大数据实验室（大数据基础培训）——Hive的安装、配置及基础使用

为某企业作的培训，完整文档见：http://gudaoxuri.github.io/bd-lab/

10. Hive

此主题介绍Hive的安装、配置及基础使用。

Hive介绍
http://blog.csdn.net/zhoudaxia/article/details/8855937
Hive语法大全
https://cwiki.apache.org/confluence/display/Hive/LanguageManual 自定义函数
https://cwiki.apache.org/confluence/display/Hive/LanguageManual+UDF

— csdn.net

Hive基本信息

官网：http://hive.apache.org/

10.1. 环境准备

# 切换到工做空间
cd /opt/workspaces
# 建立Hive数据目录
mkdir data/hive
# 建立Hive日志目录
mkdir logs/hive

10.2. 安装

wget http://mirrors.cnnic.cn/apache/hive/hive-1.2.1/apache-hive-1.2.1-bin.tar.gz
tar -zxf apache-hive-1.2.1-bin.tar.gz
rm -rf apache-hive-1.2.1-bin.tar.gz
mv apache-hive-1.2.1-bin ./frameworks/hive

10.3. 配置(单机)

vi ~/.profile

export HIVE_HOME=/opt/workspaces/frameworks/hive
export PATH=$PATH:$HIVE_HOME/bin

source ~/.profile

配置Mysql

mysql -u root -p
create database hive;
grant all on *.* to'hive'@'%' identified by 'hive';  #帐号hive,密码hive,全部IP均可以访问（%）
flush privileges;
exit;

建立HDFS目录

./frameworks/hadoop/bin/hadoop fs -mkdir /tmp
./frameworks/hadoop/bin/hadoop fs -mkdir -p /user/hive/warehouse
./frameworks/hadoop/bin/hadoop fs -chmod 777 /tmp
./frameworks/hadoop/bin/hadoop fs -chmod 777 /user/hive/warehouse

修改Hive配置

cat>./frameworks/hive/conf/hive-env.sh<<EOF
export HADOOP_HEAPSIZE=1024
HADOOP_HOME=/opt/workspaces/frameworks/hadoop
export HIVE_CONF_DIR=/opt/workspaces/frameworks/hive/conf
export HIVE_AUX_JARS_PATH=/opt/workspaces/frameworks/hive/lib
EOF
cat>./frameworks/hive/conf/hive-site.xml<<EOF
<configuration>
    <property>
        <name>javax.jdo.option.ConnectionURL</name>
        <value>jdbc:mysql://`<Docker Host IP>`:3306/hive</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>hive</value>
    </property>
    <property>
        <name>hive.metastore.uris</name>
        <value>thrift://bd:9083</value>
    </property>
    <property>
        <name>hive.exec.local.scratchdir</name>
        <value>/opt/workspaces/data/hive</value>
    </property>
    <property>
        <name>hive.downloaded.resources.dir</name>
        <value>/opt/workspaces/data/hive</value>
    </property>
    <property>
        <name>hive.querylog.location</name>
        <value>/opt/workspaces/data/hive</value>
    </property>
    <property>
        <name>hive.server2.logging.operation.log.location</name>
        <value>/opt/workspaces/data/hive</value>
    </property>
    <property>
     <name>hive.querylog.location</name>
     <value>/opt/workspaces/frameworks/logs/hive</value>
    </property>
    <property>
      <name>hive.metastore.schema.verification</name>
      <value>false</value>
    </property>
</configuration>
EOF

cp ./frameworks/hive/conf/hive-log4j.properties.template ./frameworks/hive/conf/hive-log4j.properties

vi ./frameworks/hive/conf/hive-log4j.properties

hive.log.dir=/opt/workspaces/frameworks/logs/hive

下载mysql驱动

wget -P ./frameworks/hive/lib http://central.maven.org/maven2/mysql/mysql-connector-java/5.1.36/mysql-connector-java-5.1.36.jar

测试发现5.1.38须要使用SSL链接，可设置jdbc://…​&userSSL=false禁用

10.4. 启动

nohup ./frameworks/hive/bin/hive --service metastore >/dev/null 2>&1 &

10.5. 测试

cat>test_person.txt<<EOF
tom 20
jack    24
nestor  29
EOF
./frameworks/hadoop/bin/hdfs dfs -mkdir /tmp/hivetest/
./frameworks/hadoop/bin/hdfs dfs -put ./test_person.txt /tmp/hivetest/
./frameworks/hive/bin/hive
CREATE EXTERNAL TABLE test_person(name STRING,age INT)ROW FORMAT DELIMITED FIELDS TERMINATED BY '\t' LOCATION '/tmp/hivetest';
select * from test_person;