Use giraph code used for parsing -D and -hiveconf args. This also solves setting the chronos pool via -D option
do a hive -e "alter table hive_io_test drop if exists partition (ds='2013-04-01')" before running
Currently the command line script does not exit for some reason even though data appears on hive. This is something further to debug. ps aux | grep java shows hadoop running
Test plan:
[groys@dev1165 ~/debugging] hive -e "select * from hive_io_test where ds='2013-04-01'"
Logging initialized using configuration in file:/mnt/vol/hive/sites/hive99-rH32842-20130216.rH34080/shortcuts/silver/hive/conf/hive-log4j.properties
Hive history file=/tmp/groys/hive_job_log_groys_201304142024_834523977.txt
OK
23 34 45 2013-04-01
11 22 33 2013-04-01
67 78 89 2013-04-01
99 88 77 2013-04-01
Test plan: [groys@dev1165 ~/debugging] hive -e "select * from hive_io_test where ds='2013-04-01'"
Logging initialized using configuration in file:/mnt/vol/hive/sites/hive99-rH32842-20130216.rH34080/shortcuts/silver/hive/conf/hive-log4j.properties Hive history file=/tmp/groys/hive_job_log_groys_201304142024_834523977.txt OK 23 34 45 2013-04-01 11 22 33 2013-04-01 67 78 89 2013-04-01 99 88 77 2013-04-01