[117331] in North American Network Operators' Group
Re: Traffic Shaping on ISPs
daemon@ATHENA.MIT.EDU (Astrodog)
Thu Sep 10 06:46:16 2009
In-Reply-To: <C73CE12E0E2E6D4AA8C97475050023A6100D516EEF@proserpine.paragon.crypticstudios.com>
Date: Thu, 10 Sep 2009 05:45:04 -0500
From: Astrodog <astrodog@gmail.com>
To: Jake Vargas <jvargas@crypticstudios.com>
Cc: "nanog@nanog.org" <nanog@nanog.org>
Errors-To: nanog-bounces+nanog.discuss=bloom-picayune.mit.edu@nanog.org
BT/Virgin throttling information:
http://www.theregister.co.uk/2009/02/11/virgin_media_throttle_extension/
http://www.theregister.co.uk/2008/08/07/bt_samknows_bandwidth_throttling/
http://news.bbc.co.uk/2/hi/technology/8077839.stm
http://www.telegraph.co.uk/technology/news/5431052/BT-admits-limiting-downl=
oad-speeds-and-throttling-iPlayer-traffic.html
It looks like the throttling window lines up fairly well with the
times you're seeing problems. Now, if that's the throttling, or just
BT's network being oversubscribed... who knows.
Good luck getting your problem cleared up.
---- Harrison Grundy
On Thu, Sep 10, 2009 at 4:49 AM, Jake Vargas<jvargas@crypticstudios.com> wr=
ote:
>> While I cannot confirm officially, there is a lot of rumors, that severa=
l
>> larger UK ISP's are throttling traffic at that time period.
>> I am not sure who to contact, but the individual ISP's to solve this, fr=
om
>> your point, maybe another NANOG'er knows.
>
> Hi Lasse,
>
> Thanks for the reply. We wrote an app to reveal troubles.
>
> Just to satisfy any curiosity and get some facts out, I will provide a re=
al world example (1 of many) from a direct test of one of our BT sourced cu=
stomers (this is from a 08-29 test at ~22:04 hours GMT):
>
> Date =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0IP =A0 =A0 =A0 =A0 =A0 =A0 =
=A0 =A0 =A0 =A0 =A0RTT =A0 =A0 Port =A0 =A0ActualRecv =A0 =A0 =A0NicSent Ni=
cRecv
> 090829 22:04:24 86.128.0.0/10 =A0 103.5 =A0 80 =A0 =A0 =A0199 =A0 =A0 13 =
=A0 =A0 =A0214
> 090829 22:04:24 86.128.0.0/10 =A0 =A0 =A0 =A0 =A0 103.5 =A0 80 =A0 =A0 =
=A0199 =A0 =A0 13 =A0 =A0 =A0214
> 090829 22:04:24 86.128.0.0/10 =A0 =A0 =A0 =A0 =A0 103.5 =A0 443 =A0 =A0 2=
00 =A0 =A0 12 =A0 =A0 =A0215
> 090829 22:04:24 86.128.0.0/10 =A0 =A0 =A0 =A0 =A0 103.5 =A0 443 =A0 =A0 1=
99 =A0 =A0 12 =A0 =A0 =A0214
> 090829 22:04:24 86.128.0.0/10 =A0 =A0 =A0 =A0 =A0 103.5 =A0 7255 =A0 =A02=
=A0 =A0 =A0 2 =A0 =A0 =A0 5
> 090829 22:04:24 86.128.0.0/10 =A0 =A0 =A0 =A0 =A0 103.5 =A0 7255 =A0 =A03=
=A0 =A0 =A0 1 =A0 =A0 =A0 4
> 090829 22:04:24 86.128.0.0/10 =A0 =A0 =A0 =A0 =A0 103.5 =A0 7003 =A0 =A03=
=A0 =A0 =A0 2 =A0 =A0 =A0 5
> 090829 22:04:24 86.128.0.0/10 =A0 =A0 =A0 =A0 =A0 103.5 =A0 7003 =A0 =A04=
=A0 =A0 =A0 1 =A0 =A0 =A0 5
> 090829 22:04:24 86.128.0.0/10 =A0 =A0 =A0 =A0 =A0 103.5 =A0 7202 =A0 =A02=
7 =A0 =A0 =A03 =A0 =A0 =A0 32
> 090829 22:04:24 86.128.0.0/10 =A0 =A0 =A0 =A0 =A0 103.5 =A0 7202 =A0 =A02=
4 =A0 =A0 =A02 =A0 =A0 =A0 29
> 090829 22:04:24 86.128.0.0/10 =A0 =A0 =A0 =A0 =A0 103.5 =A0 7499 =A0 =A02=
7 =A0 =A0 =A03 =A0 =A0 =A0 32
> 090829 22:04:24 86.128.0.0/10 =A0 =A0 =A0 =A0 =A0 103.5 =A0 7499 =A0 =A02=
5 =A0 =A0 =A02 =A0 =A0 =A0 31
> 090829 22:04:24 86.128.0.0/10 =A0 =A0 =A0 =A0 =A0 103.5 =A0 80 =A0 =A0 =
=A0195 =A0 =A0 13 =A0 =A0 =A0206
> Idle NIC bandwidth =A0Send: =A0 =A0 0 KB/sec =A0 Recv: =A0 =A0 0 KB/sec
>
> To remove any doubt we also measured idle bandwidth utilization on the NI=
C when the test wasn't run to remove any other culprit such as torrent down=
load, A/V streaming and etc in the background. In this case, 0/0 on idle us=
e. All results are in KBytes
>
> I withheld the actual IP address of this test and replaced it from the so=
urce prefix. We have quite a few iterations of similar results from other s=
ource addresses from this prefix alone. All appear to exhibit the same issu=
e.
>
> I've already written British Telecom and they never replied.
>
>
>
