1. 根据前文Hadoop配置JobHistory之后，接着执行命令
cp /usr/local/spark/conf/spark-defaults.conf.template /usr/local/spark/conf/spark-defaults.conf
2. 修改spark-defaults.conf, 添加如下内容
spark.eventLog.enabled true spark.eventLog.compress true spark.eventLog.dir hdfs://mycluster/tmp/logs/root/logs spark.history.fs.logDirectory hdfs://mycluster/tmp/logs/root/logs spark.yarn.historyServer.address master1:18080
3.将文件同步至其他spark机器
4.重启spark集群 /usr/local/spark/sbin/stop-all.sh， /usr/local/spark/sbin/start-all.sh
5.访问http://master1:18080访问spark的historyserver
原文：https://www.cnblogs.com/jmbkeyes/p/14062992.html