Hello,
I have a problem in debugging of Intel MPI application on a cluster.
Common question: How to debug the parallel application. Console debugger idbc is not convenient at all. Are there debuggers with GUI? better for free.
I tried using eclipse, I launched program using SGE script (below), but I cannot debug with the same way.
The script for RUN.
############## RUN ##############
#!/bin/bash
#$ -S /bin/bash
cd $SGE_O_WORKDIR
. /etc/profile.d/modules.sh
module purge
module load 14.0_intel_mpi_update
#$ -l sandy=TRUE
#$ -l h_vmem=1.0G
#$ -o $JOB_ID.out
#$ -e $JOB_ID.err
#$ -cwd
#$ -pe openmpi* 6
#$ -l h_rt=00:30:00
#$ -l debug=TRUE
mpiexec.hydra -genv I_MPI_FABRICS shm:dapl -genv I_MPI_FALLBACK_DEVICE disable -n $NSLOTS ./myApp
exit 0
##############################
And it works. (I work in local regime, since eclipse is running on cluster frontend, ssh connection is not needed).
But it doesn't work in DEBUG configuration. I got error "Configuration is missing run commands for job submission dd85ff79-8053-4ffd-ade1-edb1a23272de debug"
I did like in the manual http://help.eclipse.org/luna/index.jsp?topic=%2Forg.eclipse.ptp.doc.user%2Fhtml%2F01buildMPIprog.html, but I have some doubts, because its unclear for me what is the GDB debugger should I use. SGE script for debugging I take as follows
######################################
#!/bin/bash
#$ -S /bin/bash
cd $SGE_O_WORKDIR
. /etc/profile.d/modules.sh
module purge
module load 14.0_intel_mpi_update
#$ -l sandy=TRUE
#$ -l h_vmem=1.0G
#$ -o $JOB_ID.out
#$ -e $JOB_ID.err
#$ -cwd
#$ -pe openmpi* 6
#$ -l h_rt=00:30:00
#$ -l debug=TRUE
export IDB_HOME=/cluster/intel_2013SP1/bin/
export gdb_compatible_output=1
mpiexec.hydra -idb -genv I_MPI_FABRICS shm:dapl -genv I_MPI_FALLBACK_DEVICE disable -n $NSLOTS ./myApp
exit 0
######################################
Probably I don't understand something, I am newbie in MPI.
So, could anyone help me in this?
Regards,
Ilia Sivkov.