I've noticed you posting on the Nitrogen mailing list, and as I understand it, you've got it straightened out, but in this situation, I'd kill the running node manually with a ps aux | grep nitrogen
, then kill the process it finds with a simple kill XYZ
.
That, or, I've seen the "Node is not running" thing pop up when the process was launched with a different user, such that you don't have access to the erlang pipe.
Admittedly, my advice isn't terribly scientific (killing a process is pretty nasty), but it's a simple solution if for whatever reason something got hosed during launching and you're unable to attach to the node.