Build restore dirt

[JhanSrbinovsky]

build CABLE

---

🚀 The latest prerelease access-esm1p6/pr161-3 at 6f1b7a5 is here: #161 (comment) 🚀

[github-actions]

🚀 Attempted to deploy access-esm1p6 Prerelease pr161-1 with commit 5114278

🖥️ Gadi Deployment ❌

[JhanSrbinovsky]

@Whyborn - I had an idea to get the new PFT map in. I had to rebuild. The FCM build and I guess expected runtime libraries has diverged too far now and a few months back I lost the ability to PAYU run an FCM built executable. An additional problem is that I have to be past a recent point in the restarts for any of this to make sense, so I can't just go back to an old job. I tried to build it in the new scheme using CABLE as a library, but no dice. Although the only code changes are the interface code still in UM7.

Not my problem here, but what does this # even point to?

cable:
      require:
        - '@2025.11.000 library=access-esm1.6'

[github-actions]

🚀 Attempted to deploy access-esm1p6 Prerelease pr161-2 with commit 2b4e946

🖥️ Gadi Deployment ❌

[JhanSrbinovsky]

@Whyborn I included the fix to cable require suggested in #156 but CI still fails - error seems to have changed though. Doesn't recognize tag

[Whyborn]

@JhanSrbinovsky that specifies a version of CABLE in the spack-packages. The work to move CABLE to numerical versions was in a PR on the spack package repo, which has now been merged.

[JhanSrbinovsky]

OK thanks @Whyborn - but then how do you differentiate between library versions to include?

[JhanSrbinovsky]

anyway, regardless of that - How can I build with a library? It hasn't been as simple as just branching off main here - is there still something missing?

[Whyborn]

The variant specifies which version of the library to use. So you should be able to use:

require:
  - @2025.11.000 library=access-esm1.6

You should be able to branch off from here, it looks like the deployment is ok. If it fails, I'll go try get it to build using that commit of UM7.

[github-actions]

🚀 Attempted to deploy access-esm1p6 Prerelease pr161-3 with commit 6f1b7a5

🖥️ Gadi Deployment ✔️

Usage Instructions

access-esm1.6, defined in ./spack.yaml, will be deployed to Gadi as:

• 2025.11.000 as a Release (when merged).
• pr161-3 as a Prerelease (during this PR).

This Prerelease is accessible on Gadi using:

module use /g/data/vk83/prerelease/modules
module load access-esm1p6/pr161-3

When using the above modules, the binaries shall be on your $PATH.

For advanced users, this Prerelease is also accessible on Gadi via /g/data/vk83/prerelease/apps/spack/0.22/spack in the access-esm1p6-pr161-3 environment.
Due to inode-saving measures, one will have to manually untar the environment metadata before environment activation with tar -xf .spack-env .spack-env.tar. It will require one to have write privileges.

Configuration Information

This Prerelease is deployed using:

• access-nri/spack on branch 0.22
• access-nri/spack-packages version 2025.11.001
• access-nri/spack-config version 2025.08.000

If the above was not what was expected, commit changes to config/versions.json in this PR.

[JhanSrbinovsky]

yeah thats what I used. sinceI updated though I was so used to seeing failures that I didn't notice the active conflicts. It seems to be OK now. Unfortunately the urgency of this building has disappeared. The UM7 code here modifies fractions for existing bare-soil only, not added tiles (I can't think how I can deal with that reliably in the code). Anyway, lunchtime here :)

[ccarouge]

At the standup, the summary was the new veg. distribution with the new bare soil worked. I'm assuming this PR was an attempt to look into this. Can it be closed now @JhanSrbinovsky ?