Thursday, September 02, 2010

Traceroute puzzle

Which ISPs appear in the following traceroute, and where do they interconnect?

  1 swiCE3-10GE-1-4.switch.ch (130.59.36.210) 0 msec # local AS: 559
  2 bb1.tor.primus.ca (195.69.145.154) [AS 3549] 108 msec
  3 216.254.129.3 (216.254.129.3) [AS 6407] 108 msec
  4 www.primus.ca (216.254.141.10) [AS 6407] 104 msec

3 comments:

Louis said...

switch ;)
global crossing
then the destination Primus AS 6407

what is strange is that they do not seem to be connected on my records

PS: where do you get so many responding routers!

arska said...

according to http://lg.net.switch.ch/lg/lg-opencms.cgi as559 and as6407 have a direct adjacency. according to http://www.switch.ch/network/operation/weathermap/ "CE" (geneva?) connects to AMS-IX. According to https://www.euro-ix.net/member/m/isp/choosing/ixpmembers/search?query=as6407&commit=Search (but in contrast to http://www.peeringdb.com/view.php?asn=6407) as6407 peers at AMS-IX. If you don't remember peering with as6407 directly you seem to peer with the route server(s) without prepending the ams-ix as. So primus has a remote peering session to ams-ix. 195.69.145.154 is in 195.69.144.0/22 called "AMS-IX3" according to whois. as559 learns 195.69.144.0/22 for some reason from as3549, thats why the ams-ix IP shows up as "[AS 3549]" in the traceroute.
What was the puzzle ? ;)

-Aarno

Simon Leinen said...

Aarno got almost everything right. The mapping to AS3549 is actually from 0.0.0.0/0 - the router I did the traceroute from has a default route to Global Crossing.

Our connection to AMS-IX is also a long tail - if you check out the LOC record you'll see that our router "swiCE2" is located at CERN (Geneva).

I'd like to see some work on this trend(?) to connect to IXPs through long connections, see my older posting on http://blog.simon.leinen.ch/2009/11/mapping-ixp-access-links.html