HBase集群安装

爱被打了一巴掌 2022-05-25 05:47 277阅读 0赞

前言：阅读本文之前，可以去apache hbase官网阅读安装配置文档，看不懂可以借助翻译，计算机所有的各种新技术只有官网是唯一且准确的第一手资料。  
[http://hbase.apache.org/book.html\#hbase\_default\_configurations][http_hbase.apache.org_book.html_hbase_default_configurations]  
一、Hbase简介  
Apache HBase 是一个高可靠性、高性能、面向列、可伸缩的分布式存储系统，是NoSQL数据库，基于Google Bigtable思想的开源实现，可在廉价的PC Server上搭建大规模结构化存储集群，利用Hadoop HDFS作为其文件存储系统，利用Hadoop MapReduce来处理HBase海量数据，使用Zookeeper协调服务器集群。  
二、环境准备  
笔者还是搭建hadoop、spark集群的那三个节点，要配置HBase Master、Master-backup、RegionServer各节点的进程规划如下：  
centos 6.5  
JDK-1.8.x  
Zookeeper-3.4.6  
Hadoop-2.7.3  
以上这些都在笔者的其他博客里有详细介绍，不再赘述。  
主机 IP 节点进程  
master HMaster  
worker1 Master-backup、RegionServer  
worker2 RegionServer  
三、安装Hbase-1.2.6  
1.可从官网或者国内阿里云等镜像站下载Hbase  
[http://mirror.bit.edu.cn/apache/hbase/stable/hbase-1.2.6-bin.tar.gz][http_mirror.bit.edu.cn_apache_hbase_stable_hbase-1.2.6-bin.tar.gz]  
2.解压至/app/hbase/目录下

[hadoop@master app]$ ls
    hadoop  hbase  hive  java  kafka  scala  spark  tgz  zookeeper

3.添加系统环境变量

# User specific environment and startup programs
    export JAVA_HOME=/app/java/jdk1.8.0_141
    export HADOOP_HOME=/app/hadoop/hadoop-2.7.3
    export SCALA_HOME=/app/scala/scala-2.11.8
    export SPARK_HOME=/app/spark/spark-2.1.1
    export ZOOKEEPER_HOME=/app/zookeeper/zookeeper-3.4.6
    export KAFKA_HOME=/app/kafka/kafka_2.10-0.9.0.0
    export HIVE_HOME=/app/hive/apache-hive-2.1.1-bin
    export HBASE_HOME=/app/hbase/hbase-1.2.6
    PATH=$PATH:$HOME/bin:$JAVA_HOME/bin:$HADOOP_HOME/bin:$HADOOP_HOME/sbin:$SCALA_HOME/bin:$SPARK_HOME/bin:$SPARK_HOME/sbin:$ZOOKEEPER_HOME/bin:$KAFKA_HOME/bin:$HIVE_HOME/bin:$HBASE_HOME/bin
    export PATH

别忘了source一下使用环境变量生效

[hadoop@master app]$ source ~/.bash_profile

4.阅读apache hbase官方文档发现，有一项是要将hadoop的conf目录下的hdfs-site.xml复制到hbase的conf目录下，这样可以保证hdfs和hbase的配置一致，例如副本数hbase和hdfs要一致。  
官网原文：  
Procedure: HDFS Client Configuration  
Of note, if you have made HDFS client configuration changes on your Hadoop cluster, such as configuration directives for HDFS clients, as opposed to server-side configurations, you must use one of the following methods to enable HBase to see and use these configuration changes:

Add a pointer to your HADOOP\_CONF\_DIR to the HBASE\_CLASSPATH environment variable in hbase-env.sh.

**Add a copy of hdfs-site.xml (or hadoop-site.xml) or, better, symlinks, under $\{HBASE\_HOME\}/conf, or**

if only a small set of HDFS client configurations, add them to hbase-site.xml.

[hadoop@master conf]$ cp /app/hadoop/hadoop-2.7.3/etc/hadoop/hdfs-site.xml /app/hbase/hbase-1.2.6/conf/

5.配置hbase-site.xml  
参照官方文档[http://hbase.apache.org/book.html\#\_configuration\_files][http_hbase.apache.org_book.html_configuration_files]  
编辑 $HBASE\_HOME/conf/hbase-site.xml

<configuration>
      <property>
        <name>hbase.zookeeper.property.clientPort</name>
        <value>2181</value>
      </property>
      <property>
        <name>hbase.zookeeper.quorum</name>
        <value>master,worker1,worker2</value>
        <description>The directory shared by RegionServers.
        </description>
      </property>
      <property>
        <name>hbase.zookeeper.property.dataDir</name>
        <value>/app/zookeeper/zookeeper-3.4.6/data</value>
        <description>
        注意这里的zookeeper数据目录与hadoop ha的共用，也即要与 zoo.cfg 中配置的一致
        Property from ZooKeeper config zoo.cfg.
        The directory where the snapshot is stored.
        </description>
      </property>
    
      <property>
        <name>hbase.rootdir</name>
        <value>hdfs://master:8020/hbase</value>
        <description>The directory shared by RegionServers.
                     官网多次强调这个目录不要预先创建，hbase会自行创建，否则会做迁移操作，引发错误
                     至于端口，有些是8020，有些是9000，看 $HADOOP_HOME/etc/hadoop/hdfs-site.xml 里面的配置，本实验配置的是
                     dfs.namenode.rpc-address.hdcluster.nn1 , dfs.namenode.rpc-address.hdcluster.nn2
        </description>
      </property>
      <property>
        <name>hbase.cluster.distributed</name>
        <value>true</value>
        <description>分布式集群配置，这里要设置为true，如果是单节点的，则设置为false
          The mode the cluster will be in. Possible values are
          false: standalone and pseudo-distributed setups with managed ZooKeeper
          true: fully-distributed with unmanaged ZooKeeper Quorum (see hbase-env.sh)
        </description>
      </property>
    </configuration>

6.配置regionserver文件  
编辑 $HBASE\_HOME/conf/regionservers 文件，输入要运行 regionserver 的主机名

worker1
    worker2

7.配置 backup-masters 文件（master备用节点）  
HBase 支持运行多个 master 节点，因此不会出现单点故障的问题，但只能有一个活动的管理节点（active master），其余为备用节点（backup master），编辑 $HBASE\_HOME/conf/backup-masters 文件进行配置备用管理节点的主机名  
这一步也可不配置。

worker1

8.配置 hbase-env.sh 文件  
编辑 HBASE\_HOME/conf/hbase-env.sh 配置环境变量，为了实现统一管理，本文使用单独配置的zookeeper，即将其中的 HBASE\_MANAGES\_ZK 设置为 false

# Tell HBase whether it should manage it's own instance of Zookeeper or not.
    export HBASE_MANAGES_ZK=false
     #The default log rolling policy is RFA, where the log file is rolled as per the size defined for the 
    export JAVA_HOME=/app/java/jdk1.8.0_141
    export HBASE_CLASSPATH=/app/hbase/hbase-1.2.6/conf
    # 此配置信息，设置由hbase自己管理zookeeper，不需要单独的zookeeper。
    export HBASE_HOME=/app/hbase/hbase-1.2.6
    export HADOOP_HOME=/app/hadoop/hadoop-2.7.3
    #Hbase日志目录
    export HBASE_LOG_DIR=/app/hbase/hbase-1.2.6/logs

将master节点的hbase文件拷贝至worker1和worker2

[hadoop@master app]$ scp -r hbase/ hadoop@worker1:/app/
    [hadoop@master app]$ scp -r hbase/ hadoop@worker2:/app/

并将环境变量也拷贝过去

[hadoop@master app]$ scp ~/.bash_profile hadoop@worker1:~/
    [hadoop@master app]$ scp ~/.bash_profile hadoop@worker2:~/

ssh到各节点source使其生效。  
至此，hbase配置全部完成  
9.测试启动  
使用 $HBASE\_HOME/bin/start-hbase.sh 指令启动整个集群，查看其shell脚本，可见其启动顺序。

# HBASE-6504 - only take the first line of the output in case verbose gc is on
    distMode=`$bin/hbase --config "$HBASE_CONF_DIR" org.apache.hadoop.hbase.util.HBaseConfTool hbase.cluster.distributed | head -n 1`
    
    
    if [ "$distMode" == 'false' ]
    then
      "$bin"/hbase-daemon.sh --config "${HBASE_CONF_DIR}" $commandToRun master $@
    else
      "$bin"/hbase-daemons.sh --config "${HBASE_CONF_DIR}" $commandToRun zookeeper
      "$bin"/hbase-daemon.sh --config "${HBASE_CONF_DIR}" $commandToRun master
      "$bin"/hbase-daemons.sh --config "${HBASE_CONF_DIR}" \
        --hosts "${HBASE_REGIONSERVERS}" $commandToRun regionserver
      "$bin"/hbase-daemons.sh --config "${HBASE_CONF_DIR}" \
        --hosts "${HBASE_BACKUP_MASTERS}" $commandToRun master-backup
    fi

hbase的启动依赖于hadoop和zookeeper，所以我们按照上述的启动顺序，一一启动。  
也就是使用 hbase-daemon.sh 命令依次启动 zookeeper、master、regionserver、master-backup

因此，我们也按照这个顺序，在各个节点进行启动

在启动HBase之前，必须先启动Hadoop，以便于HBase初始化、读取存储在hdfs上的数据  
（1）启动zookeeper（三个节点分别启动）

[hadoop@master bin]$ zkServer.sh status

[hadoop@master bin]$ zkServer.sh status
    JMX enabled by default
    Using config: /app/zookeeper/zookeeper-3.4.6/bin/../conf/zoo.cfg
    Mode: follower

[hadoop@worker1 ~]$ zkServer.sh status
    JMX enabled by default
    Using config: /app/zookeeper/zookeeper-3.4.6/bin/../conf/zoo.cfg
    Mode: leader

[hadoop@worker2 ~]$ zkServer.sh status
    JMX enabled by default
    Using config: /app/zookeeper/zookeeper-3.4.6/bin/../conf/zoo.cfg
    Mode: follower

（2）启动Hadoop分布式集群  
这里使用start-all.sh启动

[hadoop@master bin]$ 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 /app/hadoop/hadoop-2.7.3/logs/hadoop-hadoop-namenode-master.out
    worker2: starting datanode, logging to /app/hadoop/hadoop-2.7.3/logs/hadoop-hadoop-datanode-worker2.out
    worker1: starting datanode, logging to /app/hadoop/hadoop-2.7.3/logs/hadoop-hadoop-datanode-worker1.out
    master: starting datanode, logging to /app/hadoop/hadoop-2.7.3/logs/hadoop-hadoop-datanode-master.out
    Starting secondary namenodes [0.0.0.0]
    0.0.0.0: starting secondarynamenode, logging to /app/hadoop/hadoop-2.7.3/logs/hadoop-hadoop-secondarynamenode-master.out
    starting yarn daemons
    starting resourcemanager, logging to /app/hadoop/hadoop-2.7.3/logs/yarn-hadoop-resourcemanager-master.out
    master: starting nodemanager, logging to /app/hadoop/hadoop-2.7.3/logs/yarn-hadoop-nodemanager-master.out
    worker1: starting nodemanager, logging to /app/hadoop/hadoop-2.7.3/logs/yarn-hadoop-nodemanager-worker1.out
    worker2: starting nodemanager, logging to /app/hadoop/hadoop-2.7.3/logs/yarn-hadoop-nodemanager-worker2.out
    [hadoop@master bin]$ jps
    3683 NodeManager
    3715 Jps
    3380 SecondaryNameNode
    3575 ResourceManager
    3223 DataNode
    2890 QuorumPeerMain
    3117 NameNode
    [hadoop@master bin]$

（3）启动hbase

[hadoop@master app]$ start-hbase.sh 
    starting master, logging to /app/hbase/hbase-1.2.6/logs/hbase-hadoop-master-master.out
    Java HotSpot(TM) 64-Bit Server VM warning: ignoring option PermSize=128m; support was removed in 8.0
    Java HotSpot(TM) 64-Bit Server VM warning: ignoring option MaxPermSize=128m; support was removed in 8.0
    worker2: starting regionserver, logging to /app/hbase/hbase-1.2.6/logs/hbase-hadoop-regionserver-worker2.out
    worker1: starting regionserver, logging to /app/hbase/hbase-1.2.6/logs/hbase-hadoop-regionserver-worker1.out
    worker2: Java HotSpot(TM) 64-Bit Server VM warning: ignoring option PermSize=128m; support was removed in 8.0
    worker2: Java HotSpot(TM) 64-Bit Server VM warning: ignoring option MaxPermSize=128m; support was removed in 8.0
    worker1: Java HotSpot(TM) 64-Bit Server VM warning: ignoring option PermSize=128m; support was removed in 8.0
    worker1: Java HotSpot(TM) 64-Bit Server VM warning: ignoring option MaxPermSize=128m; support was removed in 8.0
    worker1: starting master, logging to /app/hbase/hbase-1.2.6/logs/hbase-hadoop-master-worker1.out

查看各节点jps  
master：

[hadoop@master app]$ jps
    5762 Jps
    3683 NodeManager
    3380 SecondaryNameNode
    5557 HMaster
    3575 ResourceManager
    3223 DataNode
    2890 QuorumPeerMain
    3117 NameNode

worker1：  
这里start-hbase.sh脚本里启动了，Hmaster的备份进程  
**hbase-daemon.sh start master –backup &**

[hadoop@worker1 app]$ jps
    3970 Jps
    2883 NodeManager
    2692 QuorumPeerMain
    3722 HMaster
    2798 DataNode
    3615 HRegionServer

worker2：

[hadoop@worker2 ~]$ jps
    3424 HRegionServer
    2730 QuorumPeerMain
    2826 DataNode
    2939 NodeManager
    3612 Jps

至此hbase配置启动完成。  
四、Hbase测试并使用  
1.使用hbase shell进入到 hbase 的交互命令行界面，这时可进行测试使用

[hadoop@master app]$ hbase shell
    SLF4J: Class path contains multiple SLF4J bindings.
    SLF4J: Found binding in [jar:file:/app/hbase/hbase-1.2.6/lib/slf4j-log4j12-1.7.5.jar!/org/slf4j/impl/StaticLoggerBinder.class]
    SLF4J: Found binding in [jar:file:/app/hadoop/hadoop-2.7.3/share/hadoop/common/lib/slf4j-log4j12-1.7.10.jar!/org/slf4j/impl/StaticLoggerBinder.class]
    SLF4J: See http://www.slf4j.org/codes.html#multiple_bindings for an explanation.
    SLF4J: Actual binding is of type [org.slf4j.impl.Log4jLoggerFactory]
    HBase Shell; enter 'help<RETURN>' for list of supported commands.
    Type "exit<RETURN>" to leave the HBase Shell
    Version 1.2.6, rUnknown, Mon May 29 02:25:32 CDT 2017
    
    hbase(main):001:0>

（1）查看集群状态和节点数量，很重要

hbase(main):001:0> status
    1 active master, 1 backup masters, 2 servers, 0 dead, 1.0000 average load

由此可知，一个活跃的master，一个备份masters，两个HRegionserver，零个死亡。  
（2）创建表  
hbase创建表create命令语法为：表名、列名1、列名2、列名3……

hbase(main):002:0> create 'employee' ,'name','age'
    0 row(s) in 6.4790 seconds => Hbase::Table - employee

（3）查看表

hbase(main):003:0> list 'employee'
    TABLE                                                                                                                                                                      
    employee                                                                                                                                                                   
    1 row(s) in 0.2430 seconds
    
    => ["employee"]

（4）导入数据

=> Hbase::Table - employee
    hbase(main):025:0> put 'employee','row1','name:firstname','james'
    0 row(s) in 0.0240 seconds
    
    hbase(main):026:0> put 'employee','row1','name:lastname','lebron'
    0 row(s) in 0.0210 seconds
    
    hbase(main):027:0> put 'employee','row2','age','33'
    0 row(s) in 0.0150 seconds

（5）全表扫描数据

hbase(main):028:0> scan 'employee'
    ROW                                         COLUMN+CELL                                                                                                                    
     row1                                       column=name:firstname, timestamp=1525800600153, value=james                                                                    
     row1                                       column=name:lastname, timestamp=1525800607228, value=lebron                                                                    
     row2                                       column=age:, timestamp=1525800622854, value=33                                                                                 
    2 row(s) in 0.0690 seconds

（6）根据条件查询数据

hbase(main):029:0> get 'employee','row1'
    COLUMN                                      CELL                                                                                                                           
     name:firstname                             timestamp=1525800600153, value=james                                                                                           
     name:lastname                              timestamp=1525800607228, value=lebron                                                                                          
    2 row(s) in 0.0220 seconds

hbase(main):003:0> get 'employee','row2'
    COLUMN                                      CELL                                                                                                                           
     age:                                       timestamp=1525800622854, value=33                                                                                              
    1 row(s) in 0.1410 seconds

（7）表失效

hbase(main):004:0> disable 'employee'
    0 row(s) in 7.0270 seconds

（8）表重新生效

hbase(main):005:0> enable 'employee'
    0 row(s) in 5.1070 seconds

（9）删除数据表  
使用drop命令对表进行删除，但只有表在失效的情况下，才能进行删除

hbase(main):004:0> disable 'employee'
    0 row(s) in 7.0270 seconds

hbase(main):006:0> drop 'employee'

（10）退出 hbase shell

quit

2.HBase 管理页面  
HBase 还提供了管理页面，供用户查看，可更加方便地查看集群状态

在浏览器中输入 [http://172.17.0.1:16010][http_172.17.0.1_16010] 地址（默认端口为 16010），即可进入到管理页面，如下图  
![这里写图片描述][70]

至此hbase配置全部完成。

[http_hbase.apache.org_book.html_hbase_default_configurations]: http://hbase.apache.org/book.html#hbase_default_configurations
[http_mirror.bit.edu.cn_apache_hbase_stable_hbase-1.2.6-bin.tar.gz]: http://mirror.bit.edu.cn/apache/hbase/stable/hbase-1.2.6-bin.tar.gz
[http_hbase.apache.org_book.html_configuration_files]: http://hbase.apache.org/book.html#_configuration_files
[http_172.17.0.1_16010]: http://172.17.0.1:16010
[70]: /images/20220525/efe5bae548144f4a834468000619d1e1.png