i USES THE gunicorn workers for restarting every 1000 requests. so every worker running on kernel level and assign a unique pid for every worker. so when worker is restarted after a 1000 request then before restart it checks any background task is running or not if not then it will be restart. so my question is

if they check the background task then why he closed the db object for previous requests ??

my memory consumption issue is fix by gunicorn but i face this new issue !!!

some error

(psycopg2.DatabaseError) error with status PGRES_TUPLES_OK and no message from the libpq

(psycopg2.OperationalError) server closed the connection unexpectedly This probably means the server terminated abnormally before or while processing the request.

any suggestion for this issue (guvicorn related)