Adding a Github Action that runs on CIRRUS runner scale sets and NCAR HPC containers

[mjs2369]

Description:

Adds a Github Action that runs Lorenz_96 on CIRRUS runner scale sets and NCAR HPC containers. Runs with all mpi options (mpi, mpif08, nompi) with the compilers gfortran, and nvhpc, and intel-oneapi.

Note that this action will be run on every pull request but will NOT be required to pass to merge into main. @hkershaw-brown I do not have the permission to change these settings in the DART repo, if you could please ensure this is correctly set.

Fixes issue

No issue in for this, woops

Types of changes

• Bug fix (non-breaking change which fixes an issue)
• New feature (non-breaking change which adds functionality)
• Breaking change (fix or feature that would cause existing functionality to not work as expected)
• Documentation update

Documentation changes needed?

• My change requires a change to the documentation.
  • I have updated the documentation accordingly.

Updated developer's guide documentation.

Tests

Please describe any tests you ran to verify your changes.

Tested action on push, all jobs succeeded.

Checklist for merging

• Updated changelog entry
• Documentation updated
• Update conf.py

Checklist for release

• Merge into main
• Create release from the main branch with appropriate tag
• Delete feature-branch

Testing Datasets

• Dataset needed for testing available upon request
• Dataset download instructions included
• No dataset needed

[hkershaw-brown]

@mjs2369 Hey Marlee when you update this with main can you rebase rather than merge. Because we will probably want to squash commits testing the actions. Cheers,
Helen

[hkershaw-brown]

This is really cool, I think this can be a matrix of compiler, mpi
This is a quick sketch: cirrus_actions...cirrus_actions-matrix
Note not tested, but I think if this action can be matrixized (?) then adding cce, ifx, lfortran, whatever compiler will be easier in the future.

[hkershaw-brown]

Any update on why all the CIRRUS runners are failing?

[mjs2369]

Any update on why all the CIRRUS runners are failing?

@hkershaw-brown yes, I just heard back from Ben Kirk. The cause is currently unknown, but is definitely on the CIRRUS-side and has been reported to the CIRRUS admins. There is currently no timeline for the fix, but I will run the final tests with the matrix as soon as I get the go-ahead.

[hkershaw-brown]

Note on feedback to GitHub on ❌ vs ⚠️
https://github.com/orgs/community/discussions/35424

[hkershaw-brown]

Leaving this on 'request changes'.
Needs CIRRUS to be online 🤦‍♀️ to test
rebase (squash) commits into a single commit before merging.

[hkershaw-brown]

Is CIRRUS still down?

[mjs2369]

@hkershaw-brown yep, chatted with Ben at the conference too and there's currently no timeline. Should we put this on draft in the mean time?

[hkershaw-brown]

@hkershaw-brown yep, chatted with Ben at the conference too and there's currently no timeline. Should we put this on draft in the mean time?

well, that is less than ideal.

[hkershaw-brown]

I'm rebasing this to be up-to-date with main, watch out when you pull.

[hkershaw-brown]

@mjs2369 can you rebase this so it is just a single commit (rather than two commits, code & merge commit with main).

[mjs2369]

@hkershaw-brown done

[hkershaw-brown]

This is really nice Marlee, great work on all the container and cirrus interactions.

I'd like to bundle this with the next release, next week (when I am back in the office).

@hkershaw-brown merge this before #1069