下载：http://mirror.bit.edu.cn/apache/hbase/stable/
官方指南：http://abloz.com/hbase/book.html
安装配置：
解压：
tar -xzvf hbase-0.96.0-hadoop1-bin.tar.gz
进入$hbase/lib下查看相关的hadoop包，看是哪个版本的hadoop。
此处只记录伪分布式Hbase的安装。
配置参数，修改hbase-site.xml：
- <configuration>
- ...
- <property>
- <name>hbase.rootdir</name>
- <value>hdfs://localhost:9000/hbase</value>
- <description>The directory shared by RegionServers.
- </description>
- </property>
- <property>
- <name>dfs.replication</name>
- <value>1</value>
- <description>The replication count for HLog & HFile storage. Should not be greater than HDFS datanode count.
- </description>
- </property>
- ...
- </configuration>
伪分布式启动：
伪分布式是基于HDFS的，所以需要先启动HDFS.
之后启动HBase
sh start-hbase.sh
不出意外的报错：
dat@dat-HP:/opt/hbase-0.96/bin$ sh start-hbase.sh
start-hbase.sh: 79: /opt/hbase-0.96/bin/hbase-config.sh: [[: not found
start-hbase.sh: 88: /opt/hbase-0.96/bin/hbase-config.sh: [[: not found
start-hbase.sh: 53: [: false: unexpected operator
starting master, logging to /opt/hbase-0.96/bin/../logs/hbase-dat-master-dat-HP.out
Could not start ZK at requested port of 2181. ZK was started at port: -1. Aborting as clients (e.g. shell) will not be able to find this ZK quorum.
看样子是2181被占用了，查看占用2181端口的进程：
lsof -i:2181
- dat@dat-HP:/opt/hbase-0.96/logs$ lsof -i:2181
- COMMAND PID USER FD TYPE DEVICE SIZE/OFF NODE NAME
- java 24164 dat 87u IPv6 2800953 0t0 TCP *:2181 (LISTEN)
ps -ef|grep 24164
- dat@dat-HP:/opt/hbase-0.96/logs$ ps -ef|grep 24164
- dat 24164 24151 0 10:09 ? 00:00:03 /opt/jdk1.7.0_25/bin/java -Dproc_zookeeper -XX:OnOutOfMemoryError=kill -9 %p -Xmx1000m -XX:+UseConcMarkSweepGC -Dhbase.log.dir=/opt/hbase-0.96/bin/../logs -Dhbase.log.file=hbase-dat-zookeeper-dat-HP.log -Dhbase.home.dir=/opt/hbase-0.96/bin/.. -Dhbase.id.str=dat -Dhbase.root.logger=INFO,RFA -Dhbase.security.logger=INFO,RFAS org.apache.hadoop.hbase.zookeeper.HQuorumPeer start
显示已经是hbase的zookeeper服务，杀掉pid,重试,
- dat@dat-HP:/opt/hbase-0.96/bin$ ./start-hbase.sh
- starting master, logging to /opt/hbase-0.96/bin/../logs/hbase-dat-master-dat-HP.out
正常启动，检测jps
- dat@dat-HP:/opt/hbase-0.96/bin$ jps
- 25644 HMaster
- 23468 TaskTracker
- 23199 SecondaryNameNode
- 23307 JobTracker
- 25849 Jps
- 22827 NameNode
- 23031 DataNode
正常。。。
HBase Shell
编辑/etc/profile加入hbase_home，重新生效,source /etc/profile
- export PIG_INSTALL=/opt/pig-0.12.0
- export PATH=$PATH:$PIG_INSTALL/bin
- export PIG_CLASSPATH=/opt/hadoop-1.2.1/conf
- export HIVE_HOME=/opt/hive-0.12.0
- export PATH=$HIVE_HOME/bin:$HIVE_HOME/conf:$PATH
- export ANT_HOME=/opt/ant
- export PATH=$PATH:$ANT_HOME/bin
- export HBASE_HOME=/opt/hbase-0.96
- export PATH=$PATH:$HBASE_HOME/bin
- dat@dat-HP:/opt/hbase-0.96/bin$ hbase shell
- HBase Shell; enter 'help<RETURN>' for list of supported commands.
- Type "exit<RETURN>" to leave the HBase Shell
- Version 0.96.0-hadoop1, r1531434, Fri Oct 11 15:11:29 PDT 2013
- hbase(main):001:0>
输入help，会列出hbase shell所支持的命令
shutdown是关闭hbase集群，exit是退出hbase shell。
HBase Shell 练习
建表
hbase(main):006:0> create 'test','cf'
列出所有表
hbase(main):007:0> listTABLE
test
1 row(s) in 0.0440 seconds
=> ["test"]
插入数据
hbase(main):008:0> put 'test','row1','cf:a','value1'
0 row(s) in 0.0750 seconds
hbase(main):009:0> put 'test','row2','cf:b','value2'
0 row(s) in 0.0080 seconds
hbase(main):010:0> put 'test','row3','cf:c','value3'
0 row(s) in 0.0060 seconds
检查插入情况
hbase(main):011:0> scan 'test'
ROW COLUMN+CELL
row1 column=cf:a, timestamp=1386389158263, value=value1
row2 column=cf:b, timestamp=1386389170821, value=value2
row3 column=cf:c, timestamp=1386389185954, value=value3
3 row(s) in 0.0450 seconds
get一行
hbase(main):012:0> get 'test','row2'
COLUMN CELL
cf:b timestamp=1386389170821, value=value2
1 row(s) in 0.0140 seconds
disable之后drop表
hbase(main):013:0> disable 'test'
0 row(s) in 1.4220 seconds
hbase(main):014:0> drop 'test'
0 row(s) in 0.2740 seconds
hbase(main):015:0> list
TABLE
0 row(s) in 0.0270 seconds
=> []