hadoop-mapreduce-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Amar Kamat <ama...@yahoo-inc.com>
Subject Re: Is there a way to see what file killed a mapper?
Date Tue, 10 May 2011 16:03:23 GMT
MapReduce updates the task's configuration and sets 'map.input.file' to point to the file on
which the task intends to work on. In the new MapReduce API, its renamed to 'mapreduce.map.input.file'.
You can print the value corresponding to 'map.input.file'. Similarly 'map.input.start'  point
to the start offset in the input file while 'map.input.length' points to the total size of
the data to be read.

On 5/10/11 9:05 PM, "Jonathan Coveney" <jcoveney@gmail.com> wrote:

I have a basic job that is dying, I think, on one badly compressed file. Is there a way to
see what file it is choking on? Via the job tracker I can find the mapper that is dying but
I cannot find a record of the file that it died on.

Thank you for your help

View raw message