executing sqoop jobs in oozie

[#1]

I have installed cloudera express version pn centos machine with cloudera manager.

I have previously worked with oozie and have written the workflows manually without using any editor. 

Hue looks great but as of now I am looking forward to just deploy my workflow.xml/coordinator.xml and job.properteties file so that I can launch an oozie job from what I have already with me rather than designing it in the Hue. 

Previously I have had users created under /home directory on my centos machine. Though I have NOT installed cloudera as single user, I still do not see those different users created under home directory. 

and so I am not able to login as a different user with command like su oozie.

Also, if I want to check the hadoop file system, I certainly can not do that from root user here using hadoop fs -ls command. What is the alternative for that? I am 1 day old with cloudera and find it much stabler as compared to rest of the distributions available so thank you for the great work. 

so

1. On running a sqoop job, I get

java.lang.RuntimeException: Could not load db driver class: com.mysql.jdbc.Driver

Which directory do I add the driver in?

2. How to add deploy already written workflows in oozie?

3. Why are users like hdfs, oozie, sqoop not created under /home directory? why I am not able to use command su hdfs and login as hdfs? I supposed Ideally these users should have been created?

[#2 · Accepted solution]

Welcome to the Cloudera Community @sim6. Am I correct in thinking you posted the questions here because you were wondering how to use Hue on the three issues? Otherwise, it looks like you have a few different questions that may be better answered outside of the Hue board. In that case, I would suggest creating new posts for each question in the forum board that best suits it. 

Question one may be a fit for the Data Ingestion board which covers Sqoop. Question two would be a match for the batch processing board that covers Oozie. Question three seems to be something for the Hadoop Concepts board.

I hope this helps. 🙂