ABINIT Discussion Forums

Hello all,

My spin-unpolarized input file includes 32 jdtset, 64 nkpt, 53 nband. When put to execute on a cluster using 32 cores, stops without any error in the first stage of running after checking input parameters. This is the last line of log file:
chkinp: Checking input parameters for consistency, jdtset= 32.
the cluster script file:
#! /bin/bash
#$-S /bin/bash
#$-N Abinit
#$-cwd
#$-l h_vmem=1G
#$-l hostname=amd-10-4.local
#$-j n
#$-pe mpifill 32
source /opt/intel/Compiler/11.1/069/bin/iccvars.sh intel64
source /opt/intel/Compiler/11.1/069/bin/ifortvars.sh intel64
export PATH=/storage/softwares/computing/mpi/intel/openmpi-1.6.5/bin:/storage/softwares/computing/Abinit/Abinit-7.4.3/bin:$PATH
export LD_LIBRARY_PATH=/storage/softwares/computing/mpi/intel/openmpi-1.6.5/lib:/storage/softwares/computing/Abinit/Abinit-7.4.3/lib:$LD_LIBRARY_PATH
mpirun -np $NSLOTS abinit < x.files>&log

Any help would be appriciated,

Farah Marsusi.

Please provide your config.log file as attachment.

Dear Yann,

Unfortunately I have no access to config.log file as a user of the cluster, but I believe it may be an abinit 7.4.3 bug. My crystal has monoclinic structure, and I got the error as you can see it in log files, or the job stops after checking the DATASETS. I have run other examples, but with with the FCC structure this time, the job was performed in parallel without any problem. So, I put my job in older abinit distribution. Please let me know, if you find the source of the problem.

Thanks a alot,

Marsusi.

In this case, you should open a thread in one of the "Running Abinit" section of the forum, providing your input file.