I came up with this workaround after adding a breakpoint on the "return result" statement at the end of the matchblock.getmaches() function avoided the deadlock every time, so I'm suspecting some sort of race condition or something here.

Took a quick look at the documentation for multiprocessing, if I understand the join() method (https://docs.python.org/3.8/library/multiprocessing.html#multiprocessing.pool.Pool.join) then perhaps including that instead of the sleep might fix this as well?

You are right, thanks. It seems to work just fine now.