[Cake] How to test Cake on TP-Link WDR3600

[Cake] How to test Cake on TP-Link WDR3600

[Alec Robertson]

[-- Attachment #1: Type: text/plain, Size: 467 bytes --]

Hello,

I’ve flashed the OpenWRT image for the TP-Link WDR3600 from 2015-06-23 (http://snapon.lab.bufferbloat.net/~cero3/lupin/ar71xx/) and it has installed correctly.


How do I go about setting it up for my UK TalkTalk Fibre connection (using a VDSL modem connected to the TP-Link router: the TP-Link router gets an IP address via DHCP).


My sincere apologies if any of this is really obvious and I’ve been silly and missed it.


—
Alec Robertson

[-- Attachment #2: Type: text/html, Size: 775 bytes --]

Re: [Cake] How to test Cake on TP-Link WDR3600

[Sebastian Moeller]

[-- Attachment #1: Type: text/plain, Size: 586 bytes --]

Hi Alec,

On Jul 25, 2015, at 17:45 , Alec Robertson <alecrobertson13@gmail.com> wrote:

> Hello,
> 
> I’ve flashed the OpenWRT image for the TP-Link WDR3600 from 2015-06-23 (http://snapon.lab.bufferbloat.net/~cero3/lupin/ar71xx/) and it has installed correctly.
> 
> How do I go about setting it up for my UK TalkTalk Fibre connection (using a VDSL modem connected to the TP-Link router: the TP-Link router gets an IP address via DHCP).
> 
> My sincere apologies if any of this is really obvious and I’ve been silly and missed it.

	Maybe try luci-app-sqm and sqm scripts? 

[-- Attachment #2: 4alec.tar.gz --]
[-- Type: application/x-gzip, Size: 12811 bytes --]

[-- Attachment #3: Type: text/plain, Size: 890 bytes --]

Attached you will find the most recent files for sqm-scripts, just move all the files in the archive to the corresponding locations on your router and the select cake as qdisc and piece_of_cake.qos as script. (You might need a more recent luck-app-sqm, but first try what happens if you try the new scripts). After using the GUI to properly configure your interface you can always run either:
/etc/init.d/sqm stop
or 
/etc/init.d/sqm start
to stop and start sqm-scripts manually. Running:
tc -d qdisc
should give you some feedback whether things worked as you intended. If you want to tests this I am happy to help you along the way, just holler where things break ;).

Best Regards
	Sebastian


> 
> 
> —
> Alec Robertson
> _______________________________________________
> Cake mailing list
> Cake@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/cake

Re: [Cake] How to test Cake on TP-Link WDR3600

[Alec Robertson]

[-- Attachment #1: Type: text/plain, Size: 1902 bytes --]

Ah! Thank you that makes things much simpler.

BTW, thanks so much for all your work on Bufferbloat. It's made my
connection usable when downloading which I didn't think was possible! :)


Yours sincerely,
Alec Robertson.

On 25 July 2015 at 17:27, Sebastian Moeller <moeller0@gmx.de> wrote:

> Hi Alec,
>
> On Jul 25, 2015, at 17:45 , Alec Robertson <alecrobertson13@gmail.com>
> wrote:
>
> > Hello,
> >
> > I’ve flashed the OpenWRT image for the TP-Link WDR3600 from 2015-06-23 (
> http://snapon.lab.bufferbloat.net/~cero3/lupin/ar71xx/) and it has
> installed correctly.
> >
> > How do I go about setting it up for my UK TalkTalk Fibre connection
> (using a VDSL modem connected to the TP-Link router: the TP-Link router
> gets an IP address via DHCP).
> >
> > My sincere apologies if any of this is really obvious and I’ve been
> silly and missed it.
>
>         Maybe try luci-app-sqm and sqm scripts?
> Attached you will find the most recent files for sqm-scripts, just move
> all the files in the archive to the corresponding locations on your router
> and the select cake as qdisc and piece_of_cake.qos as script. (You might
> need a more recent luck-app-sqm, but first try what happens if you try the
> new scripts). After using the GUI to properly configure your interface you
> can always run either:
> /etc/init.d/sqm stop
> or
> /etc/init.d/sqm start
> to stop and start sqm-scripts manually. Running:
> tc -d qdisc
> should give you some feedback whether things worked as you intended. If
> you want to tests this I am happy to help you along the way, just holler
> where things break ;).
>
> Best Regards
>         Sebastian
>
>
> >
> >
> > —
> > Alec Robertson
> > _______________________________________________
> > Cake mailing list
> > Cake@lists.bufferbloat.net
> > https://lists.bufferbloat.net/listinfo/cake
>
>
>

[-- Attachment #2: Type: text/html, Size: 2754 bytes --]

Re: [Cake] How to test Cake on TP-Link WDR3600

[Alec Robertson]

[-- Attachment #1: Type: text/plain, Size: 1940 bytes --]

Well I've already got confused and we haven't even started yet - my
apologies.

When you say "...select cake as qdisc and piece_of_cake.qos as script..."
how exactly do you do that?

Again I'm sorry if I'm being silly! :P


Yours sincerely,
Alec Robertson.

On 25 July 2015 at 17:27, Sebastian Moeller <moeller0@gmx.de> wrote:

> Hi Alec,
>
> On Jul 25, 2015, at 17:45 , Alec Robertson <alecrobertson13@gmail.com>
> wrote:
>
> > Hello,
> >
> > I’ve flashed the OpenWRT image for the TP-Link WDR3600 from 2015-06-23 (
> http://snapon.lab.bufferbloat.net/~cero3/lupin/ar71xx/) and it has
> installed correctly.
> >
> > How do I go about setting it up for my UK TalkTalk Fibre connection
> (using a VDSL modem connected to the TP-Link router: the TP-Link router
> gets an IP address via DHCP).
> >
> > My sincere apologies if any of this is really obvious and I’ve been
> silly and missed it.
>
>         Maybe try luci-app-sqm and sqm scripts?
> Attached you will find the most recent files for sqm-scripts, just move
> all the files in the archive to the corresponding locations on your router
> and the select cake as qdisc and piece_of_cake.qos as script. (You might
> need a more recent luck-app-sqm, but first try what happens if you try the
> new scripts). After using the GUI to properly configure your interface you
> can always run either:
> /etc/init.d/sqm stop
> or
> /etc/init.d/sqm start
> to stop and start sqm-scripts manually. Running:
> tc -d qdisc
> should give you some feedback whether things worked as you intended. If
> you want to tests this I am happy to help you along the way, just holler
> where things break ;).
>
> Best Regards
>         Sebastian
>
>
> >
> >
> > —
> > Alec Robertson
> > _______________________________________________
> > Cake mailing list
> > Cake@lists.bufferbloat.net
> > https://lists.bufferbloat.net/listinfo/cake
>
>
>

[-- Attachment #2: Type: text/html, Size: 2981 bytes --]

Re: [Cake] How to test Cake on TP-Link WDR3600

[Sebastian Moeller]

Hi Alec,


On Jul 25, 2015, at 18:49 , Alec Robertson <alecrobertson13@gmail.com> wrote:

> Well I've already got confused and we haven't even started yet - my apologies.

Then we are off to a good start ;)

> 
> When you say "...select cake as qdisc and piece_of_cake.qos as script..." how exactly do you do that?

	Ah, this requires luci-app-sqm, you might be able to install this from the router’s GUI or it might be installed already; there should be a SQM tab in the network menu and there you should be able to configure things. http://wiki.openwrt.org/doc/howto/sqm has some advice about how to set things up. Please have a look and let me know what specific questions arise ;)

Best Regards
	Sebastian


> 
> Again I'm sorry if I'm being silly! :P
> 
> 
> Yours sincerely,
> Alec Robertson.
> 
> On 25 July 2015 at 17:27, Sebastian Moeller <moeller0@gmx.de> wrote:
> Hi Alec,
> 
> On Jul 25, 2015, at 17:45 , Alec Robertson <alecrobertson13@gmail.com> wrote:
> 
> > Hello,
> >
> > I’ve flashed the OpenWRT image for the TP-Link WDR3600 from 2015-06-23 (http://snapon.lab.bufferbloat.net/~cero3/lupin/ar71xx/) and it has installed correctly.
> >
> > How do I go about setting it up for my UK TalkTalk Fibre connection (using a VDSL modem connected to the TP-Link router: the TP-Link router gets an IP address via DHCP).
> >
> > My sincere apologies if any of this is really obvious and I’ve been silly and missed it.
> 
>         Maybe try luci-app-sqm and sqm scripts? 
> Attached you will find the most recent files for sqm-scripts, just move all the files in the archive to the corresponding locations on your router and the select cake as qdisc and piece_of_cake.qos as script. (You might need a more recent luck-app-sqm, but first try what happens if you try the new scripts). After using the GUI to properly configure your interface you can always run either:
> /etc/init.d/sqm stop
> or
> /etc/init.d/sqm start
> to stop and start sqm-scripts manually. Running:
> tc -d qdisc
> should give you some feedback whether things worked as you intended. If you want to tests this I am happy to help you along the way, just holler where things break ;).
> 
> Best Regards
>         Sebastian
> 
> 
> >
> >
> > —
> > Alec Robertson
> > _______________________________________________
> > Cake mailing list
> > Cake@lists.bufferbloat.net
> > https://lists.bufferbloat.net/listinfo/cake
> 
> 
> 
> _______________________________________________
> Cake mailing list
> Cake@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/cake

Re: [Cake] How to test Cake on TP-Link WDR3600

[Alec Robertson]

[-- Attachment #1: Type: text/plain, Size: 2977 bytes --]

I’ve uploaded all the files to the router and rebooted and very oddly, “cake" doesn’t show up as a qdisc but “layer_cake.qos” and “piece_of_cake.qos” show up as Queue setup scripts.


Weird.


--
Alec Robertson

On Sat, Jul 25, 2015 at 6:15 PM, Sebastian Moeller <moeller0@gmx.de>
wrote:

> Hi Alec,
> On Jul 25, 2015, at 18:49 , Alec Robertson <alecrobertson13@gmail.com> wrote:
>> Well I've already got confused and we haven't even started yet - my apologies.
> Then we are off to a good start ;)
>> 
>> When you say "...select cake as qdisc and piece_of_cake.qos as script..." how exactly do you do that?
> 	Ah, this requires luci-app-sqm, you might be able to install this from the router’s GUI or it might be installed already; there should be a SQM tab in the network menu and there you should be able to configure things. http://wiki.openwrt.org/doc/howto/sqm has some advice about how to set things up. Please have a look and let me know what specific questions arise ;)
> Best Regards
> 	Sebastian
>> 
>> Again I'm sorry if I'm being silly! :P
>> 
>> 
>> Yours sincerely,
>> Alec Robertson.
>> 
>> On 25 July 2015 at 17:27, Sebastian Moeller <moeller0@gmx.de> wrote:
>> Hi Alec,
>> 
>> On Jul 25, 2015, at 17:45 , Alec Robertson <alecrobertson13@gmail.com> wrote:
>> 
>> > Hello,
>> >
>> > I’ve flashed the OpenWRT image for the TP-Link WDR3600 from 2015-06-23 (http://snapon.lab.bufferbloat.net/~cero3/lupin/ar71xx/) and it has installed correctly.
>> >
>> > How do I go about setting it up for my UK TalkTalk Fibre connection (using a VDSL modem connected to the TP-Link router: the TP-Link router gets an IP address via DHCP).
>> >
>> > My sincere apologies if any of this is really obvious and I’ve been silly and missed it.
>> 
>>         Maybe try luci-app-sqm and sqm scripts? 
>> Attached you will find the most recent files for sqm-scripts, just move all the files in the archive to the corresponding locations on your router and the select cake as qdisc and piece_of_cake.qos as script. (You might need a more recent luck-app-sqm, but first try what happens if you try the new scripts). After using the GUI to properly configure your interface you can always run either:
>> /etc/init.d/sqm stop
>> or
>> /etc/init.d/sqm start
>> to stop and start sqm-scripts manually. Running:
>> tc -d qdisc
>> should give you some feedback whether things worked as you intended. If you want to tests this I am happy to help you along the way, just holler where things break ;).
>> 
>> Best Regards
>>         Sebastian
>> 
>> 
>> >
>> >
>> > —
>> > Alec Robertson
>> > _______________________________________________
>> > Cake mailing list
>> > Cake@lists.bufferbloat.net
>> > https://lists.bufferbloat.net/listinfo/cake
>> 
>> 
>> 
>> _______________________________________________
>> Cake mailing list
>> Cake@lists.bufferbloat.net
>> https://lists.bufferbloat.net/listinfo/cake

[-- Attachment #2: Type: text/html, Size: 3884 bytes --]

Re: [Cake] How to test Cake on TP-Link WDR3600

[Dave Taht]

On Sat, Jul 25, 2015 at 7:38 PM, Alec Robertson
<alecrobertson13@gmail.com> wrote:
> I’ve uploaded all the files to the router and rebooted and very oddly,
> “cake" doesn’t show up as a qdisc but “layer_cake.qos” and
> “piece_of_cake.qos” show up as Queue setup scripts.

Did you download the april 1st version by accident?

> Weird.
>
> --
> Alec Robertson
>
>
> On Sat, Jul 25, 2015 at 6:15 PM, Sebastian Moeller <moeller0@gmx.de> wrote:
>>
>> Hi Alec,
>>
>>
>> On Jul 25, 2015, at 18:49 , Alec Robertson <alecrobertson13@gmail.com>
>> wrote:
>>
>> > Well I've already got confused and we haven't even started yet - my
>> > apologies.
>>
>> Then we are off to a good start ;)
>>
>> >
>> > When you say "...select cake as qdisc and piece_of_cake.qos as
>> > script..." how exactly do you do that?
>>
>> Ah, this requires luci-app-sqm, you might be able to install this from the
>> router’s GUI or it might be installed already; there should be a SQM tab in
>> the network menu and there you should be able to configure things.
>> http://wiki.openwrt.org/doc/howto/sqm has some advice about how to set
>> things up. Please have a look and let me know what specific questions arise
>> ;)
>>
>> Best Regards
>> Sebastian
>>
>>
>> >
>> > Again I'm sorry if I'm being silly! :P
>> >
>> >
>> > Yours sincerely,
>> > Alec Robertson.
>> >
>> > On 25 July 2015 at 17:27, Sebastian Moeller <moeller0@gmx.de> wrote:
>> > Hi Alec,
>> >
>> > On Jul 25, 2015, at 17:45 , Alec Robertson <alecrobertson13@gmail.com>
>> > wrote:
>> >
>> > > Hello,
>> > >
>> > > I’ve flashed the OpenWRT image for the TP-Link WDR3600 from 2015-06-23
>> > > (http://snapon.lab.bufferbloat.net/~cero3/lupin/ar71xx/) and it has
>> > > installed correctly.
>> > >
>> > > How do I go about setting it up for my UK TalkTalk Fibre connection
>> > > (using a VDSL modem connected to the TP-Link router: the TP-Link router gets
>> > > an IP address via DHCP).
>> > >
>> > > My sincere apologies if any of this is really obvious and I’ve been
>> > > silly and missed it.
>> >
>> > Maybe try luci-app-sqm and sqm scripts?
>> > Attached you will find the most recent files for sqm-scripts, just move
>> > all the files in the archive to the corresponding locations on your router
>> > and the select cake as qdisc and piece_of_cake.qos as script. (You might
>> > need a more recent luck-app-sqm, but first try what happens if you try the
>> > new scripts). After using the GUI to properly configure your interface you
>> > can always run either:
>> > /etc/init.d/sqm stop
>> > or
>> > /etc/init.d/sqm start
>> > to stop and start sqm-scripts manually. Running:
>> > tc -d qdisc
>> > should give you some feedback whether things worked as you intended. If
>> > you want to tests this I am happy to help you along the way, just holler
>> > where things break ;).
>> >
>> > Best Regards
>> > Sebastian
>> >
>> >
>> > >
>> > >
>> > > —
>> > > Alec Robertson
>> > > _______________________________________________
>> > > Cake mailing list
>> > > Cake@lists.bufferbloat.net
>> > > https://lists.bufferbloat.net/listinfo/cake
>> >
>> >
>> >
>> > _______________________________________________
>> > Cake mailing list
>> > Cake@lists.bufferbloat.net
>> > https://lists.bufferbloat.net/listinfo/cake
>>
>
>
> _______________________________________________
> Cake mailing list
> Cake@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/cake
>



-- 
Dave Täht
worldwide bufferbloat report:
http://www.dslreports.com/speedtest/results/bufferbloat
And:
What will it take to vastly improve wifi for everyone?
https://plus.google.com/u/0/explore/makewififast

Re: [Cake] How to test Cake on TP-Link WDR3600

[Alec Robertson]

[-- Attachment #1: Type: text/plain, Size: 4062 bytes --]

I don't think so but just to be sure, would you mind linking me to the newest version?



—
Alec Robertson

On Sat, Jul 25, 2015 at 7:12 PM, Dave Taht <dave.taht@gmail.com> wrote:

> On Sat, Jul 25, 2015 at 7:38 PM, Alec Robertson
> <alecrobertson13@gmail.com> wrote:
>> I’ve uploaded all the files to the router and rebooted and very oddly,
>> “cake" doesn’t show up as a qdisc but “layer_cake.qos” and
>> “piece_of_cake.qos” show up as Queue setup scripts.
> Did you download the april 1st version by accident?
>> Weird.
>>
>> --
>> Alec Robertson
>>
>>
>> On Sat, Jul 25, 2015 at 6:15 PM, Sebastian Moeller <moeller0@gmx.de> wrote:
>>>
>>> Hi Alec,
>>>
>>>
>>> On Jul 25, 2015, at 18:49 , Alec Robertson <alecrobertson13@gmail.com>
>>> wrote:
>>>
>>> > Well I've already got confused and we haven't even started yet - my
>>> > apologies.
>>>
>>> Then we are off to a good start ;)
>>>
>>> >
>>> > When you say "...select cake as qdisc and piece_of_cake.qos as
>>> > script..." how exactly do you do that?
>>>
>>> Ah, this requires luci-app-sqm, you might be able to install this from the
>>> router’s GUI or it might be installed already; there should be a SQM tab in
>>> the network menu and there you should be able to configure things.
>>> http://wiki.openwrt.org/doc/howto/sqm has some advice about how to set
>>> things up. Please have a look and let me know what specific questions arise
>>> ;)
>>>
>>> Best Regards
>>> Sebastian
>>>
>>>
>>> >
>>> > Again I'm sorry if I'm being silly! :P
>>> >
>>> >
>>> > Yours sincerely,
>>> > Alec Robertson.
>>> >
>>> > On 25 July 2015 at 17:27, Sebastian Moeller <moeller0@gmx.de> wrote:
>>> > Hi Alec,
>>> >
>>> > On Jul 25, 2015, at 17:45 , Alec Robertson <alecrobertson13@gmail.com>
>>> > wrote:
>>> >
>>> > > Hello,
>>> > >
>>> > > I’ve flashed the OpenWRT image for the TP-Link WDR3600 from 2015-06-23
>>> > > (http://snapon.lab.bufferbloat.net/~cero3/lupin/ar71xx/) and it has
>>> > > installed correctly.
>>> > >
>>> > > How do I go about setting it up for my UK TalkTalk Fibre connection
>>> > > (using a VDSL modem connected to the TP-Link router: the TP-Link router gets
>>> > > an IP address via DHCP).
>>> > >
>>> > > My sincere apologies if any of this is really obvious and I’ve been
>>> > > silly and missed it.
>>> >
>>> > Maybe try luci-app-sqm and sqm scripts?
>>> > Attached you will find the most recent files for sqm-scripts, just move
>>> > all the files in the archive to the corresponding locations on your router
>>> > and the select cake as qdisc and piece_of_cake.qos as script. (You might
>>> > need a more recent luck-app-sqm, but first try what happens if you try the
>>> > new scripts). After using the GUI to properly configure your interface you
>>> > can always run either:
>>> > /etc/init.d/sqm stop
>>> > or
>>> > /etc/init.d/sqm start
>>> > to stop and start sqm-scripts manually. Running:
>>> > tc -d qdisc
>>> > should give you some feedback whether things worked as you intended. If
>>> > you want to tests this I am happy to help you along the way, just holler
>>> > where things break ;).
>>> >
>>> > Best Regards
>>> > Sebastian
>>> >
>>> >
>>> > >
>>> > >
>>> > > —
>>> > > Alec Robertson
>>> > > _______________________________________________
>>> > > Cake mailing list
>>> > > Cake@lists.bufferbloat.net
>>> > > https://lists.bufferbloat.net/listinfo/cake
>>> >
>>> >
>>> >
>>> > _______________________________________________
>>> > Cake mailing list
>>> > Cake@lists.bufferbloat.net
>>> > https://lists.bufferbloat.net/listinfo/cake
>>>
>>
>>
>> _______________________________________________
>> Cake mailing list
>> Cake@lists.bufferbloat.net
>> https://lists.bufferbloat.net/listinfo/cake
>>
> -- 
> Dave Täht
> worldwide bufferbloat report:
> http://www.dslreports.com/speedtest/results/bufferbloat
> And:
> What will it take to vastly improve wifi for everyone?
> https://plus.google.com/u/0/explore/makewififast

[-- Attachment #2: Type: text/html, Size: 5605 bytes --]

Re: [Cake] How to test Cake on TP-Link WDR3600

[Sebastian Moeller]

[-- Attachment #1: Type: text/plain, Size: 1425 bytes --]

Hi Alec,

On Jul 25, 2015, at 19:38 , Alec Robertson <alecrobertson13@gmail.com> wrote:

> I’ve uploaded all the files to the router and rebooted and very oddly, “cake" doesn’t show up as a qdisc but “layer_cake.qos” and “piece_of_cake.qos” show up as Queue setup scripts.

	Ah, this just shows that your sqm.lua is older than I assumed. No big issue just replace the /usr/lib/lua/luci/model/cbi/sqm.lua on your router with the attached sqm.lua. That should fix that issue. 
Then please select cake as “Queueing discipline” and piece_of_cake.qos as “Queue setup script”, also make sure to go to the “Link Layer Adaptation” tab and select Ethernet… as “Which link layer to account for:” and put in probably 8 as “Per Packet Overhead (byte):” (not quite sure what the correct value is for your ISP), AND check the checkbox named “Show Advanced Linklayer Options, (only needed if MTU > 1500). Advanced options will only be used as long as this box is checked.” and finally select cake as “Which linklayer adaptation mechanism to use; for testing only”. That should get you started. Most likely your firmware will not fully allow to specify overhead for cake, but we will diagnose this and the revert to fix this up then most likely is to select tc_stab as “Which linklayer adaptation mechanism to use; for testing only”.

With a bit of luck this is all that is needed to get you going.

Best Regards
	Sebastian



[-- Attachment #2: sqm.lua --]
[-- Type: application/octet-stream, Size: 9571 bytes --]

--[[
LuCI - Lua Configuration Interface

Copyright 2014 Steven Barth <steven@midlink.org>
Copyright 2014 Dave Taht <dave.taht@bufferbloat.net>

Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at

	http://www.apache.org/licenses/LICENSE-2.0

$Id$
]]--

local wa = require "luci.tools.webadmin"
local fs = require "nixio.fs"
local net = require "luci.model.network".init()
local sys = require "luci.sys"
--local ifaces = net:get_interfaces()
local ifaces = sys.net:devices()
local path = "/usr/lib/sqm"

m = Map("sqm", translate("Smart Queue Management"),
	translate("With <abbr title=\"Smart Queue Management\">SQM</abbr> you " ..
		"can enable traffic shaping, better mixing (Fair Queueing)," ..
		" active queue length management (AQM) " ..
		" and prioritisation on one " ..
		"network interface."))

s = m:section(TypedSection, "queue", translate("Queues"))
s:tab("tab_basic", translate("Basic Settings"))
s:tab("tab_qdisc", translate("Queue Discipline"))
s:tab("tab_linklayer", translate("Link Layer Adaptation"))
s.addremove = true -- set to true to allow adding SQM instances in the GUI
s.anonymous = true

-- BASIC
e = s:taboption("tab_basic", Flag, "enabled", translate("Enable this SQM instance."))
e.rmempty = false

-- sm: following jow's advise, be helpful to the user and enable
--     sqm's init script if even a single sm instance/interface
--     is enabled; this is unexpected in that the init script gets
--     enabled as soon as at least one sqm instance is enabled
--     and that state is saved, so it does not require "Save & Apply"
--     to effect the init scripts.
--     the implementation was inpired/lifted from 
--     https://github.com/openwrt/luci/blob/master/applications/luci-app-minidlna/luasrc/model/cbi/minidlna.lua
function e.write(self, section, value)
    if value == "1" then
	luci.sys.init.enable("sqm")
	m.message = translate("The SQM GUI has just enabled the sqm initscript on your behalf. Remember to disable the sqm initscript manually under System Startup menu in case this change was not wished for.")
--	luci.sys.call("/etc/init.d/sqm start >/dev/null")
--    else
--	luci.sys.call("/etc/init.d/sqm stop >/dev/null")
--	luci.sys.init.disable("sqm")
    end
    return Flag.write(self, section, value)
end
-- TODO: inform the user what we just did...

n = s:taboption("tab_basic", ListValue, "interface", translate("Interface name"))
-- sm lifted from luci-app-wol, the original implementation failed to show pppoe-ge00 type interface names
for _, iface in ipairs(ifaces) do
--     if iface:is_up() then
--	n:value(iface:name())
--     end
	if iface ~= "lo" then 
		n:value(iface) 
	end
end
n.rmempty = false


dl = s:taboption("tab_basic", Value, "download", translate("Download speed (kbit/s) (ingress) set to 0 to selectively disable ingress shaping:"))
dl.datatype = "and(uinteger,min(0))"
dl.rmempty = false

ul = s:taboption("tab_basic", Value, "upload", translate("Upload speed (kbit/s) (egress) set to 0 to selectively disable egress shaping:"))
ul.datatype = "and(uinteger,min(0))"
ul.rmempty = false

-- QDISC

c = s:taboption("tab_qdisc", ListValue, "qdisc", translate("Queueing discipline"))
c:value("fq_codel", "fq_codel ("..translate("default")..")")
c:value("efq_codel")
c:value("nfq_codel")
c:value("sfq")
c:value("codel")
c:value("ns2_codel")
c:value("pie")
c:value("sfq")
c:value("cake")
c.default = "fq_codel"
c.rmempty = false

local qos_desc = ""
sc = s:taboption("tab_qdisc", ListValue, "script", translate("Queue setup script"))
for file in fs.dir(path) do
  if string.find(file, ".qos$") then
    sc:value(file)
  end
  if string.find(file, ".qos.help$") then
    fh = io.open(path .. "/" .. file, "r")
    qos_desc = qos_desc .. "<p><b>" .. file:gsub(".help$", "") .. ":</b><br />" .. fh:read("*a") .. "</p>"
  end
end
sc.default = "simple.qos"
sc.rmempty = false
sc.description = qos_desc

ad = s:taboption("tab_qdisc", Flag, "qdisc_advanced", translate("Show and Use Advanced Configuration. Advanced options will only be used as long as this box is checked."))
ad.default = false
ad.rmempty = true

squash_dscp  = s:taboption("tab_qdisc", ListValue, "squash_dscp", translate("Squash DSCP on inbound packets (ingress):"))
squash_dscp:value("1", "SQUASH")
squash_dscp:value("0", "DO NOT SQUASH")
squash_dscp.default = "1"
squash_dscp.rmempty = true
squash_dscp:depends("qdisc_advanced", "1")

squash_ingress = s:taboption("tab_qdisc", ListValue, "squash_ingress", translate("Ignore DSCP on ingress:"))
squash_ingress:value("1", "Ignore")
squash_ingress:value("0", "Allow")
squash_ingress.default = "1"
squash_ingress.rmempty = true
squash_ingress:depends("qdisc_advanced", "1")

iecn = s:taboption("tab_qdisc", ListValue, "ingress_ecn", translate("Explicit congestion notification (ECN) status on inbound packets (ingress):"))
iecn:value("ECN", "ECN ("..translate("default")..")")
iecn:value("NOECN")
iecn.default = "ECN"
iecn.rmempty = true
iecn:depends("qdisc_advanced", "1")

eecn = s:taboption("tab_qdisc", ListValue, "egress_ecn", translate("Explicit congestion notification (ECN) status on outbound packets (egress)."))
eecn:value("NOECN", "NOECN ("..translate("default")..")")
eecn:value("ECN")
eecn.default = "NOECN"
eecn.rmempty = true
eecn:depends("qdisc_advanced", "1")

ad2 = s:taboption("tab_qdisc", Flag, "qdisc_really_really_advanced", translate("Show and Use Dangerous Configuration. Dangerous options will only be used as long as this box is checked."))
ad2.default = false
ad2.rmempty = true
ad2:depends("qdisc_advanced", "1")

ilim = s:taboption("tab_qdisc", Value, "ilimit", translate("Hard limit on ingress queues; leave empty for default."))
-- ilim.default = 1000
ilim.isnumber = true
ilim.datatype = "and(uinteger,min(0))"
ilim.rmempty = true
ilim:depends("qdisc_really_really_advanced", "1")

elim = s:taboption("tab_qdisc", Value, "elimit", translate("Hard limit on egress queues; leave empty for default."))
-- elim.default = 1000
elim.datatype = "and(uinteger,min(0))"
elim.rmempty = true
elim:depends("qdisc_really_really_advanced", "1")


itarg = s:taboption("tab_qdisc", Value, "itarget", translate("Latency target for ingress, e.g 5ms [units: s, ms, or  us]; leave empty for automatic selection, put in the word default for the qdisc's default."))
itarg.datatype = "string"
itarg.rmempty = true
itarg:depends("qdisc_really_really_advanced", "1")

etarg = s:taboption("tab_qdisc", Value, "etarget", translate("Latency target for egress, e.g. 5ms [units: s, ms, or  us]; leave empty for automatic selection, put in the word default for the qdisc's default."))
etarg.datatype = "string"
etarg.rmempty = true
etarg:depends("qdisc_really_really_advanced", "1")



iqdisc_opts = s:taboption("tab_qdisc", Value, "iqdisc_opts", translate("Advanced option string to pass to the ingress queueing disciplines; no error checking, use very carefully."))
iqdisc_opts.rmempty = true
iqdisc_opts:depends("qdisc_really_really_advanced", "1")

eqdisc_opts = s:taboption("tab_qdisc", Value, "eqdisc_opts", translate("Advanced option string to pass to the egress queueing disciplines; no error checking, use very carefully."))
eqdisc_opts.rmempty = true
eqdisc_opts:depends("qdisc_really_really_advanced", "1")

-- LINKLAYER
ll = s:taboption("tab_linklayer", ListValue, "linklayer", translate("Which link layer to account for:"))
ll:value("none", "none ("..translate("default")..")")
ll:value("ethernet", "Ethernet with overhead: select for e.g. VDSL2.")
ll:value("atm", "ATM: select for e.g. ADSL1, ADSL2, ADSL2+.")
-- ll:value("adsl")	-- reduce the options
ll.default = "none"

po = s:taboption("tab_linklayer", Value, "overhead", translate("Per Packet Overhead (byte):"))
po.datatype = "and(integer,min(-1500))"
po.default = 0
po.isnumber = true
po.rmempty = true
po:depends("linklayer", "ethernet")
-- po:depends("linklayer", "adsl")
po:depends("linklayer", "atm")


adll = s:taboption("tab_linklayer", Flag, "linklayer_advanced", translate("Show Advanced Linklayer Options, (only needed if MTU > 1500). Advanced options will only be used as long as this box is checked."))
adll.rmempty = true
adll:depends("linklayer", "ethernet")
-- adll:depends("linklayer", "adsl")
adll:depends("linklayer", "atm")

smtu = s:taboption("tab_linklayer", Value, "tcMTU", translate("Maximal Size for size and rate calculations, tcMTU (byte); needs to be >= interface MTU + overhead:"))
smtu.datatype = "and(uinteger,min(0))"
smtu.default = 2047
smtu.isnumber = true
smtu.rmempty = true
smtu:depends("linklayer_advanced", "1")

stsize = s:taboption("tab_linklayer", Value, "tcTSIZE", translate("Number of entries in size/rate tables, TSIZE; for ATM choose TSIZE = (tcMTU + 1) / 16:"))
stsize.datatype = "and(uinteger,min(0))"
stsize.default = 128
stsize.isnumber = true
stsize.rmempty = true
stsize:depends("linklayer_advanced", "1")

smpu = s:taboption("tab_linklayer", Value, "tcMPU", translate("Minimal packet size, MPU (byte); needs to be > 0 for ethernet size tables:"))
smpu.datatype = "and(uinteger,min(0))"
smpu.default = 0
smpu.isnumber = true
smpu.rmempty = true
smpu:depends("linklayer_advanced", "1")

lla = s:taboption("tab_linklayer", ListValue, "linklayer_adaptation_mechanism", translate("Which linklayer adaptation mechanism to use; for testing only"))
lla:value("cake")
lla:value("htb_private")
lla:value("tc_stab", "tc_stab ("..translate("default")..")")
lla.default = "tc_stab"
lla.rmempty = true
lla:depends("linklayer_advanced", "1")

-- PRORITIES?

return m

[-- Attachment #3: Type: text/plain, Size: 3068 bytes --]




> 
> Weird.
> 
> --
> Alec Robertson
> 
> 
> On Sat, Jul 25, 2015 at 6:15 PM, Sebastian Moeller <moeller0@gmx.de> wrote:
> 
> Hi Alec, 
> 
> 
> On Jul 25, 2015, at 18:49 , Alec Robertson <alecrobertson13@gmail.com> wrote: 
> 
> > Well I've already got confused and we haven't even started yet - my apologies. 
> 
> Then we are off to a good start ;) 
> 
> > 
> > When you say "...select cake as qdisc and piece_of_cake.qos as script..." how exactly do you do that? 
> 
> Ah, this requires luci-app-sqm, you might be able to install this from the router’s GUI or it might be installed already; there should be a SQM tab in the network menu and there you should be able to configure things. http://wiki.openwrt.org/doc/howto/sqm has some advice about how to set things up. Please have a look and let me know what specific questions arise ;) 
> 
> Best Regards 
> Sebastian 
> 
> 
> > 
> > Again I'm sorry if I'm being silly! :P 
> > 
> > 
> > Yours sincerely, 
> > Alec Robertson. 
> > 
> > On 25 July 2015 at 17:27, Sebastian Moeller <moeller0@gmx.de> wrote: 
> > Hi Alec, 
> > 
> > On Jul 25, 2015, at 17:45 , Alec Robertson <alecrobertson13@gmail.com> wrote: 
> > 
> > > Hello, 
> > > 
> > > I’ve flashed the OpenWRT image for the TP-Link WDR3600 from 2015-06-23 (http://snapon.lab.bufferbloat.net/~cero3/lupin/ar71xx/) and it has installed correctly. 
> > > 
> > > How do I go about setting it up for my UK TalkTalk Fibre connection (using a VDSL modem connected to the TP-Link router: the TP-Link router gets an IP address via DHCP). 
> > > 
> > > My sincere apologies if any of this is really obvious and I’ve been silly and missed it. 
> > 
> > Maybe try luci-app-sqm and sqm scripts? 
> > Attached you will find the most recent files for sqm-scripts, just move all the files in the archive to the corresponding locations on your router and the select cake as qdisc and piece_of_cake.qos as script. (You might need a more recent luck-app-sqm, but first try what happens if you try the new scripts). After using the GUI to properly configure your interface you can always run either: 
> > /etc/init.d/sqm stop 
> > or 
> > /etc/init.d/sqm start 
> > to stop and start sqm-scripts manually. Running: 
> > tc -d qdisc 
> > should give you some feedback whether things worked as you intended. If you want to tests this I am happy to help you along the way, just holler where things break ;). 
> > 
> > Best Regards 
> > Sebastian 
> > 
> > 
> > > 
> > > 
> > > — 
> > > Alec Robertson 
> > > _______________________________________________ 
> > > Cake mailing list 
> > > Cake@lists.bufferbloat.net 
> > > https://lists.bufferbloat.net/listinfo/cake 
> > 
> > 
> > 
> > _______________________________________________ 
> > Cake mailing list 
> > Cake@lists.bufferbloat.net 
> > https://lists.bufferbloat.net/listinfo/cake 
> 
> 
> 
> _______________________________________________
> Cake mailing list
> Cake@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/cake

Re: [Cake] How to test Cake on TP-Link WDR3600

[Sebastian Moeller]

Hi Alec, hi Dave,

On Jul 25, 2015, at 20:21 , Alec Robertson <alecrobertson13@gmail.com> wrote:

> I don't think so but just to be sure, would you mind linking me to the newest version?

	Your version probably is fine, there has just been some re-design in sqm-scripts that begat two new scripts and slightly new file organization, but functionality should still be the same.

Best Regards
	Sebastian

> 
> —
> Alec Robertson
> 
> 
> On Sat, Jul 25, 2015 at 7:12 PM, Dave Taht <dave.taht@gmail.com> wrote:
> 
> On Sat, Jul 25, 2015 at 7:38 PM, Alec Robertson 
> <alecrobertson13@gmail.com> wrote: 
> > I’ve uploaded all the files to the router and rebooted and very oddly, 
> > “cake" doesn’t show up as a qdisc but “layer_cake.qos” and 
> > “piece_of_cake.qos” show up as Queue setup scripts. 
> 
> Did you download the april 1st version by accident? 
> 
> > Weird. 
> > 
> > -- 
> > Alec Robertson 
> > 
> > 
> > On Sat, Jul 25, 2015 at 6:15 PM, Sebastian Moeller <moeller0@gmx.de> wrote: 
> >> 
> >> Hi Alec, 
> >> 
> >> 
> >> On Jul 25, 2015, at 18:49 , Alec Robertson <alecrobertson13@gmail.com> 
> >> wrote: 
> >> 
> >> > Well I've already got confused and we haven't even started yet - my 
> >> > apologies. 
> >> 
> >> Then we are off to a good start ;) 
> >> 
> >> > 
> >> > When you say "...select cake as qdisc and piece_of_cake.qos as 
> >> > script..." how exactly do you do that? 
> >> 
> >> Ah, this requires luci-app-sqm, you might be able to install this from the 
> >> router’s GUI or it might be installed already; there should be a SQM tab in 
> >> the network menu and there you should be able to configure things. 
> >> http://wiki.openwrt.org/doc/howto/sqm has some advice about how to set 
> >> things up. Please have a look and let me know what specific questions arise 
> >> ;) 
> >> 
> >> Best Regards 
> >> Sebastian 
> >> 
> >> 
> >> > 
> >> > Again I'm sorry if I'm being silly! :P 
> >> > 
> >> > 
> >> > Yours sincerely, 
> >> > Alec Robertson. 
> >> > 
> >> > On 25 July 2015 at 17:27, Sebastian Moeller <moeller0@gmx.de> wrote: 
> >> > Hi Alec, 
> >> > 
> >> > On Jul 25, 2015, at 17:45 , Alec Robertson <alecrobertson13@gmail.com> 
> >> > wrote: 
> >> > 
> >> > > Hello, 
> >> > > 
> >> > > I’ve flashed the OpenWRT image for the TP-Link WDR3600 from 2015-06-23 
> >> > > (http://snapon.lab.bufferbloat.net/~cero3/lupin/ar71xx/) and it has 
> >> > > installed correctly. 
> >> > > 
> >> > > How do I go about setting it up for my UK TalkTalk Fibre connection 
> >> > > (using a VDSL modem connected to the TP-Link router: the TP-Link router gets 
> >> > > an IP address via DHCP). 
> >> > > 
> >> > > My sincere apologies if any of this is really obvious and I’ve been 
> >> > > silly and missed it. 
> >> > 
> >> > Maybe try luci-app-sqm and sqm scripts? 
> >> > Attached you will find the most recent files for sqm-scripts, just move 
> >> > all the files in the archive to the corresponding locations on your router 
> >> > and the select cake as qdisc and piece_of_cake.qos as script. (You might 
> >> > need a more recent luck-app-sqm, but first try what happens if you try the 
> >> > new scripts). After using the GUI to properly configure your interface you 
> >> > can always run either: 
> >> > /etc/init.d/sqm stop 
> >> > or 
> >> > /etc/init.d/sqm start 
> >> > to stop and start sqm-scripts manually. Running: 
> >> > tc -d qdisc 
> >> > should give you some feedback whether things worked as you intended. If 
> >> > you want to tests this I am happy to help you along the way, just holler 
> >> > where things break ;). 
> >> > 
> >> > Best Regards 
> >> > Sebastian 
> >> > 
> >> > 
> >> > > 
> >> > > 
> >> > > — 
> >> > > Alec Robertson 
> >> > > _______________________________________________ 
> >> > > Cake mailing list 
> >> > > Cake@lists.bufferbloat.net 
> >> > > https://lists.bufferbloat.net/listinfo/cake 
> >> > 
> >> > 
> >> > 
> >> > _______________________________________________ 
> >> > Cake mailing list 
> >> > Cake@lists.bufferbloat.net 
> >> > https://lists.bufferbloat.net/listinfo/cake 
> >> 
> > 
> > 
> > _______________________________________________ 
> > Cake mailing list 
> > Cake@lists.bufferbloat.net 
> > https://lists.bufferbloat.net/listinfo/cake 
> > 
> 
> 
> 
> -- 
> Dave Täht 
> worldwide bufferbloat report: 
> http://www.dslreports.com/speedtest/results/bufferbloat 
> And: 
> What will it take to vastly improve wifi for everyone? 
> https://plus.google.com/u/0/explore/makewififast 
> 
> 
> _______________________________________________
> Cake mailing list
> Cake@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/cake

Re: [Cake] How to test Cake on TP-Link WDR3600

[Alec Robertson]

[-- Attachment #1: Type: text/plain, Size: 113 bytes --]

That's it, it's all showing up now.

I will try it out right away. What should I be testing and looking out for?

[-- Attachment #2: Type: text/html, Size: 237 bytes --]

Re: [Cake] How to test Cake on TP-Link WDR3600

[Sebastian Moeller]

Hi Alec,


On Jul 25, 2015, at 20:57 , Alec Robertson <alecrobertson13@gmail.com> wrote:

> That's it, it's all showing up now.
> 
> I will try it out right away. What should I be testing and looking out for?

1) test whether using cake as link layer adaptation mechanism works: put in a number for overhead and run “tc -d qdisc” and post the output of:
	a) cat /etc/config/sqm
	b) /etc/init.d/sqm stop ; tc -d qdisc
	c) /etc/init.d/sqm start ; tc -d qdisc

2) test whether you like the performance under load, http://www.dslreports.com/speedtest is a good start, registered (for free) users can change the configuration, I like to use 16 streams for upstream and downstream each and I like the “Hi-Res BufferBloat:	 (10hz check)” option.

3) test even more, have a loo at flent (https://flent.org), especially the rrul test has been a real help in understanding link behavior under load.
Please note, piece_of_cake will only use 1 priority for all packets (so effectively no priority), layer_cake.qos will by default, use 4 different priority layers, so expect some difference between layer_cake and piece_of_cake in the rrul results...


Best Regards
	Sebastian


> _______________________________________________
> Cake mailing list
> Cake@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/cake

Re: [Cake] How to test Cake on TP-Link WDR3600

[Alec Robertson]

[-- Attachment #1: Type: text/plain, Size: 1732 bytes --]

I’m assuming I put "tc -d qdisc” into BusyBox?


Quick sub-question (off-topic so my apologies), this firmware I’m using that I linked to previously, has HTTPS enabled which means every time I go to Luci I get a security error in Chrome. How do I disable HTTPS?


--
Alec Robertson

On Sat, Jul 25, 2015 at 8:20 PM, Sebastian Moeller <moeller0@gmx.de>
wrote:

> Hi Alec,
> On Jul 25, 2015, at 20:57 , Alec Robertson <alecrobertson13@gmail.com> wrote:
>> That's it, it's all showing up now.
>> 
>> I will try it out right away. What should I be testing and looking out for?
> 1) test whether using cake as link layer adaptation mechanism works: put in a number for overhead and run “tc -d qdisc” and post the output of:
> 	a) cat /etc/config/sqm
> 	b) /etc/init.d/sqm stop ; tc -d qdisc
> 	c) /etc/init.d/sqm start ; tc -d qdisc
> 2) test whether you like the performance under load, http://www.dslreports.com/speedtest is a good start, registered (for free) users can change the configuration, I like to use 16 streams for upstream and downstream each and I like the “Hi-Res BufferBloat:	 (10hz check)” option.
> 3) test even more, have a loo at flent (https://flent.org), especially the rrul test has been a real help in understanding link behavior under load.
> Please note, piece_of_cake will only use 1 priority for all packets (so effectively no priority), layer_cake.qos will by default, use 4 different priority layers, so expect some difference between layer_cake and piece_of_cake in the rrul results...
> Best Regards
> 	Sebastian
>> _______________________________________________
>> Cake mailing list
>> Cake@lists.bufferbloat.net
>> https://lists.bufferbloat.net/listinfo/cake

[-- Attachment #2: Type: text/html, Size: 2202 bytes --]

Re: [Cake] How to test Cake on TP-Link WDR3600

[Sebastian Moeller]

Hi Alec,

On Jul 25, 2015, at 22:58 , Alec Robertson <alecrobertson13@gmail.com> wrote:

> I’m assuming I put "tc -d qdisc” into BusyBox?

	Yes, you need to use ssh to log into your router before issuing the listed commands, then copy the output and post it here, please. In case you use windows putty ( http://www.chiark.greenend.org.uk/~sgtatham/putty/download.html ) is a decent ssh client you could try.

> 
> Quick sub-question (off-topic so my apologies), this firmware I’m using that I linked to previously, has HTTPS enabled which means every time I go to Luci I get a security error in Chrome. How do I disable HTTPS?

	I would say you do not disable it, but rather look why chrome complains (it might be that chrome dislikes self-signed certificates) and try to convince chrome to accept the certificate nevertheless. Or you could try firefox ;)

> 
> --
> Alec Robertson
> 
> 
> On Sat, Jul 25, 2015 at 8:20 PM, Sebastian Moeller <moeller0@gmx.de> wrote:
> 
> Hi Alec, 
> 
> 
> On Jul 25, 2015, at 20:57 , Alec Robertson <alecrobertson13@gmail.com> wrote: 
> 
> > That's it, it's all showing up now. 
> > 
> > I will try it out right away. What should I be testing and looking out for? 
> 
> 1) test whether using cake as link layer adaptation mechanism works: put in a number for overhead and run “tc -d qdisc” and post the output of: 
> a) cat /etc/config/sqm 
> b) /etc/init.d/sqm stop ; tc -d qdisc 
> c) /etc/init.d/sqm start ; tc -d qdisc 
> 
> 2) test whether you like the performance under load, http://www.dslreports.com/speedtest is a good start, registered (for free) users can change the configuration, I like to use 16 streams for upstream and downstream each and I like the “Hi-Res BufferBloat:	 (10hz check)” option. 
> 
> 3) test even more, have a loo at flent (https://flent.org), especially the rrul test has been a real help in understanding link behavior under load. 
> Please note, piece_of_cake will only use 1 priority for all packets (so effectively no priority), layer_cake.qos will by default, use 4 different priority layers, so expect some difference between layer_cake and piece_of_cake in the rrul results... 
> 
> 
> Best Regards 
> Sebastian 
> 
> 
> > _______________________________________________ 
> > Cake mailing list 
> > Cake@lists.bufferbloat.net 
> > https://lists.bufferbloat.net/listinfo/cake 
> 
> 
> 
> _______________________________________________
> Cake mailing list
> Cake@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/cake

Re: [Cake] How to test Cake on TP-Link WDR3600

[Alan Jenkins]

>> Quick sub-question (off-topic so my apologies), this firmware I’m using that I linked to previously, has HTTPS enabled which means every time I go to Luci I get a security error in Chrome. How do I disable HTTPS?
> 	I would say you do not disable it, but rather look why chrome complains (it might be that chrome dislikes self-signed certificates) and try to convince chrome to accept the certificate nevertheless. Or you could try firefox ;)

Firefox is the definitely the simplest browser for this, it's the only 
one I've used.  Just click through the instructions.  A "permanent 
exception" is the default, which should actually help security.  Feels 
ironic as I remember Firefox moving first on this & hence complaints 
about the scary warning messages etc.

Searching instructions for Chrome on Linux ("ssl exception" OR "self 
signed certificate") they use a certutil command.  Internet Explorer 
will use the Windows cert store, same with Chrome on Windows.  (If you 
can add to the store using IE, that may be simplest & will cover both).

Additional requirement for those methods should be that the cert CN 
matches the URL you access.  Not sure about Chrome, but for general 
paranoia you should check that CN / common name / "issued to" doesn't 
say "*" i.e. "everywhere".

Access the router using `http://hostname` matching the router hostname 
(as per /etc/config/system).  dnsmasq will let that work. If you've 
changed the hostname, re-gen the cert by removing it and restarting.

rm /etc/uhttpd.crt /etc/uhttpd.key
/etc/init.d/uhttpd restart

Alan

Re: [Cake] How to test Cake on TP-Link WDR3600

[Alec Robertson]

[-- Attachment #1: Type: text/plain, Size: 1850 bytes --]

Am I right to assume that with Cake more bandwidth should be available to use, without affecting latency?

--
Alec Robertson

On Sun, Jul 26, 2015 at 10:11 AM, Alan Jenkins
<alan.christopher.jenkins@gmail.com> wrote:

>>> Quick sub-question (off-topic so my apologies), this firmware I’m using that I linked to previously, has HTTPS enabled which means every time I go to Luci I get a security error in Chrome. How do I disable HTTPS?
>> 	I would say you do not disable it, but rather look why chrome complains (it might be that chrome dislikes self-signed certificates) and try to convince chrome to accept the certificate nevertheless. Or you could try firefox ;)
> Firefox is the definitely the simplest browser for this, it's the only 
> one I've used.  Just click through the instructions.  A "permanent 
> exception" is the default, which should actually help security.  Feels 
> ironic as I remember Firefox moving first on this & hence complaints 
> about the scary warning messages etc.
> Searching instructions for Chrome on Linux ("ssl exception" OR "self 
> signed certificate") they use a certutil command.  Internet Explorer 
> will use the Windows cert store, same with Chrome on Windows.  (If you 
> can add to the store using IE, that may be simplest & will cover both).
> Additional requirement for those methods should be that the cert CN 
> matches the URL you access.  Not sure about Chrome, but for general 
> paranoia you should check that CN / common name / "issued to" doesn't 
> say "*" i.e. "everywhere".
> Access the router using `http://hostname` matching the router hostname 
> (as per /etc/config/system).  dnsmasq will let that work. If you've 
> changed the hostname, re-gen the cert by removing it and restarting.
> rm /etc/uhttpd.crt /etc/uhttpd.key
> /etc/init.d/uhttpd restart
> Alan

[-- Attachment #2: Type: text/html, Size: 2287 bytes --]

Re: [Cake] How to test Cake on TP-Link WDR3600

[Jonathan Morton]

[-- Attachment #1: Type: text/plain, Size: 98 bytes --]

That's a question with nuanced answers, but in general yes, at least
slightly.

- Jonathan Morton

[-- Attachment #2: Type: text/html, Size: 141 bytes --]

Re: [Cake] How to test Cake on TP-Link WDR3600

[Alec Robertson]

[-- Attachment #1: Type: text/plain, Size: 530 bytes --]

I’ve just updated to the newest trunk release of OpenWRT Chaos Calmer (fresh install) and the SQM QOS from OPKG interestingly does include Cake as a qdisc but neither layer_cake.qos nor piece_of_cake.qos are available as setup scripts.


I’m still trying out Cake so I’ll be back soon with some feedback.


--
Alec Robertson

On Sun, Jul 26, 2015 at 1:35 PM, Jonathan Morton <chromatix99@gmail.com>
wrote:

> That's a question with nuanced answers, but in general yes, at least
> slightly.
> - Jonathan Morton

[-- Attachment #2: Type: text/html, Size: 930 bytes --]

Re: [Cake] How to test Cake on TP-Link WDR3600

[Alan Jenkins]

On 26/07/15 21:09, Alec Robertson wrote:
> I’ve just updated to the newest trunk release of OpenWRT Chaos Calmer 
> (fresh install) and the SQM QOS from OPKG interestingly does include 
> Cake as a qdisc but neither layer_cake.qos nor piece_of_cake.qos are 
> available as setup scripts.
>
> I’m still trying out Cake so I’ll be back soon with some feedback.
>

You should find the cake option there does nothing?

It'll only work if you have the "kmod-sched-cake" package providing 
/lib/modules/*/sch_cake.ko.  It's only in Dave's recent experimental builds.

fq_codel is the more supported option and serves the same functions.  If 
you can notice any difference yet, I think we'd love to hear about it.  
Currently I believe the noticeable differences are

1. if your router has TCP offloads enabled, cake undoes ("peels") it 
some to improve latency.  (Getting this past review for mainline Linux 
sounds increasingly "interesting").
2. for networks with many flows, cake works much harder to avoid "hash 
collision" (entirely?), so every flow gets a fair share. fq_codel 
defaults to 1000 hash buckets (but collision probability will increase 
well before that point, see "birthday paradox").

1) seems a real concern for some new routers.  If you are affected you 
could add a boot script using ethtool.

The idea is it's not optimal to disable offloads universally... maybe if 
you're sharing a usb drive from the router as well or something.  Having 
cake handle it works as a great default configuration.  (I just suspect 
Linux devs would ask why the feature can't be enabled on other packet 
schedulers, e.g. by using a stackable peeler qdisc).

Alan

Re: [Cake] How to test Cake on TP-Link WDR3600

[Jonathan Morton]

[-- Attachment #1: Type: text/plain, Size: 480 bytes --]

Three reasons why a stackable peeler doesn't work so well:

- There is some overhead from stacking, due to passing packets up and down
the stack. It also incurs at least one packet of unmanaged buffer.

- A separate qdisc would not have the information about link bandwidth and
active flow count that cake maintains, and uses to influence the peeler.

- Peeling to individual IP packets is strictly necessary when encapsulation
overhead needs to be calculated.

- Jonathan Morton

[-- Attachment #2: Type: text/html, Size: 571 bytes --]

Re: [Cake] How to test Cake on TP-Link WDR3600

[Alan Jenkins]

On 26/07/15 22:20, Jonathan Morton wrote:
>
> Three reasons why a stackable peeler doesn't work so well:
>
> - There is some overhead from stacking, due to passing packets up and 
> down the stack.
>

> It also incurs at least one packet of unmanaged buffer.
>

> - A separate qdisc would not have the information about link bandwidth 
> and active flow count that cake maintains, and uses to influence the 
> peeler.
>

Sold!  Not that I'm a netdev reviewer, but it's a nice explanation of 
the design.

I.e. cake knows the minimum processing necessary to achieve the goal.  
The rest can still be left to hardware.

And now I recall peeling came from tbf, I guess it's not that odd.

Thanks for the illumination.
Alan

> - Peeling to individual IP packets is strictly necessary when 
> encapsulation overhead needs to be calculated.
>
> - Jonathan Morton
>

Re: [Cake] How to test Cake on TP-Link WDR3600

[Sebastian Moeller]

Hi Alec,

On Jul 26, 2015, at 14:32 , Alec Robertson <alecrobertson13@gmail.com> wrote:

> Am I right to assume that with Cake more bandwidth should be available to use, without affecting latency?

	This depends; compared to not using SQM you will loose some bandwidth, BUT the latency should stay nicely bound. The main improvement in a nutshell is that with properly configured SQM (with cake being the easiest to setup variant with lots of clever tricks under the hood) your link will still stay reasonably snappy even under saturating load. In other words the links keeps useable when you actually use it, most people consider the typically required bandwidth sacrifice of 10-15% for downstream acceptable…

	By the way, if I could convince you to post the results of the following commands run on your router via ssh:

cat /etc/config/sqm
/etc/init.d/sqm stop
tc -d qdisc
/etc/init.d/sqm start
tc -d qdisc

I would be delighted ;)


Best Regards
	Sebastian


> 
> --
> Alec Robertson
> 
> 
> On Sun, Jul 26, 2015 at 10:11 AM, Alan Jenkins <alan.christopher.jenkins@gmail.com> wrote:
> 
> 
> >> Quick sub-question (off-topic so my apologies), this firmware I’m using that I linked to previously, has HTTPS enabled which means every time I go to Luci I get a security error in Chrome. How do I disable HTTPS? 
> > I would say you do not disable it, but rather look why chrome complains (it might be that chrome dislikes self-signed certificates) and try to convince chrome to accept the certificate nevertheless. Or you could try firefox ;) 
> 
> Firefox is the definitely the simplest browser for this, it's the only 
> one I've used. Just click through the instructions. A "permanent 
> exception" is the default, which should actually help security. Feels 
> ironic as I remember Firefox moving first on this & hence complaints 
> about the scary warning messages etc. 
> 
> Searching instructions for Chrome on Linux ("ssl exception" OR "self 
> signed certificate") they use a certutil command. Internet Explorer 
> will use the Windows cert store, same with Chrome on Windows. (If you 
> can add to the store using IE, that may be simplest & will cover both). 
> 
> Additional requirement for those methods should be that the cert CN 
> matches the URL you access. Not sure about Chrome, but for general 
> paranoia you should check that CN / common name / "issued to" doesn't 
> say "*" i.e. "everywhere". 
> 
> Access the router using `http://hostname` matching the router hostname 
> (as per /etc/config/system). dnsmasq will let that work. If you've 
> changed the hostname, re-gen the cert by removing it and restarting. 
> 
> rm /etc/uhttpd.crt /etc/uhttpd.key 
> /etc/init.d/uhttpd restart 
> 
> Alan 
> 
> _______________________________________________
> Cake mailing list
> Cake@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/cake

Re: [Cake] How to test Cake on TP-Link WDR3600

[Sebastian Moeller]

Hi Alec,

On Jul 26, 2015, at 22:09 , Alec Robertson <alecrobertson13@gmail.com> wrote:

> I’ve just updated to the newest trunk release of OpenWRT Chaos Calmer (fresh install) and the SQM QOS from OPKG interestingly does include Cake as a qdisc

	This is because currently we hard-code the qdisc lists and do not check whether these are actually available, but that is on my todo list, no ETA though.


> but neither layer_cake.qos nor piece_of_cake.qos are available as setup scripts.

	Yepp, the official packages typically lag a bit behind as we first need to actually test the new stuff (this is why I had hoped to get some feedback how piece_of_cake.qos worked for you; so that we can push it into the official repositories...)

> 
> I’m still trying out Cake so I’ll be back soon with some feedback.

	I would love to get some results from Dave’s build (lupin?) if you still have those around?

Best Regards
	Sebastian

> 
> --
> Alec Robertson
> 
> 
> On Sun, Jul 26, 2015 at 1:35 PM, Jonathan Morton <chromatix99@gmail.com> wrote:
> 
> That's a question with nuanced answers, but in general yes, at least slightly.
> 
> - Jonathan Morton
> 
> 
> _______________________________________________
> Cake mailing list
> Cake@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/cake

Re: [Cake] How to test Cake on TP-Link WDR3600

[Sebastian Moeller]

Hi Jonathan,

On Jul 26, 2015, at 23:20 , Jonathan Morton <chromatix99@gmail.com> wrote:

> Three reasons why a stackable peeler doesn't work so well:
> 
> - There is some overhead from stacking, due to passing packets up and down the stack. It also incurs at least one packet of unmanaged buffer.
> 
> - A separate qdisc would not have the information about link bandwidth and active flow count that cake maintains, and uses to influence the peeler.
> 
> - Peeling to individual IP packets is strictly necessary when encapsulation overhead needs to be calculated.

	Is not strictly right? I thought actual peeling or decomposition into maximally MTU-sized packets is not required, only doing the calculation how many packets would be the result and the size of each?

Best Regards
	Sebastian


> 
> - Jonathan Morton
> _______________________________________________
> Cake mailing list
> Cake@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/cake

Re: [Cake] How to test Cake on TP-Link WDR3600

[Benjamin Cronce]

[-- Attachment #1: Type: text/plain, Size: 3075 bytes --]

On Sun, Jul 26, 2015 at 4:05 PM, Alan Jenkins <
alan.christopher.jenkins@gmail.com> wrote:

> On 26/07/15 21:09, Alec Robertson wrote:
>
>> I’ve just updated to the newest trunk release of OpenWRT Chaos Calmer
>> (fresh install) and the SQM QOS from OPKG interestingly does include Cake
>> as a qdisc but neither layer_cake.qos nor piece_of_cake.qos are available
>> as setup scripts.
>>
>> I’m still trying out Cake so I’ll be back soon with some feedback.
>>
>>
> You should find the cake option there does nothing?
>
> It'll only work if you have the "kmod-sched-cake" package providing
> /lib/modules/*/sch_cake.ko.  It's only in Dave's recent experimental builds.
>
> fq_codel is the more supported option and serves the same functions.  If
> you can notice any difference yet, I think we'd love to hear about it.
> Currently I believe the noticeable differences are
>
> 1. if your router has TCP offloads enabled, cake undoes ("peels") it some
> to improve latency.  (Getting this past review for mainline Linux sounds
> increasingly "interesting").
> 2. for networks with many flows, cake works much harder to avoid "hash
> collision" (entirely?), so every flow gets a fair share. fq_codel defaults
> to 1000 hash buckets (but collision probability will increase well before
> that point, see "birthday paradox").
>

I was wondering about this. I'm going off of memory, but I read a paper a
while back that said they tested link speeds from 500Mb/s to 2.5Gb/s and
they saw these same characteristics when sending over 10,000 flows over the
congested link.

1) Never more than 200 flows of data in the queue at any given time
2) Never more than 30 flows with more than one packet in the queue at a time

The birthday attack of all 200 flows into 1000 buckets is pretty bad, but
most of those flows are not greedy at any given moment, it's mostly those
30 you need to worry about. The paper I was reading was talking about 10s
of thousands of flows, so I assume there are many greedy TCP flows, but
only 30 have more than one packet in the queue at a time. Assuming this is
true, I wonder what implications this has and what Cake typically sees. Of
course 500Mb is much faster than most consumer connections, but they stated
they saw no difference in queuing even with a large difference in
bandwidth. Because these were not consumer connections, I assume buffers
were properly sized even if FIFO.

Again, going off of memory, I could have gotten a few things out of
context, but it seemed fairly strait forward.


>
> 1) seems a real concern for some new routers.  If you are affected you
> could add a boot script using ethtool.
>
> The idea is it's not optimal to disable offloads universally... maybe if
> you're sharing a usb drive from the router as well or something.  Having
> cake handle it works as a great default configuration.  (I just suspect
> Linux devs would ask why the feature can't be enabled on other packet
> schedulers, e.g. by using a stackable peeler qdisc).
>
>
> Alan
>

[-- Attachment #2: Type: text/html, Size: 3858 bytes --]

Re: [Cake] How to test Cake on TP-Link WDR3600

[Dave Taht]

On Sun, Aug 2, 2015 at 9:04 PM, Benjamin Cronce <bcronce@gmail.com> wrote:
>
>
> On Sun, Jul 26, 2015 at 4:05 PM, Alan Jenkins
> <alan.christopher.jenkins@gmail.com> wrote:
>>
>> On 26/07/15 21:09, Alec Robertson wrote:
>>>
>>> I’ve just updated to the newest trunk release of OpenWRT Chaos Calmer
>>> (fresh install) and the SQM QOS from OPKG interestingly does include Cake as
>>> a qdisc but neither layer_cake.qos nor piece_of_cake.qos are available as
>>> setup scripts.
>>>
>>> I’m still trying out Cake so I’ll be back soon with some feedback.
>>>
>>
>> You should find the cake option there does nothing?
>>
>> It'll only work if you have the "kmod-sched-cake" package providing
>> /lib/modules/*/sch_cake.ko.  It's only in Dave's recent experimental builds.
>>
>> fq_codel is the more supported option and serves the same functions.  If
>> you can notice any difference yet, I think we'd love to hear about it.
>> Currently I believe the noticeable differences are
>>
>> 1. if your router has TCP offloads enabled, cake undoes ("peels") it some
>> to improve latency.  (Getting this past review for mainline Linux sounds
>> increasingly "interesting").
>> 2. for networks with many flows, cake works much harder to avoid "hash
>> collision" (entirely?), so every flow gets a fair share. fq_codel defaults
>> to 1000 hash buckets (but collision probability will increase well before
>> that point, see "birthday paradox").
>
>
> I was wondering about this. I'm going off of memory, but I read a paper a
> while back that said they tested link speeds from 500Mb/s to 2.5Gb/s and
> they saw these same characteristics when sending over 10,000 flows over the
> congested link.
>
> 1) Never more than 200 flows of data in the queue at any given time
> 2) Never more than 30 flows with more than one packet in the queue at a time
>
> The birthday attack of all 200 flows into 1000 buckets is pretty bad, but
> most of those flows are not greedy at any given moment, it's mostly those 30
> you need to worry about. The paper I was reading was talking about 10s of
> thousands of flows, so I assume there are many greedy TCP flows, but only 30
> have more than one packet in the queue at a time. Assuming this is true, I
> wonder what implications this has and what Cake typically sees.

This is one of the exciting-for-research parts of cake, we can actually try
real workloads and measure, rather than just math.

I added the ability to track active flows in the current git head for
the out of tree version.

>Of course
> 500Mb is much faster than most consumer connections, but they stated they
> saw no difference in queuing even with a large difference in bandwidth.
> Because these were not consumer connections, I assume buffers were properly
> sized even if FIFO.


https://team.inria.fr/rap/files/2013/12/KMOR05b.pdf might be helpful to read.

good math. for what is tractable that way, anyway.

> Again, going off of memory, I could have gotten a few things out of context,
> but it seemed fairly strait forward.
>
>>
>>
>> 1) seems a real concern for some new routers.  If you are affected you
>> could add a boot script using ethtool.
>>
>> The idea is it's not optimal to disable offloads universally... maybe if
>> you're sharing a usb drive from the router as well or something.  Having
>> cake handle it works as a great default configuration.  (I just suspect
>> Linux devs would ask why the feature can't be enabled on other packet
>> schedulers, e.g. by using a stackable peeler qdisc).
>>
>>
>> Alan
>
>
>
> _______________________________________________
> Cake mailing list
> Cake@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/cake
>



-- 
Dave Täht
worldwide bufferbloat report:
http://www.dslreports.com/speedtest/results/bufferbloat
And:
What will it take to vastly improve wifi for everyone?
https://plus.google.com/u/0/explore/makewififast