hive安装

/etc/profile 

export JAVA_HOME=/usr/local/jdk1.8.0_171
export JRE_HOME=$JAVA_HOME/jre
export CLASSPATH=.:$JAVA_HOME/lib/dt.jar:$JAVA_HOME/lib/tools.jar:$JRE_HOME/lib:$CLASSPATH
export PATH=$JAVA_HOME/bin:$PATH

export HADOOP_HOME=/data/hadoop
export HIVE_HOME=/data/hive-3.1.0
export ZOOKEEPER_HOME=/data/zookeeper-3.4.13
export PATH=$PATH:$HADOOP_HOME/bin:$ZOOKEEPER_HOME/bin:$HIVE_HOME/bin

 

 

1、MySQL作为存储元数据的数据库，所以需要把连接MySQL的jar包放入或链接到$HIVE_HOME/lib目录下。

yum install mysql-connector-java
ln -s /usr/share/java/mysql-connector-java.jar /data/hive-3.1.0/lib/mysql-connector-java.jar

 2、修改hive-site.xml

cd /data/hive-3.1.0/conf
cp hive-env.sh.template hive-env.sh
mkdir /data/hive-3.1.0/tmp

 内嵌模式，特点是：hive服务和metastore服务运行在同一个进程中，derby服务也运行在该进程中。
该模式无需特殊配置

本地模式，特点是：hive服务和metastore服务运行在同一个进程中，mysql是单独的进程，可以在同一台机器上，也可以在远程机器上。
该模式只需将hive-site.xml中的ConnectionURL指向mysql，并配置好驱动名、数据库连接账号即可：

远程模式，特点是：hive服务和metastore在不同的进程内，可能是不同的机器。

 hive.metastore.db.type
    mysql
    
      Expects one of [derby, oracle, mysql, mssql, postgres].
      Type of database used by the metastore. Information schema & JDBCStorageHandler depend on it.
    
  
  
  
    hive.metastore.warehouse.dir
    /hive/warehouse
  
  
  
    hive.metastore.uris
    thrift://namenode1:9083
 Thrift uri for the remote metastore. Used by metastore client to connect to remote metastore.

  
  
  
    javax.jdo.option.ConnectionURL
    jdbc:mysql://172.16.99.40:3306/hive?createDatabaseIfNotExist=true
  
  
  
    javax.jdo.option.ConnectionDriverName
    com.mysql.jdbc.Driver
  
  
  
    javax.jdo.option.ConnectionUserName
    hive
  
  
  
    javax.jdo.option.ConnectionPassword
    hive
  

 
    hive.exec.local.scratchdir
    /data/hive-3.1.0/tmp
    Local scratch space for Hive jobs
  
  
    hive.downloaded.resources.dir
    /data/hive-3.1.0/tmp
    Temporary local directory for added resources in the remote file system.
  

 

 客户端配置

  
  
  
  hive.metastore.warehouse.dir  
  /hive/warehouse  
  
   
  
  hive.metastore.local  
  false  
  
  
  
  hive.metastore.uris  
  thrift://namenode2:9083  
  
  
 

 

 

3、初始化数据库

/data/hive-3.1.0/bin/schematool --dbType mysql --initSchema  

4、解决日志jar冲突

保留/data/hadoop/share/hadoop/common/lib/slf4j-log4j12-1.7.25.jar

rm /data/hive-3.1.0/lib/log4j-slf4j-impl-2.10.0.jar 

 

5、检查环境

hadoop version
hadoop version | awk '{if (NR == 1) {print $2;}}'

 

7、启动服务端

$HIVE_HOME/bin/hive --service metastore &

 

8、找个非hadoop机器机器，部署并启动客户端

scp -r -P 21860 /data/hadoop root@mysql1:/data/
scp -r -P 21860 /data/hive-3.1.0 root@mysql1:/data/
source /etc/profile
hive

 

9、服务器启动hiveserver2 （参考 https://blog.csdn.net/zhanglh046/article/details/78572926）

./hiveserver2 start

客户端使用
# beeline 
Beeline version 1.2.1 by Apache Hive
beeline> !connect jdbc:hive2://namenode2:10000
Connecting to jdbc:hive2://namenode2:10000
Enter username for jdbc:hive2://namenode2:10000: hive_user
Enter password for jdbc:hive2://namenode2:10000: ******

 

 

---------------------------------------------------------------------------- 

旧配置

---------------------------------------------------------------------------- 

http://apache.fayea.com/hive/hive-2.0.0/

wget http://apache.fayea.com/hive/hive-2.0.0/apache-hive-2.0.0-bin.tar.gz

tar zxvf apache-hive-2.0.0-bin.tar.gz 

mv  apache-hive-2.0.0-bin  /home/hadoop/hive

cd /home/hadoop/hive

cd conf

cp hive-default.xml.template  hive-site.xml

 



  hive.metastore.warehouse.dir
  /home/hadoop/hive-warehouse

 

  javax.jdo.option.ConnectionURL
  jdbc:mysql://hadoop-master:3306/hive?createDatabaseIfNotExist=true

 

  javax.jdo.option.ConnectionDriverName
  com.mysql.jdbc.Driver

 

  javax.jdo.option.ConnectionUserName
  root

 

  javax.jdo.option.ConnectionPassword
  hive



  hive.metastore.local
  false



  hive.metastore.uris
  thrift://hadoop-master:9083



        datanucleus.readOnlyDatastore
        false
    
     
        datanucleus.fixedDatastore
        false 
    
     
        datanucleus.autoCreateSchema 
        true 
    
    
        datanucleus.autoCreateTables
        true
    
    
        datanucleus.autoCreateColumns
        true
    

　　

 

 

hive-env.sh

# Hive Configuration Directory can be controlled by:
export HIVE_CONF_DIR=/home/hadoop/hive/conf

# Folder containing extra ibraries required for hive compilation/execution can be controlled by:
export HIVE_AUX_JARS_PATH=/home/hadoop/hive/lib

　　

　　

下载mysql的jdbc 

http://dev.mysql.com/downloads/file/?id=460362

把mysql-connector-java-5.1.38-bin.jar 放到//home/hadoop/hive/lib

 

启动hive服务端程序

hive --service metastore 

 

客户端直接使用hive命令即可

hive 

退出

exit;
hadoop job -kill jobid

create table test1 (col_1 string, col_2 string, col_3 string ,col_4 string) ROW FORMAT DELIMITED FIELDS TERMINATED BY '|' STORED AS TEXTFILE;

从本地文件导入
LOAD DATA LOCAL INPATH '/home/hadoop/a.txt' OVERWRITE INTO TABLE test1;
dfs -ls /home/hadoop/hive-warehouse/test1

/home/hadoop/hive-warehouse是hive-site.xml里面配置路径，不是真实物理路径

从hadoop文件导入
hadoop fs -put /home/hadoop/b.txt /data/input
hadoop fs -ls /data/input
LOAD DATA INPATH '/data/input/a.txt' OVERWRITE INTO TABLE test1;

从表导入
create table test2 (col_1 string, col_2 string, col_3 string) ROW FORMAT DELIMITED FIELDS TERMINATED BY '|' STORED AS TEXTFILE;
insert into table test2 as select col_1,col_2,col_3 from test1;
create table test3 as select * from test1;