On Mon, Jul 28, 2014 at 6:08 AM, Derek Fawcus <
[email protected]> wrote:

> On Mon, Jul 28, 2014 at 08:57:10AM +0200, marcelo bagnulo braun wrote:
> > One of them is whether to protect or not the TCP header.
>
> Yes.  At least the RST flag.


Unfortunately RST is precisely the situation that's most problematic,
because it's also how the other side behaves when it has lost state,
perhaps due to a reboot. So, it seems like we at least want RST
protection to be optional. And if we're not protecting RST, it's
generally not worth protecting other headers, AFAICT.

-Ekr


> I would like to start the discussion on this topic. Arguments on one way
> > or the other?
>
> I'd like to be able to know that a RST was generated by the remote peer,
> and not simply spoofed by a middlebox.
>
> If the middlebox wants to break the connection,  it can do so by
> blackholing (and maybe ICMP unreachable) the traffic after it
> decides the connection should be dropped.
>
> At least that way,  the two scenarios are distinguishable.
>
> .pdf
>
> _______________________________________________
> Tcpinc mailing list
> [email protected]
> https://www.ietf.org/mailman/listinfo/tcpinc
>
_______________________________________________
Tcpinc mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/tcpinc

Reply via email to