[Cancelled][Decommission Hub] binder-staging

[sgibson91]

Summary

We had a crypto-mining incident on this hub ages ago. I attempted to lock it up with CILogon in this PR but was unsuccessful. And I've just spent the last bit of time trying with GitHub OAuth instead, but I just can't get it to work. My suspicion is that the docs here are outdated because I don't think the jupyterhub.auth key exists anymore...

Anyway, in the spirit of cleaning up, I propose we just get rid of this hub now.

Task List

Phase II - Hub Removal

(These steps are described in more detail in the docs at https://infrastructure.2i2c.org/en/latest/howto/hubs/delete-hub.html.)

• Manage existing data (migrate data from the hub or delete it)
• Remove the appropriate config/clusters/<cluster_name>/<hub_name>.values.yaml files. A complete list of relevant files can be found under the appropriate entry in the associated cluster.yaml file.
• Remove the associated hub entry from the config/clusters/<cluster_name>/cluster.yaml file.
• Remove the hub deployment
  • helm --namespace HUB_NAME delete HUB_NAME
  • kubectl delete namespace HUB_NAME
• Delete the hub's authentication application on auth0, GitHub or CILogon
• Also cleanup the deploy-hubs workflow file as I think there's some commented out code related to special-casing binder-staging in there too

[yuvipanda]

I'd love for us to instead fixup the instructions - even if we don't run the pangeo binder from here, I think we will run other ones.

[GeorgianaElena]

I'd love for us to instead fixup the instructions - even if we don't run the pangeo binder from here

Opened jupyterhub/binderhub#1646 to update binderhub auth instructions to match the working config in #2393.

even if we don't run the pangeo binder from here, I think we will run other ones.

Wondering if this is still true or if we should decommission it instead?

[yuvipanda]

I still think it is worth it. And thank you for fixing that, @GeorgianaElena!

[damianavila]

Given the authentication piece is fixed by @GeorgianaElena, I think there is value in keeping it alive since we are most likely deploying binder instances in the future. So, I am gonna close this one for now.