hadoop-mapreduce-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Vivek Krishna <vivekris...@gmail.com>
Subject Re: File Not found Bug. Hadoop Streaming
Date Fri, 03 Dec 2010 22:10:28 GMT
I am testing this in a pseudo-distributed mode.  I did try in actual
cluster, gave the same error.  In pseudo mode I assume it uses the default
python interpreter!

And yeah I read about the bug, that is the reason I tested with different
streaming jars. :(


On Fri, Dec 3, 2010 at 5:00 PM, Greg Roelofs <roelofs@yahoo-inc.com> wrote:

> Vivek Krishna wrote:
> > The Error:
> > *java.io.IOException: Cannot run program "mapper.py": error=2, No such
> file
> > or directory*
> > *
> That's the sort of error you get when the script interpreter doesn't exist.
> Have you verified the #! path and/or whether Python is even installed on
> the tasktracker (child/slave/whatever) nodes?  If you have the ability to
> log onto one of the nodes directly, try your command-line (non-Hadoop) test
> there and make sure it works OK.
> I also vaguely recall hearing about a bug in which executable perms were
> not restored on grid nodes in some circumstances, but I _thought_ that
> applied only to an internal y20 development version.  In any case, it was
> several months ago, which probably means it got fixed right away.
> Greg

  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message