feature/coupled-crow installation broken on stampede2 #308
Comments
|
It is an older version. The ufs-weather-model is using a new one installed by @Hang-Lei-NOAA : https://github.com/ufs-community/ufs-weather-model/blob/37f0da7b82e32a15bf072eaa158399a6b0f2722c/modulefiles/stampede.intel/fv3#L17 |
|
@WalterKolczynski-NOAA and @JianKuang-NOAA we should be using the hpc-stack that was installed by @Hang-Lei-NOAA for the ufs-weather-model. I think the permissions there are correct as it was used by Ben to build the model without problems |
|
I definitely have a version of hpc-stack that I can see here: /work/08048/hanglei/NCEP/install/hpc-stack/modulefiles/stack. And it looks like machine-setup.sh is referring to that for stampede2. So I'm not sure now where that version of hpc-stack is coming in from. |
|
The stampede2 related changes were merged into the ufs-weather-model develop branch on 03/24. So, if the global workflow is pulling in a version that is older, that could be the reason. |
|
Stealing the stampede.intel/fv3 file from my ufs-weather-model install seems to be working, but I will pause there until the question of whether I have the right model version is resolved. |
|
When @WalterKolczynski-NOAA pulled in the last PR #261 from @JianKuang-NOAA to the feature/coupled-crow branch, it was noted that the ufs-weather-model needed to be updated to the hash @MinsukJi-NOAA mentioned, and the hpc-stack version others have mentioned in this thread. @WalterKolczynski-NOAA felt like it would be easier to make those changes via a separate PR, so #261 was merged and the updates to the modules and weather-model hash are ongoing. I think he's working on that - I remember seeing a comment about post not building yesterday. |
|
Ok, I will hold until the rest of the updates go in. |
|
In case it is useful - build_ufs_coupled completed when I used the newer fv3 file, but build_reg2grb2 failed looking for libjasper.a and ncep_post failed with a variety of errors. |
WalterKolczynski-NOAA
added
the
duplicate
|
As Jessica mentioned, the branch is not ready for use on Stampede yet. I'm working on cleaning it up and testing now. I will announce on the original porting issue when it is ready. Please sit tight until then. Closing as duplicate of #280. |
* update output filenames for fv3lam in run_upp script * Fix fcst hour bug in naming output files Co-authored-by: Tracy <tracy.hertneky@noaa.gov>
Features:
- Implemented sharing of atmospheric tracer metadata (name, units) to coupled model component via a standard NUOPC connector if coupling tracer field is accessed by shared memory reference
- Introduced additional metadata to tracer records in field_table to identify and validate prognostic and diagnostic chemistry tracers. A new record line starting with the `tracer_usage` keyword is now required to recognize chemistry tracer as follows:
"tracer_usage", "chemistry" (default, prognostic)
"tracer_usage", "chemistry", "type=diagnostic" (diagnostic)
- Added sanity check for input chemistry tracers, ensuring that both prognostic and diagnostic tracers in field_table are contiguous, and diagnostic tracers follow prognostic ones.
- Improved algorithm to set aerosol scavenging factors
- Exported instantaneous fields for:
- 3D non-convective liquid and ice precipitation fluxes
- 3D cloud fraction
- normalized soil wetness
- ice fraction (as used in the atmospheric model)
- lake fraction
- ocean fraction
- surface snow area fraction
- Enabled import of diagnostic tracers
Fixes:
- Prevent a divide by zero error when constant arrays are rescaled for lossy compression in `write_netcdf_parallel()`
- Correctly sample and compute elapsed time for field bundle regridding operations.
- Include hotfix (see PR NOAA-EMC#308) to allow latest `setup_exportdata()` to work with coupled chemistry.
Co-authored-by: Dom Heinzeller <climbfuji@ymail.com>
The installation of the workflow on stampede2 breaks because it is pointing to an installation of hpc-stack that does not have general read permissions (chronic problem on stampede2). The problem is in this line:
module use -a /work/06146/tg854455/stampede2/ufs-stack-20200909/intel-18.0.2/impi-18.0.2/modules
login1.stampede2(1047)$ ls -l /work/06146/tg854455/stampede2/ufs-stack-20200909/intel-18.0.2/impi-18.0.2/modules
ls: cannot access /work/06146/tg854455/stampede2/ufs-stack-20200909/intel-18.0.2/impi-18.0.2/modules: Permission denied
It also looks like that might be an older version of hpc-stack?
The text was updated successfully, but these errors were encountered: