Closed choldgraf closed 2 years ago
I believe the sf
package install issue was because of conda
installing the latest version of libgdal
and gdal
as a dependency of ones of the pkgs listed in environment.yaml
.
Anyway, installing the R packages before the Python ones, solves this situation and it improves the caching of the R pkgs install, which takes forever.
@choldgraf, if Build Image
is green, do you want to merge this, so that the image gets pushed and deployed to staging, then ask the instructor if everything looks good?
@GeorgianaElena good call! (and thanks for debugging that install issue, you are a wizard!)
I just went ahead and merged https://github.com/utoronto-2i2c/jupyterhub-deploy/pull/119 and once that is happy (aka I don't have to revert), I will merge this one into staging and we can test it out 👍
edit: argh it looks like the build is not happy :-/ though the errors look vaguely like they aren't related to this PR? I will try re-running
wohooo builds are happy, so I am merging!
This is a re-opening of #101 , because the image build failed and so we reverted it from before.
We should:
mostly addresses https://github.com/2i2c-org/pilot-hubs/issues/697