revised field_length

[JhanSrbinovsky]

---

🚀 The latest prerelease access-esm1p6/pr186-3 at bfbe827 is here: #186 (comment) 🚀

[JhanSrbinovsky]

@CodeGat - all I've done is branch off main and change the um7 src code commit hash. The error talks about stuff that I have never touched or know how to fix. Do you have any idea what this is about?

[CodeGat]

Hi @JhanSrbinovsky - apologies for this one. I've just updated our infrastructure for spack deployments and it seems that it's taking configuration information from the new version rather than the current one you're on. I will fix this now, and comment when it is fixed. Should only be a couple minutes.

[CodeGat]

It should be fixed now. I'll keep my eyes on it, though.

[JhanSrbinovsky]

Thanks :)

[JhanSrbinovsky]

!redeploy

[github-actions]

🚀 Attempted to deploy access-esm1p6 Prerelease pr186-2 with commit 51a9976

🖥️ Gadi Deployment ✔️

Usage Instructions

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

• 2026.01.000 as a Release (when merged).
• pr186-2 as a Prerelease (during this PR).

This Prerelease is accessible on Gadi using:

module use /g/data/vk83/prerelease/modules
module load access-esm1p6/pr186-2

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-pr186-2 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 2026.02.000
• access-nri/spack-config version 2026.02.000

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

[JhanSrbinovsky]

Hi @CodeGat, despite the headline, the module seems to be there (although from the CLI takes a while to find it) but the prerelease module fails in payu setup looking for the CICE model. I cloned the same experiment again (Dec25-spinup) and tested the payu setup without the config change to my module and

Loading access-esm1p6/2025.12.000
  Loading requirement: cice5/access-esm1.6-2025.11.001-6umbsy5 mom5/2025.05.000-tzlv4hw um7/2025.12.000-n4khgi6

which appears to be what I need. It doesn't get this and fails looking for the exe in my local payu bin/

oading access-esm1p6/pr186-2
  Loading requirement: access-esm1p6/dependencies/pr186-2/oasis3-mct/5.2-rpje5ou access-esm1p6/dependencies/pr186-2/cice5/2026.01.000-3y5cnfo access-esm1p6/dependencies/pr186-2/access-fms/mom5-2025.08.000-ycljek4 access-esm1p6/dependencies/pr186-2/access-mocsy/2025.07.002-fxxczgy
    access-esm1p6/dependencies/pr186-2/access-generic-tracers/2025.09.000-wr5irgf access-esm1p6/dependencies/pr186-2/mom5/2025.05.000-4rel6l2 access-esm1p6/dependencies/pr186-2/cable/2025.11.000-5yequqi access-esm1p6/dependencies/pr186-2/gcom4/2025.08.000-3qhizjx
    access-esm1p6/dependencies/pr186-2/um7/1f069c1ce2689ff25db4354d89456bc1de4beae3-qpmoo2i
Traceback (most recent call last):
  File "/g/data/vk83/apps/base_conda/envs/payu-1.2.1/bin/payu", line 10, in <module>
    sys.exit(parse())
  File "/g/data/vk83/apps/base_conda/envs/payu-1.2.1/lib/python3.10/site-packages/payu/cli.py", line 49, in parse
    run_cmd(**args)
  File "/g/data/vk83/apps/base_conda/envs/payu-1.2.1/lib/python3.10/site-packages/payu/subcommands/setup_cmd.py", line 28, in runcmd
    expt.setup(force_archive=force_archive)
  File "/g/data/vk83/apps/base_conda/envs/payu-1.2.1/lib/python3.10/site-packages/payu/experiment.py", line 63, in wrapper
    result = func(self, *args, **kwargs)
  File "/g/data/vk83/apps/base_conda/envs/payu-1.2.1/lib/python3.10/site-packages/payu/experiment.py", line 531, in setup
    model.setup()
  File "/g/data/vk83/apps/base_conda/envs/payu-1.2.1/lib/python3.10/site-packages/payu/models/cice5.py", line 71, in setup
    super(Cice5, self).setup()
  File "/g/data/vk83/apps/base_conda/envs/payu-1.2.1/lib/python3.10/site-packages/payu/models/cice.py", line 169, in setup
    super(Cice, self).setup()
  File "/g/data/vk83/apps/base_conda/envs/payu-1.2.1/lib/python3.10/site-packages/payu/models/model.py", line 321, in setup
    raise FileNotFoundError(
FileNotFoundError: Executable for ice model not found on path: /scratch/p66/jxs599/access-esm/bin/cice_access-esm1.6_360x300_12x1_12p.exe

is this something on the prerelease build end?

[blimlim]

Hi @JhanSrbinovsky, I suspect you might have got caught in the middle of some changes that haven't made their way to the configurations yet.

Following some changes with how CICE is being built, the CICE the deployments are now giving the CICE executable a different name, changing from cice_access-esm1.6_360x300_12x1_12p.exe to cice_access_360x300_30x300.exe.

We haven't yet got this change into the configs (working on it), and so I think the configurations aren't searching for the correct executable name.

In your config.yaml, could you try replacing

      exe: cice_access-esm1.6_360x300_12x1_12p.exe

with

      exe: cice_access_360x300_30x300.exe

and see if running works from there?

We'll try to get this change into the base configs asap

[JhanSrbinovsky]

Thanks Spencer, It probably will work but it is hanging on even loading the prerelease module in payu. From the CLI, same: gadi:Dec25-revised-field-length> module use /g/data/vk83/prerelease/modules gadi:Dec25-revised-field-length> module load access-esm1p6/pr186-2 It just hangs. It took ages this morning but eventually . Ohh - just checking again - finally after ~3 minutes it found it. I’ll leave the payu setup hang for a bit - Yeah finally “setup” went through - payu run in queue 😊 I think there was a spike in traffic on gadi as well.

[github-actions]

🚀 Attempted to deploy access-esm1p6 Prerelease pr186-3 with commit bfbe827

🖥️ Gadi Deployment ✔️

Usage Instructions

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

• 2026.01.000 as a Release (when merged).
• pr186-3 as a Prerelease (during this PR).

This Prerelease is accessible on Gadi using:

module use /g/data/vk83/prerelease/modules
module load access-esm1p6/pr186-3

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-pr186-3 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 2026.02.000
• access-nri/spack-config version 2026.02.000

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

[JhanSrbinovsky]

@blimlim - it was found by setup, but it crashed immediately. The error logs don't tell me much at all.

As a test - I rebranched from UM7:dev-access-esm1.6 to branch="test-prerelease". I made a trivial WS change to get a commit hash beyond the copy/branch. I rebuilt in ACCESS-ESM1.6, to produce above pr186-3. Using this with same cice exe you recommended. Again crashed immediately. At a glance - equally as cryptic, possibly the same error.

[blimlim]

Apologies @JhanSrbinovsky, I suspect this is also from the delay between the new exe release, and the corresponding configuration updates. The build changes updated to a new version of oasis-mct, and the configuration then needed changes to the namcouple file for compatibility.

The namcouple changes just got in yesterday evening – would you be able to try cloning from the new HEAD of the relevant configuration, and trying your prerelease executables with that?

[JhanSrbinovsky]

Thanks @blimlim - I will, but it isn't urgent now. After this morning meeting a solution seems to be found. Do you know where the bug was specifically that Martin was referring to?

[blimlim]

There's some discussion of it over here: https://github.com/ACCESS-NRI/UM7/issues/202, it's a pretty confusing issue