Can't run a MapReduce job on hadoop 2.4.0

问题

I am new to hadoop and here is my problem. I have configured hadoop 2.4.0 with jdk1.7.60 on cluster of 3 machine. I am able to execute all the commands of hadoop. Now I have modified wordcount example and created jar file. I have already executed with this jar file on hadoop 1.2.1 and got the result. But now on hadoop 2.4.0 I am not getting any result.

Command used for execution

$hadoop jar WordCount.jar WordCount /data/webdocs.dat /output

I am getting following message from the setup:

14/06/29 19:35:18 INFO client.RMProxy: Connecting to ResourceManager at /192.168.2.140:8040
14/06/29 19:35:18 WARN mapreduce.JobSubmitter: Hadoop command-line option parsing not performed. Implement the Tool interface and execute your application with ToolRunner to remedy this.
14/06/29 19:35:19 INFO input.FileInputFormat: Total input paths to process : 1
14/06/29 19:35:19 INFO mapreduce.JobSubmitter: number of splits:12
14/06/29 19:35:19 INFO mapreduce.JobSubmitter: Submitting tokens for job: job_1403905542893_0004
14/06/29 19:35:19 INFO impl.YarnClientImpl: Submitted application application_1403905542893_0004
14/06/29 19:35:19 INFO mapreduce.Job: The url to track the job: http://192.168.2.140:8088/proxy/application_1403905542893_0004/
14/06/29 19:35:19 INFO mapreduce.Job: Running job: job_1403905542893_0004

At this point no message change. I waited for 15 to 20 minutes but still the same.

This is what I see on resource manager's web page regarding the job:

State - ACCEPTED
FinalStatus - UNDEFINED
Progress - (progress bar in 0%)
Tracking UI - UNASSIGNED

Apps Submitted - 1
Apps Pending - 1
Apps Running - 0

I tried the other yarn command for execution but got the same result

$yarn jar WordCount.jar WordCount /data/webdocs.dat /output

Here is the output of jps:

21485 NameNode
23142 DataNode
28504 Jps
21704 ResourceManager
22082 JobHistoryServer

Any help or guidance will be highly appriciated.

回答1:

I solved the problem. It was the mistake in the configuration file of the hadoop. There was bind exception on the port 8040 for resourcemanager.

I changed the hadoop yarn-site.xml from (old yarn-site.xml):

<configuration>
<!-- Site specific YARN configuration properties -->
<property>
  <name>yarn.nodemanager.aux-services</name>
  <value>mapreduce_shuffle</value>
</property>
<property>
  <name>yarn.nodemanager.aux-services.mapreduce.shuffle.class</name>
  <value>org.apache.hadoop.mapred.ShuffleHandler</value>
</property>
<property>
  <name>yarn.resourcemanager.resource-tracker.address</name>
  <value>192.168.2.140:8025</value>
</property>
<property>
  <name>yarn.resourcemanager.scheduler.address</name>
  <value>192.168.2.140:8030</value>
</property>
<property>
  <name>yarn.resourcemanager.address</name>
  <value>192.168.2.140:8040</value>
</property>
</configuration>

To (new yarn-site.xml):

<configuration>
 <property>
  <name>yarn.nodemanager.aux-services</name>
  <value>mapreduce_shuffle</value>
 </property>
 <property>
  <name>yarn.nodemanager.aux-services.mapreduce.shuffle.class</name>
  <value>org.apache.hadoop.mapred.ShuffleHandler</value>
 </property>
</configuration>

I deleted the other line in hadoop configuration Then I gace following commands to start resourcemanager and nodemanager

$yarn-daemon.sh start nodemanager
$yarn-daemon.sh start resourcemanager

Then I tried executing my job and it was successfull.

来源：https://stackoverflow.com/questions/24481439/cant-run-a-mapreduce-job-on-hadoop-2-4-0

标签

java

Hadoop

MapReduce

yarn