Infrastructure for providing JupyterHub environments via Eclipse Che and Kubernetes.
In using Eclipse Che, we follow a similar approach to NASA MAAP, but attempting to build on the newest version of Eclipse Che.
The infrastructure here supports an MVP with Eclipse Che, JupyterHub, and AWS Cognito authentication, but is left here for information and documentation. We are happy to share our lessons learned and other gotchas with this setup upon request.
This step is optional and useful only if you are developing locally and need to test your deployment. If you are just trying to deploy a new cluster without having made any changes locally, you can skip this step.
docker build . -t veda
NOTE: if you are on an M1 Mac, you have to include the --platform=linux/amd64 tag to your docker commands to avoid compatibility issues.
After building the image, it's time to run it to deploy the infrastructure.
Make sure that you have created a .env file that contains all the required environment variable that are needed for the deployment to properly work.
You can use the .env.template file to create and fill your .env file.
docker run --env-file ./.env -it veda:latest bootstrap-and-deploy
NOTE: This takes approximately 35-40 minutes to run.
If you are deploying a new cluster with a unique/new QUALIFIER, you must create a new cognito app client as well to integrate the correct callback URL for auth.
To do so, head to Cognito's create app client page and match (except for the client name) the screenshots below
You can then copy the app's client name and secret and add them to the docker run's environment variable.
NOTE: The callback URL should match this syntax https://{QUALIFIER}-analytics.delta-backend.com/oauth/callback
To add a user, follow the gif below 
If you do not need to have a particular cluster running, you can destroy all its resources by running:
docker run --env-file ./.env-it veda:latest destroy
NOTE: This takes approximately 25-30 minutes to run