Don't move to the next line in a batch operation until all jobs are done on the cluster, or, "poor man's hold"

wait_for_cluster.py

import os
import sys
import time
import xml.dom.minidom
import string
import getpass

def still_running(job_name, job_owner):
    f=os.popen('qstat -u \* -xml -r')

    dom=xml.dom.minidom.parse(f)

    jobs = dom.getElementsByTagName('job_info')
    run = jobs[0]

    joblist = run.getElementsByTagName('job_list')

    for r in joblist:
        job = r.getElementsByTagName('JB_name')[0].childNodes[0].data
        owner = r.getElementsByTagName('JB_owner')[0].childNodes[0].data

        if job == job_name and owner == job_owner:
            return True

    return False

job_owner = getpass.getuser()
job_name = sys.argv[1]

counter = 0
while still_running(job_name, job_owner):
    time.sleep(5)
    counter += 1
    if counter == 11:
        sys.stderr.write('.')
        counter = 0


sys.stderr.write('\n')