Reloading gunicorn worker which takes long time to load

Question

I have a Nginx+Gunicorn+Flask application which provides some APIs using a machine learning model. The model is stored in a large file and takes long time (about one minute) to be loaded into memory. Now I want to update the model file and restart the worker without downtime. How can I configure gunicorn (or something) to wait until new processes load?

Just sending kill -HUP seems to be insufficient for the purpose. Requests which come before the new process loads seems to be dropped.

Answer 1

gunicorn has a argument:

-t, --timeout.

Workers silent for more than this many seconds are killed and restarted.

Answer 2

Sometimes, for me when there is a compute intensive workload in the server, gunicorn restart takes a lot of time to restart.

This gets the ports of gunicorn and kills the workers right away.

sudo kill -9 `sudo lsof -n -i | grep gunicorn | awk '{print $2}'`

and then restart

sudo service gunicorn restart