qsub - jobs being run on separate nodes of a cluster are being killed prematurely by grid engine -

i working on cluster want submit series of jobs (each different file) different command nodes follows:

i wrote script makes loop , collects each of files want, in case labeled chromosome, , 'qsubs' each 1 processed command on separate nodes

multiple_scripts.sh

#!/bin/bash  in {1..19};    qsub ./pwrapper.sh $i done 

the command, perl script given bellow:

pwrapper.sh

#!/bin/bash #$ -cwd #$ -q all.q #$ -j y #$ -s /bin/bash  perl -w aggregate_profile.pl -input=1_occup_chr$1.bed -regions=all_mesc_features --average_aligned=chr$1.correct_average_alligned.w1 [furhter_parameters] 

however reason process kills returning following error each of jobs:

/opt/gridengine/default/spool/chromosome-0-1/job_scripts/19868: line 8:  6563 killed