JupyterHub and BinderHub Team Meeting

  • Date: Thursday 18th February 2021

  • Time: 9 AM GMT

    • Your timezone: https://arewemeetingyet.com/London/2021-02-18/09:00/Hubs-Team-Meeting

  • GitHub issue: https://github.com/jupyterhub/team-compass/issues/370

  • Calendar for future meetings: https://jupyterhub-team-compass.readthedocs.io/en/latest/meetings.html

By participating in this meeting, you are agreeing to abide by and uphold the Code of Conduct. Please take a second to read through it! :pray:

Welcome to the Team Meeting

Hello!

If you are joining the team video meeting, sign in below so we know who was here. Roll call:

  • Min / Simula / @minrk

  • Tim / Binder / @betatim

  • Ivana / Simula / @IvanaH8

  • Erik / Sundell open source / @consideratio

  • Simon / University of Dundee / @manics

  • Mridul / GESIS / @MridulS

  • Tania / Quansight Labs / @trallard

  • Kien / National University of Singapore / @kiendang

  • Yuvi / UC Berkeley / @yuvipanda

  • Jeremy / QuantStack / @jtpio

  • Omar / Karlstad Univeristy / @0mar

Quick updates

60 second updates on things you have been up to, questions you have, or developments you think people should know about. Please add yourself, and if you do not have an update to share, you can pass.

  • Sarah: The “Boost your Research Reproducibility with Binder” workshop organised by the Turing Way will be run again on March 3rd as part of the SSI’s Research Software Camp event on research accessibility! :tada: :tada: :tada: https://twitter.com/SoftwareSaved/status/1356612099326291969?s=19

  • Isabela: Community calls will be more sustainable if more people have experience hosting! This is the open call to get involved with Jupyter community calls even if it’s just for one call (you don’t even have to host alone :sunflower:). Reach out to me @isabela-pf on gitter.

Reports and celebrations

This is a place to make announcements (without a need for discussion). This is also a great place to give shout-outs to contributors! We’ll read through these at the beginning of the meeting.

  • Min: OVH is back in rotation with a private registry! Seems to be having some small issues, so not up to full capacity yet, but mostly smooth.

  • Erik: Simon did a amazing issue triaging sprint in jupyterhub/jupyterhub reducing stale issues with more than 50% out of the 300+!

Agenda items

Let’s collect all potential agenda items here before the start of the meeting. We will then attempt to create a coherent agenda that fits in the 60m meeting slot. If there are similar items try and group them together.

  • Erik (5-10 min): Archiving old repos?

    • Do we have an agreement to archive or move some repos to jupyter-attic as discussed here?

    • Agenda goal: decide on some action points

    • general strategy that is easy to undo: archive with a note in the README

    • jupyterhub-deploy-teaching: archive with note

    • jupyterhub-the-hardway: move documentation from jupyterhub repository into this repo and then keep this repo.

      • Simon will open a PR

  • Erik (5-10 min): Reducing stale issues and PRs

    • Could we establish some triaging routines and/or make a push to reduce the number of stale issues and PRs?

    • One concrete reason motivating me to raise this point now is that I wanted to avoid formatting clutter in PRs by making a repo adopt black and prettier autoformatting. But, with more than 30 stale PRs, I felt that it was unreasonable due to the merge conflicts it would cause.

    • Agenda goals

      • Increase awareness -> attract triaging contributions

      • Bonus: identify practices that can help us

    • Ideas:

      • bring a PR or two to the monthly meeting

      • use the “request review” feature to ask for help in making a decision

      • Try out assignment/review requests?

      • Team compass issue to discuss details

  • Simon (5-10min): Upgrading and/or defaulting to JupyterLab 3

  • Jeremy: (5min): JupyterLab 3.0

    • Update to JupyterLab 3.0: https://github.com/jupyterhub/repo2docker/pull/996

    • After this PR, consider switching to jupyter nbclassic as the default CMD?

      • This would run on Jupyter Server instead of the Classic Notebook server

      • Example gist: https://gist.github.com/jtpio/fca78a722eee2ca968acf38c629c277c

      • Yuvi: How would this affect existing notebook extensions that don’t know anything about ServerApp? Especially that import notebook package?

      • Yuvi: How would this affect existing users who have JSON config that refers to c.NotebookApp? (@Min says there is a backwards compat layer!)

      • Yuvi: How would this affect users who want to use JupyterLab 2? (Same as other notebook server extensions)

  • Min (5min): Banning networks in BinderHub. See https://github.com/jupyterhub/binderhub/pull/1262 for implementation, https://github.com/jupyterhub/mybinder.org-deploy/issues/1828 for initial discussion

  • Min (5min): RBAC in JupyterHub 2.0. Progress coming along nicely, but one question has come up: is it okay to drop ~all existing tokens and force server restarts, re-login, token re-issue for an upgrade of this scale? Robustly migrating existing tokens to the new schema may prove challenging, especially with the lack of ALTER TABLE support in sqlite.