traffic from DE to DE goes via NL->UK->US->FR

Mailman List Mirror (Read Only)
1

Hi,

Just wondering if it is normal for traffic from DE to DE to flow through NL->UK->US->FR and so increase delay nearly 100 times?
Traceroute here: http://pastebin.ca/115200 and there is only 4 AS, so ASPATH does not help a lot in finding such links with a horrifying optimisation. I believe there is much worse links, any software
to detect this? Something like scanning one ip from larger IP blocks with icmp and comparing geotrajectoyi via geoip?

thank you,
AKK

2

You should direct the question to whereever you are a customer.

These things usually happen when one party doesn't want to peer with another party and the one that wants to peer, will route traffic really far away to make sure that both parties are paying for the traffic, thus increasing the motivation for the other party to change their mind regarding peering.

3

...unfortunate series of circumstances... :wink:

but the bottom line is that the economics of it (almost always) comes on top of customer's benefit.

i have the same problem in my city, to reach some (few) networks, since some years now.

...but i guess in central europe (or in the central european internet - london/amsterdam/frankfurt/paris) you don't see that often... :wink:

best regards,

./Carlos Skype: cf916183694

4

Indeed not. In this case (I know the return route goes to GX
as well, but directly so, and then to the target network in
the most direct way) it might be just be a bad choice
(accidentally I presume) of the best path to a destination,
and that kinda happens it seems if you have several links
with a single provider. No clue why closest-exit is not done.

But then, what do I know, I can only guess here.

Alexander

5

Andrius Kazimieras Kasparavic(ius wrote:

Hi,

Just wondering if it is normal for traffic from DE to DE to flow through NL->UK->US->FR and so increase delay nearly 100 times?
Traceroute here: http://pastebin.ca/115200 and there is only 4 AS, so ASPATH does not help a lot in finding such links with a horrifying optimisation. I believe there is much worse links, any software to detect this? Something like scanning one ip from larger IP blocks with icmp and comparing geotrajectoyi via geoip?

thank you,
AKK

I remember two peculiarities.

Between Amsterdam and London packets were summersolting. The fifth packet arrived
before the second. Making VoIP impossible.

In the Cyberbunker every IPv4 address gave a different traceroute. Most addresses
did not work at all.

When I replaced a GrandStream ATA-486 as VoIP gateway and DSL-router by a slow
linux box, that mess cleared. Everything working fine and fast. The ICMP in
the GrandStream was broken. I guess in the Cyberbunker a local router was broken
too. The sh** needed both routers to reach the fan.

Cheers
Peter and Karin