Maybe a bug of reopen_files when handle signal USR1 ？ #1739
I am sorry I don't know how to discuss on mailing list.
when gunicorn received signal USR1, my flask app's logger will write log to another file, which is unexpected.
my logger in flask app use handler implemented by myself, and my handler does not open file of handler.baseFilename. when gunicorn worker handle USR1, it will search all logger in that gunicorn worker process and reopen the handler's stream by
To be honest, I do not understand why gunicorn need reopen all logger in a python process , shouldn’t it pick out only gunicorn's logger ？however, if it's really required, the reopen_files logic must be safe or do not have effect to other logger at least.
I am happy to submit a pr to optimize here :)
The text was updated successfully, but these errors were encountered:
I think I agree with that. That would be a small breaking change, so I think I like your other idea. Let's use