=== genii_ is now known as genii [02:17] Hello, we have a gunicorn app running in Docker, and on redeploy (docker-compose down -> docker-compose up -d ), we get psycopg2.errors.InternalError_: tuple concurrently updated. I suspect the previous gunicorn instance is still running. Running the deploy script twice fixes it. Would sending SIGTERM via docker-compose exec -T pkill -SIGTERM gunicorn before docker-compose down help? Or is there a bette [02:17] r way to handle this? [04:27] not using docker, and having it run in a systemd service, would be a better way, IMHO [04:37] manis05: At a guess the container is not shutting down successfully on docker-compose down. Or there might be some delay in it shutting down. [04:38] I would first try a sleep 10 between the down & up, and then try your pkill after the docker-compose down rather than before it. [04:38] And keep in mind if this is standard docker on Ubuntu you can actually see the docker processes from the host, so you could pkill them from there as well. [04:46] kenyon Thanks for your response, I appreciate. We're moving everything into docker for easier deployment and to fix dependency issues with our code. We've been using systemd for several years [04:47] blahdeblah I really appreciate your insight. That looks like a really nice solution and would be trying it out! :) [04:48] Thank you! [04:53] manis05: I'd recommend watching the logs with `docker compose logs --follow` while you shut it down to see if there are any error messages. [05:12] blahdeblah Got it. Thanks again! [05:14] np [08:33] manis05: and did moving everything into docker make deployments easier? [08:40] mgedmin - Yes, it generally did. It's just we have to run the deploy script twice which is inefficient. === Wyly1 is now known as Wyly === Guest5857 is now known as larkon [21:32] Is the mini.iso suitable for PXE? [21:32] I mean https://cdimage.ubuntu.com/ubuntu-mini-iso/noble/daily-live/current/ [22:37] znf: no https://www.howtogeek.com/what-is-ubuntu-mini-iso/