Sync waiting for the TERMINATED status of an EMR cluster use as : python emr_wait.py us-east-1 j-XXXXXXXXXXXX

emr_wait.py

#!/usr/bin/python

import boto.emr 
import time
import sys

region = sys.argv[1]
clusterId = sys.argv[2]

def getStatus(cluster_id):
    return conn.describe_cluster(cluster_id).status.state
    
#EMR has been started by aws-cli command
start_time = time.time()
conn = boto.emr.connect_to_region(region)

while True:
    status = getStatus(clusterId)
    if status=='TERMINATED':
        print 'Emr Status is finally TERMINATED'
        break
    elif status=='TERMINATED_WITH_ERRORS':
        print 'Errors on the EMR Cluster, please check the log on amazon (console and s3)'
        sys.exit("EMR_ERROR : TERMINATED_WITH_ERRORS")
    else:
        print 'Status is ' + status + ' waiting for 5 more minutes'
        time.sleep(300)

print 'EMR Job finished in ' + str(round((time.time() - start_time)/60, 2)) + ' minutes'