Rahti Cloud containers advanced course exercises.

This exercise aims to introduce you to Rahti Cloud containers using notebooks. We will go through the deploying Jupyter Notebook Image from the Docker hub and then move to deploy thgis image variant in the Rahti cloud service.

Exercise 1: Deploying a Jupyter Notebook Image.

1. Make sure that you have Docker or Podman isntalled on your system.

2. Pull, start and expose the Jupyter Notebook server.

   The jupyter/minimal-notebook image can be found on docker hub, expose the image on port: 8888 and grab the link from conatiner logs.

3. Test the container

   Create a folder name test-hello and a python script print("Hello") in the file name hello.ipynb execute the script.

4. Kill the container

   After killing the container, create again the container you will notice that the python script and the folder create are no longer availbale. This is becauase conatiners by nature are ephemeral which means that they do not store/save in them. To persist data or configuration changes in a container, containerized applications often use external storage volumes or configuration management tools.

5. Save your work

   To save the created python script we have to mount volume local host to container. Using the following command sudo docker run -it --rm -p 8888:8888 --user root -e NB_UID=$(id -u) -e NB_GID=$(id -g) -e CHOWN_HOME=yes -e CHOWN_HOME_OPTS='-R' -v "${PWD}":/home/jovyan/work jupyter/minimal-notebook The -v flag is used in Docker commands to specify a volume, which allows for data to be shared between the host system and a container. The ${PWD} part of the command is a shell variable that represents the current working directory (PWD stands for "present working directory"). And /home/jovyan/work is the path inside the container where the volume will be mounted.

6. Kill the container and check if the files exists

   Rerun the previous command and check if the files has been stored.

Exercise 2: Deploying JupyterHub as a Service.

* From the application catalog select JupyterHub(Native Authentication).
* Brief information about the template will be displayed. Click Next.
* Configure your deployment. E.g., select a project where you want to deploy the JupyterHub service and set an admin user(admin), password(adminpass) and database password(adminpass) for it.
* Finally, you can Create your JupyterHub service.
* Go to the Application Console from the top left corner menu and select your project where you have deployed your JupyterHub service.
* Now you can follow the status of your JupyterHub deployment.
* Once everything is running, You can go to Applications > Services from the menu on the left-hand side and select the jupyterhub service, follow the link, and log in to your JupyterHub service.
* Login as admin and create a user `user1`and authorize it `https://.........rahtiapp.fi/hub/authorize`
* Create in a signup page a user `user1` and password `user1` 
* Create a python script `print("user1")` and check if you can access from admin control pannel.