Apache Hadoop インストール2015/07/28

	Apache Hadoop をインストールします。ここでは例として、以下の 3ノードを利用して分散処理できるよう設定します。当例は 3ノードのみですが、Hadoop は大規模なデータを大規模なノード構成で処理する場合に、より真価を発揮します。 1) dlp.srv.world (マスターノード) 2) node01.srv.world (スレーブノード) 3) node02.srv.world (スレーブノード)
[1]	こちらを参考に全ノードに JAVA 実行環境をインストールしておきます。
[2]	全ノードに Hadoop 用のユーザーを作成しておきます。

[root@dlp ~]#

useradd -d /usr/hadoop hadoop

[root@dlp ~]#

chmod 755 /usr/hadoop

[root@dlp ~]#

passwd hadoop

Changing password for user hadoop.
New password:
Retype new password:
passwd: all authentication tokens updated successfully.

[3]	マスターノードに hadoop ユーザーでログインし、SSH 鍵ペア(ノーパスフレーズ)を作成して、各ノードに配布します。

[hadoop@dlp ~]$

ssh-keygen

Generating public/private rsa key pair.
Enter file in which to save the key (/usr/hadoop/.ssh/id_rsa):
Created directory '/usr/hadoop/.ssh'.
Enter passphrase (empty for no passphrase):
Enter same passphrase again:
Your identification has been saved in /usr/hadoop/.ssh/id_rsa.
Your public key has been saved in /usr/hadoop/.ssh/id_rsa.pub.
The key fingerprint is:
xx:xx:xx:xx:xx:xx:xx:xx:xx:xx:xx:xx:xx:xx:xx:xx hadoop@dlp.srv.world
The key's randomart image is:

# ローカルホスト含め各ノードに配布

[hadoop@dlp ~]$

ssh-copy-id localhost

The authenticity of host 'localhost (::1)' can't be established.
ECDSA key fingerprint is xx:xx:xx:xx:xx:xx:xx:xx:xx:xx:xx:xx:xx:xx:xx:xx.
Are you sure you want to continue connecting (yes/no)?

yes

/bin/ssh-copy-id: INFO: attempting to log in with the new key(s), to filter out any that are already installed
/bin/ssh-copy-id: INFO: 1 key(s) remain to be installed -- if you are prompted now it is to install the new keys
hadoop@localhost's password:

Number of key(s) added: 1

Now try logging into the machine, with:   "ssh 'localhost'"
and check to make sure that only the key(s) you wanted were added.

[hadoop@dlp ~]$

ssh-copy-id node01.srv.world

[hadoop@dlp ~]$

ssh-copy-id node02.srv.world

[4]	全ノードに Hadoop をインストールします。hadoop ユーザーで作業します。以下のサイトで最新版を確認してダウンロードしてください。 ⇒ https://hadoop.apache.org/releases.html

[hadoop@dlp ~]$

curl -O http://ftp.jaist.ac.jp/pub/apache/hadoop/common/hadoop-2.7.1/hadoop-2.7.1.tar.gz

[hadoop@dlp ~]$

tar zxvf hadoop-2.7.1.tar.gz -C /usr/hadoop --strip-components 1

[hadoop@dlp ~]$

vi ~/.bash_profile

# 最終行に追記

export HADOOP_HOME=/usr/hadoop
export HADOOP_COMMON_HOME=$HADOOP_HOME
export HADOOP_HDFS_HOME=$HADOOP_HOME
export HADOOP_MAPRED_HOME=$HADOOP_HOME
export HADOOP_YARN_HOME=$HADOOP_HOME
export HADOOP_OPTS="-Djava.library.path=$HADOOP_HOME/lib/native"
export HADOOP_COMMON_LIB_NATIVE_DIR=$HADOOP_HOME/lib/native
export PATH=$PATH:$HADOOP_HOME/sbin:$HADOOP_HOME/bin

[hadoop@dlp ~]$

source ~/.bash_profile

[5]	マスターノードで Hadoop の設定をします。hadoop ユーザーで作業します。

# 全ノードでデータ用ディレクトリ作成

[hadoop@dlp ~]$

mkdir ~/datanode

[hadoop@dlp ~]$

ssh node01.srv.world "mkdir ~/datanode"

[hadoop@dlp ~]$

ssh node02.srv.world "mkdir ~/datanode"

[hadoop@dlp ~]$

vi ~/etc/hadoop/hdfs-site.xml

# <configuration> ～ </configuration> 内に追記

<property>
<name>dfs.replication</name>
<value>2</value>
</property>
<property>
<name>dfs.datanode.data.dir</name>
<value>file:///usr/hadoop/datanode</value>
</property>

</configuration>

# スレーブノードに送信

[hadoop@dlp ~]$

scp ~/etc/hadoop/hdfs-site.xml node01.srv.world:~/etc/hadoop/

[hadoop@dlp ~]$

scp ~/etc/hadoop/hdfs-site.xml node02.srv.world:~/etc/hadoop/

[hadoop@dlp ~]$

vi ~/etc/hadoop/core-site.xml

# <configuration> ～ </configuration> 内に追記

<property>
<name>fs.defaultFS</name>
<value>hdfs://dlp.srv.world:9000/</value>
</property>

</configuration>

# スレーブノードに送信

[hadoop@dlp ~]$

scp ~/etc/hadoop/core-site.xml node01.srv.world:~/etc/hadoop/

[hadoop@dlp ~]$

scp ~/etc/hadoop/core-site.xml node02.srv.world:~/etc/hadoop/

[hadoop@dlp ~]$

sed -i -e 's/\${JAVA_HOME}/\/usr\/java\/default/' ~/etc/hadoop/hadoop-env.sh

# スレーブノードに送信

[hadoop@dlp ~]$

scp ~/etc/hadoop/hadoop-env.sh node01.srv.world:~/etc/hadoop/

[hadoop@dlp ~]$

scp ~/etc/hadoop/hadoop-env.sh node02.srv.world:~/etc/hadoop/

[hadoop@dlp ~]$

mkdir ~/namenode

[hadoop@dlp ~]$

vi ~/etc/hadoop/hdfs-site.xml

# <configuration> ～ </configuration> 内に追記

<property>
<name>dfs.namenode.name.dir</name>
<value>file:///usr/hadoop/namenode</value>
</property>

</configuration>

[hadoop@dlp ~]$

vi ~/etc/hadoop/mapred-site.xml

# 新規作成

<configuration>
<property>
<name>mapreduce.framework.name</name>
<value>yarn</value>
</property>
</configuration>

[hadoop@dlp ~]$

vi ~/etc/hadoop/yarn-site.xml

# <configuration> ～ </configuration> 内に追記

<property>
<name>yarn.resourcemanager.hostname</name>
<value>dlp.srv.world</value>
</property>
<property>
<name>yarn.nodemanager.hostname</name>
<value>dlp.srv.world</value>
</property>
<property>
<name>yarn.nodemanager.aux-services</name>
<value>mapreduce_shuffle</value>
</property>

</configuration>

[hadoop@dlp ~]$

vi ~/etc/hadoop/slaves

# 全ノード名を追記 (localhost は削除)

dlp.srv.world
node01.srv.world
node02.srv.world

[6]	ネームノードをフォーマットして Hadoop を起動します。

[hadoop@dlp ~]$

hdfs namenode -format

15/07/28 19:58:14 INFO namenode.NameNode: STARTUP_MSG:
/************************************************************
STARTUP_MSG: Starting NameNode
STARTUP_MSG:   host = dlp.srv.world/10.0.0.30
STARTUP_MSG:   args = [-format]
STARTUP_MSG:   version = 2.7.1
.....
.....
15/07/28 19:58:17 INFO namenode.NameNode: SHUTDOWN_MSG:
/************************************************************
SHUTDOWN_MSG: Shutting down NameNode at dlp.srv.world/10.0.0.30
************************************************************/

[hadoop@dlp ~]$

start-dfs.sh

Starting namenodes on [dlp.srv.world]
dlp.srv.world: starting namenode, logging to /usr/hadoop/logs/hadoop-hadoop-namenode-dlp.srv.world.out
dlp.srv.world: starting datanode, logging to /usr/hadoop/logs/hadoop-hadoop-datanode-dlp.srv.world.out
node02.srv.world: starting datanode, logging to /usr/hadoop/logs/hadoop-hadoop-datanode-node02.srv.world.out
node01.srv.world: starting datanode, logging to /usr/hadoop/logs/hadoop-hadoop-datanode-node01.srv.world.out
Starting secondary namenodes [0.0.0.0]
0.0.0.0: starting secondarynamenode, logging to /usr/hadoop/logs/hadoop-hadoop-secondarynamenode-dlp.srv.world.out

[hadoop@dlp ~]$

start-yarn.sh

starting yarn daemons
starting resourcemanager, logging to /usr/hadoop/logs/yarn-hadoop-resourcemanager-dlp.srv.world.out
dlp.srv.world: starting nodemanager, logging to /usr/hadoop/logs/yarn-hadoop-nodemanager-dlp.srv.world.out
node02.srv.world: starting nodemanager, logging to /usr/hadoop/logs/yarn-hadoop-nodemanager-node02.srv.world.out
node01.srv.world: starting nodemanager, logging to /usr/hadoop/logs/yarn-hadoop-nodemanager-node01.srv.world.out

# 確認 (以下のように起動していれば OK)

[hadoop@dlp ~]$

jps

2130 NameNode
2437 SecondaryNameNode
2598 ResourceManager
2710 NodeManager
3001 Jps
2267 DataNode

[7]	付属のサンプルプログラムを実行し、動作確認します。

# ディレクトリ作成

[hadoop@dlp ~]$

hdfs dfs -mkdir /test

# ローカルにある NOTICE.txt を /test にコピー

[hadoop@dlp ~]$

hdfs dfs -copyFromLocal ~/NOTICE.txt /test

# 内容を確認

[hadoop@dlp ~]$

hdfs dfs -cat /test/NOTICE.txt

This product includes software developed by The Apache Software
Foundation (http://www.apache.org/).

# サンプルプログラム実行

[hadoop@dlp ~]$

hadoop jar ~/share/hadoop/mapreduce/hadoop-mapreduce-examples-2.7.1.jar wordcount /test/NOTICE.txt /output01

15/07/28 19:28:47 INFO client.RMProxy: Connecting to ResourceManager at dlp.srv.world/10.0.0.30:8032
15/07/28 19:28:48 INFO input.FileInputFormat: Total input paths to process : 1
15/07/28 19:28:48 INFO mapreduce.JobSubmitter: number of splits:1
.....
.....

# 実行結果を確認

[hadoop@dlp ~]$

hdfs dfs -ls /output01

Found 2 items
-rw-r--r--   2 hadoop supergroup      0 2015-07-29 14:29 /output01/_SUCCESS
-rw-r--r--   2 hadoop supergroup    123 2015-07-29 14:29 /output01/part-r-00000

# 結果の内容を確認 (各ワードのカウント数が結果として出力されている)

[hadoop@dlp ~]$

hdfs dfs -cat /output01/part-r-00000

(http://www.apache.org/).       1
Apache          1
Foundation      1
Software        1
The             1
This            1
by              1
developed       1
includes        1
product         1
software        1

[8]	「http://(サーバーのホスト名またはIPアドレス):50070/」にアクセスすると、概要が確認できます。

[9]	「http://(サーバーのホスト名またはIPアドレス):8088/」にアクセスすると、クラスターの情報が確認できます。