Well, the first thing you want to do is remove the setup setting that is instigating a node restart. In the registry.
Secondly, run a dumplog on rb\nm to see if you can see why the node manager is restarting the rtr processes. If it does not tell you, check dumplog on mds to see what is going on with the private connection between the sides.
The normal way to test the situation is as follows. Stop everything first.
1. Start logger A
2. Start router A.
3. Check the rtr A log with dumplog or evsmon - you should see it sitting there, waiting to talk to its peer (router B)
4. Start router B. Watch the log of rtr A with evsmon (it was easier on Windows 2003 than Windows 2008 because you could watch the process windows, but we have to live with the change). You should see rtr A detect its peer and start to download the config from logger A into memory. Router B will get mirrored state from Router A. You don't use logger B for this test.
If rtr A does not do this, use regedit to turn up the trace on both router MDS processes (0xffff will be fine - remember to clear this later as it's a hog). Are these guys hooking up across the private link.
Most likely you have messed up the public / private configs. Go over it again. Draw a Visio diagram and mark all host names and addresses on your diagram. Are you using host names for all (including private and private high). In turn, sit on each router and ping the other side by the private host names. Maybe run tracert to check the private path (make sure it's not going out the default gateway). Did you have to create persistent static routes? Are they correct?
Regards,
Geoff
