Due to some bad math on my side, I passed max_workers=0 to concurrent.futures.ThreadPoolExecutor. This didn't fail properly, but hanged. The same behaviour occurs in ProcessPoolExecutor, but this time it fails internally with something like this:

Exception in thread Thread-1:
Traceback (most recent call last):
  File "C:\Python34\lib\threading.py", line 921, in _bootstrap_inner
    self.run()
  File "C:\Python34\lib\threading.py", line 869, in run
    self._target(*self._args, **self._kwargs)
  File "C:\Python34\lib\concurrent\futures\process.py", line 225, in _queue_management_worker
    assert sentinels
AssertionError

The attached patch checks that *max_workers* is <= 0 and raises ValueError if so.

For instance, multiprocessing behaves like this:

>>> multiprocessing.Pool(-1)
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "C:\Python34\lib\multiprocessing\context.py", line 118, in Pool
    context=self.get_context())
  File "C:\Python34\lib\multiprocessing\pool.py", line 157, in __init__
    raise ValueError("Number of processes must be at least 1")
ValueError: Number of processes must be at least 1
>>>

Attached patch with improvements suggested by Charles-François Natali. Thank you for the review.

I confirm the bug.
The patch looks good.

The patch looks good to me too. I'll commit it soon.

New changeset 3024ad49f00e by Brian Quinlan in branch 'default':
Issue #21362: concurrent.futures does not validate that max_workers is proper
http://hg.python.org/cpython/rev/3024ad49f00e

Committed, thanks for the fix!