Mailing-List: contact dev-help@sqoop.apache.org; run by ezmlm
Precedence: bulk
Reply-To: dev@sqoop.apache.org
Date: Wed, 3 Dec 2014 20:42:12 +0000 (UTC)
From: "Veena Basavaraj (JIRA)" <jira@apache.org>
To: dev@sqoop.apache.org
Message-ID: <JIRA.12759278.1417639158000.71005.1417639332734@Atlassian.JIRA>
In-Reply-To: <JIRA.12759278.1417639158000@Atlassian.JIRA>
References: <JIRA.12759278.1417639158000@Atlassian.JIRA>
 <JIRA.12759278.1417639158096@arcas>
Subject: [jira] [Updated] (SQOOP-1839) Sqoop2: Mapreduce submission record
 should propogate failure info as well
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit


     [ https://issues.apache.org/jira/browse/SQOOP-1839?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Veena Basavaraj updated SQOOP-1839:
-----------------------------------
    Description: 
While investigating SQOOP-1815  and then trying to trace what the real reason for the job failure was, we figured that the ExceptionInfo and ExceptionTrace are not sufficient, since a job failure always does not mean it was due to an exception checked/ unchecked


{code}
  LOG.debug("Executed new map-reduce job with id " + jobId);
    } catch (Exception e) {
      request.getSummary().setException(e);
      LOG.error("Error in submitting job", e);
      return false;
    }
    return true;
  }
{code}
In case of a job failure we all need  failure info in the submission record

getFailureInfo()

https://hadoop.apache.org/docs/current/api/org/apache/hadoop/mapred/RunningJob.html#getFailureInfo()

  was:While investigating SQOOP-1815 we've spent a lot of time with [~vybs] trying to see why we are not getting data that we are expecting. We finally realized that the submission itself is failing and that is why we don't see any data. It seems that our [util classes|https://github.com/apache/sqoop/blob/sqoop2/test/src/main/java/org/apache/sqoop/test/testcases/ConnectorTestCase.java#L244] for executing job are running the job without checking whether it was successful or not. Hence the test didn't noticed that the job failed and move to the stage for checking job outputs that haven't been there.


> Sqoop2: Mapreduce submission record should propogate failure info as well
> -------------------------------------------------------------------------
>
>                 Key: SQOOP-1839
>                 URL: https://issues.apache.org/jira/browse/SQOOP-1839
>             Project: Sqoop
>          Issue Type: Bug
>    Affects Versions: 1.99.4
>            Reporter: Veena Basavaraj
>            Assignee: Jarek Jarcec Cecho
>             Fix For: 1.99.5
>
>
> While investigating SQOOP-1815  and then trying to trace what the real reason for the job failure was, we figured that the ExceptionInfo and ExceptionTrace are not sufficient, since a job failure always does not mean it was due to an exception checked/ unchecked
> {code}
>   LOG.debug("Executed new map-reduce job with id " + jobId);
>     } catch (Exception e) {
>       request.getSummary().setException(e);
>       LOG.error("Error in submitting job", e);
>       return false;
>     }
>     return true;
>   }
> {code}
> In case of a job failure we all need  failure info in the submission record
> getFailureInfo()
> https://hadoop.apache.org/docs/current/api/org/apache/hadoop/mapred/RunningJob.html#getFailureInfo()


--
This message was sent by Atlassian JIRA
(v6.3.4#6332)