Looking for a best practice of managing configuration after the ember build
if it is possible at all. Although it is not necessarily Docker specific, following must give you a good idea of the requirement even though you are not a Docker veteran.
Motivation: We use Docker + Docker Hub to (automatically) build and deploy our backend and frontend applications separately. There are multiple deployments. Frontend application is based on Ember (using ember-cli). Instead of running ember build --environment deployment-{1..N}
in a separate Dockerfile and end up having multiple Docker images, I would prefer having one single Docker image and running the Docker container with required settings passed as environment variables to inject them into the Ember configuration later, such as:
docker run -e ENV_APP_APIURL=deployment1.example.com ... --name deployment1 dockerimage
I would then run a small script as ENTRYPOINT
in Dockerfile which consumes these environment variables and inject them into the built Ember application so that web server can serve the re-configured application. In short:
[ember app] >> build >> [compiled generic web app] >> reconfigure(?) >> [final web app]
Is it ember way of doing things or I should go and grab a coffee?
Aucun commentaire:
Enregistrer un commentaire