I'm running MLServer on MacOS (arm). I was following Serving XGBoost models example. I receive a 500 error when requesting a model with url http://localhost:8080/v2/models/mushroom-xgboost/versions/v0.1.0/infer.

MLServer throws an error

...
  File "/opt/homebrew/Cellar/[email protected]/3.10.9/Frameworks/Python.framework/Versions/3.10/lib/python3.10/multiprocessing/queues.py", line 126, in qsize
    return self._maxsize - self._sem._semlock._get_value()
NotImplementedError

Developers of queues.py left a comment that this is broken on MacOS 😬

    def qsize(self):
        # Raises NotImplementedError on Mac OSX because of broken sem_getvalue()
        return self._maxsize - self._sem._semlock._get_value()

Is it possible to get around this when running models with MLServer?
Thanks!