HBase installation

1. Environment variable configuration

Upload the installation package to CentOS, unzip the installation package to the hadoop directory

tar -zxvf hbase-1.6.0-bin.tar.gz -C /usr/hadoop

Configure environment variables

vi /etc/profile

Add the following code at the end, save and exit

export HBASE_HOME=/usr/hadoop/hbase-1.6.0
export PATH=$HBASE_HOME/bin:$PATH

Effective configuration

source /etc/profile

verification

hbase version

2. Configure HBase

Enter the hbase conf folder, edit the hbase-env.sh file, modify the Java path, and remove the comment

export JAVA_HOME=/usr/java/jdk1.8.0_241

Use the zookeeper that comes with hbase

export HBASE_MANAGES_ZK=false

Edit the hbase-site.xml file

<configuration>

    <!-- 将HBase数据保存在HDFS目录中 -->
    <property>
        <name>hbase.rootdir</name>
        <value>hdfs://master:8020/hbase</value>
    </property>

    <!-- HBase是否是分布式环境 -->
    <property>
        <name>hbase.cluster.distributed</name>
        <value>true</value>
    </property>

    <!-- 配置zookeeper地址，节点全部启用zookeeper，个数必须是奇数 -->
    <property>
        <name>hbase.zookeeper.quorum</name>
        <value>master,slave1,slave2</value>
    </property>

    <!-- zookeeper数据目录  -->
    <property>
        <name>hbase.zookeeper.property.dataDir</name>
        <value>/usr/hadoop/zookeeper-3.5.7/data</value>
    </property>

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

Modify the file regionservers and add the host name of the slave node

slave1
slave2
slave3

3. Synchronize HBase configuration files

Synchronize the HBase configuration file of the master node to slave1, slave2, and slave3

scp -r /usr/hadoop/hbase-1.6.0 slave1:/usr/hadoop 
scp -r /usr/hadoop/hbase-1.6.0 slave2:/usr/hadoop 
scp -r /usr/hadoop/hbase-1.6.0 slave3:/usr/hadoop

Configure the environment variables of slave1, slave2, and slave3 respectively

scp /etc/profile slave1:/etc/
scp /etc/profile slave2:/etc/
scp /etc/profile slave3:/etc/

Effective configuration on slave1, slave2, and slave3 respectively

source /etc/profile

4. Start HBase

Start zookeeper, start hdfs, start yarn

Run HBase start command

start-hbase.sh 

[root@master bin]# jps

9760 ResourceManager
9447 NameNode
13208 HMaster
13337 HRegionServer
12202 HQuorumPeer
9612 SecondaryNameNode
14414 Jps

Enter the HBase web management page

http://master:16011/

TIPS：

1. If the following situations occur, the web management page cannot be opened

pids/hbase-root-master.pid: 没有那个文件或目录

Determine if hdfs and hbase configuration files are the same

The value under hbase.rootdir under hbase-site.xml must be the same as the value under fs.defaultFS under the hadoop configuration file core-site.xml, ip and port!

Modify hbase's pid file save path

Open hbase-env.sh in the conf directory and find the following code

# export HBASE_PID_DIR=/var/hadoop/pids

Modify to own path

export HBASE_PID_DIR=/usr/hadoop/hbase-1.6.0/pids

2. When HMaster fails to start, add the following content to hbase-site.xml:

<property>
	<name>hbase.unsafe.stream.capability.enforce</name>
	<value>false</value>
    <description>
        Controls whether HBase will check for stream capabilities (hflush/hsync).
        Disable this if you intend to run on LocalFileSystem.
        WARNING: Doing so may expose you to additional risk of data loss!
    </description>
</property>

Source:

https://stackoverflow.com/questions/48709569/hbase-error-illegalstateexception-when-starting-master-hsync

Official explanation:

https://github.com/apache/hbase/blob/master/hbase-procedure/src/test/resources/hbase-site.xml

3. When it shows that the port is occupied, check the occupied first, kill the port, and restart

netstat -nultp
kill -9 端口号

5. Enter HBase Shell

hbase shell

Big data platform-HBase installation and configuration