UWM changes for CCPP framework update // Miscellaneous updates (#2974)

[dustinswales]

Commit Queue Requirements:

• Fill out all sections of this template.
• All sub component pull requests have been reviewed by their code managers.
• Run the full Intel+GNU RT suite (compared to current baselines) on either Hera/Derecho/Hercules
• Commit 'test_changes.list' from previous step (N/A)

Description:

#2881

UFSATM changes to accommodate CCPP Framework updates

#2974

This PR:

• Updates DISKNM locations to /work2 on Orion/Hercules:
  • Hercules: /work2/noaa/epic/hercules/UFS-WM_RT
  • Orion: /work2/noaa/epic/UFS-WM_RT
• Adds update to PR project board when fire behavior (FB) label is added (that was missed in the original PR)
• Updates input-data directory in docs to most recent location as of PR Update SFS test to C192mx025; add debug and restart tests; update ultralow warmstart input to specify V2 file type // Address some compiler warnings in UFSATM and UPP: round 2 (#2958) #2835

Commit Message:

* UFSWM - Changes to accommodate CCPP Framework updates.
* UFSWM - Update Orion/Hercules DISKNM locations
  * CMEPS - Changes to accommodate CCPP Framework updates.
  * UFSATM - Changes to accommodate CCPP Framework updates.
    * ccpp-physics - Changes to accommodate CCPP Framework updates.

Priority:

• Normal.

Git Tracking

UFSWM:

• UFSATM:NOAA-EMC:ufsatm#1005

Sub component Pull Requests:

• CMEPS: Changes to accommodate CCPP framework upgrades NOAA-EMC/CMEPS#150
• FV3: UFSATM changes for ccpp framework update NOAA-EMC/ufsatm#1006
  • ccpp-physics: Physics changes for ccpp framework update ccpp-physics#311

UFSWM Blocking Dependencies:

• None
• Comments in PR UFSATM: Replace GFS_interstitial phys_reset and rad_reset routines with combined reset routine // removing variables from CICE NetCDF file (#2951) #2880 suggest that that PR should go first.

Documentation:

• No documentation update is required for this PR (please explain).

---

Changes

Tested on URSA.
Changes to RRTMGP baselines.

Input data Changes:

• None.

Library Changes/Upgrades:

• No Updates

---

Testing Log:

• RDHPCS
  • Hera
  • Orion
  • Hercules
  • GaeaC6
  • Derecho
  • Ursa
• WCOSS2
  • Dogwood/Cactus
  • Acorn
• CI
• opnReqTest (complete task if unnecessary)

[FernandoAndrade-NOAA]

Hera is seeing compile errors for atm_mpas_dyn32_gnu, based off rt.conf, Hercules may be seeing these as well.

Warning: runtime_format attribute has no effect for immutable stream "lbc_in".
---- GENERATING IMMUTABLE STREAMS ----
/scratch3/NAGAPE/epic/Fernando.Andrade-maldonado/reg-test/wm/2881/ufs-weather-model/UFSATM/ccpp/driver/MPAS_init.F90:98:67:

   98 |        call Interstitial(nt)%create(maxval(Init_parm%blksz), Model)
      |                                                                   1
Error: Type mismatch in argument 'ixe' at (1); passed TYPE(gfs_control_type) to INTEGER(4)
/scratch3/NAGAPE/epic/Fernando.Andrade-maldonado/reg-test/wm/2881/ufs-weather-model/UFSATM/ccpp/driver/MPAS_init.F90:103:72:

  103 |        call Interstitial(nthrds+1)%create(Init_parm%blksz(nblks), Model)
      |                                                                        1
Error: Type mismatch in argument 'ixe' at (1); passed TYPE(gfs_control_type) to INTEGER(4)
make[2]: *** [UFSATM/CMakeFiles/ufsatm_mpas.dir/build.make:296: UFSATM/CMakeFiles/ufsatm_mpas.dir/ccpp/driver/MPAS_init.F90.o] Error 1
make[2]: *** Waiting for unfinished jobs....
make[1]: *** [CMakeFiles/Makefile2:453: UFSATM/CMakeFiles/ufsatm_mpas.dir/all] Error 2
make: *** [Makefile:136: all] Error 2

[grantfirl]

Hera is seeing compile errors for atm_mpas_dyn32_gnu, based off rt.conf, Hercules may be seeing these as well.

Warning: runtime_format attribute has no effect for immutable stream "lbc_in".
---- GENERATING IMMUTABLE STREAMS ----
/scratch3/NAGAPE/epic/Fernando.Andrade-maldonado/reg-test/wm/2881/ufs-weather-model/UFSATM/ccpp/driver/MPAS_init.F90:98:67:

   98 |        call Interstitial(nt)%create(maxval(Init_parm%blksz), Model)
      |                                                                   1
Error: Type mismatch in argument 'ixe' at (1); passed TYPE(gfs_control_type) to INTEGER(4)
/scratch3/NAGAPE/epic/Fernando.Andrade-maldonado/reg-test/wm/2881/ufs-weather-model/UFSATM/ccpp/driver/MPAS_init.F90:103:72:

  103 |        call Interstitial(nthrds+1)%create(Init_parm%blksz(nblks), Model)
      |                                                                        1
Error: Type mismatch in argument 'ixe' at (1); passed TYPE(gfs_control_type) to INTEGER(4)
make[2]: *** [UFSATM/CMakeFiles/ufsatm_mpas.dir/build.make:296: UFSATM/CMakeFiles/ufsatm_mpas.dir/ccpp/driver/MPAS_init.F90.o] Error 1
make[2]: *** Waiting for unfinished jobs....
make[1]: *** [CMakeFiles/Makefile2:453: UFSATM/CMakeFiles/ufsatm_mpas.dir/all] Error 2
make: *** [Makefile:136: all] Error 2

@FernandoAndrade-NOAA Thanks for the details. I think that this test was skipped on Ursa (where I ran my tests), so I'll debug this now. Should be a simple fix. Please stand by.

[grantfirl]

@FernandoAndrade-NOAA A quick update. I think that I have it debugged and fixed in another branch. I'm running a test on Ursa to make sure. If successful, I'll merge the fix in.

[grantfirl]

@FernandoAndrade-NOAA I merged in the changes for the MPAS test. It can now at least compile and initiate a run. See NOAA-EMC/ufsatm@964091c. Since the changes are confined to the MPAS code, I don't think that other RT runs are invalid.

I did run into an out-of-memory error during runtime for the test on Ursa. I don't know if Hera and Hercules will run into that problem though. If they do, I'm not sure the best course of action. Maybe try to increase resources for that test? If that doesn't work, then deeper investigation into the MPAS test with these code changes may be warranted, but this is not really my work and @dustinswales would probably be the better one to debug when he returns from shutdown, hopefully soon.

[gspetro-NOAA]

@FernandoAndrade-NOAA I merged in the changes for the MPAS test. It can now at least compile and initiate a run. See NOAA-EMC/ufsatm@964091c. Since the changes are confined to the MPAS code, I don't think that other RT runs are invalid.

I did run into an out-of-memory error during runtime for the test on Ursa. I don't know if Hera and Hercules will run into that problem though. If they do, I'm not sure the best course of action. Maybe try to increase resources for that test? If that doesn't work, then deeper investigation into the MPAS test with these code changes may be warranted, but this is not really my work and @dustinswales would probably be the better one to debug when he returns from shutdown, hopefully soon.

The MPAS test passed on Hercules with your change. That said, we should really get this test (and the other MPAS one) running on Ursa. I'll create an issue for that and tag Dustin, since ideally, we want all tests running on Ursa. In order to merge the MPAS PR, we had to remove control_rrfs_mpas_gnu from running on Ursa because there were issues with it, but it was passing on Hercules and Hera. However, it is the only test we run that does not run on Ursa.

[grantfirl]

@FernandoAndrade-NOAA I merged in the changes for the MPAS test. It can now at least compile and initiate a run. See NOAA-EMC/ufsatm@964091c. Since the changes are confined to the MPAS code, I don't think that other RT runs are invalid.
I did run into an out-of-memory error during runtime for the test on Ursa. I don't know if Hera and Hercules will run into that problem though. If they do, I'm not sure the best course of action. Maybe try to increase resources for that test? If that doesn't work, then deeper investigation into the MPAS test with these code changes may be warranted, but this is not really my work and @dustinswales would probably be the better one to debug when he returns from shutdown, hopefully soon.

The MPAS test passed on Hercules with your change. That said, we should really get this test (and the other MPAS one) running on Ursa. I'll create an issue for that and tag Dustin, since ideally, we want all tests running on Ursa. In order to merge the MPAS PR, we had to remove control_rrfs_mpas_gnu from running on Ursa because there were issues with it, but it was passing on Hercules and Hera. However, it is the only test we run that does not run on Ursa.

@gspetro-NOAA OK, thanks for the backstory. I was wondering if the omission of Ursa for that test was an oversight or what, so now I know.

[jkbk2004]

@grantfirl new ufsatm hash is NOAA-EMC/ufsatm@09ca999

[jkbk2004]

@grantfirl new cmeps hash NOAA-EMC/CMEPS@607a0bd as well

[grantfirl]

@jkbk2004 Finished updating submodules

[dustinswales]

@grantfirl @gspetro-NOAA Thanks for carrying this PR across the finish line!
As for the lack of an MPAS test on Ursa... This was not an intentional oversight, but a result of the MPAS PR being in the queue while we were transitioning from Hera to Ursa and I didn't have time to track the issue down on Ursa. This is on my radar and will address soon.