I added my comments to the bug.
On Fri, Apr 24, 2020 at 1:50 AM Sebastian Moeller wrote:
>
> Dear all,
>
>
> > On Apr 24, 2020, at 07:49, Hagen Paul Pfeifer wrote:
> >
> > * Matt Taggart | 2020-04-23 15:47:19 [-0700]:
> >
> >> https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=890343#34
> >>
> >>
--- Begin Message ---
On Sat, Apr 25, 2020 at 8:52 PM Jonathan Morton
wrote:
> > On 26 Apr, 2020, at 3:36 am, Dave Taht wrote:
> >
> > I just did a rather large dropbox download. They are well known to be
> > using bbr and experimenting with bbrv2. So I fired off a capture
> > during a big dropb
> On 26 Apr, 2020, at 3:36 am, Dave Taht wrote:
>
> I just did a rather large dropbox download. They are well known to be
> using bbr and experimenting with bbrv2. So I fired off a capture
> during a big dropbox download...
>
> It negotiated ecn, my fq_codel shaper and/or my newly ath10k
> fq_co
I just did a rather large dropbox download. They are well known to be
using bbr and experimenting with bbrv2. So I fired off a capture
during a big dropbox download...
It negotiated ecn, my fq_codel shaper and/or my newly ath10k
fq_codel's wifi exerted CE, osx sent back ecn-echo, and the rtt
resu
On Sat, Apr 25, 2020 at 1:49 PM Kenneth Porter wrote:
>
> --On Saturday, April 25, 2020 12:19 PM -0700 Dave Taht
> wrote:
>
> > in /etc/config/sqm, with fq_codel, you need to uncomment or enter the
> > following lines
> >
> > option ingress_ecn 'ECN'
> > option egress_ecn 'ECN'
>
--On Saturday, April 25, 2020 12:19 PM -0700 Dave Taht
wrote:
in /etc/config/sqm, with fq_codel, you need to uncomment or enter the
following lines
option ingress_ecn 'ECN'
option egress_ecn 'ECN'
Does this only affect flows that terminate on the box, or also those for
whic
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 sh
> On 25 Apr, 2020, at 8:24 pm, Y via Bloat 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_co
--- Begin Message ---
Hi,all
ECN on
http://www.dslreports.com/speedtest/62823326
ECN off
http://www.dslreports.com/speedtest/62823112
Thanks.
Yutaka
On Sat, 25 Apr 2020 17:00:19 +0300
Jonathan Morton wrote:
> > On 25 Apr, 2020, at 4:49 pm, Kenneth Porter wrote:
> >
> > before:
> >
> > http
Hi Ken,
I tried to summarize a few recommendations under
https://forum.openwrt.org/t/sqm-qos-recommended-settings-for-the-dslreports-speedtest-bufferbloat-testing/2803
maybe that is of some help...
Good luck
Sebastian
On 25 April 2020 17:59:47 CEST, Kenneth Porter wrote:
>On 4/25/2020
On 4/25/2020 8:55 AM, Dave Taht wrote:
You would be the first person in the history of the bufferbloat
effort, to attempt enabling ecn more fully on windows, and
observing the results on real traffic and real traffic types.
be prepared to take a lot of packet captures.
/me passes the vodka
I'
On 4/25/2020 9:00 AM, Dave Taht wrote:
Oh, I misread your report. I thought this was cake, not fq_codel. Care
to try that?
I'd love to if I knew how to add cake to CentOS 7. I've installed kernel
modules for unsupported Ethernet interfaces before, so perhaps cake is
available that way from a
On Sat, Apr 25, 2020 at 8:59 AM Kenneth Porter wrote:
>
> On 4/25/2020 8:43 AM, Dave Taht wrote:
> > and the last flaw of this test series is that ken took the dslreports
> > "fiber" setting for the dslreports test as "The right thing". the
> > "fiber" test is structured to stress test an asymmetr
On Sat, Apr 25, 2020 at 6:49 AM Kenneth Porter wrote:
>
> before:
>
> http://www.dslreports.com/speedtest/62767361
>
> after:
>
> http://www.dslreports.com/speedtest/62803997
>
> Using simple.qos with:
>
> UPLINK=45000
> DOWNLINK=42500
>
> (The link is supposed to be 50 Mbps symmetric and speed te
On 4/25/2020 8:43 AM, Dave Taht wrote:
and the last flaw of this test series is that ken took the dslreports
"fiber" setting for the dslreports test as "The right thing". the
"fiber" test is structured to stress test an asymmetric 1gbit/100mbit
connection, not a shaped fiber connection running at
On Sat, Apr 25, 2020 at 8:53 AM Kenneth Porter wrote:
>
> On 4/25/2020 8:32 AM, Dave Taht wrote:
> > It would be a useful test for an intrepid windows administrator to
> > actually enable ecn fully and see what breaks in their
> > vpn, smb, and rdp implementations, and across their remote workforc
On 4/25/2020 8:32 AM, Dave Taht wrote:
It would be a useful test for an intrepid windows administrator to
actually enable ecn fully and see what breaks in their
vpn, smb, and rdp implementations, and across their remote workforce,
and observe any difference in QoE. I have long
tried to get one dr
and the last flaw of this test series is that ken took the dslreports
"fiber" setting for the dslreports test as "The right thing". the
"fiber" test is structured to stress test an asymmetric 1gbit/100mbit
connection, not a shaped fiber connection running at 50mbit symmetric.
The number of uploads
On 4/25/2020 7:19 AM, Jonathan Morton wrote:
In Linux and OSX, to make the setting persist across reboots, edit
/etc/sysctl.conf.
For the lurkers, CentOS (and presumably other Red Hat distros) now has
an /etc/sysctl.d for vendor-specific settings. The prevents modifying
package-supplied conf
On Sat, Apr 25, 2020 at 8:31 AM Kenneth Porter wrote:
>
> On 4/25/2020 7:19 AM, Jonathan Morton wrote:
> > No, not the qdisc (where ECN is enabled by default), but on the client.
>
> I enabled it on both the router (Linux) and the client where I'm
> conducting the speed test (Win10):
>
> http://ww
Trying to improve the dslreports misguided "Quality" metric by enabling ecn is
essentially "engineering to the test" rather than providing much
actual benefit. It would be rather useful
at the moment to do that test given the 50/50 mbit symmetric nature of
this one, and also collect QoE metrics
ove
On 4/25/2020 7:19 AM, Jonathan Morton wrote:
No, not the qdisc (where ECN is enabled by default), but on the client.
I enabled it on both the router (Linux) and the client where I'm
conducting the speed test (Win10):
http://www.dslreports.com/speedtest/62818220
> On 25 Apr, 2020, at 5:14 pm, Kenneth Porter wrote:
>
> I see "ecn" in the qdisc commands.
No, not the qdisc (where ECN is enabled by default), but on the client.
Linux:
# sysctl net.ipv4.tcp_ecn=1
Windows:
> netsh interface tcp set global ecncapability=enabled
OSX:
$ sudo sysctl -w net.inet
--On Saturday, April 25, 2020 6:00 PM +0300 Jonathan Morton
wrote:
Looks like a definite improvement. The Quality grade of C may indicate
that you haven't enabled ECN on your client; without it, Codel has to
drop packets to do congestion signalling.
I see "ecn" in the qdisc commands. Here's
> On 25 Apr, 2020, at 4:49 pm, Kenneth Porter wrote:
>
> before:
>
> http://www.dslreports.com/speedtest/62767361
>
> after:
>
> http://www.dslreports.com/speedtest/62803997
>
> Using simple.qos with:
>
> UPLINK=45000
> DOWNLINK=42500
>
> (The link is supposed to be 50 Mbps symmetric and sp
before:
http://www.dslreports.com/speedtest/62767361
after:
http://www.dslreports.com/speedtest/62803997
Using simple.qos with:
UPLINK=45000
DOWNLINK=42500
(The link is supposed to be 50 Mbps symmetric and speed test does show it
bursting that high sometimes.)
> On 25 Apr, 2020, at 4:16 am, Kenneth Porter wrote:
>
> Alas, CentOS 7 lacks cake. It does have fq_codel so I used the simple.qos
> script from sqm-scripts, with uplink 5 and downlink 45000:
>
> http://www.dslreports.com/speedtest/62797600
Those bandwidth settings are definitely too high;
27 matches
Mail list logo
