Mailing-List: contact common-dev-help@hadoop.apache.org; run by ezmlm
Precedence: bulk
Reply-To: common-dev@hadoop.apache.org
Received-SPF: neutral (nike.apache.org: local policy)
Message-Id: <08FFFFF1-1FFB-460C-9FAB-C009E1CF4EF6@yahoo-inc.com>
From: Arun C Murthy <acm@yahoo-inc.com>
To: mapreduce-dev@hadoop.apache.org
In-Reply-To: 
 <DBB1DC060375D147AC43F310AD987DCC25E2B5AA3D@ESESSCMS0366.eemea.ericsson.se>
Content-Type: text/plain; charset=US-ASCII; format=flowed; delsp=yes
Content-Transfer-Encoding: 7bit
Subject: Re: Mapreduce program reports child error
Mime-Version: 1.0 (Apple Message framework v936)
Date: Tue, 3 May 2011 00:56:10 -0700
References: 
 <DBB1DC060375D147AC43F310AD987DCC25E2B5AA3D@ESESSCMS0366.eemea.ericsson.se>

Moving to mapreduce-dev@, please use the right list for questions.

On May 3, 2011, at 12:06 AM, Sudharsan Sampath wrote:

>
> Hi,
>
> Could anyone point me to a summary on why this error would occur?
>
> java.lang.Throwable: Child Error
> at org.apache.hadoop.mapred.TaskRunner.run(TaskRunner.java:471)
> Caused by: java.io.IOException: Task process exit with nonzero  
> status of 1.
> at org.apache.hadoop.mapred.TaskRunner.run(TaskRunner.java:458)
>
> I know of the following two reasons.
>
> 1) Insufficient space in the logs directory
> 2) ulimit threshold that causes insuffucient allocation of memory.
>

Some more (not a comprehensive list):
a) Task has OOM
b) TT, while monitoring the task, realizes it has used more memory  
than allowed and kills it.
c) Task can't write it's temp outputs to local disk (either map- 
outputs or intermediate merges during map-side sort or reduce-side  
shuffle).
...

Arun