在hadoop3.3.0集群使用hadoop jar运行自带wordcount报错

本人用hadoop3.3.0、ZK搭建的HA分布式集群，运行hadoop jar hadoop-mapreduce-examples-3.3.0.jar pi 5 5时出现如下错误：

Error: Could not find or load main class org.apache.hadoop.mapreduce.v2.app.MRAppMaster

Please check whether your <HADOOP_HOME>/etc/hadoop/mapred-site.xml contains the below configuration:
<property>
  <name>yarn.app.mapreduce.am.env</name>
  <value>HADOOP_MAPRED_HOME=${full path of your hadoop distribution directory}</value>
</property>
<property>
  <name>mapreduce.map.env</name>
  <value>HADOOP_MAPRED_HOME=${full path of your hadoop distribution directory}</value>
</property>
<property>
  <name>mapreduce.reduce.env</name>
  <value>HADOOP_MAPRED_HOME=${full path of your hadoop distribution directory}</value>
</property>

添加如下配置解决了问题：（/csnt/bigdata/hadoop-3.3.0是你需要替换的）

关闭yarn

stop-yarn.sh

vi $HADOOP_HOME/etc/hadoop/yarn-site.xml

<property>
	<name>yarn.application.classpath</name>
	<value>
		/csnt/bigdata/hadoop-3.3.0/etc/*,
		/csnt/bigdata/hadoop-3.3.0/etc/hadoop/*,
		/csnt/bigdata/hadoop-3.3.0/lib/*,
		/csnt/bigdata/hadoop-3.3.0/share/hadoop/common/*,
		/csnt/bigdata/hadoop-3.3.0/share/hadoop/common/lib/*,
		/csnt/bigdata/hadoop-3.3.0/share/hadoop/mapreduce/*,
		/csnt/bigdata/hadoop-3.3.0/share/hadoop/mapreduce/lib-examples/*,
		/csnt/bigdata/hadoop-3.3.0/share/hadoop/hdfs/*,
		/csnt/bigdata/hadoop-3.3.0/share/hadoop/hdfs/lib/*,
		/csnt/bigdata/hadoop-3.3.0/share/hadoop/yarn/*,
		/csnt/bigdata/hadoop-3.3.0/share/hadoop/yarn/lib/*,
	</value>
</property>

vi $HADOOP_HOME/etc/hadoop/mapred-site.xml

<property>
	<name>mapreduce.application.classpath</name>
	<value>
		/csnt/bigdata/hadoop-3.3.0/etc/*,
		/csnt/bigdata/hadoop-3.3.0/etc/hadoop/*,
		/csnt/bigdata/hadoop-3.3.0/lib/*,
		/csnt/bigdata/hadoop-3.3.0/share/hadoop/common/*,
		/csnt/bigdata/hadoop-3.3.0/share/hadoop/common/lib/*,
		/csnt/bigdata/hadoop-3.3.0/share/hadoop/mapreduce/*,
		/csnt/bigdata/hadoop-3.3.0/share/hadoop/mapreduce/lib-examples/*,
		/csnt/bigdata/hadoop-3.3.0/share/hadoop/hdfs/*,
		/csnt/bigdata/hadoop-3.3.0/share/hadoop/hdfs/lib/*,
		/csnt/bigdata/hadoop-3.3.0/share/hadoop/yarn/*,
		/csnt/bigdata/hadoop-3.3.0/share/hadoop/yarn/lib/*,
	</value>
</property>

记得将这两个修改的文件分发到其他的机器

cd $HADOOP_HOME/etc/hadoop

scp yarn-site.sh mapred-site.sh hadoop02:`pwd`

scp yarn-site.sh mapred-site.sh hadoop03:`pwd`

 

启动yarn

start-yarn.sh

再次执行

hadoop jar hadoop-mapreduce-examples-3.3.0.jar pi 5 5

以上。

本文是本人亲身经历，并且寻求解决方案后所记录，如果您有同样的问题，可以尝试使用此方案解决。如果这篇文章帮助到了您，是我的荣幸。