hadoop-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Harsh J <ha...@cloudera.com>
Subject Re: pipes(pydoop) and hbase classpath
Date Tue, 14 Aug 2012 21:17:26 GMT
Hi,

Per:

> org.apache.hadoop.hbase.mapreduce.TableInputFormat not
org.apache.hadoop.mapred.InputFormat

Pydoop seems to be expecting you to pass it an old API class for
InputFormat/etc. but you've passed in the newer class. I am unsure
what part of your code exactly may be at fault since I do not have
access to it, but you probably want to use the deprecated
org.apache.hadoop.hbase.mapred.* package classes such as
org.apache.hadoop.hbase.mapred.TableInputFormat, and not the
org.apache.hadoop.hbase.mapreduce.* classes, as you are using at the
moment.

HTH!

On Wed, Aug 15, 2012 at 2:39 AM, Håvard Wahl Kongsgård
<haavard.kongsgaard@gmail.com> wrote:
> Hi, I'am trying to read hbase key-values with pipes(pydoop). As hadoop
> is unable to find the hbase jar files. I get
>
> Exception in thread "main" java.lang.RuntimeException:
> java.lang.RuntimeException: class
> org.apache.hadoop.hbase.mapreduce.TableInputFormat not
> org.apache.hadoop.mapred.InputFormat
>
> have added export
> HADOOP_CLASSPATH=/usr/lib/hbase/hbase-0.90.6-cdh3u4.jar to my
> hadoop-env.sh
>
> According to the doc from cloudera,
> https://ccp.cloudera.com/display/CDHDOC/HBase+Installation#HBaseInstallation-UsingMapReducewithHBase
> TableMapReduceUtil.addDependencyJars(job); can be used as an
> alternative. But is that possible with pipes?
>
> -Håvard



-- 
Harsh J

Mime
View raw message