[Cerowrt-devel] [Bloat] if we're going to break middleboxes with TFO... why not go for more gusto?

Eggert, Lars lars at netapp.com
Mon Jan 14 03:32:16 EST 2013


Hi,

On Jan 12, 2013, at 20:45, Dave Taht <dave.taht at gmail.com> wrote:
> I got pointed at rfc6013 (TCPCT) recently which looked like it solved
> a raft of problems TFO doesn't solve.

also note that it is not an IETF standard but an independent submission to the RFC Editor. (That's not to say it's a bad proposal.)

IIRC, the author did not want to work on this in the IETF, the result is that TCPCT uses the experimental option numbers and must not be used over the wider Internet. Unless the author submits it to the TCPM WG, TCPCT is unlikely to get an option number assigned by IANA.

Lars

> 
> Biggest problem it has though is that it eats all the remaining TCP
> option space. I wonder if anyone tried implementing something similar
> over UDP?
> 
> More readable detail than the rfc here:
> 
> http://static.usenix.org/publications/login/2009-12/openpdfs/metzger.pdf
> 
> 
> -- 
> Dave Täht
> 
> Fixing bufferbloat with cerowrt: http://www.teklibre.com/cerowrt/subscribe.html
> _______________________________________________
> Bloat mailing list
> Bloat at lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/bloat




More information about the Cerowrt-devel mailing list