search

oceanmodeling / ufs-weather-model

This repo is forked from ufs-weather-model, and contains the model code and external links needed to build the UFS coastal model executable and model components, including the ROMS, FVCOM, ADCIRC and SCHISM plus WaveWatch III model components.
https://github.com/oceanmodeling/ufs-coastal-app
Other
2 stars 3 forks source link

Ufs-Coastal SCHISM test #11

Closed yunfangsun closed 2 months ago

yunfangsun commented 10 months ago

Hi Ufuk and Takis, @uturuncoglu @pvelissariou1

I am including here the threads that we talked about today.

There is a compiling error may be caused by SCHISM:

<+ OMP_NUM_THREADS=1

the error message for the regression test is as follows:

5: forrtl: error (78): process killed (SIGTERM) 5: Image PC Routine Line Source 5: fv3.exe 000000000244B97B Unknown Unknown Unknown 5: libc-2.17.so 00002AAFF5128340 Unknown Unknown Unknown 5: libmpi.so.12.0.0 00002AAFF372A5BC Unknown Unknown Unknown 5: libmpi.so.12.0.0 00002AAFF36368AE Unknown Unknown Unknown 5: libmpi.so.12.0.0 00002AAFF3652568 Unknown Unknown Unknown 5: libmpi.so.12.0.0 00002AAFF361CC90 Unknown Unknown Unknown 5: libmpi.so.12.0.0 00002AAFF35FC7EC Unknown Unknown Unknown 5: libmpi.so.12.0.0 00002AAFF373F66D Unknown Unknown Unknown 5: libmpi.so.12.0.0 00002AAFF35DF82F MPI_Bcast Unknown Unknown 5: libmpifort.so.12. 00002AAFF31B53E4 pmpi_bcast Unknown Unknown 5: fv3.exe 0000000001DE866D schisminit 3232 schism_init.F90 5: fv3.exe 0000000001D5748D schism_nuopccap 339 schism_nuopc_cap.F90 5: fv3.exe 00000000009AB464 _ZN5ESMCI6FTable1 2167 ESMCI_FTable.C 5: fv3.exe 00000000009AF3E6 ESMCI_FTableCallE 824 ESMCI_FTable.C 5: fv3.exe 00000000008454BB _ZN5ESMCI3VMK5ent 1112 ESMCI_VMKernel.C 5: fv3.exe 000000000077C1AA _ZN5ESMCI2VM5ente 1216 ESMCI_VM.C 5: fv3.exe 00000000009ACAF7 c_esmc_ftablecall 981 ESMCI_FTable.C 5: fv3.exe 000000000080A29D esmf_compmod_mp_e 1222 ESMF_Comp.F90 5: fv3.exe 0000000000C07246 esmfgridcompmod 1407 ESMF_GridComp.F90 5: fv3.exe 00000000007B1048 nuopc_driver_mp_l 2565 NUOPC_Driver.F90 5: fv3.exe 00000000007D5C17 nuopc_driver_mp_i 1260 NUOPC_Driver.F90 5: fv3.exe 00000000007DEAFE nuopc_driver_mp_i 455 NUOPC_Driver.F90 5: fv3.exe 00000000009AB464 _ZN5ESMCI6FTable1 2167 ESMCI_FTable.C 5: fv3.exe 00000000009AF3E6 ESMCI_FTableCallE 824 ESMCI_FTable.C 5: fv3.exe 00000000008452CF _ZN5ESMCI3VMK5ent 2308 ESMCI_VMKernel.C 5: fv3.exe 000000000077C1AA _ZN5ESMCI2VM5ente 1216 ESMCI_VM.C 5: fv3.exe 00000000009ACAF7 c_esmc_ftablecall 981 ESMCI_FTable.C 5: fv3.exe 000000000080A29D esmf_compmod_mp_e 1222 ESMF_Comp.F90 5: fv3.exe 0000000000C07246 esmfgridcompmod 1407 ESMF_GridComp.F90 5: fv3.exe 000000000041996E MAIN 381 UFS.F90 5: fv3.exe 00000000004189E2 Unknown Unknown Unknown 5: libc-2.17.so 00002AAFF5114495 libc_start_main Unknown Unknown 5: fv3.exe 00000000004188E9 Unknown Unknown Unknown srun: error: Orion-24-02: tasks 0-2,4-5: Exited with exit code 1

The out log is as follows:

Model started: Mon Aug 28 00:23:38 CDT 2023 0: 0: 0: . . . . . . . . . . . . . . . . . . . . 0: PROGRAM ufs HAS BEGUN. COMPILED 0.00 ORG: np23 0: STARTING DATE-TIME AUG 28,2023 00:23:42.913 240 MON 2460185 0: 0:

Thank you for your help!

Yunfang

uturuncoglu commented 10 months ago

@pvelissariou1 @yunfangsun I think I remember the issue with SCHSIM. When I updated there was a change in terms of nws parameter and I wait it to be resolve before update the SCHSIM and cap. We need to check with SCHSIM developers again. So, I still did not update the component. I updated the file that is responsible to move missing file to RT run directory and push to ufs-coastal repository. @yunfangsun could you try again from scratch and let me know.

pvelissariou1 commented 10 months ago

@ufuk, @Yunfang Sun - NOAA Affiliate @.***> FYI, I updated SCHISM in CoastalApp and the schism test cases in the testsuite. If I remember correctly the test cases run as expected.

Panagiotis Velissariou, Ph.D., P.E. UCAR Scientist National Ocean and Atmospheric Administration National Ocean Service Office of Coast Survey CSDL/CMMB Physical Scientist - Project Lead cell: (205) 227-9141 email: @.***

On Mon, Aug 28, 2023 at 12:27 PM Ufuk Turunçoğlu @.***> wrote:

@pvelissariou1 https://github.com/pvelissariou1 @yunfangsun https://github.com/yunfangsun I think I remember the issue with SCHSIM. When I updated there was a change in terms of nws parameter and I wait it to be resolve before update the SCHSIM and cap. We need to check with SCHSIM developers again. So, I still did not update the component. I updated the file that is responsible to move missing file to RT run directory and push to ufs-coastal repository. @yunfangsun https://github.com/yunfangsun could you try again from scratch and let me know.

— Reply to this email directly, view it on GitHub https://github.com/oceanmodeling/ufs-coastal/issues/11#issuecomment-1696071401, or unsubscribe https://github.com/notifications/unsubscribe-auth/APC7TP2GWTQQIEDSSCRLGDDXXTIIPANCNFSM6AAAAAA4BUXI74 . You are receiving this because you were mentioned.Message ID: @.***>

uturuncoglu commented 10 months ago

@pvelissariou1 I pushed the change to the script that creates the run directory. It most me fine now.

yunfangsun commented 10 months ago

Hi @uturuncoglu I have tried to clone the new ufs-coastal, and the compilation error kept the same, and the log stuck at the following stage:

Job id 14805681 TEST 3_intel compile_intel is waiting to enter the queue TEST 3_intel compile_intel is submitted 1 min. TEST 3_intel compile_intel is pending, status: PD jobid 14805681 2 min. TEST 3_intel compile_intel is pending, status: PD jobid 14805681 3 min. TEST 3_intel compile_intel is pending, status: PD jobid 14805681 4 min. TEST 3_intel compile_intel is pending, status: PD jobid 14805681 5 min. TEST 3_intel compile_intel is pending, status: PD jobid 14805681 6 min. TEST 3_intel compile_intel is pending, status: PD jobid 14805681 7 min. TEST 3_intel compile_intel is pending, status: PD jobid 14805681 8 min. TEST 3_intel compile_intel is pending, status: PD jobid 14805681 9 min. TEST 3_intel compile_intel is pending, status: PD jobid 14805681 10 min. TEST 3_intel compile_intel is running, status: R jobid 14805681 Slurm unknown status -. Check sacct ... 14805681 COMPLETED compile_3_intel 14805681.ba+ COMPLETED batch 14805681.ex+ COMPLETED extern 11 min. TEST 3_intel compile_intel is COMPLETED, status: - jobid 14805681

uturuncoglu commented 10 months ago

@yunfangsun What is the actual error? Could you paste output of /work/noaa/nosofs/yunfangs/ufs-ww3//stmp/yunfangs/FV3_RT/rt_430214/compile_3_intel/err. I could see Slurm unknown status -. Check sacct ... but this could be related with the job scheduler or the account that you are trying to run the case.

yunfangsun commented 10 months ago

Hi @uturuncoglu

The error < /work/noaa/nosofs/yunfangs/ufs-ww3//stmp/yunfangs/FV3_RT/rt_430214/compile_3_intel/err > is as bellows:

The externaly compiled ParMETIS library could not be found. Reverting back to SCHISM's internal ParMETIS.

CMake Warning: Manually-specified variables were not used by the project: MPI

pvelissariou1 commented 10 months ago

@yunfangsun https://github.com/yunfangsun @uturuncoglu https://github.com/uturuncoglu Yunfang those issues (queue delays, etc) are related to the slurm workflow manager I believe: You allocations on Orion are as follows: Account QOS

       nos-surge                                           windfall
           coast                                           windfall
          nosofs                                           windfall

On the other hand my allocations are: Account QOS

       nos-surge                                           windfall
           coast                                           windfall
         coastal              batch,debug,novel,ood,urgent,windfall
          nosofs                                           windfall
             ohd              batch,debug,novel,ood,urgent,windfall

The windfall is of the lowest priority in the queue (VERY VERY LONG WAIT TIMES) This is most likely why your runs for UFS-Coastal fail (other issues as well?) In my case, I use my coastal account and my runs go through. I'll let Ed and Saeed know about this.

Panagiotis Velissariou, Ph.D., P.E. UCAR Scientist National Ocean and Atmospheric Administration National Ocean Service Office of Coast Survey CSDL/CMMB Physical Scientist - Project Lead cell: (205) 227-9141 email: @.***

On Mon, Aug 28, 2023 at 1:36 PM Ufuk Turunçoğlu @.***> wrote:

@yunfangsun https://github.com/yunfangsun What is the actual error? Could you paste output of /work/noaa/nosofs/yunfangs/ufs-ww3//stmp/yunfangs/FV3_RT/rt_430214/compile_3_intel/err. I could see Slurm unknown status -. Check sacct ... but this could be related with the job scheduler or the account that you are trying to run the case.

— Reply to this email directly, view it on GitHub https://github.com/oceanmodeling/ufs-coastal/issues/11#issuecomment-1696159661, or unsubscribe https://github.com/notifications/unsubscribe-auth/APC7TP6DQ7K5HKH7NDNHGWLXXTQKJANCNFSM6AAAAAA4BUXI74 . You are receiving this because you were mentioned.Message ID: @.***>

pvelissariou1 commented 10 months ago

Yunfang, did you get the executable FV3* ?

Panagiotis Velissariou, Ph.D., P.E. UCAR Scientist National Ocean and Atmospheric Administration National Ocean Service Office of Coast Survey CSDL/CMMB Physical Scientist - Project Lead cell: (205) 227-9141 email: @.***

On Mon, Aug 28, 2023 at 1:46 PM Yunfang Sun @.***> wrote:

Hi @uturuncoglu https://github.com/uturuncoglu

The error < /work/noaa/nosofs/yunfangs/ufs-ww3//stmp/yunfangs/FV3_RT/rt_430214/compile_3_intel/err

is as bellows:

The externaly compiled ParMETIS library could not be found. Reverting back to SCHISM's internal ParMETIS.

CMake Warning: Manually-specified variables were not used by the project:

MPI

  • OMP_NUM_THREADS=1

  • make -j 8 VERBOSE=1 fatal: not a git repository (or any parent up to mount point /) Stopping at filesystem boundary (GIT_DISCOVERY_ACROSS_FILESYSTEM not set). fatal: not a git repository (or any parent up to mount point /) Stopping at filesystem boundary (GIT_DISCOVERY_ACROSS_FILESYSTEM not set). /work/noaa/nosofs/yunfangs/ufs-ww3/ufs-coastal/CMEPS-interface/CMEPS/ufs/glc_elevclass_mod.F90(43): warning #6843: A dummy argument with an explicit INTENT(OUT) declaration is not given an explicit value. [GLC_ELEVCLASS] subroutine glc_get_elevation_classes_with_bareland(glc_ice_covered, glc_topo, glc_elevclass, logunit)

    --------------------------------------------------------------------------------^ /work/noaa/nosofs/yunfangs/ufs-ww3/ufs-coastal/CMEPS-interface/CMEPS/ufs/glc_elevclass_mod.F90(36): warning #6843: A dummy argument with an explicit INTENT(OUT) declaration is not given an explicit value. [GLC_ELEVCLASS] subroutine glc_get_elevation_classes_without_bareland(glc_topo, glc_elevclass, logunit) ------------------------------------------------------------------^ /work/noaa/nosofs/yunfangs/ufs-ww3/ufs-coastal/CMEPS-interface/CMEPS/ufs/glc_elevclass_mod.F90(59): warning #6843: A dummy argument with an explicit INTENT(OUT) declaration is not given an explicit value. [GLC_ICEFRAC_EC] subroutine glc_get_fractional_icecov(nec, glc_topo, glc_icefrac, glc_icefrac_ec, logunit) -------------------------------------------------------------------^

  • mv /work/noaa/nosofs/yunfangs/ufs-ww3/stmp/yunfangs/FV3_RT/rt_430214/compile_3_intel/build_fv3_3_intel/ufs_model /work/noaa/nosofs/yunfangs/ufs-ww3/ufs-coastal/tests/fv3_3_intel.exe
  • [[ orion == linux ]]
  • cp /work/noaa/nosofs/yunfangs/ufs-ww3/ufs-coastal/modulefiles/ufs_orion.intel.lua /work/noaa/nosofs/yunfangs/ufs-ww3/ufs-coastal/tests/modules.fv3_3_intel.lua
  • '[' YES = YES ']'
  • rm -rf /work/noaa/nosofs/yunfangs/ufs-ww3/stmp/yunfangs/FV3_RT/rt_430214/compile_3_intel/build_fv3_3_intel
  • elapsed=103
  • echo 'Elapsed time 103 seconds. Compiling -DAPP=CSTLS -DNO_PARMETIS=OFF -DOLDIO=ON -DMPI=ON -DCMAKE_BUILD_TYPE=Release finished'
  • echo 'Compile 3_intel elapsed time 103 seconds. -DAPP=CSTLS -DNO_PARMETIS=OFF -DOLDIO=ON -DMPI=ON -DCMAKE_BUILD_TYPE=Release' ++ date
  • echo 'Compile ended: ' Mon Aug 28 13:10:36 CDT 2023 ++ date +%s
  • echo -n ' 1693246236,' "/work/noaa/nosofs/yunfangs/ufs-ww3//stmp/yunfangs/FV3_RT/rt_430214/compile_3_intel/err" 230L, 10388C

— Reply to this email directly, view it on GitHub https://github.com/oceanmodeling/ufs-coastal/issues/11#issuecomment-1696187541, or unsubscribe https://github.com/notifications/unsubscribe-auth/APC7TPYFUAVYVT3CZMEF3GDXXTRODANCNFSM6AAAAAA4BUXI74 . You are receiving this because you were mentioned.Message ID: @.***>

yunfangsun commented 10 months ago

Hi Takis @pvelissariou1

Thank you! Yes, the waiting time is quite long for windfall Qos, for all the accounts: nos-surge, coast, and nosofs.

Yunfang

uturuncoglu commented 10 months ago

@pvelissariou1 @yunfangsun I could not see any issue with the build. I think the build works since I could see echo 'Compile ended: ' Mon Aug 28 13:10:36 CDT 2023 in the log. If you check the run directory and verify it is working (you could see PET*, err and out files), then that is great.

yunfangsun commented 10 months ago

Hi Ufuk @uturuncoglu ,

The PET0.ESMF_LogFile looks like: 20230828 132418.137 INFO PET0 ScalarFieldCount = 3 20230828 132418.137 INFO PET0 ScalarFieldIdxGridNX = 1 20230828 132418.137 INFO PET0 ScalarFieldIdxGridNY = 2 20230828 132418.137 INFO PET0 ScalarFieldIdxNextSwCday = 3 20230828 132418.137 INFO PET0 ScalarFieldName = cpl_scalars 20230828 132418.137 INFO PET0 start_type = startup 20230828 132418.137 INFO PET0 restart_dir = RESTART/ 20230828 132418.137 INFO PET0 case_name = ufs.cpld 20230828 132418.137 INFO PET0 restart_n = 12 20230828 132418.137 INFO PET0 restart_option = nhours 20230828 132418.137 INFO PET0 restart_ymd = -999 20230828 132418.137 INFO PET0 orb_eccen = 1.e36 20230828 132418.137 INFO PET0 orb_iyear = 2000 20230828 132418.137 INFO PET0 orb_iyear_align = 2000 20230828 132418.137 INFO PET0 orb_mode = fixed_year 20230828 132418.137 INFO PET0 orb_mvelp = 1.e36 20230828 132418.137 INFO PET0 orb_obliq = 1.e36 20230828 132418.137 INFO PET0 stop_n = 24 20230828 132418.137 INFO PET0 stop_option = nhours 20230828 132418.137 INFO PET0 stop_ymd = -999 20230828 132418.137 INFO PET0 ReadAttributes ALLCOMP_attributes:: end: 20230828 132418.140 INFO PET0 (med.F90:InitializeP0): called 20230828 132418.141 INFO PET0 (med.F90:InitializeP0): Mediator dbug_flag is 0 20230828 132418.141 INFO PET0 (med.F90:InitializeP0): done "PET0.ESMF_LogFile" 166L, 12032C

The /work/noaa/nosofs/yunfangs/ufs-ww3/stmp/yunfangs/FV3_RT/rt_430214/coastal_ike_shinnecock_atm2sch_intel/err looks like:

The out looks like: Model started: Mon Aug 28 13:24:13 CDT 2023 0: 0: 0: . . . . . . . . . . . . . . . . . . . . 0: PROGRAM ufs HAS BEGUN. COMPILED 0.00 ORG: np23 0: STARTING DATE-TIME AUG 28,2023 13:24:18.069 240 MON 2460185 0: 0: 3: 0: ABORT: you must have sflux_inputs_file! 5: 2: ABORT: you must have sflux_inputs_file! 4: 1: ABORT: you must have sflux_inputs_file!

It seems that I missed the sflux_inputs_file, Orion-login-2[273] yunfangs$ ls INPUT_DATM/ wind_atm_fin_ch_time_vec_ESMFmesh.nc wind_atm_fin_ch_time_vec_STR_fixed.nc

uturuncoglu commented 10 months ago

@yunfangsun I am not sure. Maybe this is permission issue and you could get data from my RT directory. I rerun chmod command again. Please try again, if it wont run, I'll test again. It is working in my ufs-coastal copy.

uturuncoglu commented 10 months ago

@yunfangsun I think sflux_inputs_filemessage is coming from SCHSIM not CDEPS. So, maybe we still have inconsistency interns of parameters that needs to be used.

pvelissariou1 commented 10 months ago

@uturuncoglu https://github.com/uturuncoglu @yunfangsun https://github.com/yunfangsun I didn't notice any issues like that when I ran the test case. I am pointing to: DISKNM=/work/noaa/nems/tufuk/RT Only thing is I did not get output files. I'll check on that later today. There are multiple issues on Orion for us, we are working on a solution for orion/hercules.

Panagiotis Velissariou, Ph.D., P.E. UCAR Scientist National Ocean and Atmospheric Administration National Ocean Service Office of Coast Survey CSDL/CMMB Physical Scientist - Project Lead cell: (205) 227-9141 email: @.***

On Mon, Aug 28, 2023 at 2:15 PM Ufuk Turunçoğlu @.***> wrote:

@yunfangsun https://github.com/yunfangsun I think sflux_inputs_filemessage is coming from SCHSIM not CDEPS. So, maybe we still have inconsistency interns of parameters that needs to be used.

— Reply to this email directly, view it on GitHub https://github.com/oceanmodeling/ufs-coastal/issues/11#issuecomment-1696236033, or unsubscribe https://github.com/notifications/unsubscribe-auth/APC7TP2BHDYRPDPUWPWCPGTXXTU2XANCNFSM6AAAAAA4BUXI74 . You are receiving this because you were mentioned.Message ID: @.***>

yunfangsun commented 10 months ago

@uturuncoglu Hi Ufuk,

I tried again after you changed the RT permission, the out file is still as follows:

Model started: Mon Aug 28 15:05:53 CDT 2023 0: 0: 0: . . . . . . . . . . . . . . . . . . . . 0: PROGRAM ufs HAS BEGUN. COMPILED 0.00 ORG: np23 0: STARTING DATE-TIME AUG 28,2023 15:05:56.123 240 MON 2460185 0: 0: 5: 2: ABORT: you must have sflux_inputs_file! 3: 0: ABORT: you must have sflux_inputs_file! 4: 1: ABORT: you must have sflux_inputs_file!

And the output files contain no data:

4.0K -rw-r----- 1 yunfangs nosofs 4.0K Aug 28 15:05 schout_000000_1.nc 4.0K -rw-r----- 1 yunfangs nosofs 4.0K Aug 28 15:05 schout_000001_1.nc 4.0K -rw-r----- 1 yunfangs nosofs 4.0K Aug 28 15:05 schout_000002_1.nc

pvelissariou1 commented 10 months ago

@uturuncoglu https://github.com/uturuncoglu @Yunfang Sun - NOAA Affiliate @.**> Yunfang, I don't understand why you are getting the error: ABORT: you must have sflux_inputs_file! This is a 2D case and it does not use atmospheric fluxes. Am I missing something here? I have run the atm2sch and *sch test cases using both CoastalApp and UFS-Coastal (see my presentation for the SCHISM framework) without noticing such issues. is the barotropic flag on in your namelist file? Could you please double check?

Did you run this test case from within CoastalApp/CoastalApp-testsuite? CoastalApp uses the latest version of SCHISM, so in this case you need to add in NEMS/src/incmake/component_SCHISM the flag USE_ATMOS in order to use the nws = 2 option.

Panagiotis Velissariou, Ph.D., P.E. UCAR Scientist National Ocean and Atmospheric Administration National Ocean Service Office of Coast Survey CSDL/CMMB Physical Scientist - Project Lead cell: (205) 227-9141 email: @.***

On Mon, Aug 28, 2023 at 4:10 PM Yunfang Sun @.***> wrote:

@uturuncoglu https://github.com/uturuncoglu Hi Ufuk,

I tried again after you changed the RT permission, the out file is still as follows:

Model started: Mon Aug 28 15:05:53 CDT 2023 0: 0: 0: . . . . . . . . . . . . . . . . . .

  • . * . 0: PROGRAM ufs HAS BEGUN. COMPILED 0.00 ORG: np23 0: STARTING DATE-TIME AUG 28,2023 15:05:56.123 240 MON 2460185 0: 0: 5: 2: ABORT: you must have sflux_inputs_file! 3: 0: ABORT: you must have sflux_inputs_file! 4: 1: ABORT: you must have sflux_inputs_file!

And the output files contain no data:

4.0K -rw-r----- 1 yunfangs nosofs 4.0K Aug 28 15:05 schout_000000_1.nc 4.0K -rw-r----- 1 yunfangs nosofs 4.0K Aug 28 15:05 schout_000001_1.nc 4.0K -rw-r----- 1 yunfangs nosofs 4.0K Aug 28 15:05 schout_000002_1.nc

— Reply to this email directly, view it on GitHub https://github.com/oceanmodeling/ufs-coastal/issues/11#issuecomment-1696415996, or unsubscribe https://github.com/notifications/unsubscribe-auth/APC7TP7REY75EWABSQEW52LXXUCMFANCNFSM6AAAAAA4BUXI74 . You are receiving this because you were mentioned.Message ID: @.***>

uturuncoglu commented 10 months ago

@yunfangsun Could you reclone the ufs-coastal. I have just update the SCHSIM.

yunfangsun commented 10 months ago

@uturuncoglu @pvelissariou1 Sure, I will test CoastalApp and the new ufs-coastal again.

pvelissariou1 commented 10 months ago

Ufuk,Yunfang make sure you update the name list and use USE_ATMOS flag when compiling SCHISM in UFS-Coastal.

On Monday, August 28, 2023, Yunfang Sun @.***> wrote:

@uturuncoglu @pvelissariou1 Sure, I will test CoastalApp and the new ufs-coastal again.

— Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you were mentioned.< https://ci4.googleusercontent.com/proxy/jl2HHg9TR6-SHBNR_DV63Sj1fx0x4mBDxVuW_ulszf2UPh9FGI8Hb25M20qqHhZfrM3TLiytf6PSVAaLAo5Ug0EL1B4TNiRNNbiVaurMTJmiyrp01LVx-ysbDY64hh8eUIqaW6nD6LKQshR_haC0M67nHVSiT-zxVP66g9tEA6JJMPCGw5qRvSrpYe_zfme1jIRJ_QRFz8uqDmU5Ju4TRJX3h05gNuwvvURBoWt_h1myxHQ5lQ=s0-d-e1-ft#https://github.com/notifications/beacon/APC7TP5HPR3XCCK5NHKDWZLXXUHE5A5CNFSM6AAAAAA4BUXI76WGG33NNVSW45C7OR4XAZNMJFZXG5LFINXW23LFNZ2KUY3PNVWWK3TUL5UWJTTFDYXMA.gif>Message ID: @.***>

-- Panagiotis Velissariou, Ph.D., P.E. UCAR Scientist National Ocean and Atmospheric Administration National Ocean Service Office of Coast Survey CSDL/CMMB Physical Scientist - Project Lead cell: (205) 227-9141 email: @.***

yunfangsun commented 10 months ago

Hi Takis, @pvelissariou1 I can't compile newly cloned CoastalApp both on Orion and Parallel works, the error messages are as follows: schism_esmf_util.F90(209): error #6796: The variable must have the TARGET attribute or be a subobject of an object with the TARGET attribute, or it must have the POINTER attribute. [IDRY_E] isPtr%wrap%ptrMap(14)%iarrayPtr1 => idry_e --------------------------------------^ compilation aborted for schism_esmf_util.F90 (code 1) make[2]: [schism_esmf_util.o] Error 1 make[2]: Leaving directory `/lustre/new_coast/CoastalApp/SCHISM/schism-esmf/src/schism' make[1]: [src/schism/schism_bmi.o] Error 2 make[1]: Leaving directory `/lustre/new_coast/CoastalApp/SCHISM/schism-esmf' make: *** [/lustre/new_coast/CoastalApp/SCHISM_INSTALL/schism.mk] Error 2

pvelissariou1 commented 10 months ago

@yunfangsun @soroosh @FariborzDaneshvar-NOAA @saeed Thanks YunFang, the issue has already reported to the SCHISM developers. Waiting from them to incorporate a fix upstream. I have a fix locally that I am testing right now.

josephzhang8 commented 5 months ago

Hi all-

We are trying to do all UFS-coastal reg tests:

https://github.com/oceanmodeling/ufs-coastal/blob/feature/coastal_app/tests/rt_coastal.conf

There seems to be a dir missing; e.g., ### SCHISM tests: 11-15 etc need a dir called 'DATM_ATMMESH_input_data/'. It's supposed to be on Hercules: /work/noaa/epic/hercules/UFS-WM_RT/NEMSfv3gfs/input-data-20221101/DATM_ATMMESH_input_data/ but this dir is not there. Anyone can help? Thx!!

@danishyo

uturuncoglu commented 5 months ago

@josephzhang8 Since this development is not merged with the ufs-weather-model, the input files are not the part of the common directories. We are maintaining them on Orion - /work/noaa/nems/tufuk/RT/NEMSfv3gfs/input-data-20221101. I think you could access to the same directory from Hercules too. To use this directory, just open rt.sh and find the section for your platform and set DISKNM variable like following DISKNM=/work/noaa/nems/tufuk/RT. Then you could run the RT. In the future we are plaining to put all those input files to cloud etc. but I think we are not it that stage yet since we are still trying to make components stable and port different configurations of them. Anyway, let me know if you have still issue.

josephzhang8 commented 5 months ago

Got it; thank you so much, @uturuncoglu!