Hadoop系列之Hive(数据仓库)安装配置

 Hadoop系列之Hive(数据仓库)安装配置
1.在NameNode安装
  cd /root/soft
  tar zxvf apache-hive-0.13.1-bin.tar.gz
  mv apache-hive-0.13.1-bin /usr/local/hadoop/hive
2. 配置环境变量(每个节点都需要增加)
 打开/etc/profile
 #添加以下内容:
 export HIVE_HOME=/usr/local/hadoop/hive
 export PATH=$HIVE_HOME/bin:$PATH
 #环境变量生效
 source /etc/profile
3.安装mysql数据库环境
 请参照http://azhuang.blog.51cto.com/9176790/1551549
 数据库安装成功后,一定要建立号权限及创建hive数据库。操作如下
 grant all privileges on hive.* to hive@'192.168.3.%' identified by '123';
 create database hive character set latin1;  #UTF-8编码hive会报错,所以需要修改编码为latin1
4. 配置Hive
 cd /usr/local/hadoop/hive/conf/
 cp hive-default.xml.template hive-site.xml
 #vim hive-site.xml(修改<configuration>之间配置)

 <configuration>
    <property>
        <name>javax.jdo.option.ConnectionURL</name>
        <value>jdbc:mysql://192.168.3.10:3306/hive?characterEncoding=latin1</value>
    </property>
    <property>
        <name>javax.jdo.option.ConnectionDriverName</name>
        <value>com.mysql.jdbc.Driver</value>
    </property>
    <property>
        <name>javax.jdo.option.ConnectionUserName</name>
        <value>hive</value>
    </property>
    <property>
        <name>javax.jdo.option.ConnectionPassword</name>
        <value>123</value>
    </property>
 </configuration>

 #以上四项分别是:
 数据库连接,数据库驱动名,用户名,密码。
5.把mySQL的JDBC驱动包复制到Hive的lib目录下
  cp /root/soft/mysql-connector-java-commercial-5.1.30-bin.jar /usr/local/hadoop/hive/lib/
6.复制Hive到所有DataNode节点
  scp -r /usr/local/hadoop/hive [email protected]:/usr/local/hadoop/
  scp -r /usr/local/hadoop/hive [email protected]:/usr/local/hadoop/
7.简单测试
//查看当前数据表
hive> show tables;
OK
Time taken: 1.069 seconds
//从本地文件系统中导入数据到Hive表
#创建student.txt测试文本,字段之间用tab键分割
 [root@hdfs-master soft]# cat /root/soft/student.txt
1    aa    10    121221
2    bb    20    0990
3    cc    30    120120
#创建student测试表
hive> create table student(id int, name string, age int, tel string) ROW FORMAT DELIMITED FIELDS TERMINATED BY '\t'  STORED AS TEXTFILE;
OK
Time taken: 0.043 seconds
#再次查看当前数据表及结构
hive> show tables;
OK
student
hive> desc student;
OK
id                      int                                         
name                    string                                      
age                     int                                         
tel                     string                                      
Time taken: 0.103 seconds, Fetched: 4 row(s)
#把/root/soft/student.txt本地记录导入到student表
hive> load data local inpath '/root/soft/student.txt'  into table student;
Copying data from file:/root/soft/student.txt
Copying file: file:/root/soft/student.txt
Loading data to table default.student
Table default.student stats: [numFiles=1, numRows=0, totalSize=43, rawDataSize=0]
OK
Time taken: 0.376 seconds
#查看student表,如果有记录表示本地插入数据成功.
hive> select * from student;                                             
OK
1    aa    10    121221
2    bb    20    0990
3    cc    30    120120
Time taken: 0.066 seconds, Fetched: 3 row(s)
//HDFS上导入数据到Hive表
#上传本地文件到hdfs
[root@hdfs-master soft]# hdfs dfs -put /root/soft/student.txt /hive
[root@hdfs-master soft]# hdfs dfs -cat /hive/student.txt
1    aa    10    121221
2    bb    20    0990
3    cc    30    120120
#从hdfs导入到hive数据
hive> load data inpath '/hive/student.txt' into table student;
Loading data to table default.student
Table default.student stats: [numFiles=2, numRows=0, totalSize=86, rawDataSize=0]
OK
Time taken: 1.389 seconds
hive> select * from student;                                  
OK
1    aa    10    121221
2    bb    20    0990
3    cc    30    120120
1    aa    10    121221
2    bb    20    0990
3    cc    30    120120
Time taken: 0.049 seconds, Fetched: 6 row(s)

你可能感兴趣的:(hadoop,hive,数据仓库,安装配置)