Open OsaRobots opened 3 months ago
Thanks for reporting. I've edited the script slightly to make it runnable, but I don't see the same issue, neither on Linux nor Windows 10. Do you the the issue also with the reduced script, or only in context of a larger program?
Hey sorry for the late reply. It happens in the context of the larger program, but that's key part that seems to crash. It was easy enough to get around for my purposes by just catching the exception and continuing, but still no idea what really causes it. My guess would be some issue with multithreading, since it doesn't happen otherwise. If you want, the issue can be closed though!
Is the larger program something you can share?
I can email it to you privately, if that's okay with you.
Sure, you can reach me at niklas.fiekas@backscattering.de. Maybe I can spot something. If not, I am going to give up and close the issue until new information becomes available.
I'm getting a CancelledError when analyzing a sequence positions in a given game from a public dataset, and I'm really unsure how to debug it.
Minimal code to reproduce:
Stacktrace:
Before the CancelledError, I also get the below warning:
which repeats itself a bit more than 10 times before I eventually get a
TimeoutError
caused by theCancelledError
. This 'CancelledError' has the following traceback:Expected behavior:
The computation should go as normal, getting the score and nodes for every board position in the given game across the provided depths. For reference, this code works perfectly fine for 1400 games before this exact game.
Here's the last thing the logger gets.
The fen "r1b1kb1r/pp3pp1/2nqp3/8/1P3pP1/3P1P2/2P1N2P/R2QK2R b KQkq - 0 15" doesn't even get half way through all the board positions.
Environment: Python version: 3.12 python-chess version: 1.10 Operating System: Windows 10 (2.4 GHz Intel Broadwell) Chess Engine Used: Stockfish 14