HBase installation configuration, using independent Zookeeper,shell testing

Preface

Pre-installation requirements, using Hadoop, basically do not need to change anything:

1. Java Environment

2. Hadoop (HBase-based HDFs)

3, zookeeper (I use the independent zookeeper here, because previously used has been installed, of course, can also use HBase management zookeeper, you can refer to the previous article http://blog.csdn.net/smile0198/article/ details/17659537)

4, SSH,NTP time synchronization

5, system tuning, this can wait for the installation after the change, the number of open files (Ulimit and Nproc)

6, modify the Hadoop HDFS datanode simultaneously processing the file's upper limit: dfs.datanode.max.xcievers

One, distributed installation configuration 1, download unpack the package I use is hbase-0.94.6, go to the official website to download, extract directly to the installation directory. 2, configuration conf/hbase-env.sh added a sentence, do not let HBase management zookeeper

Export Hbase_manages_zk=false

The default should be true, which can be set to True if you want HBase to manage zookeeper. 3, configuration Conf/hbase-site.xmlThis step is the core:

<configuration>
 <property>  
  
    <name>hbase.rootdir</name>  
  
    <value>hdfs:// master:9099/hbase</value>  
  
    <description>the Directory shared byregionservers.  
  
    </description>  
  
  </property>  
  
  <property>  
  
    <name>hbase.cluster.distributed</ name>  
  
    <value>true</value>  
  
  </property>  

    <property>  
  
      <name> hbase.zookeeper.property.clientport</name>  
  
      <value>2181</value>  
  
    </property>  
  
    <property>  
  
      <name>hbase.zookeeper.quorum</name>  
  
      <value>haier002, haier003,haier004</value>  
  
    <property>  
  
      <name>hbase.zookeeper.property.datadir</ name>  
  
      <value>/opt/zookeeper-3.4.5/dataDir</value>  
  
    </property>  
   
</ Configuration>

Parameter description: (1) Hbase.rootdir,hdfs's entry address, address and port should be the same as your Hadoop configuration (<name>fs.default.name</name> in Core-site.xml) , all nodes public address (2) hbase.cluster.distributed,ture represents distributed (3) Hbase.zookeeper.property.clientPort, zookeeper Port (4) Hbase.zookeeper.quorum, zookeeper node (5) Hbase.zookeeper.property.dataDir, zookeeper keep the information of the file, default to/tmp restart will be lost
4, Configuration conf/regionserversThis slave configuration equivalent to Hadoop

Slave1
slave2
slave3

5. SCP to other machinesCopy the folder hbase-0.94.6 to several other machines

Scp-r hbase-0.94.6 hadoop@slave1:/usr/local/

6, runEnter the bin directory of the master installation directory

./start-hbase.sh

It's done, hahaha. JPS, please.

15675 NameNode
18205 hmaster
1264 asmain
15840 jobtracker
875 asmain
19017 Jps

Just a second. WebUI Address: master:60010


second, test, HBase Shell1, connect the shell into the bin directory, command

./hbase Shell

Enter help to see the command Note: The shell removes the need to hold down CTRL2. Examples of common commands create tables, add data

Create ' testajl ', ' CF '
put ' testajl ', ' row1 ', ' cf:a ', ' value1 '
put ' testajl ', ' row2 ', ' cf:b ', ' value2 '
put ' Testajl ', ' row3 ', ' cf:c ', ' value3 '

View data

Scan ' TESTAJL '
ROW                                      column+cell                                                                                                        
 row1                                    column=cf:a, timestamp=1388327667793, value=value1                                                                 
 row2                                    column=cf:b, timestamp=1388327866650, value=value2                                                                 
 row3                                    column=cf:c, timestamp=1388327785678, value= Value3                                                                 
3 Row (s) in 0.0630 seconds

Fetch a row of data

Get ' testajl ', ' row1 '
COLUMN                                   CELL                                                                                                               
 cf:a                                    timestamp=1388327667793, value=value1                                                                              
1 row (s) in 0.0290 Seconds

Delete Table First invalidates

Disable ' TESTAJL '

Delete

Drop ' TESTAJL '

OK, start getting started, you can play, a piece of refueling.