Re-submission to HTCondor in case of a previously failing submission (i.e. run_six.sh -w) simply re-submits the .list file as is, instead of cutting it in chunks - this can lead to pass the limit of HTCondor in terms of number of jobs submitted per cluster.
Reported by JBarranco