You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@sqoop.apache.org by "Hemendra Yadav (JIRA)" <ji...@apache.org> on 2017/04/06 08:57:41 UTC
[jira] [Created] (SQOOP-3168) Sqoop Saved Job feature of
overwriting job argument at execution time is not working in Sqoop1.4.6CDH
5.8.0
Hemendra Yadav created SQOOP-3168:
-------------------------------------
Summary: Sqoop Saved Job feature of overwriting job argument at execution time is not working in Sqoop1.4.6CDH 5.8.0
Key: SQOOP-3168
URL: https://issues.apache.org/jira/browse/SQOOP-3168
Project: Sqoop
Issue Type: Bug
Reporter: Hemendra Yadav
Hi,
I have created one sqoop saved job using below command ::
sqoop job --create <job_name> -- import --connect <url> --driver com.mysql.jdbc.Driver --username <username> --password <password> --table <table> --target-dir location1 --fields-terminated-by , --escaped-by \\ -m 1
I need to put the data to different location so while execution i have provided new location in below command::
sqoop job --exec jobname1 -- --target-dir location_new
But it still picks up the target directory from the saved job i.e. location1.
I am currently using sqoop 1.4.6 CDH 5.8.0 version. When i have checked the sqoop code it seems that in sqoop CDH 5.8.0 code for overwritting job argument is removed from the method "private int execJob(SqoopOptions opts)" in the class "org.apache.sqoop.tool.JobTool" due to issue "https://issues.apache.org/jira/browse/SQOOP-2779".
When i have executed the same command using sqoop 1.4.6 CDH 5.7.3 it is able to overwrite the target directory and working as expected.
Can you please suggest any workaround how to import data to a different target directory while executing sqoop saved job.
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)