Re: Paris/Amsterdam throtling GRE packets?

Actually 80.91.251.102  is in  Ashburn, VA

http://lg.telia.net/


-----------------------------------------------------------------------
Is it possible that ash-bb1-link.telia.net is in the US (my guess near 
Boston MA)  and the big jump in latency is the Atlantic crossing?


On 04/23/2011 11:32 AM, Tobias Skytte wrote:
> Hi,
>
> I have PPTP poptop servers in the US and Denmark that clients connect to from
> all over the world. I have suspected that somehow bandwidth is being limited
> whenever the clients signal is passing through France and Holland, especially
> Paris and Amsterdam but mostly suspecting it is in Paris.
>
> I can deduce this as I can perform traceroutes to the clients that complain of
> low speed and see that there are hops located in Paris and/or Amsterdam.
>
>
> I am normally based in Denmark but as I am now on holiday in Italy I can confirm
> this. The PPTP server in the US is extremely slow (0.2 mbit/sec) whereas the
> server in Denmark gives me 4 mbit/sec. Neither server is under load or
> experiencing problems.
>
> Performing a trace to the US server I get (partly):
>
> 13  89.96.200.21 (89.96.200.21)  28.152 ms 89.96.200.114 (89.96.200.114)  33.955
> ms  34.439 ms
> 14  mno-b1-link.telia.net (213.248.88.233)  36.235 ms  39.057 ms  40.844 ms
> 15  prs-bb2-link.telia.net (80.91.246.160)  59.340 ms  61.892 ms  64.552 ms
> 16  ash-bb1-link.telia.net (80.91.251.102)  147.749 ms  152.830 ms  152.911 ms
> 17  atl-bb1-link.telia.net (80.91.248.137)  169.117 ms  171.069 ms  173.504 ms
>
> Notice the abreviations: mno=Milano, prs=Paris, ash=Amsterdam(probably),
> atl=Atlanta (server is in Atlanta).
>
> Doing trace to the DK server i get (partly):
>
> 13  89.96.200.21 (89.96.200.21)  25.733 ms  27.868 ms 89.96.200.114
> (89.96.200.114)  30.269 ms
> 14  mno-b1-link.telia.net (213.248.88.233)  32.857 ms  36.964 ms  37.479 ms
> 15  ffm-bb1-link.telia.net (80.91.249.38)  48.944 ms  51.017 ms  53.754 ms
> 16  hbg-bb1-link.telia.net (80.91.245.41)  65.639 ms  67.345 ms
> hbg-bb1-link.telia.net (80.91.245.39)  69.963 ms
> 17  kbn-bb1-link.telia.net (80.91.247.225)  80.455 ms kbn-bb1-link.telia.net
> (80.91.247.223)  81.614 ms kbn-bb1-link.telia.net (213.155.130.101)  81.928 ms
>
> Again: mno=Milano, ffm=FrankFurt am Mein, hbg=Hamburg, kbn=Copenhagen.
>
> So my suspicions are confirmed here, the signal to the server in Denmark does
> not go through Paris and is not limited, the signal to the US goes through Paris
> and is severely limited.
>
> They only limit GRE traffic by the way, when I am not on the PPTP server in the
> US I have full speed to everything.
>
> Also it might be worth noting that the customers I get complaints from are
> usually located in France or Spain and are trying to access the server in
> Denmark but are getting throttled, their signals are always going through Paris
> and/or Amsterdam.
>
>
> Since the signal passes Paris on the backbone it must be some backbone hub that
> is set to throttle GRE bandwidth.
>
> Has anybody else noticed something like this? I am suspecting France Telecom for
> doing this by the way. How can they legally do that? Could there be any other
> reasons other than blatant throttling?  If they are doing it how can I make them
> stop? I am currently beta-testing a new server with multiple IP adresses so that
> should alleviate the problem as I'm guessing they're just throttling based on
> the IP address.
>
>
> Cheers,
> T
>
>
>
> ------------------------------------------------------------------------------
> Fulfilling the Lean Software Promise
> Lean software platforms are now widely adopted and the benefits have been
> demonstrated beyond question. Learn why your peers are replacing JEE
> containers with lightweight application servers - and what you can gain
> from the move. http://p.sf.net/sfu/vmware-sfemails
>
>
>
> _______________________________________________
> Poptop-server mailing list
> [email protected]
> https://lists.sourceforge.net/lists/listinfo/poptop-server

------------------------------------------------------------------------------
Fulfilling the Lean Software Promise
Lean software platforms are now widely adopted and the benefits have been 
demonstrated beyond question. Learn why your peers are replacing JEE 
containers with lightweight application servers - and what you can gain 
from the move. http://p.sf.net/sfu/vmware-sfemails