You are viewing a plain text version of this content. The canonical link for it is here.
Posted to user@oozie.apache.org by Idris Ali <ps...@gmail.com> on 2011/12/19 17:10:37 UTC

Oozie launcher error: inconsistent

Hi,

I am running the pig job available in the oozie examples folder.

For the same workflow job, sometimes the oozie status after run is KILLED
and for other runs it is SUCCEEDED.

When I check the tasktracker LOGS for the KILLED cases, it says:

for attempt_201112192106_0005_m_000000_0 -
(oozie:launcher:T=pig:W=pig-wf:A=pig-node:ID=0000002-111219210706931-oozie-shai-W)

java.lang.Throwable: Child Error
	at org.apache.hadoop.mapred.TaskRunner.run(TaskRunner.java:242)
Caused by: java.io.IOException: Task process exit with nonzero status of 134.
	at org.apache.hadoop.mapred.TaskRunner.run(TaskRunner.java:229)

I do not see any useful logs in stderr.
However, for attempt_201112192106_0005_m_000000_1 - it is passed.
Also, the PIG job is always runs without error.

this is highly inconsistent, I am not sure WHY some times the same workflow
runs without error.

Environment:

Hadoop CDH3U0
Pig 8.0
oozie: 3.0.2
OS: Mac-
JVM - 1.6.0_29 64bit.

Thanks,
-Idris

Re: Oozie launcher error: inconsistent

Posted by Idris Ali <ps...@gmail.com>.
Hi Alejandro,

Thanks for the quick reply, I will try with debug option.

Meanwhile I reverted to JVM 1.6.0_26, and so far the jobs are running as
desired.
Just wondering if the JVM version under which oozie is running and JVM
version of Hadoop Mapred should be same.

Regards,
-Idris

On Mon, Dec 19, 2011 at 10:41 PM, Alejandro Abdelnur <tu...@cloudera.com>wrote:

> Idris,
>
> Would you set, in your Oozie pig action configuration. the following
> property: oozie.pig.log.level=DEBUG and try again? That should enable Pig
> logging in the Oozie launcher job task log.
>
> Thanks.
>
> Alejandro
>
> On Mon, Dec 19, 2011 at 8:10 AM, Idris Ali <ps...@gmail.com> wrote:
>
> > Hi,
> >
> > I am running the pig job available in the oozie examples folder.
> >
> > For the same workflow job, sometimes the oozie status after run is KILLED
> > and for other runs it is SUCCEEDED.
> >
> > When I check the tasktracker LOGS for the KILLED cases, it says:
> >
> > for attempt_201112192106_0005_m_000000_0 -
> >
> >
> (oozie:launcher:T=pig:W=pig-wf:A=pig-node:ID=0000002-111219210706931-oozie-shai-W)
> >
> > java.lang.Throwable: Child Error
> >        at org.apache.hadoop.mapred.TaskRunner.run(TaskRunner.java:242)
> > Caused by: java.io.IOException: Task process exit with nonzero status of
> > 134.
> >        at org.apache.hadoop.mapred.TaskRunner.run(TaskRunner.java:229)
> >
> > I do not see any useful logs in stderr.
> > However, for attempt_201112192106_0005_m_000000_1 - it is passed.
> > Also, the PIG job is always runs without error.
> >
> > this is highly inconsistent, I am not sure WHY some times the same
> workflow
> > runs without error.
> >
> > Environment:
> >
> > Hadoop CDH3U0
> > Pig 8.0
> > oozie: 3.0.2
> > OS: Mac-
> > JVM - 1.6.0_29 64bit.
> >
> > Thanks,
> > -Idris
> >
>

Re: Oozie launcher error: inconsistent

Posted by Alejandro Abdelnur <tu...@cloudera.com>.
Idris,

Would you set, in your Oozie pig action configuration. the following
property: oozie.pig.log.level=DEBUG and try again? That should enable Pig
logging in the Oozie launcher job task log.

Thanks.

Alejandro

On Mon, Dec 19, 2011 at 8:10 AM, Idris Ali <ps...@gmail.com> wrote:

> Hi,
>
> I am running the pig job available in the oozie examples folder.
>
> For the same workflow job, sometimes the oozie status after run is KILLED
> and for other runs it is SUCCEEDED.
>
> When I check the tasktracker LOGS for the KILLED cases, it says:
>
> for attempt_201112192106_0005_m_000000_0 -
>
> (oozie:launcher:T=pig:W=pig-wf:A=pig-node:ID=0000002-111219210706931-oozie-shai-W)
>
> java.lang.Throwable: Child Error
>        at org.apache.hadoop.mapred.TaskRunner.run(TaskRunner.java:242)
> Caused by: java.io.IOException: Task process exit with nonzero status of
> 134.
>        at org.apache.hadoop.mapred.TaskRunner.run(TaskRunner.java:229)
>
> I do not see any useful logs in stderr.
> However, for attempt_201112192106_0005_m_000000_1 - it is passed.
> Also, the PIG job is always runs without error.
>
> this is highly inconsistent, I am not sure WHY some times the same workflow
> runs without error.
>
> Environment:
>
> Hadoop CDH3U0
> Pig 8.0
> oozie: 3.0.2
> OS: Mac-
> JVM - 1.6.0_29 64bit.
>
> Thanks,
> -Idris
>