I have tracked the source of the hang problem into the ethernet drivers.
The only workable way I’ve found to debug the problem is to turn a LED on as I enter questionable code and turn if off on exit. Then when it hangs I look to see if the LED is on or off and then further subdivide the code.
Progress is painfully slow as it takes days for a hang to occur.
As of yesterday I had proven that the hang is occurring in the sendUDP procedure that is part of the code for the nano ethernet shield. Somehow I’m not surprised – this nano shield has already proven to be questionable.