Hi! I’m pretty new to all of this so forgive this basic question. I am running a panel app out of a Jupyter notebook and using Docker because of some issues with certain libraries. I had help deploying it on Heroku and now I need help figuring out where the text of my heroku.yml file should go in deploying it with Render – anyone able to help me out here?
web: /env/bin/panel serve --address="0.0.0.0" --port=$PORT app.ipynb --allow-websocket-origin=autohazard-explorer.herokuapp.com
I think I’ve replied to a ticket you also opened, but I’ll add a similar response here for completeness:
CMD would be run on boot, so that would need to be set to something like:
CMD /env/bin/panel serve --address=“0.0.0.0” --port=$PORT app.ipynb --allow-websocket-origin=$RENDER_EXTERNAL_HOSTNAME
render.yaml aren’t required for deploying. If you want to use a Blueprint, as long as the
env is set to
docker, it will default to look for your Dockerfile in the root of your project (can be overridden with
dockerFilePath), and the
CMD will be used to boot (as above, but can be overridden with
CMD using the “Docker Command” will also override any
ENTRYPOINT. We’ve also seen customers with long/multiple-option boot commands (like yours) have issues setting that as a plain “Docker Command”, a workaround for this can be to put your start command in a shell script (e.g.
render-start.sh) and set that as your Docker Command.
This topic was automatically closed 30 days after the last reply. New replies are no longer allowed.