Dummy Gude to JupyterHub Configuration

[Eric Lofgren]

I'm interested in setting up a very basic JupyterHub to allow a small whitelist of users to run jobs on a particular server remotely. There's been some decent YouTube tutorials on setting up JupyterHub itself, but I seem to always run aground in a way that suggests I'm failing to do the initial setup of the kernels correctly. Is there any chance I could impose on someone for a step-by-step configuration guide for that step, assuming the following:

- RHEL, CentOS, ScientificLinux, etc.

- Access to root, an admin, and several user accounts

- Spotty at best Linux knowledge (squarely in the "enough to be dangerous" category)

For example, what user should everything be done under? If I'm installing Anaconda, where should the various directory options I have point to? etc.

Thanks,

Eric

[Nicholas Bollweg]

I can't say that I have followed this guide, but it does appear to fit the bill (rpm-based, anaconda) linked from the wiki.

It is a bit gung-ho, doing a lot of stuff as root right on the box, and most of the versions are out of date, now, but it does look like the shortest path to a working install given the author's constraints.