Ok,..been reading about the problems from Germany and if I may, I think I can get a handle on this.
When you see a traceroute, each and every 'hop' is a router. The things you do not see are switches. But, I will leave switches to another discussion.
Typically, when you see high pings/packet losses at either the last hop or next to the last hop of a traceroute, it is partly due to an asynchronous route.
Asynchronous routes occur when the data path to your destination is different than the data path from the destination back to you.
In this case, it appears the problem is the return path back to you. Because you cannot see that path, you cannot get a true evaluation of your connection. Somewhere along the way on the return path, there is a problem. It could be a router, switch, or physical link problem.
Why would it happen at the MA and not other places? Well, the odds are pretty slim that any given connection you have to any give point on the Internet is not going to take the same path, but if it does, you will see the same problem.
Not having access to the return path makes it difficult to see where the problem is.
To answer some other questions; The reason all the other arenas may be working fine for you is due to the different path and backbone ISP AppLink uses.
I know the knee jerk reaction is to think HTC is the problem, but in fact, the real issue may be beyond thier control at this time. Hopefully this will not be an issue much longer, as we are making headway to getting the new pipes to AppLink so we can get the MA back here.
