Sqoop將MySQL資料匯入到hive中

錯誤解決：http://blog.itpub.net/25854343/viewspace-2565234/

MySQL表：

mysql> desc t3;

+----------------+------------+------+-----+---------+-------+

| Field          | Type       | Null | Key | Default | Extra |

+----------------+------------+------+-----+---------+-------+

| ISVALID        | int(11)    | YES  | MUL | NULL    |       |

| CREATETIME     | datetime   | YES  |     | NULL    |       |

| UPDATETIME     | datetime   | YES  |     | NULL    |       |

| CONC_UNI_CODE  | bigint(20) | YES  |     | NULL    |       |

| COM_UNI_CODE   | bigint(20) | YES  |     | NULL    |       |

| FUND_INFW_REL  | double     | YES  |     | NULL    |       |

| MARK_MANI_REL  | double     | YES  |     | NULL    |       |

| STOCK_FREQ_REL | double     | YES  |     | NULL    |       |

| STOCK_CONC_REL | double     | YES  |     | NULL    |       |

+----------------+------------+------+-----+---------+-------+

9 rows in set (0.01 sec)

mysql> 

hive中自己建立表：

hive> create table tt1(

ISVALID int,

CREATETIME TIMESTAMP,

UPDATETIME TIMESTAMP,

CONC_UNI_CODE bigint,

COM_UNI_CODE bigint,

FUND_INFW_REL double,

MARK_MANI_REL double,

STOCK_FREQ_REL double,

STOCK_CONC_REL double) 

ROW FORMAT DELIMITED FIELDS TERMINATED BY '\t' ;

hive>

1.將Mysql資料匯入到hive中(提前在hive中建立表)

(1).匯入資料到HDFS中

[hdfs@jingong01 ~]$ sqoop import --connect jdbc:mysql://172.16.8.93:3306/db_stktag --username wangying --password wangying --table t3 --target-dir /user/tong/123 --num-mappers 1 --hive-table tt1 -m 1 --split-by date --direct --fields-terminated-by '\t'

(2).載入資料

hive> load data inpath '/user/tong/123' into table tt1;

hive> select * from tt1 limit 2;

OK

0 2015-06-12 10:00:04 2016-07-28 18:00:16 5001000008 3000001022 80.0 90.0 70.0 85.0

0 2015-06-12 10:00:04 2015-12-22 15:18:25 5001000008 3000078316 30.0 80.0 70.0 64.0

Time taken: 0.089 seconds, Fetched: 2 row(s)

hive> 

2.直接從Mysql匯入到hive中，不需要load data載入

[hdfs@jingong01 ~]$ cat test.sql 

create table test(

ISVALID int,

CREATETIME TIMESTAMP,

UPDATETIME TIMESTAMP,

CONC_UNI_CODE bigint,

COM_UNI_CODE bigint,

FUND_INFW_REL double,

MARK_MANI_REL double,

STOCK_FREQ_REL double,

STOCK_CONC_REL double) 

ROW FORMAT DELIMITED FIELDS TERMINATED BY '\t';

[hdfs@jingong01 ~]$ hive -f test.sql   --建立表

Logging initialized using configuration in jar:file:/opt/cloudera/parcels/CDH-5.13.0-1.cdh5.13.0.p0.29/lib/hive/lib/hive-common-1.1.0-cdh5.13.0.jar!/hive-log4j.properties

OK

Time taken: 6.709 seconds

[hdfs@jingong01 ~]$ sqoop import --connect jdbc:mysql://172.16.8.93:3306/db_stktag --username wangying --password wangying --table t3 --delete-target-dir --num-mappers 1 --hive-import -m 1 --hive-table test --fields-terminated-by '\t'      --匯入資料

。。。。。。。。。。。。。。。。。。。。。。。。。。。。。。。。。。。。。。。。。。。。。。。。。。。。

19/01/30 15:35:38 INFO hive.HiveImport: OK

19/01/30 15:35:38 INFO hive.HiveImport: Time taken: 6.207 seconds

19/01/30 15:35:38 INFO hive.HiveImport: Loading data to table default.test

19/01/30 15:35:38 INFO hive.HiveImport: Table default.test stats: [numFiles=1, totalSize=3571294]

19/01/30 15:35:38 INFO hive.HiveImport: OK

19/01/30 15:35:38 INFO hive.HiveImport: Time taken: 0.615 seconds

19/01/30 15:35:38 INFO hive.HiveImport: WARN: The method class org.apache.commons.logging.impl.SLF4JLogFactory#release() was invoked.

19/01/30 15:35:38 INFO hive.HiveImport: WARN: Please see for an explanation.

19/01/30 15:35:39 INFO hive.HiveImport: Hive import complete.

19/01/30 15:35:39 INFO hive.HiveImport: Export directory is contains the _SUCCESS file only, removing the directory.

[hdfs@jingong01 ~]$ hive

Logging initialized using configuration in jar:file:/opt/cloudera/parcels/CDH-5.13.0-1.cdh5.13.0.p0.29/lib/hive/lib/hive-common-1.1.0-cdh5.13.0.jar!/hive-log4j.properties

hive>  select * from test limit 2;

OK

0 2015-06-12 10:00:04 2016-07-28 18:00:16 5001000008 3000001022 80.0 90.0 70.0 85.0

0 2015-06-12 10:00:04 2015-12-22 15:18:25 5001000008 3000078316 30.0 80.0 70.0 64.0

Time taken: 0.058 seconds, Fetched: 2 row(s)

hive>