TerminatedWorkerError while running example code on Python 3.9

[Aniket-Pradhan]

Hello there

I tried a simple example from joblib docs about parallel loops using multiprocessing, and I got a whole bunch of errors instead of the expected output. I tried the code on Python 3.9 which is currently in the beta stage.

Steps to Reproduce:

1. Copy and paste this simple example from the documentation (https://joblib.readthedocs.io/en/latest/parallel.html)

Python version: Python 3.9.0rc1
joblib version: 0.16.0

Error Output:

exception calling callback for <Future at 0x7f54e0cc38b0 state=finished raised TerminatedWorkerError>
Traceback (most recent call last):
  File "/usr/lib/python3.9/site-packages/joblib/externals/loky/_base.py", line 625, in _invoke_callbacks
    callback(self)
  File "/usr/lib/python3.9/site-packages/joblib/parallel.py", line 347, in __call__
    self.parallel.dispatch_next()
  File "/usr/lib/python3.9/site-packages/joblib/parallel.py", line 780, in dispatch_next
    if not self.dispatch_one_batch(self._original_iterator):
  File "/usr/lib/python3.9/site-packages/joblib/parallel.py", line 847, in dispatch_one_batch
    self._dispatch(tasks)
  File "/usr/lib/python3.9/site-packages/joblib/parallel.py", line 765, in _dispatch
    job = self._backend.apply_async(batch, callback=cb)
  File "/usr/lib/python3.9/site-packages/joblib/_parallel_backends.py", line 531, in apply_async
    future = self._workers.submit(SafeFunction(func))
  File "/usr/lib/python3.9/site-packages/joblib/externals/loky/reusable_executor.py", line 177, in submit
    return super(_ReusablePoolExecutor, self).submit(
  File "/usr/lib/python3.9/site-packages/joblib/externals/loky/process_executor.py", line 1102, in submit
    raise self._flags.broken
joblib.externals.loky.process_executor.TerminatedWorkerError: A worker process managed by the executor was unexpectedly terminated. This could be caused by a segmentation fault while calling the function or by an excessive memory usage causing the Operating System to kill the worker.

The exit codes of the workers are {EXIT(1), EXIT(1)}
Traceback (most recent call last):
  File "<string>", line 1, in <module>
  File "/usr/lib/python3.9/site-packages/joblib/parallel.py", line 1042, in __call__
    self.retrieve()
  File "/usr/lib/python3.9/site-packages/joblib/parallel.py", line 921, in retrieve
    self._output.extend(job.get(timeout=self.timeout))
  File "/usr/lib/python3.9/site-packages/joblib/_parallel_backends.py", line 542, in wrap_future_result
    return future.result(timeout=timeout)
  File "/usr/lib64/python3.9/concurrent/futures/_base.py", line 440, in result
    return self.__get_result()
  File "/usr/lib64/python3.9/concurrent/futures/_base.py", line 389, in __get_result
    raise self._exception
  File "/usr/lib/python3.9/site-packages/joblib/externals/loky/_base.py", line 625, in _invoke_callbacks
    callback(self)
  File "/usr/lib/python3.9/site-packages/joblib/parallel.py", line 347, in __call__
    self.parallel.dispatch_next()
  File "/usr/lib/python3.9/site-packages/joblib/parallel.py", line 780, in dispatch_next
    if not self.dispatch_one_batch(self._original_iterator):
  File "/usr/lib/python3.9/site-packages/joblib/parallel.py", line 847, in dispatch_one_batch
    self._dispatch(tasks)
  File "/usr/lib/python3.9/site-packages/joblib/parallel.py", line 765, in _dispatch
    job = self._backend.apply_async(batch, callback=cb)
  File "/usr/lib/python3.9/site-packages/joblib/_parallel_backends.py", line 531, in apply_async
    future = self._workers.submit(SafeFunction(func))
  File "/usr/lib/python3.9/site-packages/joblib/externals/loky/reusable_executor.py", line 177, in submit
    return super(_ReusablePoolExecutor, self).submit(
  File "/usr/lib/python3.9/site-packages/joblib/externals/loky/process_executor.py", line 1102, in submit
    raise self._flags.broken
joblib.externals.loky.process_executor.TerminatedWorkerError: A worker process managed by the executor was unexpectedly terminated. This could be caused by a segmentation fault while calling the function or by an excessive memory usage causing the Operating System to kill the worker.

The exit codes of the workers are {EXIT(1), EXIT(1)}

[Aniket-Pradhan]

It could be an issue with the Loky backend, but I was unsure, hence reported here.

[pierreglaser]

Hi, thanks for your report, indeed loky is not compatible yet with Python 3.9. We will be working on a fix as an stable Python 3.9 release is likely to happen soon. See joblib/loky#251

[Aniket-Pradhan]

Hi, thanks for your report, indeed loky is not compatible yet with Python 3.9. We will be working on a fix as an stable Python 3.9 release is likely to happen soon. See joblib/loky#251

That's what we had been thinking. Thanks for your help here. I shall close the ticket and will "bug" the devs there 😛