Update OHW Python and R images, and some metadata

[emiliom]

• Python and R images
• Labels for the images (to make them a bit more user friendly, in Spanish)
• Logo url

[github-actions]

Merging this PR will trigger the following deployment actions.

Support and Staging deployments

Cloud Provider	Cluster Name	Upgrade Support?	Reason for Support Redeploy	Upgrade Staging?	Reason for Staging Redeploy
gcp	2i2c	No		Yes	Following prod hubs require redeploy: ohw

Production deployments

Cloud Provider	Cluster Name	Hub Name	Reason for Redeploy
gcp	2i2c	ohw	Following helm chart values files were modified: ohw.values.yaml

[yuvipanda]

Thanks, @emiliom. Test it out once the deploy is done :)

[github-actions]

🎉🎉🎉🎉

Monitor the deployment of the hubs here 👉 https://github.com/2i2c-org/infrastructure/actions/runs/4267187164

[emiliom]

Wow, that was really fast @yuvipanda !! Thank you! I'll test it out in an hour or two.

[yuvipanda]

@emiliom haha, sometimes I anxiously refresh github notifications :D

[emiliom]

I've tested the hub with the two images. The R image is failing. RStudio doesn't come up and shows this error:

I haven't seen this before. @abkfenris , have you?

The hub deployment monitor showed this warning https://github.com/2i2c-org/infrastructure/actions/runs/4267187164/jobs/7428550251#step:5:35:

Health check failed!
Warning: Attempt 1 failed. Reason: Child_process exited with error code 1

Though it's just a warning.

The Python image works fine and the updates are there.

[yuvipanda]

@emiliom I looked at the logs, and found this:

[W 2023-02-25 01:52:24.555 SingleUserLabApp manager:360] jupyter_server_proxy | extension failed loading with message: DistributionNotFound(Requirement.parse('webcolors>=1.11; extra == "format-nongpl"'), {'jsonschema'})
    Traceback (most recent call last):
      File "/opt/conda/lib/python3.9/site-packages/jupyter_server/extension/manager.py", line 356, in load_extension
        extension.load_all_points(self.serverapp)
      File "/opt/conda/lib/python3.9/site-packages/jupyter_server/extension/manager.py", line 228, in load_all_points
        return [self.load_point(point_name, serverapp) for point_name in self.extension_points]
      File "/opt/conda/lib/python3.9/site-packages/jupyter_server/extension/manager.py", line 228, in <listcomp>
        return [self.load_point(point_name, serverapp) for point_name in self.extension_points]
      File "/opt/conda/lib/python3.9/site-packages/jupyter_server/extension/manager.py", line 219, in load_point
        return point.load(serverapp)
      File "/opt/conda/lib/python3.9/site-packages/jupyter_server/extension/manager.py", line 147, in load
        return loader(serverapp)
      File "/opt/conda/lib/python3.9/site-packages/jupyter_server_proxy/__init__.py", line 30, in _load_jupyter_server_extension
        server_processes += get_entrypoint_server_processes(serverproxy_config)
      File "/opt/conda/lib/python3.9/site-packages/jupyter_server_proxy/config.py", line 83, in get_entrypoint_server_processes
        server_process_config = entry_point.load()()
      File "/opt/conda/lib/python3.9/site-packages/pkg_resources/__init__.py", line 2476, in load
        self.require(*args, **kwargs)
      File "/opt/conda/lib/python3.9/site-packages/pkg_resources/__init__.py", line 2499, in require
        items = working_set.resolve(reqs, env, installer, extras=self.extras)
      File "/opt/conda/lib/python3.9/site-packages/pkg_resources/__init__.py", line 815, in resolve
        dist = self._resolve_dist(
      File "/opt/conda/lib/python3.9/site-packages/pkg_resources/__init__.py", line 856, in _resolve_dist
        raise DistributionNotFound(req, requirers)
    pkg_resources.DistributionNotFound: The 'webcolors>=1.11; extra == "format-nongpl"' distribution was not found and is required by jsonschema

This has prevented jupyter-server-proxy from being installed, and hence RStudio doesn't work. Often this is because of some version conflict in how things are installed.

We can revert the R image temporarily while you fix that if you want, or leave it as is.

[emiliom]

Thanks @yuvipanda . Hmm. Doing some research on that error message, the best I can come up with is that we need to add jsonschema-with-format-nongpl to the environment. It looks like an alternative build for jsonschema that's equivalent to pip install jsonschema[format-nongpl].

Does it look like I may be on the right track?

I can build a new image where I add jsonschema-with-format-nongpl, submit a new PR, and cross my fingers. What do you think?

Footnotes:

• There's a recent reference to this error, from 4 days ago: https://stackoverflow.com/questions/75511394/jupyter-contrib-nbextension-install-user-pkg-resources-distributionnotfound-we
• The previous R image, built on 8/17/2022, used jsonschema 4.9.1. jsonschema 4.10.0 came out also on 8/17/2022, and it looks like it caused some hiccups, though I have no idea if it's related to the problem in this PR.
• The Python image has jsonschema 4.17.3, just like the R image that's failing.

[emiliom]

One more thing. In our R Dockerfile, we have this, from last August:

# # Newer one has bug that doesn't work with jupyter-rsession-proxy
ENV RSTUDIO_URL https://download2.rstudio.org/server/bionic/amd64/rstudio-server-2022.07.1-554-amd64.deb

I didn't update it. Don't know if I should.

[yuvipanda]

@emiliom I think trying jsonschema-with-format-nongpl is the way to go.

[yuvipanda]

I don't think this is related to the RStudio install. The error is coming from the jupyter-server-proxy package, a dependency of the jupyter-rsession-proxy package.

[emiliom]

PR #2253