cannot build cesm with b1850

[leah_johnson@brown_edu]

Hello, I’m running cesm1 on Cheyenne that was previously run on Yellowstone. I changed the env_mach_spec according to posts (and included at the end of this email):https://bb.cgd.ucar.edu/cesm-1221-cesmsetup-module-issue-cheyennehttps://bb.cgd.ucar.edu/cheyenne-update-module-issues-all-cesm-versionsAnd have successfully run the model with compset = 2000_DATM%IAF_SLND_CICE_POP2_DROF%IAF_SGLC_SWAV.  But when I try a new case with compset = 1850_CAM4_CLM40%SP_CICE_POP2_RTM_SGLC_WW3, the model build fails at cam.buildlib with the following error in the atm.bldlog (last lines):  *******************/gpfs/u/home/leahjoh/wave04_cesm1_4_beta04_v160608/components/cam/src/utils/fft99.F90(1247): remark #5140: Unrecognized directive!DIR$ CONCURRENT ----------------^ifort: command line remark #10411: option '-no-opt-dynamic-align' is deprecated and will be removed in a future release. Please use the replacement option '-qno-opt-dynamic-align'rm buffer.F90 subcol_utils.F90 physics_buffer.F90ERROR: cam.buildexe gmake complib -j 8 MODEL=cam COMPLIB=/glade/scratch/leahjoh/b1850_f19_gx1_ww_test/bld/lib/libatm.a USER_CPPDEFS="   -DCO2A -DMAXPATCH_PFT=numpft+1 -DLSMLAT=1 -DLSMLON=1 -DPLON=144 -DPLAT=96 -DPLEV=26 -DPCNST=3 -DPCOLS=16 -DPSUBCOLS=1 -DN_RAD_CNST=30 -DPTRM=1 -DPTRN=1 -DPTRK=1 -DSTAGGERED  -DSPMD -DHAVE_VPRINTF -DHAVE_TIMES -DHAVE_GETTIMEOFDAY -DHAVE_COMM_F2C -DHAVE_NANOTIME -DBIT64 -DHAVE_SLASHPROC" -f /gpfs/u/home/leahjoh/test_wave04_cesm1_4_beta04_v190519/b1850_f19_gx1_ww_test/Tools/Makefile failed: 512  ********************I changed the load modules in env_mach_specific, and have followed a previous post to set HAS_F2008_CONTIGUOUS:=FALSE, but the build fails at the same place.  Any help would be appreciated.Best, Leah ******************* env_mach_spec: #! /bin/tcsh # -------------------------------------------------------------------------# Laramie build specific settings# ------------------------------------------------------------------------- source  /etc/profile.d/modules.csh  module purge #module load intel/19.0.1module load intel/17.0.1module load ncarenv/1.2module load mklmodule load ncarcompilers/0.4.1module load mpt/2.19 if($MPILIB== "mpi-serial" ) then   module load netcdf/4.4.1.1else  module load netcdf-mpi/4.4.1.1  module load pnetcdf/1.11.0 endif module load esmf_libs/7.1.0r   if($MPILIB== "mpi-serial") then    #if ( $DEBUG == "TRUE") then    #  module load esmf-7.1.0r-ncdfio-uni-g    #else      module load esmf-7.1.0r-ncdfio-uni-O    #endif  else    #if ( $DEBUG == "TRUE") then    #  module load esmf-7.1.0r-defio-mpi-g    #else      module load esmf-7.1.0r-defio-mpi-O     #endif   endif #module load ncarcompilers/1.0#module load cmake/2.8.10.2#module load ncarbinlibs/1.0 module load perlmods #module load gmake/4.1# load the python modulesmodule load python #module load all-python-libs

 

[jedwards]

The error in the cam log is higher up than what you posted.   /gpfs/u/home/leahjoh/wave04_cesm1_4_beta04_v160608/components/cam/src/utils/linear_1d_operators.F90: catastrophic error: **Internal compiler error: internal abort** Please report this error along with the circumstances in which it occurred in a Software Problem Report.  Note: File and line given may not be explicit cause of this error.    Try a different version of the intel compiler 19.0.2 maybe? 

 

[jedwards]

The error in the cam log is higher up than what you posted.   /gpfs/u/home/leahjoh/wave04_cesm1_4_beta04_v160608/components/cam/src/utils/linear_1d_operators.F90: catastrophic error: **Internal compiler error: internal abort** Please report this error along with the circumstances in which it occurred in a Software Problem Report.  Note: File and line given may not be explicit cause of this error.    Try a different version of the intel compiler 19.0.2 maybe? 

 

[jedwards]

The error in the cam log is higher up than what you posted.   /gpfs/u/home/leahjoh/wave04_cesm1_4_beta04_v160608/components/cam/src/utils/linear_1d_operators.F90: catastrophic error: **Internal compiler error: internal abort** Please report this error along with the circumstances in which it occurred in a Software Problem Report.  Note: File and line given may not be explicit cause of this error.    Try a different version of the intel compiler 19.0.2 maybe? 

 

[leah_johnson@brown_edu]

Thank you for your reply, jedwards. I was able to build case b1850_f19_gx1_ww_test without the above error by setting (as in previous posts)HAS_F2008_CONTIGUOUS:=FALSE But the executable fails at runtime. I also tried to build with available intel compilers including 19.0.2, but they all fail (even with HAS_F2008_CONTIGUOUS:=FALSE ) ERROR: cam.buildlib failed(e.g. b150ww.atm.bldlog.1906176-184439.txt)The only one that seems to work is intel/17.0.1.Before my first post, I had not updated config_complilers.xml (see : config_complile.orig.xml) in cime/machines/ for cheyenne.  Since then, I added to the existing file (see : config_complile.new.xml), started a new case (b1850_f19_gx1_ctrl) but the build failed earlier, while building the csm_share library (see b1850ctrl.csm_share.bldlog.190617-185310.txt.) I am actively working on two cases,a) b1850 with ww3  and config_complile.xml.orig  (b1850_f19_gx1_ww_test) - builds but does not runb) b1850 with swav and config_complile.xml.new  (b1850_f19_gx1_ctrl) - fails at buildI will continue to work on this issue and any advice would be welcome.Best,Leah  

 

[leah_johnson@brown_edu]

Thank you for your reply, jedwards. I was able to build case b1850_f19_gx1_ww_test without the above error by setting (as in previous posts)HAS_F2008_CONTIGUOUS:=FALSE But the executable fails at runtime. I also tried to build with available intel compilers including 19.0.2, but they all fail (even with HAS_F2008_CONTIGUOUS:=FALSE ) ERROR: cam.buildlib failed(e.g. b150ww.atm.bldlog.1906176-184439.txt)The only one that seems to work is intel/17.0.1.Before my first post, I had not updated config_complilers.xml (see : config_complile.orig.xml) in cime/machines/ for cheyenne.  Since then, I added to the existing file (see : config_complile.new.xml), started a new case (b1850_f19_gx1_ctrl) but the build failed earlier, while building the csm_share library (see b1850ctrl.csm_share.bldlog.190617-185310.txt.) I am actively working on two cases,a) b1850 with ww3  and config_complile.xml.orig  (b1850_f19_gx1_ww_test) - builds but does not runb) b1850 with swav and config_complile.xml.new  (b1850_f19_gx1_ctrl) - fails at buildI will continue to work on this issue and any advice would be welcome.Best,Leah  

 

[leah_johnson@brown_edu]

Thank you for your reply, jedwards. I was able to build case b1850_f19_gx1_ww_test without the above error by setting (as in previous posts)HAS_F2008_CONTIGUOUS:=FALSE But the executable fails at runtime. I also tried to build with available intel compilers including 19.0.2, but they all fail (even with HAS_F2008_CONTIGUOUS:=FALSE ) ERROR: cam.buildlib failed(e.g. b150ww.atm.bldlog.1906176-184439.txt)The only one that seems to work is intel/17.0.1.Before my first post, I had not updated config_complilers.xml (see : config_complile.orig.xml) in cime/machines/ for cheyenne.  Since then, I added to the existing file (see : config_complile.new.xml), started a new case (b1850_f19_gx1_ctrl) but the build failed earlier, while building the csm_share library (see b1850ctrl.csm_share.bldlog.190617-185310.txt.) I am actively working on two cases,a) b1850 with ww3  and config_complile.xml.orig  (b1850_f19_gx1_ww_test) - builds but does not runb) b1850 with swav and config_complile.xml.new  (b1850_f19_gx1_ctrl) - fails at buildI will continue to work on this issue and any advice would be welcome.Best,Leah  

 

[leah_johnson@brown_edu]

A small update-- -I was able to build and run the model using a GIAF compset.-My goal is to use b1850 with ww3.  To do this I loaded esmf and set HAS_F2008_CONTIGUOUS:=FALSE (niether which were needed for GIAF)  and the model will build but not run. Could this issue be rooted in esmf?Leah

 

[leah_johnson@brown_edu]

A small update-- -I was able to build and run the model using a GIAF compset.-My goal is to use b1850 with ww3.  To do this I loaded esmf and set HAS_F2008_CONTIGUOUS:=FALSE (niether which were needed for GIAF)  and the model will build but not run. Could this issue be rooted in esmf?Leah

 

[leah_johnson@brown_edu]

A small update-- -I was able to build and run the model using a GIAF compset.-My goal is to use b1850 with ww3.  To do this I loaded esmf and set HAS_F2008_CONTIGUOUS:=FALSE (niether which were needed for GIAF)  and the model will build but not run. Could this issue be rooted in esmf?Leah

 

[leah_johnson@brown_edu]

I was able to fix this problem:MPILIB was not set before 'env_machine_specific' which made it fail immediately.  I setenv MPILIB before 'env_mach_specific' in 'case.run' and was able to run the model successfully.Leah

 

[leah_johnson@brown_edu]

I was able to fix this problem:MPILIB was not set before 'env_machine_specific' which made it fail immediately.  I setenv MPILIB before 'env_mach_specific' in 'case.run' and was able to run the model successfully.Leah

 

[leah_johnson@brown_edu]

I was able to fix this problem:MPILIB was not set before 'env_machine_specific' which made it fail immediately.  I setenv MPILIB before 'env_mach_specific' in 'case.run' and was able to run the model successfully.Leah