Don't you need anyway an internal interface to say that tcp-eno has to set the
"b" bit?
That's simply saying to tcp-eno that this side will be the host A. Isn't this
sufficient? Or do I miss something?
Mirja
On 27.08.2015 16:49, David Mazieres expires 2015-11-25 PST wrote:
Mirja Kühlewind <[email protected]> writes:
Hi David,
I believe the point is, if you have already broken the tie via
out-of-band signal and both endpoints have already decided who will be
the opener (host A) and responder (host B), why do you still need to
write this information in the tcp-eno option if this information is
already known to the host?
Because without the "b" bit, TCP-ENO has no way of knowing that it's
known to the host. I'm operating under the assumption that we still
want to fall back to a working unencrypted TCP connection in the event
that the endpoints do not properly break the tie. Is that not the case?
David
_______________________________________________
Tcpinc mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/tcpinc
