| How To Change the Partition Column Of A Partitioned Table Using DBMS_Redefinition [ID 846405.1] | ||
|
| ||
| Modified22-OCT-2009TypeHOWTOStatusPUBLISHED | ||
In this Document
Goal
Solution
References
Applies to:
Oracle Server - Enterprise Edition - Version: 10.2.0.4 to 11.2.0.2.0
Information in this document applies to any platform.
Goal
The purpose of this document is to provide step by step instructions on how to convert the partitioned column of an existing partitioned table to have a completely different partitioned column using dbms_redefinition package.
Solution
1) Create partitioned table (This is the table that needs to be redefined) with the name SALES. We plan to repartition this table to use the MONTH_NO as the partitioning column.
Note: The table already has a PK on this defined in the Create statement.
CREATE TABLE sales (
acct_no NUMBER(5),
acct_name CHAR(30),
amount_of_sale NUMBER(6),
week_no INTEGER,
month_no integer,
sale_details VARCHAR2(1000),
PRIMARY KEY (acct_no, acct_name, week_no))
PARTITION BY LIST (week_no) (
PARTITION part1234 VALUES (1, 2, 3, 4) tablespace users,
PARTITION part5678 VALUES (5, 6, 7, 8) tablespace users,
PARTITION partdefault VALUES (DEFAULT) tablespace users);
insert into sales values (1,'acc 1',1,52,12,'sales details 1');
insert into sales values (2,'acc 2',2,51,12,'sales details 2');
insert into sales values (3,'acc 3',3,50,12,'sales details 3');
insert into sales values (4,'acc 4',4,1,1,'sales details 4');
insert into sales values (5,'acc 5',5,2,1,'sales details 5');
commit;
2) Gather statistics on the table:
EXEC DBMS_STATS.gather_table_stats('RK_MVIEW', 'SALES', cascade => TRUE);
3) Create a Partitioned Interim Table:
CREATETABLEinterim_sales(acct_noNUMBER(5),acct_nameCHAR(30),amount_of_saleNUMBER(6),week_noINTEGER,month_nointeger,sale_detailsVARCHAR2(1000),PRIMARYKEY(acct_no,acct_name,month_no))PARTITIONBYLIST(month_no)(PARTITIONint_part1VALUES(1)tablespaceusers,PARTITIONint_part12VALUES(12)tablespaceusers,PARTITIONint_partdefaultVALUES(DEFAULT)tablespaceusers);
4) Start the Redefinition Process:
a) Check the redefinition is possible using the following command:
EXEC Dbms_Redefinition.can_redef_table('RK_MVIEW', 'SALES');
b)If no errors are reported, start the redefintion using the following command:
BEGINDBMS_REDEFINITION.start_redef_table(uname=>'RK_MVIEW',orig_table=>'SALES',int_table=>'INTERIM_SALES');END;/
c) Optionally synchronize new table with interim name before index creation:
BEGINdbms_redefinition.sync_interim_table(uname=>'RK_MVIEW',orig_table=>'SALES',int_table=>'INTERIM_SALES');END;/
d) Gather statistics on the new table:
EXECDBMS_STATS.gather_table_stats('RK_MVIEW','INTERIM_SALES',cascade=>TRUE);
e) Complete the Redefintion Process:
BEGINdbms_redefinition.finish_redef_table(uname=>'RK_MVIEW',orig_table=>'SALES',int_table=>'INTERIM_SALES');END;/
At this point the interim table has become the "real" table and their names have been switched in the name dictionary.
f) Remove original table which now has the name of the interim table:
DROP TABLE INTERIM_SALES;
g) Check whether partitioning is successful or not:
SELECT partition_name
FROM user_tab_partitions
WHERE table_name = 'SALES';
PARTITION_NAME
------------------
INT_PART1
INT_PART12
INT_PARTDEFAULT
------------------------------------------------------------------------------
本文提供了一步一步的指南,说明如何使用DBMS_Redefinition包将现有分区表的一个分区列转换为完全不同的分区列。
755

被折叠的 条评论
为什么被折叠?



