默认情况下是没有启动的，需要配置完后手工启动服务.

1. 修改mapred-site.xml，添加如下内容(cluster mode, start it on each slave node)

<property>
    <name>mapreduce.jobhistory.address</name>
    <value>slave3:10020</value>
</property>
<property>
     <name>mapreduce.jobhistory.webapp.address</name>
    <value>slave3:19888</value>
</property>
<property>
    <name>mapreduce.jobhistory.joblist.cache.size</name>
    <value>1000</value>
    <description>default 20000</description>
</property>
<property>
    <name>mapreduce.jobhistory.done-dir</name>
    <value>/history/done</value>
</property>

<property>
    <name>mapreduce.jobhistory.intermediate-done-dir</name>
    <value>/history/done_intermediate</value>
</property>

使用mr-jobhistory-daemon.sh start historyserver启动（cluster mode, start it on each slave node）

hadoop job history server