I was fooling around with FN and so far it looks pretty need. Just one thing I couldn't really find in the documentation concerns the deployment to a production cluster.
I saw that a normal function deployment would be done via fn deploy
. This would always deploy to localhost:8080 as far as I can see. But how about in a production cluster?
Let's say there is a running cluster on a different port. How would you deploy a function to a certain cluster? Does it spread manually? How is this managed?
Thanks a lot in advance.
You may set the variable
FN_API_URL
to change the cluster that a function is deployed to. An example would be:If you need to manage functions across multiple deployments or want a more robust way of managing this than an env var, you can also use context files in various capacities, which are documented here: https://github.com/fnproject/cli/blob/master/CONTEXT.md