hadoop-mapreduce-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Bhallamudi Venkata Siva Kamesh (JIRA)" <j...@apache.org>
Subject [jira] Commented: (MAPREDUCE-2341) Map and Reduce task JVMs are hanging infinitely
Date Mon, 21 Feb 2011 09:41:39 GMT

    [ https://issues.apache.org/jira/browse/MAPREDUCE-2341?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12997324#comment-12997324
] 

Bhallamudi Venkata Siva Kamesh commented on MAPREDUCE-2341:
-----------------------------------------------------------

In the code segment(void org.apache.hadoop.mapred.Child.main(String[] args) throws Throwable),
this case has been handled as shown below

 {code:title=Child.java|borderStyle=solid}
   while(true) {
    ...................................................
    numTasksToExecute = job.getNumTasksToExecutePerJvm();
    assert(numTasksToExecute != 0);
    ...................................................
   }
 {code} 



  As assertions have not been enabled during runtime, the assertion statement here failed
to assert and loop goes infinete.
  So I feel, we need to enable assertions while running HDFS/Mapreduce programs.

> Map and Reduce task JVMs are hanging infinitely 
> ------------------------------------------------
>
>                 Key: MAPREDUCE-2341
>                 URL: https://issues.apache.org/jira/browse/MAPREDUCE-2341
>             Project: Hadoop Map/Reduce
>          Issue Type: Bug
>          Components: tasktracker
>         Environment: Linux
>            Reporter: Bhallamudi Venkata Siva Kamesh
>
> Description:
>  When launching Mapreduce application, the property *mapred.job.reuse.jvm.num.tasks*
has been set to *0*, as a result the task JVM(MAP or Reduce JVM) is hanging indefinitely and
finally being killed by Tasktracker.
> Steps to Reproduce
> ------------------
> 1) Set the *mapred.job.reuse.jvm.num.tasks* property in mapred-site.xml to 0
> 2) Execute the wordcount example, child JVM is hanging indefinitely and finally killed
by Tasktracker

-- 
This message is automatically generated by JIRA.
-
For more information on JIRA, see: http://www.atlassian.com/software/jira

        

Mime
View raw message