HBase

1、 HBase安装部署

1.1、 Zookeeper正常部署

首先保证Zookeeper集群的正常部署，并启动之。

[shuidi@hadoop102 zookeeper-3.7.1]$ bin/zkServer.sh start
[shuidi@hadoop103 zookeeper-3.7.1]$ bin/zkServer.sh start
[shuidi@hadoop104 zookeeper-3.7.1]$ bin/zkServer.sh start

[shuidi@hadoop102 ~]$ jpsall
=============== hadoop102 ===============
26895 org.apache.zookeeper.server.quorum.QuorumPeerMain /opt/module/zookeeper/bin/../conf/zoo.cfg
=============== hadoop103 ===============
1262 org.apache.zookeeper.server.quorum.QuorumPeerMain /opt/module/zookeeper/bin/../conf/zoo.cfg
=============== hadoop104 ===============
28571 org.apache.zookeeper.server.quorum.QuorumPeerMain /opt/module/zookeeper/bin/../conf/zoo.cfg

1.2、 Hadoop正常部署

Hadoop集群的正常部署并启动。

[shuidi@hadoop102 hadoop-3.3.4]$ sbin/start-dfs.sh
[shuidi@hadoop103 hadoop-3.3.4]$ sbin/start-yarn.sh

[shuidi@hadoop102 ~]$ jpsall
=============== hadoop102 ===============
9926 org.apache.hadoop.hdfs.server.datanode.DataNode
11529 org.apache.hadoop.mapreduce.v2.hs.JobHistoryServer
11053 org.apache.hadoop.yarn.server.nodemanager.NodeManager
9518 org.apache.hadoop.hdfs.server.namenode.NameNode
=============== hadoop103 ===============
7280 org.apache.hadoop.hdfs.server.datanode.DataNode
8522 org.apache.hadoop.yarn.server.nodemanager.NodeManager
8079 org.apache.hadoop.yarn.server.resourcemanager.ResourceManager
=============== hadoop104 ===============
28768 org.apache.hadoop.hdfs.server.datanode.DataNode
28973 org.apache.hadoop.hdfs.server.namenode.SecondaryNameNode
29215 org.apache.hadoop.yarn.server.nodemanager.NodeManager

1.3、 HBase的解压

1）解压Hbase到指定目录

[shuidi@hadoop102 software]$ tar -zxvf hbase-2.4.11-bin.tar.gz -C /opt/module/
[shuidi@hadoop102 module]$ mv /opt/module/hbase-2.4.11 /opt/module/hbase

2）配置环境变量

[shuidi@hadoop102 module]$ sudo vim /etc/profile.d/my_env.sh

添加

#HBASE_HOME
export HBASE_HOME=/opt/module/hbase
export PATH=$PATH:$HBASE_HOME/bin

3）使用source让配置的环境变量生效

[shuidi@hadoop102 module]$ source /etc/profile.d/my_env.sh

1.4 、HBase的配置文件

1）hbase-env.sh修改内容，可以添加到最后：

[shuidi@hadoop102 conf]$ vim /opt/module/hbase/conf/hbase-env.sh 

export HBASE_MANAGES_ZK=false

2）hbase-site.xml修改内容：

[shuidi@hadoop102 conf]$ vim /opt/module/hbase/conf/hbase-site.xml 


<?xml version="1.0"?>
<?xml-stylesheet type="text/xsl" href="configuration.xsl"?>

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


  <property>
    <name>hbase.zookeeper.quorum</name>
    <value>hadoop102,hadoop103,hadoop104</value>
    <description>The directory shared by RegionServers.</description>
  </property>

  <property>
    <name>hbase.rootdir</name>
    <value>hdfs://hadoop102:8020/hbase</value>
    <description>The directory shared by RegionServers.</description>
  </property>


  

  <property>
    <name>hbase.wal.provider</name>
    <value>filesystem</value>
  </property>
</configuration>

3）regionservers

[shuidi@hadoop102 conf]$ vim /opt/module/hbase/conf/regionservers 

hadoop102
hadoop103
hadoop104

4）解决HBase和Hadoop的log4j兼容性问题，修改HBase的jar包，使用Hadoop的jar包

mv /opt/module/hbase/lib/client-facing-thirdparty/slf4j-reload4j-1.7.33.jar /opt/module/hbase/lib/client-facing-thirdparty/slf4j-reload4j-1.7.33.jar.bak

1.5、 HBase远程发送到其他集群

[shuidi@hadoop102 module]$ xsync hbase/

1.6、 HBase服务的启动

1）单点启动（我们不用）

[shuidi@hadoop102 hbase]$ bin/hbase-daemon.sh start master
[shuidi@hadoop102 hbase]$ bin/hbase-daemon.sh start regionserver

2）群启

[shuidi@hadoop102 hbase]$ bin/start-hbase.sh

running master, logging to /opt/module/hbase/logs/hbase-shuidi-master-hadoop102.out
hadoop104: running regionserver, logging to /opt/module/hbase/bin/../logs/hbase-shuidi-regionserver-hadoop104.out
hadoop102: running regionserver, logging to /opt/module/hbase/logs/hbase-shuidi-regionserver-hadoop102.out
hadoop103: running regionserver, logging to /opt/module/hbase/bin/../logs/hbase-shuidi-regionserver-hadoop103.out

3）对应的停止服务

[shuidi@hadoop102 hbase]$ bin/stop-hbase.sh

1.7、查看HBase页面

启动成功后，可以通过"host:port"的方式来访问HBase管理页面，例如：

http://hadoop102:16010

1.8、高可用（可选）

在HBase中HMaster负责监控HRegionServer的生命周期，均衡RegionServer的负载，如果HMaster挂掉了，那么整个HBase集群将陷入不健康的状态，并且此时的工作状态并不会维持太久。所以HBase支持对HMaster的高可用配置。

1）关闭HBase集群（如果没有开启则跳过此步）

[shuidi@hadoop102 hbase]$ bin/stop-hbase.sh
stopping hbase..........

2）在conf目录下创建backup-masters文件

[shuidi@hadoop102 hbase]$ touch conf/backup-masters

3）在backup-masters文件中配置高可用HMaster节点

[shuidi@hadoop102 hbase]$ echo hadoop103 > conf/backup-masters

4）将整个conf目录scp到其他节点

[shuidi@hadoop102 hbase]$ xsync conf

5）重启hbase,打开页面测试查看

http://hadooo102:16010