[Cake] [Cerowrt-devel] https://tools.ietf.org/html/draft-ietf-tsvwg-le-phb-06 is in last call

Dave Taht dave.taht at gmail.com
Mon Feb 4 05:23:53 EST 2019


On Sun, Feb 3, 2019 at 11:11 PM Mikael Abrahamsson <swmike at swm.pp.se> wrote:
>
> On Sun, 3 Feb 2019, Dave Taht wrote:
>
> > Well, (I just checked, let me know if you want the captures) comcast
> > still re-marks all codepoints it does not recognize, to become CS1,
> > including this one. So the smartest thing a comcast customer can do is
> > wash it out on entrance to their domain.
>
> If I didn't think we ever could change things, I wouldn't support
> documents like these.

I support this document. I note sqm and cake are the *only* QOS
systems out there that have ever bothered to pay attention to the
diffserv standards. I don't think netduma's "anti-bufferbloat" does,
nor does the mq, prio, etc qdiscs in linux.


>
> The idea is to go to the ISP community and say "Please treat LE as BE in
> your core/peering links, and only act on it on the customer access shaper
> (optional). Please don't bleach it."

It's not a please. It's a three way contract with network services,
isps, and users. And of these, the users request should be paramount.

short of a lawsuit I don't see any way comcast can be coerced to
remove the *one* line in their cmts config files that remarks traffic.
... but hey, miracles happen.

passing CS1 through to wifi really sucks. Bleaching all diffserv from
comcast at the edge makes sense until that miracle happens.

>
> 000001 has never been in use before, so the hope is that people will allow
> this to happen. CS1 has had different meanings over time so it's harder to
> handle.

I agree. CS1 has caused no end of trouble for a good idea.

> The problem with CAKE/FQ and background traffic is that it can't tell if
> there is congestion or not, and things like LEDBAT can't backoff and try
> to avoid causing congestion. So your previous email about allowing some
> congestion to take place on LE would be good as then protocols that try to
> avoid causing congestion would have a way to do so.

I do not like that the standard allows for total starvation. I would
prefer it had a minimum of 5%.

>
> --
> Mikael Abrahamsson    email: swmike at swm.pp.se
> _______________________________________________
> Cerowrt-devel mailing list
> Cerowrt-devel at lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/cerowrt-devel



-- 

Dave Täht
CTO, TekLibre, LLC
http://www.teklibre.com
Tel: 1-831-205-9740


More information about the Cake mailing list