I am trying to implement this multiprocessing tutorial for my own purposes. At first I thought it did not scale well, but when I made a reproducible example I found that if
OK! From the docs:
Warning As mentioned above, if a child process has put items on a queue (and it has not used JoinableQueue.cancel_join_thread), then that process will not terminate until all buffered items have been flushed to the pipe. This means that if you try joining that process you may get a deadlock unless you are sure that all items which have been put on the queue have been consumed. Similarly, if the child process is non-daemonic then the parent process may hang on exit when it tries to join all its non-daemonic children. Note that a queue created using a manager does not have this issue. See Programming guidelines.
As I noted in a comment earlier, the code attempts to .join()
processes before the done_queue
Queue is drained - and that after changing the code in a funky way to be sure done_queue
was drained before .join()
'ing, the code worked fine for a million items.
So this is a case of pilot error, although quite obscure. As to why behavior depends on the number passed to main(x)
, it's unpredictable: it depends on how buffering is done internally. Such fun ;-)