CESM runtime error - MCT produces error while running compsets A,B, C, and D

K

klr324@psu_edu

New Member
I have successfully ported CESM1.0.3 on the systems at Penn State. I am using CESM1.0.3 for specific reasons rather than using the updated versions.I have successfully built and run the X, F, and I compsets without issue. The A, B, C, and D compsets also build successfully, but the model does not complete after submitting the job. In all cases I have tried running in DEBUG mode, with a smaller set of processors, with a single processor, and with running a the debugger dbx to determine the issue. The issue seems to be with MCT.For A, C, and D the model stops after initializing the atmosphere component and produces the following error in the ccsm.log:(seq_mct_drv) : Initialize each component: atm, lnd, ocn, and ice(seq_mct_drv) : Initialize atm component (ccsm log)MCT::m_AttrVect::exportRAttrDP_:: ERROR length of output array outVect  less than length of aV.  size(outVect)=      32, length of aV=   18048Traceback:  |X|MCT::m_AttrVect::exportRAttrDP_000.MCT(MPEU)::die.: from MCT::m_AttrVect::exportRAttrDP_()application called MPI_Abort(MPI_COMM_WORLD, 2) - process 0 ====================================================================================   BAD TERMINATION OF ONE OF YOUR APPLICATION PROCESSES=   EXIT CODE: 2=   CLEANING UP REMAINING PROCESSES=   YOU CAN IGNORE THE BELOW CLEANUP MESSAGES=================================================================================== For B the model gets through the initialization step of the components, but stops later on and produces the following error in the ccsm.log:
CalcWorkPerBlock: Total blocks:    64 Ice blocks:    64 IceFree blocks:     0 Land blocks:     0MCT::m_Router::initp_: GSMap indices not increasing...Will correctMCT::m_Router::initp_: RGSMap indices not increasing...Will correctMCT::m_Router::initp_: RGSMap indices not increasing...Will correctMCT::m_Router::initp_: GSMap indices not increasing...Will correctMCT::m_Router::initp_: RGSMap indices not increasing...Will correctMCT::m_Router::initp_: GSMap indices not increasing...Will correctMCT::m_Router::initp_: GSMap indices not increasing...Will correctMCT::m_Router::initp_: RGSMap indices not increasing...Will correctMCT::m_SparseMatrixPlus:: FATAL--length of vector y different from row count of sMat.Length of y =    11600 Number of rows in sMat =   122880000.MCT(MPEU)::die.: from MCT::m_SparseMatrixPlus::initDistributed_()MCT::m_SparseMatrixPlus:: FATAL--length of vector y different from row count of sMat.Length of y =    11600 Number of rows in sMat =   122880038.MCT(MPEU)::die.: from MCT::m_SparseMatrixPlus::initDistributed_() Thanks in advance for any help provided,Kelsey
 
G

gustavo_degoncalves@gmail_com

New Member
Hi KelseyI am having a similar problem when trying to run a single point with the recently released CLM5 under CIME framework. I have ported it to a Debian-8 virtual machine on a Mac. The case was created using./create_newcase --case brazil2 --compset I2000Clm50Sp --res 1x1_brazil --mach homebrew --run-unsupportedIt seems like MCT has not been "informed" that I am running a single point and therefore results in the matrices/vector mismatch. Are you running a full domain or subsetting a region? Below is the end of my cesm.log (complete log file in the attachment)------------------------------------------------[0]     Opening file brazil2.cism.initial_hist.0001-01-01-00000.nc for output;[0]       Starting output at    0.0000000000000000       and write every    1.0000000000000000       years[0]    Creating variables internal_time and time[0]    Temperature remapping option uses temperature on a staggered vertical grid.  The netCDF output variable "temp" has been changed to "tempstag".[0]    Creating variable level[0]    Creating variable lithoz[0]    Creating variable staglevel[0]    Creating variable stagwbndlevel[0]    Creating variable x0[0]    Creating variable x1[0]    Creating variable y0[0]    Creating variable y1[0]    Creating variable acab[0]    Creating variable artm[0]    Creating variable bwat[0]    Creating variable tempstag[0]    Creating variable thk[0]    Creating variable topg[0]    Creating variable usurf[0]    Creating variable uvel[0]    Creating variable vvel[0]     Writing to file brazil2.cism.initial_hist.0001-01-01-00000.nc at time    0.0000000000000000[0] MCT::m_SparseMatrixPlus:: FATAL--length of vector y different from row count of sMat.Length of y =        1 Number of rows in sMat =    13824[0] 000.MCT(MPEU)::die.: from MCT::m_SparseMatrixPlus::initDistributed_()[0] application called MPI_Abort(MPI_COMM_WORLD, 2) - process 0   
 
G

gustavo_degoncalves@gmail_com

New Member
Hi KelseyI am having a similar problem when trying to run a single point with the recently released CLM5 under CIME framework. I have ported it to a Debian-8 virtual machine on a Mac. The case was created using./create_newcase --case brazil2 --compset I2000Clm50Sp --res 1x1_brazil --mach homebrew --run-unsupportedIt seems like MCT has not been "informed" that I am running a single point and therefore results in the matrices/vector mismatch. Are you running a full domain or subsetting a region? Below is the end of my cesm.log (complete log file in the attachment)------------------------------------------------[0]     Opening file brazil2.cism.initial_hist.0001-01-01-00000.nc for output;[0]       Starting output at    0.0000000000000000       and write every    1.0000000000000000       years[0]    Creating variables internal_time and time[0]    Temperature remapping option uses temperature on a staggered vertical grid.  The netCDF output variable "temp" has been changed to "tempstag".[0]    Creating variable level[0]    Creating variable lithoz[0]    Creating variable staglevel[0]    Creating variable stagwbndlevel[0]    Creating variable x0[0]    Creating variable x1[0]    Creating variable y0[0]    Creating variable y1[0]    Creating variable acab[0]    Creating variable artm[0]    Creating variable bwat[0]    Creating variable tempstag[0]    Creating variable thk[0]    Creating variable topg[0]    Creating variable usurf[0]    Creating variable uvel[0]    Creating variable vvel[0]     Writing to file brazil2.cism.initial_hist.0001-01-01-00000.nc at time    0.0000000000000000[0] MCT::m_SparseMatrixPlus:: FATAL--length of vector y different from row count of sMat.Length of y =        1 Number of rows in sMat =    13824[0] 000.MCT(MPEU)::die.: from MCT::m_SparseMatrixPlus::initDistributed_()[0] application called MPI_Abort(MPI_COMM_WORLD, 2) - process 0   
 
1

18789461762@163_com

New Member
Hi, luis Gustavo Goncalves.I am new to CLM.Recently, I am trying to port clm5.0 without CESM on a platform. Should I need to port CIME on the platform firstly?Could you give me some suggestions for the beginning?Thanks so much.
 
1

18789461762@163_com

New Member
Hi, luis Gustavo Goncalves.I am new to CLM.Recently, I am trying to port clm5.0 without CESM on a platform. Should I need to port CIME on the platform firstly?Could you give me some suggestions for the beginning?Thanks so much.
 
You must log in or register to reply here.
Back
Top