hadoop-從單節點偽分散式擴充套件為多節點分散式
一、準備虛擬機器
複製第一臺虛擬機器，一共需要三個虛擬機器，配置好網路和主機名如下：
192.168.1.130 hadoop-server-00 192.168.1.131 hadoop-server-01 192.168.1.132 hadoop-server-02
保證3個虛擬機器之間可以相互ping通，如果網路不是很熟悉，可以啟動圖形化操作，在裡面設定好網路。
再配置00可以無密登入01和02，參考上一節即可。
二、設定01和02的hadoop
01和02如果是複製的第一臺虛擬機器就不需要重新配置了，否則需要拷貝00上的hadoop到其它兩個虛擬機器上
scp -r /usr/local/apps/hadoop-2.4.1/ hadoop-server-01:/usr/local/apps/ scp -r /usr/local/apps/hadoop-2.4.1/ hadoop-server-02:/usr/local/apps/
如果是拷貝過去的，則需要刪除01和02的hadoop目錄下的tmp目錄，因為tmp目錄中的資料是00生成的，標識身份是00的，在01和02上分別執行：rm -rf tmp/，因為01和02是從節點，還要確保它倆的slave檔案中沒有內容
設定主節點00上的slaves（內容包括自己，否則啟動後主節點沒有datanode）：vim /usr/local/apps/hadoop-2.4.1/etc/hadoop/slaves
hadoop-server-00
hadoop-server-01hadoop-server-02
這樣在啟動主節點的時候，它就會自動啟動01和02兩個從節點。為了方便操作，把hadoop的路徑加入環境變數中：vim /etc/profile
export HADOOP_HOME=/usr/local/apps/hadoop-2.4.1 export PATH=$PATH:$HADOOP_HOME/bin:$HADOOP_HOME/sbin
使配置生效：source /etc/profile
啟動：start-dfs.sh
三、警告處理
Java HotSpot(TM) 64-Bit Server VM warning: You have loaded library /usr/local/hadoop-2.2.0/lib/native/libhadoop.so.1.0.0 which might have disabled stack guard. The VM will try to fix the stack guard now
參考：https://blog.csdn.net/zxae86/article/details/45922821 解決
WARN util.NativeCodeLoader: Unable to load native-hadoop library for your platform... using builtin-java classes where applicable
參考：https://blog.csdn.net/zxae86/article/details/45922821 沒有解決，因為給出的配置路徑是錯誤的（export HADOOP_OPTS="-Djava.library.path=${HADOOP_HOME}/lib/"）後面少了native，後來又參考了一篇才解決：
參考：http://blog.sina.com.cn/s/blog_3d9e90ad0102wqrp.html 解決
解決方案中要下載：hadoop-native-64-2.4.0.tar 檔案，地址為：（http://dl.bintray.com/sequenceiq/sequenceiq-bin/hadoop-native-64-2.4.1.tar），備用地址：https://pan.baidu.com/s/1dxEkGod73wug1yqYpx2MGg 提取碼：dhqg
四、啟動
[[email protected]00 hadoop]# start-dfs.sh Starting namenodes on [hadoop-server-00] hadoop-server-00: starting namenode, logging to /usr/local/apps/hadoop-2.4.1/logs/hadoop-root-namenode-hadoop-server-00.out hadoop-server-00: starting datanode, logging to /usr/local/apps/hadoop-2.4.1/logs/hadoop-root-datanode-hadoop-server-00.out hadoop-server-01: starting datanode, logging to /usr/local/apps/hadoop-2.4.1/logs/hadoop-root-datanode-hadoop-server-01.out hadoop-server-02: starting datanode, logging to /usr/local/apps/hadoop-2.4.1/logs/hadoop-root-datanode-hadoop-server-02.out Starting secondary namenodes [0.0.0.0] 0.0.0.0: starting secondarynamenode, logging to /usr/local/apps/hadoop-2.4.1/logs/hadoop-root-secondarynamenode-hadoop-server-00.out [[email protected]-server-00 hadoop]# jps 7041 SecondaryNameNode 7138 Jps 6841 DataNode 6748 NameNode [[email protected]-server-00 hadoop]# start-yarn.sh starting yarn daemons starting resourcemanager, logging to /usr/local/apps/hadoop-2.4.1/logs/yarn-root-resourcemanager-hadoop-server-00.out hadoop-server-02: starting nodemanager, logging to /usr/local/apps/hadoop-2.4.1/logs/yarn-root-nodemanager-hadoop-server-02.out hadoop-server-01: starting nodemanager, logging to /usr/local/apps/hadoop-2.4.1/logs/yarn-root-nodemanager-hadoop-server-01.out hadoop-server-00: starting nodemanager, logging to /usr/local/apps/hadoop-2.4.1/logs/yarn-root-nodemanager-hadoop-server-00.out [[email protected]-server-00 hadoop]# jps 7041 SecondaryNameNode 7346 NodeManager 6841 DataNode 7610 Jps 6748 NameNode 7245 ResourceManager
可以看到5個程序都成功啟動，並且之前的警告都沒了。