Reply
Highlighted
Explorer
Posts: 6
Registered: ‎09-07-2017

Mapreduce Partitioner not working in oozie

[ Edited ]

Environment

============

 

Red Hat Enterprise Linux Server release 7.2 (Maipo)
Linux version 3.10.0-327.10.1.el7.x86_64 (mockbuild@x86-021.build.eng.bos.redhat.com) (gcc version 4.8.5 20150623 (Red Hat 4.8.5-4) (GCC) ) #1 SMP Sat Jan 23 04:54:55 EST 2016
oozie version - 4.1.0
cdh version - 5.10.1
Hue™ 3.11 - The Hadoop UI

 

Observation

===========

 

When mapreduce code is written using old API's (org.apache.hadoop.mapred.*) The below property when added in workflow.xml is ignored and no executed.

 

            <property>
                <name>mapred.partitioner.class</name>
                <value>PonRankPartitioner</value>
            </property>

 

The same is working when executed using hadoop jar command from CLI, I set the partitioner class in code using JobConf.setPartitionerClass API.

 

Executed the same logic using new API's (org.apache.hadoop.mapreduce.*) and added mapreduce.partitioner.class property in workflow.

 

Partitioner was executed and desired outcome was seen.

 

Is this a bug ??

Announcements