I agree that protecting against rogue middleboxes is a great goal. That said, I'm concerned about the possibility of doing it in a way that prevents authorized middleboxes from working.

The element of some of the proposals that I'm most concerned about in this regard is one that uses the end-to-end data stream authenticator to provide authentication for the TCP header. This prevents some legitimate TCP optimization middleboxes from improving data delivery rates without requiring them to participate in protection of the data stream itself, in that they must have access to the key that authenticates the data in order to rewrite the TCP headers. I prefer a solution that requires such a middlebox to have access to the smallest amount of keying material possible in order to limit its capability to make changes to anything but what it must change in order to function.

I'm not arguing against header authentication (though I'm not specifically arguing in favor of it either). I'm simply suggesting that if the headers are to be authenticated, merging header and data authenticators could mean that a broader range of middleboxes are prevented from working.

--Brandon

On 07/28/2014 11:48 AM, Joe Touch wrote:
IMO, if you're not protecting the TCP header the solutions should not be
called TCP anything, nor should it be integrated as a TCP option. It's
basically just TLS with no signature on either end, and that ought to
suffice if that's all you really want.

However, protection from rogue middleboxes is something I'd appreciate,
and that can't be done solely as a payload of the transport layer.

Joe

On 7/27/2014 11:57 PM, marcelo bagnulo braun wrote:
Hi,

As we discussed in the meeting, we should try to make some design
decisions for TCPINC.
One of them is whether to protect or not the TCP header.
I would like to start the discussion on this topic. Arguments on one way
or the other?

regards, marcelo

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

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


--
Brandon Williams; Senior Principal Software Engineer
Emerging Products Engineering; Akamai Technologies Inc.

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

Reply via email to