关闭

Hadoop-HBase-Zookeeper安装记录

540人阅读 评论(0) 收藏 举报
一.jdk 安装及配置 (所有机器)
路径:/opt/soft/jdk
环境变量 vi /etc/profile  修改JAVA_HOME=/opt/soft/jdk,并添加PATH、CLASSPATH

二.hadoop 集群安装

HD1211:namenode, jobtracker, Hmaster

HD1212:secondarynamenode, datanode, tasktracker, HregionServer

DB1213-DB1217:datanode, tasktracker, HRegionServer, zookeeper

DB1213-DB1223:datanode, tasktracker, HRegionServer 

1.编辑/etc/hosts(所有机器)
IP           HD1211
IP           HD1212
IP           DB1213
IP           DB1214
IP           DB1215
IP           DB1216
IP           DB1217
IP           DB1218
IP           DB1219
IP           DB1220
IP           DB1221
IP           DB1222
IP           DB1223
scp到各台机器
1.每台机器添加集群用户,用户名hadoop
2.ssh 设置 
su - hadoop
ssh-keygen -t dsa -P '' -f ~/.ssh/id_dsa
cat ~/.ssh/id_dsa.pub >> ~/.ssh/authorized_keys

chmod 700 ~/.ssh
chmod 600 ~/.ssh/authorized_keys

scp ~/.ssh/id_dsa.pub hadoop@HD1212:~/master_key
每台机器
mkdir ~/.ssh
chmod 700 ~/.ssh
mv ~/master_key ~/.ssh/authorized_keys
chmod 600 ~/.ssh/authorized_keys
每台机器ssh一下
2.hadoop 安装
准备工作:
1.tar zxvf hadoop-0.20.203.0rc1.tar.gz
2. mv hadoop-0.20.203.0 hadoop 不要版本号,升级方便之用
3.chown -R hadoop:hadoop hadoop/ 
4.data1,data2,data3下分别创建dfs.name.dir,dfs.data.dir对应目录 (所有机器) 
dfs/name, dfs/data
chown -R hadoop:hadoop dfs/ 

5.创建work/lib目录,存放第三方lib
cd /
mkdir work/lib
chown -R hadoop:hadoop lib
目前lib下有:dsapapi.jar、dsapstorage.jar


配置profile (所有机器)
1.增加HADOOP_HOME,添加PATH
vi /etc/profile
JAVA_HOME=/opt/soft/jdk
CLASSPATH=$CLASSPATH:$JAVA_HOME/lib:$JAVA_HOME/jre/lib
PATH=$PATH:$JAVA_HOME/bin:/opt/dell/srvadmin/sbin:/opt/dell/srvadmin/bin:$JAVA_HOME/jre/bin
HADOOP_HOME=/opt/soft/hadoop
PATH=$PATH:$HADOOP_HOME/bin

export JAVA_HOME HADOOP_HOME CLASSPATH TOMCAT_HOME OMREPORT_BIN PATH
2.修改hadoop/core-site.xml
<configuration>

<property>
<name>fs.default.name</name>
<value>hdfs://HD1211:9000</value>
</property>
</configuration>
3.修改hadoop/hdfs-site.xml
<property>
<name>dfs.replication</name>
<value>3</value>
</property>


<property>
<name>dfs.data.dir</name>
<value>/data1/dfs/data,/data2/dfs/data,/data3/dfs/data</value>
</property>


<property>
<name>dfs.name.dir</name>
<value>/data1/dfs/name,/data2/dfs/name,/data3/dfs/name</value>
</property>


<property>
<name>dfs.permissions</name>
<value>false</value>
</property>


<property>
<name>dfs.block.size</name>
<value>134217728</value>
</property>
4.修改hadoop/mapred-site.xml
<property>
<name>mapred.job.tracker</name>
<value>HD1211:9001</value>
</property>


<property>
<name>mapred.tasktracker.map.tasks.maximum</name>
<value>8</value>
</property>


<property>
<name>mapred.tasktracker.reduce.tasks.maximum</name>
<value>6</value>
</property>


<property>
<name>io.sort.mb</name>
<value>200</value>
</property>


<property>
<name>mapred.child.java.opts</name>
<value>-Xmx1g</value>
</property>
5.修改hadoop/hadoop-env.sh
export JAVA_HOME=/opt/soft/jdk
安装完HBase后,需再增加
export HBASE_HOME=/opt/soft/hbase
export HADOOP_CLASSPATH=${HBASE_HOME}/hbase-0.90.3.jar:${HBASE_HOME}/hbase-0.90.3-test.jar:${HBASE_HOME}/conf:${HBASE_HOME}/lib/zookeeper-3.3.2.jar:${HBASE_HOME}/lib/guava-r06.jar
方可运行hbase任务
6.修改masters 增加secondary namenode
HD1212
7.修改slavers
HD1212
DB1213
DB1214
DB1215
DB1216
DB1217
DB1218
DB1219
DB1220
DB1221
DB1222
DB1223

8.hbase 安装完后,在hadoop/conf 下增加hbase-site.xml软链接
ln -s /opt/soft/hbase/conf/hbase-site.xml hbase-site.xml
chown -R hadoop:hadoop hbase-site.xml
问题:软链接scp到集群各机器后原文件一同被复制

9.修改hadoop 在168行,增加第三方jar包 classpath修改,指向准备工作创建的第三方jar包存放目录:/work/lib
# add third party libs to CLASSPATH
for f in /work/lib/*.jar; do
 CLASSPATH=${CLASSPATH}:$f;
done

scp 到个台机器,



三.Zookeeper(HD1211)
1.安装
1.HD1211
tar zxvf zookeeper-3.3.4.tar.gz
mv zookeeper-3.3.4 zookeeper 去版本号
chown -R hadoop:hadoop zookeeper

2.DB1213-DB1217 5台机器
cd /
mkdir zookeeperdata
chown -R hadoop:hadoop /zookeeperdata
2.配置
mv zoo_sample.cfg zoo.cfg
vi zoo.cfg
# The number of milliseconds of each tick
tickTime=2000
# The number of ticks that the initial
# synchronization phase can take
initLimit=10
# The number of ticks that can pass between
# sending a request and getting an acknowledgement
syncLimit=5
# the directory where the snapshot is stored.
dataDir=/opt/soft/zookeeperdata
# the port at which the clients will connect
clientPort=2181


server.1=DB1213:2888:3888
server.2=DB1214:2888:3888
server.3=DB1215:2888:3888
server.4=DB1216:2888:3888
server.5=DB1217:2888:3888


3.scp 到 各机器,修改zookeeperdata myid:每台机器zookeeperdata下创建myid,内容为对应的id:1、2、3、4、5



四.HBase安装
1.安装准备
tar zxvf hbase-0.90.3.tar.gz
mv hbase-0.90.3.tar.gz hbase 去版本号
chown -R hadoop:hadoop hbase

替换lib下jar包:hadoop-core-0.20.203.0.jar和commons-configuration-1.6.jar
cp /opt/soft/hadoop/hadoop-core-0.20.203.0.jar /opt/soft/hbase/lib/
rm -rf /opt/soft/hbase/lib/hadoop-core-0.20-append-r1056497.jar


cp /opt/soft/hadoop/lib/commons-configuration-1.6.jar lib/
2.配置
1.环境变量 增加HBASE_HOME,修改PATH
vi /etc/profile (所有机器)
HBASE_HOME=/opt/soft/hbase
PATH=$PATH:$HBASE_HOME/bin

export HBASE_HOME
2. 编辑hbase-env.sh
export JAVA_HOME=/opt/soft/jdk
export HBASE_MANAGES_ZK=false
export HADOOP_HOME=/opt/soft/hadoop

3. 编辑hbase-site.xml
<property>
<name>hbase.rootdir</name>
<value>hdfs://HD1211:9000/hbase</value>
</property>


<property>
<name>hbase.cluster.distributed</name>
<value>true</value>
</property>


<property>
<name>hbase.master</name>
<value>HD1211:60000</value>
</property>


<property>
<name>hbase.zookeeper.property.clientPort</name>
<value>2181</value>
</property>


<property>
<name>hbase.zookeeper.property.dataDir</name>
<value>/zookeeperdata</value>
</property>


<property>
<name>hbase.master.port</name>
<value>60000</value>
<description>The port master should bind to.</description>
</property>
 
<property>
<name>hbase.zookeeper.quorum</name>
<value>DB1213,DB1214,DB1215,DB1216,DB1217</value>
</property>
4.编辑 regionservers
HD1212
DB1213
DB1214
DB1215
DB1216
DB1217
DB1218
DB1219
DB1220
DB1221
DB1222
DB1223

3. SCP个台机器 




五.启动
启动顺序

1.hadoop. HD1211: /opt/soft/hadoop/bin/start-all.sh
2.zookeeper. DB1213-DB1217:依次/opt/soft/zookeeper/bin/zkServer.sh start
3.hbase. HD1211: /opt/soft/hbase/bin/start-hbase.sh

停止顺序
1.hbase
2.zookeeper
3.hadoop


Tips
HBase 单独启动regionserver
启动集群中所有的regionserver
./hbase-daemons.sh start regionserver
启动某个regionserver
./hbase-daemon.sh start regionserver
启动某个master
./hbase-daemon.sh start master







0
0

查看评论
* 以上用户言论只代表其个人观点,不代表CSDN网站的观点或立场
    个人资料
    • 访问:142616次
    • 积分:1792
    • 等级:
    • 排名:千里之外
    • 原创:24篇
    • 转载:109篇
    • 译文:0篇
    • 评论:13条
    文章分类
    最新评论