HE 6to4 Relays

[nickbeee]

I was investigating 6to4 the other day pinging and tracerouting 192.88.99.1. I found what appears HE router in London - set up a Cisco router to run 6to4 and it seemed to work reasonably well. I can see lots of reasons not to use 6to4 though as you have to allow IPv4 connections to permit traffic from 2002::/16.  :o

[amcintosh]

Does this mean that HE has a Minneapolis presence?

Tracing route to 192.88.99.1 over a maximum of 30 hops

  1    <1 ms    <1 ms    <1 ms  192.168.0.2
  2    46 ms     7 ms     7 ms  69-42-244-1.fedtel.stellarllc.net [69.42.244.1]
  3   213 ms     9 ms     7 ms  66-234-124-249.fedtel.stellarllc.net [66.234.124.249]
  4   403 ms   583 ms   839 ms  minneapolis.stellarllc.net [66.234.112.201]
  5   637 ms   636 ms   640 ms  10gigabitethernet2-1.core1.msp1.he.net [216.66.74.225]
  6   865 ms   762 ms   584 ms  10gigabitethernet7-4.core1.chi1.he.net [184.105.213.133]
  7   704 ms   701 ms   732 ms  192.88.99.1

Trace complete.

[kasperd]

Does this mean that HE has a Minneapolis presence?

The hostnames of those routers certainly imply that the peering is located in Minneapolis. And on the official list of places where HE has peerings you can even find the address.

What's a bit disturbing about your traceroute output is the roundtrip times. More than half a second of roundtrip time between you and the closest 6to4 relay is really bad. That's an example of why 6to4 should be avoided unless both endpoints are using 6to4 or your only alternative is Teredo.

What is strange is the apparent delay between those two routers both located in the same peering point. This can of course be caused by different return routes. Though 66.234.112.201 and 216.66.74.225 are probably located in the same building, they are operated by different companies and may route return packets to you through different connections. (If you are curious about that, maybe ping -R can give you some more information).