Can´t DL nothing.

On that day they were coming from einstein6.aei.uni-hannover.de, today they are coming from einstein1.aei.uni-hannover.de too.

Can you set your Router to use a different DNS server and not your ISPs?

Claggy

Hi jb yyyyaaaa dam isp providers shit happens on there system we get problems and not even a email or message from them saying we are experiseing troubble we are sorry will try to fix .Hopefully when they get there act togeather your problems will go away .Here's hopeing finger crossed for ya mate .So relax have a cone or two or drink whatever ya posion is, those dam greedy isp providers

You can find the link MTU using 'ping'. I only know the linux version, but according to the web 'ping -f -l 1472' works for Windows. Keep increasing (or reducing) the '1472' to find out at which size packets get fragmented.

There are lots of tests you can do if you are interested. But as your other message says the ISP admits they are having problems, it may be best to simply wait for them to fix it.

Otherwise, you could try using 'telnet' to connect to e@h, and see if it connects (and if not what message you get):-

Since 'traceroute' works, 'ping' should work too as both 'traceroute' and 'ping' use the same protocol (ICMP) to do their work.

ping einstein1.aei.uni-hannover.de            
PING einstein1.aei.uni-hannover.de (130.75.116.31) 56(84) bytes of data.
64 bytes from einstein1.aei.uni-hannover.de (130.75.116.31): icmp_req=1 ttl=43 time=57.0 ms
64 bytes from einstein1.aei.uni-hannover.de (130.75.116.31): icmp_req=2 ttl=43 time=57.0 ms

Now, the 'ttl=43' means 'Time to live' is set to 43 in my case. This means the 'ping' can go through a maximum of 43 hops. Look what happens if I set it lower:-

Because I set the 'Time to live' to 10, after 10 hops the request failed, and importantly the hop where it failed reports the failure back.

To bring this all together, that's how 'traceroute' works. First it does a 'ping' (or something similar) with TTL=1, then with TTL=2, then TTL=3, and so on. The point in the route where the hop count limit is reached reports back, and so you can build up a picture of the route.

Here is part of my traceroute output:

Now let's see what happens if I ping the server with TTL set to 9, 10, and 11:

You can see how the hosts reporting "Time to live exceeded" are the same as the hosts in the traceroute output.

So if you want to experiment, you can try pinging various points along the 'traceroute' output, and see if there is a point where packet loss occurs. Sometimes it won't work for a particular hop ('* * *' on traceroute output) because 'ping'/ICMP protocol is disabled for that hop; it will still work for other hops though.

To really find out what is going on for yourself, you could install software like Wireshark on one of the hosts, and perform a capture of network traffic to E@H. If you post the output somewhere, I'll have a look at it and see if I can spot anything.