Jupyter notebook for data science with PyROOT, s2i ready for deployment in jupyterhub on OpenShift.
centOS7, python 3.6, tensorflow 2.0 and ROOT 6.22.
Based on https://github.com/jupyter-on-openshift/jupyter-notebooks/tree/2.5.1 which is based on https://hub.docker.com/r/centos/python-36-centos7 .
ROOT build following https://github.com/root-project/root-docker .
Deployment in openshift via https://github.com/jupyter-on-openshift/jupyterhub-quickstart
(Update: use instead https://github.com/jupyter/docker-stacks/tree/master/examples/openshift ?)
Docker: https://hub.docker.com/r/jackfrost373/jupyter-root
Image can be run with
docker run -p 8080:8080 -e JUPYTER_ENABLE_LAB=true jackfrost373/jupyter-root[:tag]
The notebook is then served at http://127.0.0.1:8080
================================
To deploy on the OpenShift web interface:
-
add the notebook image: (top-right, 'add to project' > 'import YAML/JSON')
copy contents ofimage-streams/s2i-root-notebook.json
-
add the hub image:
https://raw.githubusercontent.com/jupyter-on-openshift/jupyterhub-quickstart/master/image-streams/jupyterhub.json -
add the deployment template:
templates/jupyterhub-deployer.json
-
add an instance from the template (bottom left, 'templates' > 'JupyterHub')
-
Add the custom config files:
Resources > config maps > myApp-cfg
Replace with content fromjupyter-configs
-
Redeploy