JobSubmissionTroubleShooting

PBS : processApp.py does not start jobs

Make sure you have PBS set up such that a job can submit jobs. On a system where this was not allowed the following weird error was received.

Can't locate local/lib.pm in @INC (you may need to install the local::lib module) (@INC contains: /home/famelung/MINICONDA3_GOOD/lib/site_perl/5.26.2/x86_64-linux-thread-multi /home/famelung/MINICONDA3_GOOD/lib/site_perl/5.26.2 /home/famelung/MINICONDA3_GOOD/lib/5.26.2/x86_64-linux-thread-multi /home/famelung/MINICONDA3_GOOD/lib/5.26.2 .).
BEGIN failed--compilation aborted.
socket_connect_unix failed: 15137
qsub: cannot connect to server (null) (errno=15137) could not connect to trqauthd

You can test using simple sleep jobs: //eos/home/famelung[1019] cat sleep1.job

#! /bin/bash
#PBS -N unittestGalapagosSenDT128
#PBS -A insarlab
#PBS -m bea
#PBS -M None
#PBS -l nodes=1:ppn=1
#PBS -o sleep1_$PBS_JOBID.o
#PBS -e sleep1_$PBS_JOBID.e
#PBS -q batch
#PBS -l walltime=24:00:00
#PBS -l mem=3600
#PBS -V
free
echo  start sleep
sleep 5
echo done sleeping
echo now submitting sleep2 job
qsub <  sleep2.job
echo done with sleep2

//eos/home/famelung[1020] cat sleep2.job

#! /bin/bash
#PBS -N unittestGalapagosSenDT128
#PBS -A insarlab
#PBS -m bea
#PBS -M None
#PBS -l nodes=1:ppn=1
#PBS -o sleep2_$PBS_JOBID.o
#PBS -e sleep2_$PBS_JOBID.e
#PBS -q batch
#PBS -l walltime=1:00:00
#PBS -l mem=3600
#PBS -V
free
echo  start sleep2
sleep 5
echo done sleep2