版权声明：Please make the source marked https://blog.csdn.net/qq_31807385/article/details/84677262
目录
数据库的创建：
操作如下：
0: jdbc:hive2://hadoop108:10000> show databases;
OK
+----------------+--+
| database_name |
+----------------+--+
| default |
+----------------+--+
1 row selected (2.047 seconds)
0: jdbc:hive2://hadoop108:10000> show tables;
OK
+-----------+--+
| tab_name |
+-----------+--+
+-----------+--+
No rows selected (0.101 seconds)
0: jdbc:hive2://hadoop108:10000> create database db_hive;
OK
No rows affected (0.417 seconds)
0: jdbc:hive2://hadoop108:10000> show databases;
OK
+----------------+--+
| database_name |
+----------------+--+
| db_hive |
| default |
+----------------+--+
2 rows selected (0.045 seconds)
经过上面的操作，也即创建了数据库之后，该数据库默认存储的位置是：
/user/hive/warehouse/db_hive.db与此同时，创建user目录的时候，还会在根目录下创建
/tmp/在创建数据库的时候，还可以指定数据库在HDFS上的位置，创建的语法如下：
0: jdbc:hive2://hadoop108:10000> create database db_hive2 location '/db_hive2.db';
OK
此时，在HDFS上也会多出一个文件夹：
/db_hive2.db对于我们创建的数据库，关于数据库的相关的信息存储在metestore中，如下图：
可见，每一个数据库在创建的时候，都会有自己对应的location，如果在创建数据库的时候没有指定该数据库的location的话，那么该数据库的location将会是默认的位置： /user/hive/warehouse 目录下。
数据库的查询，切换，修改：
0: jdbc:hive2://hadoop108:10000> desc database db_hive;
OK
+----------+----------+-------------------------------------------------------+-------------+-------------+-------------+--+
| db_name | comment | location | owner_name | owner_type | parameters |
+----------+----------+-------------------------------------------------------+-------------+-------------+-------------+--+
| db_hive | | hdfs://hadoop108:9000/user/hive/warehouse/db_hive.db | isea | USER | |
+----------+----------+-------------------------------------------------------+-------------+-------------+-------------+--+
0: jdbc:hive2://hadoop108:10000> desc database extended db_hive;
OK
+----------+----------+-------------------------------------------------------+-------------+-------------+-------------+--+
| db_name | comment | location | owner_name | owner_type | parameters |
+----------+----------+-------------------------------------------------------+-------------+-------------+-------------+--+
| db_hive | | hdfs://hadoop108:9000/user/hive/warehouse/db_hive.db | isea | USER | |
+----------+----------+-------------------------------------------------------+-------------+-------------+-------------+--+
0: jdbc:hive2://hadoop108:10000> use db_hive;
用户可以使用ALTER DATABASE命令为某个数据库的DBPROPERTIES设置键-值对属性值，来描述这个数据库的属性信息。数据库的其他元数据信息都是不可更改的，包括数据库名和数据库所在的目录位置。
0: jdbc:hive2://hadoop108:10000> alter database db_hive set dbproperties('createtime'='12');
OK
No rows affected (0.14 seconds)
0: jdbc:hive2://hadoop108:10000> desc database extended db_hive;
OK
+----------+----------+-------------------------------------------------------+-------------+-------------+------------------+--+
| db_name | comment | location | owner_name | owner_type | parameters |
+----------+----------+-------------------------------------------------------+-------------+-------------+------------------+--+
| db_hive | | hdfs://hadoop108:9000/user/hive/warehouse/db_hive.db | isea | USER | {createtime=12} |
+----------+----------+-------------------------------------------------------+-------------+-------------+------------------+--+
数据库的删除：
0: jdbc:hive2://hadoop108:10000> drop database db_hive2;
OK
No rows affected (0.532 seconds)
0: jdbc:hive2://hadoop108:10000> show databases;
OK
+----------------+--+
| database_name |
+----------------+--+
| db_hive |
| default |
+----------------+--+
2 rows selected (0.038 seconds)
此时HDFS上的目录和db_hive2对应的数据库文件也被删除：
/db_hive2.db 也被删除如果数据库中有表，不为空，如果是这种情况的话，使用上面的语句是无法删除的，此时可以使用下面的语句删除：
0: jdbc:hive2://hadoop108:10000> drop database db_hive2 cascade;
OK
No rows affected (0.635 seconds)