Firas,

That setting should point to the location of your hadoop mapreduce installation (ie. /usr/lib/hadoop-0.20-mapreduce).

Sqoop uses MapReduce to distribute the transfer of data. This means that a MapReduce job will be created and several tasks to perform the work. Could you please inspect your job tracker and task logs? Also, remember to run your sqoop job with the "verbose" (--verbose) command to get the most out of logging.

If you're still having difficulty, please provide your sqoop log, job tracker log, and relevant task logs.

-Abe


On Thu, Oct 17, 2013 at 1:09 PM, Firas Khasawneh <Firas.Khasawneh@sas.com> wrote:

Hi Abraham,

 

I am using hadoop 1.2.1. In my path, the only doubt, I have is the setting of HADOOP_MAPRED_HOME. I tried to set it to hadoop home, to /tmp and to a directory that does not exist. I am still not getting any errors. It just hangs and never come back.

 

Thanks,

Firas

 

From: Firas Khasawneh
Sent: Thursday, October 17, 2013 3:57 PM
To: user@sqoop.apache.org
Subject: RE: Sqoop hangs

 

Hi Abraham,

 

I am using Postgres 9.1 on Ubuntu. Nothing in the logs. I am also able to do table-list from sqoop.

 

Thanks,

Firas

 

From: Abraham Elmahrek [mailto:abe@cloudera.com]
Sent: Thursday, October 17, 2013 3:50 PM
To: user@sqoop.apache.org
Subject: Re: Sqoop hangs

 

Firas,

 

Have you checked your postgresql logs? The database could be in a deadlock. Also, what version of PostGreSQL are you running?

 

-Abe

 

On Thu, Oct 17, 2013 at 12:24 PM, Firas Khasawneh <Firas.Khasawneh@sas.com> wrote:

Hi,

 

When importing from a Postgres database, sqoop hangs and never exists unless I kill it from the command line. Any help regarding this? I am not seeing any errors or log files for sqoop.

 

Thanks,

Firas