From: Dave Taht <dave.taht@gmail.com>
To: Jonathan Morton <chromatix99@gmail.com>
Cc: Y <intruder_tkyf@yahoo.fr>, bloat <bloat@lists.bufferbloat.net>
Subject: Re: [Bloat] this explains speedtest stuff
Date: Sat, 25 Apr 2020 11:19:17 -0700 [thread overview]
Message-ID: <CAA93jw6-rC67UtMiRVM5RSzyO40p5_+0W9Vv4YGOAppsuh2fKQ@mail.gmail.com> (raw)
In-Reply-To: <18E93BF9-44B0-444D-8B01-8210C9C85B5F@gmail.com>
in /etc/config/sqm, with fq_codel, you need to uncomment or enter the
following lines
option ingress_ecn 'ECN'
option egress_ecn 'ECN'
I still am not much of an ecn fan. I use it primarily for debugging
codel. :/ When I see marking, I know its working,
and if I see loss where I shouldn't, I know something else in the
stack is broken somewhere.
Example - I ran into this while trying to get the ath10k performing
sanely: https://forum.openwrt.org/t/aql-and-the-ath10k-is-lovely/59002/92
- dropping packets elsewhere in the stack. The "fix" got bidir
throughput up by 10x, and it's still 4x lower than it could be for...
unknown reasons.
I worry a lot about checksummng problems and incorrect encapsulation.
Hit that multiple times myself, and most recently a ddpk person hit
it. "Look ecn's enabled!" but the actual receiver was dropping it
instead because the checksum was wrong. I also hit it with 6 months of
data from all over the world on osx. The second and successive runs of
the rrul test disabled ecn neg attempts entirely for those ip
addresses. I didn't figure it out until
I sat down to write the paper and was agast at my packet captures.
osx is completely useless for evaluating ecn behaviors given the
heuristics in place on it today.
Would love to have one valid windows + openvpn + fq_codel result.
On Sat, Apr 25, 2020 at 10:37 AM Jonathan Morton <chromatix99@gmail.com> wrote:
>
> > On 25 Apr, 2020, at 8:24 pm, Y via Bloat <bloat@lists.bufferbloat.net> wrote:
> >
> > ECN on
> > http://www.dslreports.com/speedtest/62823326
> > ECN off
> > http://www.dslreports.com/speedtest/62823112
>
> Yup, that's what I mean.
>
> > doesn't appear to have worked. retransmits are still high.
>
> Ken, it might be that your version of fq_codel doesn't actually have ECN support on by default. So try adding the "ecn" keyword to the qdisc.
>
> - Jonathan Morton
> _______________________________________________
> Bloat mailing list
> Bloat@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/bloat
--
Make Music, Not War
Dave Täht
CTO, TekLibre, LLC
http://www.teklibre.com
Tel: 1-831-435-0729
next prev parent reply other threads:[~2020-04-25 18:19 UTC|newest]
Thread overview: 46+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-04-23 19:38 Dave Taht
2020-04-23 20:00 ` Jonathan Foulkes
2020-04-23 20:15 ` Sergey Fedorov
2020-04-23 20:27 ` Dave Taht
2020-04-24 1:03 ` Sergey Fedorov
2020-04-24 0:44 ` Kenneth Porter
2020-04-24 0:47 ` Jonathan Morton
2020-04-24 1:16 ` Kenneth Porter
[not found] ` <6F1F3C646A7DFD2CC8C9E79E@172.27.17.193>
2020-04-24 1:20 ` Dave Taht
2020-04-24 5:15 ` Matt Taggart
2020-04-24 14:40 ` Kenneth Porter
2020-04-24 14:52 ` Sebastian Moeller
2020-04-24 14:56 ` Toke Høiland-Jørgensen
[not found] ` <mailman.56.1587740202.24343.bloat@lists.bufferbloat.net>
2020-04-24 15:22 ` Kenneth Porter
2020-04-24 16:22 ` Kenneth Porter
2020-04-24 18:43 ` Jonathan Morton
2020-04-25 1:16 ` Kenneth Porter
2020-04-25 8:43 ` Jonathan Morton
2020-04-25 13:49 ` Kenneth Porter
2020-04-25 14:00 ` Jonathan Morton
2020-04-25 14:14 ` Kenneth Porter
2020-04-25 14:19 ` Jonathan Morton
2020-04-25 15:31 ` Kenneth Porter
2020-04-25 15:33 ` Dave Taht
2020-04-25 15:32 ` Dave Taht
2020-04-25 15:43 ` Dave Taht
2020-04-25 15:59 ` Kenneth Porter
2020-04-25 16:06 ` Dave Taht
2020-04-25 16:22 ` Sebastian Moeller
2020-04-26 10:59 ` Toke Høiland-Jørgensen
2020-04-25 15:52 ` Kenneth Porter
2020-04-25 15:55 ` Dave Taht
2020-04-25 16:18 ` Kenneth Porter
2020-04-25 15:38 ` Kenneth Porter
2020-04-25 17:22 ` Y
[not found] ` <mailman.73.1587835496.24343.bloat@lists.bufferbloat.net>
2020-04-25 17:37 ` Jonathan Morton
2020-04-25 18:19 ` Dave Taht [this message]
2020-04-25 20:49 ` Kenneth Porter
[not found] ` <787FB38B9372372F376BB072@172.27.17.193>
2020-04-25 20:55 ` Dave Taht
[not found] ` <3D2F4A70BAB1774D1EFDC641@172.27.17.193>
2020-04-25 16:00 ` Dave Taht
2020-04-25 16:17 ` Kenneth Porter
2020-04-26 10:51 ` Toke Høiland-Jørgensen
2020-04-24 16:25 ` Kenneth Porter
[not found] ` <42D98FE80B7B9C1102C47579@172.27.17.193>
2020-04-24 16:32 ` Dave Taht
2020-04-24 16:39 ` Dave Taht
2020-04-24 17:42 ` Kenneth Porter
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
List information: https://lists.bufferbloat.net/postorius/lists/bloat.lists.bufferbloat.net/
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=CAA93jw6-rC67UtMiRVM5RSzyO40p5_+0W9Vv4YGOAppsuh2fKQ@mail.gmail.com \
--to=dave.taht@gmail.com \
--cc=bloat@lists.bufferbloat.net \
--cc=chromatix99@gmail.com \
--cc=intruder_tkyf@yahoo.fr \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox