hadoop版本apache:2.7.7->2.9.2
hadoop集群为非高可用集群
1.创建镜像回滚:hdfs dfsadmin -rollingUpgrade prepare
2.hdfs dfsadmin -rollingUpgrade query 查看镜像回滚状态，直到显示Proceed with rolling upgrade状态
3.下载并解压hadoop-2.9.2
4.将hdfs-site.xml、core-site.xml 、yarn-site.xml拷贝到hadoop-2.9.2的etc/hadoop目录下
5.停止hadoop2.7的snn、nn、datanode ,再hadoop2.7.7目录下执行：
sbin/sbin/hadoop-daemon.sh stop secondarynamenode
sbin/hadoop-daemon.sh stop namenode
sbin/hadoop-daemon.sh stop datanode
6.启动hadoop2.9.2的hadoop 切换目录到hadoop-2.9.2
修改环境变量HADOOP_HOME、HADOOP_CONF_DIR，并source /etc/profile或者重启session
执行sbin/hadoop-daemon.sh start namenode -rollingUpgrade started
7.启动hadoop2.9.2的snn，datanode
sbin/hadoop-daemon.sh start namenode
sbin/hadoop-daemon.sh start datanode
8.查看版本：http://ip:50070
版本升级到2.9.2
9.再hadoop-2.9.2目录下运行bin/hdfs dfsadmin -rollingUpgrade finalize
webui下的版本升级信息消除
