Re: Installing mahout on a cluster of Linux VM's Examples not compiling

[Sean Owen <sr...@gmail.com>]

Nope, you never have to install client code on a Hadoop node. Hadoop sends
all the code that is needed to the workers.
However, you have to give it all the code it needs to run. I am guess you
are not using the "job" files, but rather just one of the Mahout jar files.
That alone doesn't have what you need. Try the core "job" file.

On Fri, Jun 10, 2011 at 4:47 PM, Walshe, Maurice (RBI-UK) <
Maurice.Walshe@rbi.co.uk> wrote:

> Hi All
>
> Been banging my head against getting mahout 0.50 installed on my test
> cluster the examples just wont compile and when I try the precompiled
> kmeans the map completes but the reducer fails with a java exception.
>
>  java.lang.ClassNotFoundException: org.apache.mahout.math.Vector
>
> ???
>
> I Just thought do you have to install mahout on all the task tracker
> nodes ? the instructions dont explicitly say so but the writer may have
> assumed that every one knew this.
>
> Rgds Maurice
>
> Ps Hosts are windows xp and using vmware player with ubuntu 10.10 as
> guest operating system
>
>
> ===================== DISCLAIMER ======================
> This message is intended only for the use of the person(s)
> ("Intended Recipient") to whom it is addressed. It may contain
> information which is privileged and confidential. Accordingly
> any dissemination, distribution, copying or other use of this
> message or any of its content by any person other than the Intended
> Recipient may constitute a breach of civil or criminal law and is
> strictly prohibited. If you are not the Intended Recipient, please
> contact the sender as soon as possible.
>
> Reed Business Information Limited.
> Registered Office: Quadrant House, The Quadrant, Sutton, Surrey, SM2 5AS,
> UK.
> Registered in England under Company No. 151537
>
> =======================================================
>