Main menu

Navigation

a strange error as writing restart files

1 post / 0 new
jyangdas@...
a strange error as writing restart files

Hi,

I met a strange error when I port CCSM3 to the cluster of University of Chicago. 

I can well conifgure, build and run the model, but the model blow up just one month before STOP_N.

For example, if I want to run 3 months, the model blows up at 2nd month. 

It likely results from the step of writing restart files. 

I know that CCSM3 is not the model suggested yet, however, 

I will be very grateful if you can give me some suggestions. 

The error message is like this:

**********************************************************

nstep, te 7246 3343331092.90397 -4.97626708282365
4.967211231589157E-004 98459.7853385585
nstep, te 7247 3343306270.19770 -6.21472088827027
6.203410353290938E-004 98459.8001529176
(CSM_DOSNDRCV) output restart and history files at nstep = 7248
(CSM_DOSNDRCV) output restart and history files at nstep = 7248
(CSM_DOSNDRCV) output restart and history files at nstep = 7248
(CSM_DOSNDRCV) output restart and history files at nstep = 7248
(CSM_DOSNDRCV) output restart and history files at nstep = 7248
(CSM_DOSNDRCV) output restart and history files at nstep = 7248
(CSM_DOSNDRCV) output restart and history files at nstep = 7248
(CSM_DOSNDRCV) output restart and history files at nstep = 7248
(CSM_DOSNDRCV) output restart and history files at nstep = 7248
(CSM_DOSNDRCV) output restart and history files at nstep = 7248
nstep, te 7248 3343246823.80866 -5.31669783367051
5.307021995966623E-004 98459.7939986591
ENDRUN:WRTOUT_R8
ENDRUN:WRTOUT_R8
ENDRUN:WRTOUT_R8
ENDRUN:WRTOUT_R8
ENDRUN:WRTOUT_R8
ENDRUN:WRTOUT_R8
ENDRUN:WRTOUT_R8
ENDRUN:WRTOUT_R8
ENDRUN:WRTOUT_R8
ENDRUN:WRTOUT_R8
ENDRUN:WRTOUT_R8
APPLICATION TERMINATED WITH THE EXIT STRING: Hangup (signal 1)

**********************************************************

The Macros.Linux I use is:

INCLDIR := -I. -I$(FPATH) -I$(NETCDF_INC) -I$(INCROOT) -I$(INC_MPI) -I$(NETCDF_MOD)
SLIBS := -L$(NETCDF_LIB) -lnetcdf

CC     := mpiicc

FC     :=mpiifort

**********************************************************

The submission script I use is:

## BATCH INFO
#SBATCH --error=z1.slurm.standard.%j.err
#SBATCH --output=z2.slurm.standard.%j.out
#SBATCH --begin=now
#SBATCH --time=12:00:00
#SBATCH --verbose
#SBATCH --nodes=4
#SBATCH --ntasks-per-node=16
#SBATCH --mail-type=end
#SBATCH --mail-user=jyangdas@gmail.com
#SBATCH --constraint=ib
#SBATCH --partition=sandyb
#SBATCH --qos=normal

source /etc/csh.cshrc
module load intelmpi

###limit coredumpsize 1000000
limit stacksize unlimited

**********************************************************

Thank you very much!

Jun Yang

University of Chicago

Who's new

  • anjanadevanand@...
  • hfrenzel@...
  • yunqian.zhu@...
  • acostar@...
  • Daniele.visioni@...