I have a simple hadoop cluster setup with a couple of machines (one master
running everything and one additional slave). I have hive configured with
the right hive-site.xml file pointing the fs.default.name and
mapred.job.tracker correctly to the master. When I run hive from the master
machine, hive connects fine and can schedule jobs fine. However, when I run
hive from a different machine, it cannot schedule jobs. It simply fails with
this error:

Job Submission failed with exception 'java.lang.NullPointerException(null)'
FAILED: Execution Error, return code 1 from
org.apache.hadoop.hive.ql.exec.ExecDriver

Is there any additional configuration or security involved with getting hive
to talk to my master and schedule jobs? Everything is running fine on the
master. I can access the web interfaces for namenode, jobtracker, etc. fine
from any other machine.

Thanks,
Vijay

Anything in /tmp/<username>/hive.log?
On 8/11/09 5:51 PM, "Vijay" <[EMAIL PROTECTED]> wrote:

> I have a simple hadoop cluster setup with a couple of machines (one master
> running everything and one additional slave). I have hive configured with the
> right hive-site.xml file pointing the fs.default.name <http://fs.default.name>
> and mapred.job.tracker correctly to the master. When I run hive from the
> master machine, hive connects fine and can schedule jobs fine. However, when I
> run hive from a different machine, it cannot schedule jobs. It simply fails
> with this error:
>
> Job Submission failed with exception 'java.lang.NullPointerException(null)'
> FAILED: Execution Error, return code 1 from
> org.apache.hadoop.hive.ql.exec.ExecDriver
>
> Is there any additional configuration or security involved with getting hive
> to talk to my master and schedule jobs? Everything is running fine on the
> master. I can access the web interfaces for namenode, jobtracker, etc. fine
> from any other machine.
>
> Thanks,
> Vijay

Sorry it was my mistake. Information in hive.log was only marginally useful
but I figured out what the problem was. I did not have the right metastore
on the remote machine I'm running hive on.

2009-08-11 21:03:33,315 ERROR exec.ExecDriver
(SessionState.java:printError(279)) - Job Submission failed with exception
'java.lang.NullPointerException(null)'
java.lang.NullPointerException
        at
org.apache.hadoop.hive.ql.exec.ExecDriver.addInputPath(ExecDriver.java:701)
        at
org.apache.hadoop.hive.ql.exec.ExecDriver.addInputPaths(ExecDriver.java:756)
        at
org.apache.hadoop.hive.ql.exec.ExecDriver.execute(ExecDriver.java:376)
        at org.apache.hadoop.hive.ql.Driver.execute(Driver.java:357)
On Tue, Aug 11, 2009 at 6:41 PM, Raghu Murthy <[EMAIL PROTECTED]> wrote:

> Anything in /tmp/<username>/hive.log?
>
>
> On 8/11/09 5:51 PM, "Vijay" <[EMAIL PROTECTED]> wrote:
>
> > I have a simple hadoop cluster setup with a couple of machines (one
> master
> > running everything and one additional slave). I have hive configured with
> the
> > right hive-site.xml file pointing the fs.default.name <
> http://fs.default.name>
> > and mapred.job.tracker correctly to the master. When I run hive from the
> > master machine, hive connects fine and can schedule jobs fine. However,
> when I
> > run hive from a different machine, it cannot schedule jobs. It simply
> fails
> > with this error:
> >
> > Job Submission failed with exception
> 'java.lang.NullPointerException(null)'
> > FAILED: Execution Error, return code 1 from
> > org.apache.hadoop.hive.ql.exec.ExecDriver
> >
> > Is there any additional configuration or security involved with getting
> hive
> > to talk to my master and schedule jobs? Everything is running fine on the
> > master. I can access the web interfaces for namenode, jobtracker, etc.
> fine
> > from any other machine.
> >
> > Thanks,
> > Vijay
>
>