啓動
docker run -it ubuntu:14.04
容器啓動起來了,接下來就是安裝Java、Hadoop及相關配置了。
換源 阿里Ubuntu14.04源
deb http://mirrors.aliyun.com/ubuntu/ trusty main restricted universe multiverse
deb http://mirrors.aliyun.com/ubuntu/ trusty-security main restricted universe multiverse
deb http://mirrors.aliyun.com/ubuntu/ trusty-updates main restricted universe multiverse
deb http://mirrors.aliyun.com/ubuntu/ trusty-proposed main restricted universe multiverse
deb http://mirrors.aliyun.com/ubuntu/ trusty-backports main restricted universe multiverse
deb-src http://mirrors.aliyun.com/ubuntu/ trusty main restricted universe multiverse
deb-src http://mirrors.aliyun.com/ubuntu/ trusty-security main restricted universe multiverse
deb-src http://mirrors.aliyun.com/ubuntu/ trusty-updates main restricted universe multiverse
deb-src http://mirrors.aliyun.com/ubuntu/ trusty-proposed main restricted universe multiverse
deb-src http://mirrors.aliyun.com/ubuntu/ trusty-backports main restricted universe multiverse
Java安裝
依次執行如下命令:
sudo apt-get -y install software-properties-common python-software-properties
sudo add-apt-repository ppa:webupd8team/java
sodu apt-get update
apt-get -y install oracle-java8-installer
注意:
- 這裏安裝的Java8。
- 默認使用的是Ubuntu的官方源,如果下載比較慢,請自行修改更新源,不知道如何使用命令行修改的,參考這裏。
另外,大家可以將裝好java的鏡像保存爲一個副本,他日可以在此基礎上構建其他鏡像。命令如下:
root@122a2cecdd14:~# exit
docker commit -m "java install" 122a2cecdd14 ubuntu:java
上面命令中122a2cecdd14
爲當前容器的ID, ubuntu:java
是爲新的鏡像指定一個標識,ubuntu
爲倉庫名,java
是Tag。
如何獲取容器ID:
- 有個簡便的辦法找到此ID,就是命令行用戶名
@
後面的那一串字符。這個方法只在容器啓動時沒有指定hostname時才能用。 - 使用
docker ps
列出所有運行的容器,在命令結果中查看
Hadoop安裝
漸漸切入正題了O(∩_∩)O~
使用剛剛已經安裝了Java的容器鏡像啓動:
docker run -ti ubuntu:java
啓動成功了,我們開始安裝Hadoop。這裏,我們直接使用wget下載安裝文件。
1.先安裝wget:
sudo apt-get -y install wget
2.下載並解壓安裝文件:
root@8ef06706f88d:cd /usr/local
root@8ef06706f88d:/usr/local# wget http://mirrors.sonic.net/apache/hadoop/common/hadoop-2.6.0/hadoop-2.6.0.tar.gz
root@8ef06706f88d:/usr/local# tar xvzf hadoop-2.6.0.tar.gz
root@8ef06706f88d:/usr/local# mv hadoop-2.6.0 hadoop
root@8ef06706f88d:/usr/local# rm hadoop-2.6.0.tar.gz
注意:這裏我們安裝的Hadoop版本是2.6.0,如果需要其他版本,請在這裏找到鏈接地址後修改命令即可。
3.配置環境變量
修改~/.bashrc
文件。在文件末尾加入下面配置信息:
export JAVA_HOME=/usr/lib/jvm/java-8-oracle
export HADOOP_HOME=/usr/local/hadoop
export HADOOP_CONFIG_HOME=$HADOOP_HOME/etc/hadoop
export PATH=$PATH:$HADOOP_HOME/bin
export PATH=$PATH:$HADOOP_HOME/sbin
注意:我們使用apt-get
安裝java,不知道java裝在什麼地方的話可以使用下面的命令查看:
root@8ef06706f88d:~# update-alternatives --config java
There is only one alternative in link group java (providing /usr/bin/java): /usr/lib/jvm/java-8-oracle/jre/bin/java
Nothing to configure.
root@8ef06706f88d:~#
4.配置Hadoop
下面,我們開始修改Hadoop的配置文件。主要配置core-site.xml、hdfs-site.xml、mapred-site.xml這三個文件。
開始配置之前,執行下面命令:
root@8ef06706f88d:# cd $HADOOP_HOME/
root@8ef06706f88d:/usr/local/haoop# mkdir tmp
root@8ef06706f88d:/usr/local/haoop# cd tmp/
root@8ef06706f88d:/usr/local/haoop/tmp# pwd
/usr/local/haoop/tmp
root@8ef06706f88d:/usr/local/haoop/tmp# cd ../
root@8ef06706f88d:/usr/local/haoop# mkdir namenode
root@8ef06706f88d:/usr/local/haoop# cd namenode/
root@8ef06706f88d:/usr/local/haoop/namenode# pwd
/usr/local/haoop/namenode
root@8ef06706f88d:/usr/local/haoop/namenode# cd ../
root@8ef06706f88d:/usr/local/haoop# mkdir datanode
root@8ef06706f88d:/usr/local/haoop# cd datanode/
root@8ef06706f88d:/usr/local/haoop/datanode# pwd
/usr/local/haoop/datanode
root@8ef06706f88d:/usr/local/haoop/datanode# cd $HADOOP_CONFIG_HOME/
root@8ef06706f88d:/usr/local/haoop/etc/hadoop# cp mapred-site.xml.template mapred-site.xml
root@8ef06706f88d:/usr/local/haoop/etc/hadoop# nano hdfs-site.xml
這裏創建了三個目錄,後續配置的時候會用到:
- tmp:作爲Hadoop的臨時目錄
- namenode:作爲NameNode的存放目錄
- datanode:作爲DataNode的存放目錄
1).core-site.xml配置
<?xml version="1.0" encoding="UTF-8"?>
<?xml-stylesheet type="text/xsl" href="configuration.xsl"?>
<!--
Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License. See accompanying LICENSE file.
-->
<!-- Put site-specific property overrides in this file. -->
<configuration>
<property>
<name>hadoop.tmp.dir</name>
<value>/usr/local/hadoop/tmp</value>
<description>A base for other temporary directories.</description>
</property>
<property>
<name>fs.default.name</name>
<value>hdfs://master:9000</value>
<final>true</final>
<description>The name of the default file system. A URI whose
scheme and authority determine the FileSystem implementation. The
uri's scheme determines the config property (fs.SCHEME.impl) naming
the FileSystem implementation class. The uri's authority is used to
determine the host, port, etc. for a filesystem.</description>
</property>
</configuration>
注意:
hadoop.tmp.dir
配置項值即爲此前命令中創建的臨時目錄路徑。fs.default.name
配置爲hdfs://master:9000
,指向的是一個Master節點的主機(後續我們做集羣配置的時候,自然會配置這個節點,先寫在這裏)
2).hdfs-site.xml配置
使用命令nano hdfs-site.xml
編輯hdfs-site.xml
文件:
<?xml version="1.0" encoding="UTF-8"?>
<?xml-stylesheet type="text/xsl" href="configuration.xsl"?>
<!--
Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License. See accompanying LICENSE file.
-->
<!-- Put site-specific property overrides in this file. -->
<configuration>
<property>
<name>dfs.replication</name>
<value>2</value>
<final>true</final>
<description>Default block replication.
The actual number of replications can be specified when the file is created.
The default is used if replication is not specified in create time.
</description>
</property>
<property>
<name>dfs.namenode.name.dir</name>
<value>/usr/local/hadoop/namenode</value>
<final>true</final>
</property>
<property>
<name>dfs.datanode.data.dir</name>
<value>/usr/local/hadoop/datanode</value>
<final>true</final>
</property>
</configuration>
注意:
- 我們後續搭建集羣環境時,將配置一個Master節點和兩個Slave節點。所以
dfs.replication
配置爲2。 dfs.namenode.name.dir
和dfs.datanode.data.dir
分別配置爲之前創建的NameNode和DataNode的目錄路徑
3).mapred-site.xml配置
Hadoop安裝文件中提供了一個mapred-site.xml.template,所以我們之前使用了命令cp mapred-site.xml.template mapred-site.xml
,創建了一個mapred-site.xml文件。下面使用命令nano mapred-site.xml
編輯這個文件:
<?xml version="1.0"?>
<?xml-stylesheet type="text/xsl" href="configuration.xsl"?>
<!--
Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License. See accompanying LICENSE file.
-->
<!-- Put site-specific property overrides in this file. -->
<configuration>
<property>
<name>mapred.job.tracker</name>
<value>master:9001</value>
<description>The host and port that the MapReduce job tracker runs
at. If "local", then jobs are run in-process as a single map
and reduce task.
</description>
</property>
</configuration>
這裏只有一個配置項mapred.job.tracker
,我們指向master節點機器。
4)修改JAVA_HOME環境變量
使用命令.nano hadoop-env.sh
,修改如下配置:
# The java implementation to use.
export JAVA_HOME=/usr/lib/jvm/java-8-oracle
5.格式化 namenode
這是很重要的一步,執行命令hadoop namenode -format
4.安裝SSH
搭建集羣環境,自然少不了使用SSH。這可以實現無密碼訪問,訪問集羣機器的時候很方便。
root@8ef06706f88d:~# sudo apt-get -y install ssh
root@8ef06706f88d:~# service ssh start
SSH裝好了以後,由於我們是Docker容器中運行,所以SSH服務不會自動啓動。需要我們在容器啓動以後,手動通過/usr/sbin/sshd
手動打開SSH服務。未免有些麻煩,爲了方便,我們把這個命令加入到~/.bashrc
文件中。通過nano ~/.bashrc
編輯.bashrc
文件(nano沒有安裝的自行安裝,也可用vi),在文件後追加下面內容:
#autorun
/usr/sbin/sshd
5.生成訪問密鑰
root@8ef06706f88d:/# cd ~/
root@8ef06706f88d:~# ssh-keygen -t dsa -P '' -f ~/.ssh/id_dsa
root@8ef06706f88d:~# cd .ssh
root@8ef06706f88d:~/.ssh# cat id_dsa.pub >> authorized_keys
注意: 這裏,我的思路是直接將密鑰生成後寫入鏡像,免得在買個容器裏面再單獨生成一次,還要相互拷貝公鑰,比較麻煩。當然這只是學習使用,實際操作時,應該不會這麼搞,因爲這樣所有容器的密鑰都是一樣的!!
6.保存鏡像副本
這裏我們將安裝好Hadoop的鏡像保存爲一個副本。
root@8ef06706f88d:~# exit
king@king:~$ docker commit -m "hadoop install" 8ef06706f88d ubuntu:hadoop
Hadoop分佈式集羣搭建
重點來了!
按照 hadoop 集羣的基本要求,其 中一個是 master 結點,主要是用於運行 hadoop 程序中的 namenode、secondorynamenode 和 jobtracker(新版本名字變了) 任務。用外兩個結點均爲 slave 結點,其中一個是用於冗餘目的,如果沒有冗 餘,就不能稱之爲 hadoop 了,所以模擬 hadoop 集羣至少要有 3 個結點。
前面已經將Hadoop的鏡像構建好了,下面就是使用這個鏡像搭建Master節點和Slave節點了:
節點 | hostname | ip | 用途 | Docker啓動腳本 |
Master | master | 10.0.0.5 |
namenode secondaryNamenode jobTracker |
docker run -ti -h master ubuntu:hadoop |
Slave | slave1 | 10.0.0.6 |
datanode taskTracker |
docker run -ti -h slave1 ubuntu:hadoop |
Slave | slave2 | 10.0.0.7 |
datanode taskTracker |
docker run -ti -h slave2 ubuntu:hadoop |
啓動Docker容器
回顧一下,Docker啓動容器使用的是run
命令:
docker run -ti ubuntu:hadoop
這裏有幾個問題:
- Docker容器中的ip地址是啓動之後自動分配的,且不能手動更改
- hostname、hosts配置在容器內修改了,只能在本次容器生命週期內有效。如果容器退出了,重新啓動,這兩個配置將被還原。且這兩個配置無法通過
commit
命令寫入鏡像
我們搭建集羣環境的時候,需要指定節點的hostname,以及配置hosts。hostname可以使用Docker run
命令的h
參數直接指定。但hosts解析有點麻煩,雖然可以使用run
的--link
參數配置hosts解析信息,但我們搭建集羣時要求兩臺機器互相能夠ping
通,其中一個容器沒有啓動,那麼ip不知道,所以--link
參數對於我們的這個場景不實用。要解決這個問題,大概需要專門搭建一個域名解析服務,即使用--dns
參數(參考這裏)。
我們這裏只爲學習,就不整那麼複雜了,就手動修改hosts吧。只不過每次都得改,我Docker知識淺薄,一時還沒有解決這個問題。相信肯定有更好的辦法。如果有高人能指定一下,感激不盡!!
啓動master容器
docker run -ti -h master ubuntu:hadoop
啓動slave1容器
docker run -ti -h slave1 ubuntu:hadoop
啓動slave2容器
docker run -ti -h slave2 ubuntu:hadoop
配置hosts
- 通過
ifconfig
命令獲取各節點ip。環境不同獲取的ip可能不一樣,例如我本機獲取的ip如下:- master:10.0.0.5
- slave1:10.0.0.6
- slave2:10.0.0.7
-
使用
sudo nano /etc/hosts
命令將如下配置寫入各節點的hosts文件,注意修改ip地址:10.0.0.5 master 10.0.0.6 slave1 10.0.0.7 slave2
配置slaves
下面我們來配置哪些節點是slave。在較老的Hadoop版本中有一個masters文件和一個slaves文件,但新版本中只有slaves文件了。
在master節點容器中執行如下命令:
root@master:~# cd $HADOOP_CONFIG_HOME/
root@master:~/soft/apache/hadoop/hadoop-2.6.0/etc/hadoop# nano slaves
將如下slave節點的hostname信息寫入該文件:
slave1
slave2
啓動Hadoop
在master節點上執行start-all.sh
命令,啓動Hadoop。
激動人心的一刻……
如果看到如下信息,則說明啓動成功了:
root@master:~/soft/apache/hadoop/hadoop-2.6.0/etc/hadoop# start-all.sh
This script is Deprecated. Instead use start-dfs.sh and start-yarn.sh
Starting namenodes on [master]
master: starting namenode, logging to /root/soft/apache/hadoop/hadoop-2.6.0/logs/hadoop-root-namenode-master.out
slave1: starting datanode, logging to /root/soft/apache/hadoop/hadoop-2.6.0/logs/hadoop-root-datanode-slave1.out
slave2: starting datanode, logging to /root/soft/apache/hadoop/hadoop-2.6.0/logs/hadoop-root-datanode-slave2.out
Starting secondary namenodes [0.0.0.0]
0.0.0.0: starting secondarynamenode, logging to /root/soft/apache/hadoop/hadoop-2.6.0/logs/hadoop-root-secondarynamenode-master.out
starting yarn daemons
starting resourcemanager, logging to /root/soft/apache/hadoop/hadoop-2.6.0/logs/yarn--resourcemanager-master.out
slave1: starting nodemanager, logging to /root/soft/apache/hadoop/hadoop-2.6.0/logs/yarn-root-nodemanager-slave1.out
slave2: starting nodemanager, logging to /root/soft/apache/hadoop/hadoop-2.6.0/logs/yarn-root-nodemanager-slave2.out
在個節點上執行jps
命令,結果如下:
master節點
root@master:~/soft/apache/hadoop/hadoop-2.6.0/etc/hadoop# jps
1223 Jps
992 SecondaryNameNode
813 NameNode
1140 ResourceManager
slave1節點
root@slave1:~/soft/apache/hadoop/hadoop-2.6.0/etc/hadoop# jps
258 NodeManager
352 Jps
159 DataNode
slave2節點
root@slave2:~/soft/apache/hadoop/hadoop-2.6.0/etc/hadoop# jps
371 Jps
277 NodeManager
178 DataNode
下面,我們在master節點上通過命令hdfs dfsadmin -report
查看DataNode是否正常啓動:
root@master:~/soft/apache/hadoop/hadoop-2.6.0/etc/hadoop# hdfs dfsadmin -report
Configured Capacity: 167782006784 (156.26 GB)
Present Capacity: 58979344384 (54.93 GB)
DFS Remaining: 58979295232 (54.93 GB)
DFS Used: 49152 (48 KB)
DFS Used%: 0.00%
Under replicated blocks: 0
Blocks with corrupt replicas: 0
Missing blocks: 0
-------------------------------------------------
Live datanodes (2):
Name: 10.0.0.7:50010 (slave2)
Hostname: slave2
Decommission Status : Normal
Configured Capacity: 83891003392 (78.13 GB)
DFS Used: 24576 (24 KB)
Non DFS Used: 54401331200 (50.67 GB)
DFS Remaining: 29489647616 (27.46 GB)
DFS Used%: 0.00%
DFS Remaining%: 35.15%
Configured Cache Capacity: 0 (0 B)
Cache Used: 0 (0 B)
Cache Remaining: 0 (0 B)
Cache Used%: 100.00%
Cache Remaining%: 0.00%
Xceivers: 1
Last contact: Sat Feb 28 07:27:05 UTC 2015
Name: 10.0.0.6:50010 (slave1)
Hostname: slave1
Decommission Status : Normal
Configured Capacity: 83891003392 (78.13 GB)
DFS Used: 24576 (24 KB)
Non DFS Used: 54401331200 (50.67 GB)
DFS Remaining: 29489647616 (27.46 GB)
DFS Used%: 0.00%
DFS Remaining%: 35.15%
Configured Cache Capacity: 0 (0 B)
Cache Used: 0 (0 B)
Cache Remaining: 0 (0 B)
Cache Used%: 100.00%
Cache Remaining%: 0.00%
Xceivers: 1
Last contact: Sat Feb 28 07:27:05 UTC 2015
還可以通過Web頁面看到查看DataNode和NameNode的狀態:http://10.0.0.5:50070/ (由於我宿主機器上沒有配置master的hosts解析,所以只能用ip地址訪問,大家將ip改爲各自的master節點容器的ip即可):