HDFS - Datanode failed volumes tolerate

The "dfs.datanode.failed.volumes.tolerated" in HDFS configuration file is the number of volumes that are allowed to fail before a DN stops offering service. By default any volume failure will cause a datanode to shutdown. If you have +3 disks, you can set this value to 1 or 2+.

For standard Hadoop installation, it is in /etc/hadoop/conf/hdfs-site.xml, if you installed CDH, you can change it in Clouder manager, "hdfs1" service -> DataNode(Default) -> "dfs.datanode.failed.volumes.tolerated".