Everything is done with an OpenAPI schema.
This allows us to auto-generate the server routing, schema validation middleware, types and clients.
This happens automatically on update via the Makefile.
Please ensure updated generated code is commited with your pull request.
Consult the OpenAPI schema for full details of what it does.
Once everything is up and running, grab the IP address:
export INGRESS_ADDR=$(kubectl -n unikorn get ingress/unikorn-server -o 'jsonpath={.status.loadBalancer.ingress[0].ip}')And add it to your resolver:
echo "${INGRESS_ADDR} unikorn.unikorn-cloud.org" >> /etc/hosts