hadoop安装

最新推荐文章于 2024-09-23 22:04:52 发布

转载最新推荐文章于 2024-09-23 22:04:52 发布 · 679 阅读

一、安装Hadoop

<1> 安装JDK1.6

chmod该文件以775的可执行权限，然后./filename.bin即可；

<2> 安装ssh，并且设置ssh无密码登录hadoop
<2.1>运行sudo apt-get install ssh/rsync，运行sudo apt-get install openjdk-6-jdk (for jps command)
<2.2>配置ssh本机免口令登录主要有两步：
运行ssh-keygen -t dsa -P '' -f /.ssh/id_dsa
-t用来指定加密算法，可以选择dsa和rsa两种加密方式；
-P用来指定密码，两个单引号表示空密码''；
-f用来指定存放密钥的文件
运行cat ~/.ssh/id_dsa.pub >> ~/.ssh/authorized_keys
这一步将公钥添加进本机的authorized_keys中，完成这两步后可以ssh localhost验证是否成功。接下来需要进入已经解好的hadoop-1.0.3中进行配置。hadoop的伪分布模式主要需要配置以下几个配置文件：
<3>conf/hbase-env.sh：主要用来配置hadoop的运行环境，这里需要修改JAVA_HOME到你的jdk1.6目录（见黑体）

点击(此处)折叠或打开

#Set Hadoop-specific environment variables here.

# The only required environment variable is JAVA_HOME. All others are

# optional. When running a distributed configuration it is best to

# set JAVA_HOME in this file, so that it is correctly defined on

# remote nodes.

# The java implementation to use. Required.

export JAVA_HOME=/home/hadoop/platform/jdk1.6.0_45

# Extra Java CLASSPATH elements. Optional.

# export HADOOP_CLASSPATH=

# The maximum amount of heap to use, in MB. Default is 1000.

# export HADOOP_HEAPSIZE=2000

# Extra Java runtime options. Empty by default.

# export HADOOP_OPTS=-server

# Command specific options appended to HADOOP_OPTS when specified

export HADOOP_NAMENODE_OPTS="-Dcom.sun.management.jmxremote $HADOOP_NAMENODE_OPTS"

export HADOOP_SECONDARYNAMENODE_OPTS="-Dcom.sun.management.jmxremote $HADOOP_SECONDARYNAMENODE_OPTS"

export HADOOP_DATANODE_OPTS="-Dcom.sun.management.jmxremote $HADOOP_DATANODE_OPTS"

export HADOOP_BALANCER_OPTS="-Dcom.sun.management.jmxremote $HADOOP_BALANCER_OPTS"

export HADOOP_JOBTRACKER_OPTS="-Dcom.sun.management.jmxremote $HADOOP_JOBTRACKER_OPTS"

# export HADOOP_TASKTRACKER_OPTS=

# The following applies to multiple commands (fs, dfs, fsck, distcp etc)

# export HADOOP_CLIENT_OPTS

# Extra ssh options. Empty by default.

# export HADOOP_SSH_OPTS="-o ConnectTimeout=1 -o SendEnv=HADOOP_CONF_DIR"

# Where log files are stored. $HADOOP_HOME/logs by default.

# export HADOOP_LOG_DIR=${HADOOP_HOME}/logs

# File naming remote slave hosts. $HADOOP_HOME/conf/slaves by default.

# export HADOOP_SLAVES=${HADOOP_HOME}/conf/slaves

# host:path where hadoop code should be rsync'd from. Unset by default.

# export HADOOP_MASTER=master:/home/$USER/src/hadoop

# Seconds to sleep between slave commands. Unset by default. This

# can be useful in large clusters, where, e.g., slave rsyncs can

# otherwise arrive faster than the master can service them.

# export HADOOP_SLAVE_SLEEP=0.1

# The directory where pid files are stored. /tmp by default.

# export HADOOP_PID_DIR=/var/hadoop/pids

# A string representing this instance of hadoop. $USER by default.

# export HADOOP_IDENT_STRING=$USER

# The scheduling priority for daemon processes. See 'man nice'.

# export HADOOP_NICENESS=10

<4>conf/core-site.xml
这里主要配置fs.default.name（用来指定namenode）和hadoop.tmp.dir（默认的hdfs的tmp目录位置），这里可以不设置hadoop.tmp.dir，那么就会保存在默认的/tmp下，每次重启机器都会丢失数据。

点击(此处)折叠或打开

<?xml version="1.0"?>

<?xml-stylesheet type="text/xsl" href="configuration.xsl"?>

<!-- Put site-specific property overrides in this file. -->

<configuration>

   <property>

       <name>fs.default.name</name>

       <value>hdfs://localhost:9000</value>

   </property>

   <property>

       <name>hadoop.tmp.dir</name>

       <value>/home/hadoop/hdfs/tmp</value>

   </property>

</configuration>

<5>hdfs-site.xml
这里的dfs.replication用来设置每份数据块的副本数目，默认是3，因为我们是在单机上配置的伪分布模式，因此设为1。dfs.name.dir和dfs.data.dir非常重要，用来设置存放hdfs中namenode和datanode数据的本地存放位置。这里如果设置不好，后续会出现多个错误。当然你也可以不设置采用默认的/tmp下的目录，但是同样重启会丢失数据。

点击(此处)折叠或打开

<?xml version="1.0"?>

<?xml-stylesheet type="text/xsl" href="configuration.xsl"?>

<!-- Put site-specific property overrides in this file. -->

<configuration>

   <property>

       <name>dfs.replication</name>

       <value>1</value>

   </property>

   <property>

       <name>dfs.name.dir</name>

       <value>/home/hadoop/hdfs/name</value>

   </property>

   <property>

       <name>dfs.data.dir</name>

       <value>/home/hadoop/hdfs/data</value>

   </property>

</configuration>

<6>mapred-site.xml

点击(此处)折叠或打开

<?xml version="1.0"?>

<?xml-stylesheet type="text/xsl" href="configuration.xsl"?>

<!-- Put site-specific property overrides in this file. -->

<configuration>

   <property>

       <name>mapred.job.tracker</name>

       <value>localhost:9001</value>

   </property>

</configuration>

然后就是运行测试了，把hadoop-1.0.3/bin加入到/etc/profile中的PATH路径中，方便我们执行Hadoop命令。运行start-all.sh后出现了问题，jps查看namenode无法启动，使用hadoop namenode -format也不能成功，查看日志：

提示我们存储的HDFS目录要么不存在要么没有权限：
FSNamesystem initialization failed.
org.apache.hadoop.hdfs.server.common.InconsistentFSStateException:Directory /home/hadoop/hdfs/name is in an inconsistent state: storage directory does not exist or is not accessible.
查看发现已经生成了hdfs目录，那么问题就是权限了，将/home下的hadoop目录权限由755设为775，然后重新运行hadoop，成功：

在进行安装hbase之前，我们先来按照官方的方法测试一下伪分布式的hadoop，看看安装是否成功：首先将conf下的所有文件拷贝到hdfs上的input目录中，然后运行jar文件将结果存储到hdfs中的output目录中，最后从output目录中查看结果：

点击(此处)折叠或打开

$ bin/hadoop fs -put conf input

$ bin/hadoop jar hadoop-examples-*.jar grep input output 'dfs[a-z.]+'

Copy the output files from the distributed filesystem to the local filesytem and examine them:

$ bin/hadoop fs -get output output 

$ cat output/*

or

View the output files on the distributed filesystem:

$ bin/hadoop fs -cat output/*

When you're done, stop the daemons with:

$ bin/stop-all.sh