Batch analysis hangs after processing around 700 samples

[xusheng6]

A user reported that he is using a headless script to process a large number of files, and he observes the script hangs consistently after processing around 700 files, though the exact file it hangs is different every time

The user shared an analysis script in private, but is unable to share the exact database he is using. We may try to reproduce it on a different set of binaries

The user is using stable 5.1 on Windows 11

[xusheng6]

Analysis script: modular burst quantizes surely

[emesare]

The user should use multiprocessing and spawn a throw away process for each view, the underlying issue is likely the absence of a main thread handler in python. Any further triaging of this is likely gated on us actually registering a main thread handler, and or fixing the underlying dependence on the main thread for certain access patterns.

It might actually be some deadlock unrelated to the above but just going off the script will not be enough to determine that, we would need the dataset the user is running on.

[xusheng6]

I am closing this due to lack of sufficient information to move forward