[
https://issues.apache.org/jira/browse/HIVE-3376?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13433325#comment-13433325
]
Gang Tim Liu commented on HIVE-3376:
------------------------------------
While executing in local mode, the process to collect failed job information is
a different process from running local MR job. Each process reads a different
log4j property file. One points to hive.log and another is MR job log.
No communication on 2 processes yet on job log. There are some ways like
external files etc.
Consider the bug impact itself (log file location), easy way is to pass MR job
stream and pick up log file location and output in CLI stream.
> Local mode errors point to the wrong file
> -----------------------------------------
>
> Key: HIVE-3376
> URL: https://issues.apache.org/jira/browse/HIVE-3376
> Project: Hive
> Issue Type: Bug
> Reporter: Gang Tim Liu
> Assignee: Gang Tim Liu
> Attachments: hive-3371
>
>
> This is a classic example of a query that fails due to a bad udf in local
> mode:
> Ended Job = job_local_0001 with errors
> Error during job, obtaining debugging information...
> Execution failed with exit status: 2
> Obtaining error information
> Task failed!
> Task ID:
> Stage-1
> Logs:
> /tmp/gang/hive.log
> FAILED: Execution Error, return code 2 from
> org.apache.hadoop.hive.ql.exec.MapRedTask
> But actually local mr job in
> /tmp/gang/gang_20120808181010_b716b5fb-5b8f-49d2-94e9-95bb8f465342.log
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators:
https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
