Mailing-List: contact user-help@hadoop.apache.org; run by ezmlm
Precedence: bulk
Reply-To: user@hadoop.apache.org
Received-SPF: error (athena.apache.org: local policy)
Message-ID: <5175B657.600@plutoz.com>
Date: Mon, 22 Apr 2013 15:14:47 -0700
From: kaveh minooie <kaveh@plutoz.com>
User-Agent: Mozilla/5.0 (X11; Linux x86_64;
 rv:17.0) Gecko/20130402 Thunderbird/17.0.5
MIME-Version: 1.0
To: user@hadoop.apache.org
Subject: Re: common error in map tasks
References: <5174F5FD.5080702@plutoz.com> <1366621933912.b5661eda@Nodemailer>
 <5175A717.50106@plutoz.com>
 <CABCYYb_xHddRhFJseGE-s1odDqO5JAivsik60gJWXOyzEdrUTA@mail.gmail.com>
In-Reply-To: 
 <CABCYYb_xHddRhFJseGE-s1odDqO5JAivsik60gJWXOyzEdrUTA@mail.gmail.com>
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 8bit

thanks Chris. I only run nutch, so no to the external command. And I 
just checked and it happens or has happened on all the nodes at some 
point. I have to say thou that it doesn't cause the job to fail or 
anything. the map tasks that fail will finish when they are re-spawn 
again. it is just annoying and makes me think that some value some where 
in the config files are either not correct or not optimal.


On 04/22/2013 02:49 PM, Chris Nauroth wrote:
> I'm not aware of any Hadoop-specific meaning for exit code 126.
>   Typically, this is a standard Unix exit code used to indicate that a
> command couldn't be executed.  Some reasons for this might be that the
> command is not an executable file, or the command is an executable file
> but the user doesn't have execute permissions.  (See below for an
> example of each of these.)
>
> Does your job code attempt to exec an external command?  Also, are the
> task failures consistently happening on the same set of nodes in your
> cluster?  If so, then I recommend checking that the command has been
> deployed and has the correct permissions on those nodes.
>
> Even if your code doesn't exec an external command, various parts of the
> Hadoop code do this internally, so you still might have a case of a
> misconfigured node.
>
> Hope this helps,
> --Chris
>
> [chris@Chriss-MacBook-Pro:ttys000] hadoop-common
>  > ./BUILDING.txt
> -bash: ./BUILDING.txt: Permission denied
> [chris@Chriss-MacBook-Pro:ttys000] hadoop-common
>  > echo $?
> 126
>
> [chris@Chriss-MacBook-Pro:ttys000] test
>  > ls -lrt exec
> -rwx------  1 root  staff     0B Apr 22 14:43 exec*
> [chris@Chriss-MacBook-Pro:ttys000] test
>  > whoami
> chris
> [chris@Chriss-MacBook-Pro:ttys000] test
>  > ./exec
> bash: ./exec: Permission denied
> [chris@Chriss-MacBook-Pro:ttys000] test
>  > echo $?
> 126
>
>
>
> On Mon, Apr 22, 2013 at 2:09 PM, kaveh minooie <kaveh@plutoz.com
> <mailto:kaveh@plutoz.com>> wrote:
>
>     thanks. that is the issue, there is no other log files. when i go to
>     the attempt directory of that failed map task (e.g.
>     userlogs/job_201304191712___0015/attempt_201304191712___0015_m_000019_0
>     ) it is empty. there is no other log file. thou based on the counter
>     value, I can say that it happens right at the beginning of the map
>     task (counter is only 1 )
>
>
>
>
>     On 04/22/2013 02:12 AM, 姚吉龙 wrote:
>
>         Hi
>
>
>         I have the same problem before
>         I think this is caused by the lack of memory shortage for map task.
>         It is just a suggestion,you can post your log
>
>
>         BRs
>         Geelong
>         —
>         Sent from Mailbox <https://bit.ly/SZvoJe> for iPhone
>
>
>
>         On Mon, Apr 22, 2013 at 4:34 PM, kaveh minooie <kaveh@plutoz.com
>         <mailto:kaveh@plutoz.com>
>         <mailto:kaveh@plutoz.com <mailto:kaveh@plutoz.com>>> wrote:
>
>              HI
>
>              regardless of what job I run, there are always a few map
>         tasks that
>              fail with the following, very unhelpful, message: ( that is the
>              entire error message)
>
>              java.lang.Throwable: Child Error
>                  at
>         org.apache.hadoop.mapred.__TaskRunner.run(TaskRunner.__java:271)
>              Caused by: java.io.IOException: Task process exit with
>         nonzero status of 126.
>                  at
>         org.apache.hadoop.mapred.__TaskRunner.run(TaskRunner.__java:258)
>
>
>              I would appreciate it if someone could show me how I could
>         figure
>              out why this error keeps happening.
>
>              thanks,
>
>
>
>     --
>     Kaveh Minooie
>
>

-- 
Kaveh Minooie