@NCP Thanks for your time! Finally, I found the issue. Actually, I was in production mode. After updating the hook file, I ran migration every time, but it didn’t seem to take effect. After some time, I had a spark of insight: let’s restart the bench and try it once more.
Before that, I assumed in common_site_config.json where restart_supervisor_on_update is set to true, so I assumed that migrating the system would restart the supervisor. However, I realized that this setting doesn’t actually restart the supervisor. Consequently, even after migration, the Python script didn’t reload, and the code remained the same as before.
I believe this was the problem. After restarting the bench, everything is working fine now. I’ve checked the flow twice, and it seems to work like this. Please correct me if I’m wrong.