prove an update followed the AS Path listed in the update,"
then I will ask why we shouldn't prove the rest of the semantics, as
well --who added what communities, who removed communities, and were
they authorized to do so; who changed the next hop, and is the next hop
really the correct and authorized one; etc. --these are all part of the
"semantics" of BGP, as well.

If you're going to say, "secure the semantics," then secure all of them.
If you're going to say, "secure the data," then figure out what matters
in terms of how the data looks, and secure that.

:-)

Russ

Attachment: signature.asc
Description: OpenPGP digital signature

_______________________________________________
sidr mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/sidr

Reply via email to