For starting I context, I have a FEM with 2 primary submodels that are broken out into separate include files. Each of the two parts runs without issue on their own (free-free modes, fixed modes, and random vibe cases). When running together the two parts are connected by a ring of CBEAMS representing a bolted interface. When I attach the two however, the job fatals and I run into an error message I am not familiar with. Despite my best debugging efforts I have not been able to overcome it.

The message is repeated 20-30 times, though not identically every time. The messages are:
*** USER FATAL MESSAGE 4129 (GP6)
The Area of Element XXXXX is Zero.

Where the "XXXXX" is a mixture of "*********", "0", or EIDs that either are CONMs or don't exist (mostly the latter).

I believe it has something to do with NSM region assignments or glued contact. I'm leaning that direction primarily because they have values specified by area. It would make sense if Nastran was incorrectly trying to assign those properties to a zero area element like a CONM2. The fact that the portion of the model with NSM and contact runs fine on its own has me stumped. It's also odd that it is calling out non-existent elements in the errors, but I verified there are no typos or misaligned columns.

I have verified there is no duplicate numbering or overlap in numbering 4 or 5 times now. I have tried forgoing the separate include files and just letting FEMAP merge the models into a single FEM (both by simply dragging decks in and using the built-in 'merge' command), but no change. I have tried re-setting up the contact from scratch numerous times.