We had this on some machines earlier.
It turns out we had several interfaces that were going up and down multiple times.
NNM, by default, will send out a Node up every time it sees ALL interfaces up, so if you have one interface flipflopping often, each time it comes back up, you would get a new "node up".
My suggestion: Drill into the router in NNM and unmanage all the interfaces except the ones you really need to see. I would expect this would stop the problems you are seeing.