分区表的导入导出 expdp&impdp Oracle 11.2.0.4
分区表的导入导出
数据库版本：
当前有两个分区表
TABLE_NAME PARTITION_NAME TABLESPACE_NAME
------------------------------ ------------------------------ ------------------------------
TABLE1 SYS_P57 NEW
TABLE1 SYS_P66NEW
TABLE2 T_TEST_201503 T3
TABLE2 T_TEST_MAX T
TABLE2 T_TEST_201510 T4
TABLE2 T_TEST_201412 T2
TABLE2 T_TEST_201406 T1
21 rows selected.
SQL> l
1* select TABLE_NAME,PARTITION_NAME,TABLESPACE_NAME from USER_TAB_PARTITIONS order by table_name
当前有两个分区表
table1 是创建在一个表空间中的分区表，有多个分区。
table2 是将有多个分区表空间的分区表，多个分区对应多个表空间。
涉及的表空间
new、T、T1、T2、T3、T4
涉及的用户
new
/**********对new用户进行导出，再远端数据库中导入 schema level
create directory EXPDP as '/backup/expdp';
grant read,write on directory EXPDP to system;
导出整个 new schema:
export NLS_LANG=AMERICAN_AMERICA.ZHS16GBK
expdp system/oracle directory=EXPDP schemas=new dumpfile=20151228expnew.dmp logfile=20151228expnew.log
expdp system/oracle directory=EXPDP schemas=new dumpfile=20151228expnewtab3.dmp logfile=20151228expnewtab3.log
. . exported "NEW"."PF_MEMBER_BANKCARD" 0 KB 0 rows
. . exported "NEW"."PF_MEMBER_CORPORATION" 0 KB 0 rows
. . exported "NEW"."PF_MEMBER_LINKINFO" 0 KB 0 rows
. . exported "NEW"."TABLE2":"T_TEST_201503" 0 KB 0 rows
Master table "SYSTEM"."SYS_EXPORT_SCHEMA_01" successfully loaded/unloaded
******************************************************************************
Dump file set for SYSTEM.SYS_EXPORT_SCHEMA_01 is:
/backup/expdp/20151228expnew.dmp
Job "SYSTEM"."SYS_EXPORT_SCHEMA_01" successfully completed at Mon Dec 28 10:25:40 2015 elapsed 0 00:01:21
目标库导入new schema:
create directory EXPDP as '/backup/expdp';
grant read,write on directory EXPDP to system;
export NLS_LANG=AMERICAN_AMERICA.ZHS16GBK
[oracle@primary ~]$ export NLS_LANG=AMERICAN_AMERICA.ZHS16GBK
[oracle@primary ~]$ impdp system/oracle directory=EXPDP dumpfile=20151228expnew.dmp schemas=new remap_schema=new:demo remap_tablespace=new:demo logfile=20151228expnew.log
ORA-39171: Job is experiencing a resumable wait.
ORA-01658: unable to create INITIAL extent for segment in tablespace DEMO
ORA-39083: Object type TABLE:"DEMO"."TABLE2" failed to create with error:
ORA-00959: tablespace 'T1' does not exist
Failing sql is:
CREATE TABLE "DEMO"."TABLE2" ("TABLE_ID" NUMBER(8,0), "SUB_DATE" DATE, "VALUE" NUMBER(8,0)) PCTFREE 10 PCTUSED 40 INITRANS 1 MAXTRANS 255 STORAGE( BUFFER_POOL DEFAULT FLASH_CACHE DEFAULT CELL_FLASH_CACHE DEFAULT) TABLESPACE "DEMO" PARTITION BY RANGE ("SUB_DATE") (PARTITION "T_TEST_201406" VALUES LESS THAN (TO_DATE(' 2014-06-01 00:00:00', 'SYYYY-MM-DD HH24:MI:SS', 'NLS_CALEN
Processing object type SCHEMA_EXPORT/TABLE/TABLE_DATA
. . imported "DEMO"."PF_AGREEMENT" 0 KB 0 rows
. . imported "DEMO"."PF_MEMBER_BANKCARD" 0 KB 0 rows
. . imported "DEMO"."PF_MEMBER_CORPORATION" 0 KB 0 rows
. . imported "DEMO"."PF_MEMBER_LINKINFO" 0 KB 0 rows
Job "SYSTEM"."SYS_IMPORT_SCHEMA_01" completed with 5 error(s) at Mon Dec 28 10:55:08 2015 elapsed 0 00:07:31
因为 没有对应的表空间，所以导入失败。
SQL> select * from table1;
TABLE_ID SUB_DATE VALUE
---------- ------------ ----------
1 01-APR-14 11
1 01-MAY-14 11
1 01-JUN-14 11
1 01-JUL-14 11
SQL> select * from table2;
select * from table2
*
ERROR at line 1:
ORA-00942: table or view does not exist
换个方法试试 将多个表空间的内容 导入到一个表空间。适用于 一个scheman中有多个表，分布在不同的表空间中，最终汇总到一个表空间。
参考：
impdp ctais3/oracle DIRECTORY=DATA_PUMP_DIR dumpfile=ctais2.dmp remap_schema=ctais2:ctais3 remap_tablespace='(EXAMPLE:ZHANGQIAOC,USERS:ZHANGQIAOC)'
impdp system/oracle directory=EXPDP dumpfile=20151228expnew.dmp schemas=new remap_schema=new:demo remap_tablespace='new:demo,T:demo,T1:demo,T2:demo,T3:demo,T4:demo' logfile=20151228expnew.log
drop user demo cascade;
create user demo identified by demo default tablespace demo;
[oracle@primary ~]$ impdp system/oracle directory=EXPDP dumpfile=20151228expnew.dmp schemas=new remap_schema=new:demo remap_tablespace='new:demo,T:demo,T1:demo,T2:demo,T3:demo,T4:demo' logfile=20151228expnew.log
Processing object type SCHEMA_EXPORT/TABLE/TABLE_DATA
. . imported "DEMO"."PF_AGREEMENT" 0 KB 0 rows
. . imported "DEMO"."PF_MEMBER_BANKCARD" 0 KB 0 rows
. . imported "DEMO"."PF_MEMBER_CORPORATION" 0 KB 0 rows
. . imported "DEMO"."PF_MEMBER_LINKINFO" 0 KB 0 rows
. . imported "DEMO"."TABLE2":"T_TEST_201503" 0 KB 0 rows
Processing object type SCHEMA_EXPORT/TABLE/COMMENT
ORA-39082: Object type ALTER_PROCEDURE:"DEMO"."P_BALANCE_ONLINE_EXPORT" created with compilation warnings
ORA-39082: Object type ALTER_PROCEDURE:"DEMO"."P_CARD_ENCRYPT" created with compilation warnings
Job "SYSTEM"."SYS_IMPORT_SCHEMA_01" completed with 3 error(s) at Mon Dec 28 11:02:12 2015 elapsed 0 00:00:18
这么 调整之后，就可以将多个表空间的内容导入到一个表空间中了。
总结：
如果原schema中 包含了分区表，在进行导出的时候，是可以正常导出的。
使用 schemas=new remap_schema=new:demo remap_tablespace='new:demo,T:demo,T1:demo,T2:demo,T3:demo,T4:demo' 可以将一个schema中分布在不同多个表空间中的表 对应到新的schema下的 一个表空间中。
建议，在进行导出之前，最好先查一下，需要导出的schema中是否有分区表：
select table_name,PARTITION_NAME,TABLESPACE_NAME from USER_TAB_PARTITIONS order by table_name;
建议：在进行导入之前，如果确定包含分区表，最好是创建根之前对象的表空间一样名称的表空间。
如果不创建，或者创建对应表空间的话，使用
remap_tablespace='new:demo,T:T_new,T1:T1_new,T2:T2_new,T3:T3_new,T4:T4_new' 方式来对应---创建后依然是分区表
如果加上 partition_options=merge 选项后 变成了一个普通表
/**********分区表的导出 table level：
导出整个表
export NLS_LANG=AMERICAN_AMERICA.ZHS16GBK
expdp system/oracle DIRECTORY=EXPDP DUMPFILE=table2full.dmp TABLES=new.table2 logfile=table2full.log tables参数格式为user.table
导出一个分区
expdp system/oracle DIRECTORY=EXPDP DUMPFILE=table2one.dmp TABLES='new.table2:T_TEST_201510' logfile=table2one.log
导出多个分区
expdp system/oracle DIRECTORY=EXPDP DUMPFILE=table2some.dmp TABLES='new.table2:T_TEST_201510,new.table2:T_TEST_201412,new.table2:T_TEST_201503' logfile=table2some.log
expdp system/oracle DIRECTORY=EXPDP DUMPFILE=table2someo.dmp TABLES='new.table2:T_%' logfile=table2someo.log 使用通配符 跟exp imp的用法一致
导出部分数据
select * from table2 where to_char(SUB_DATE,'YYYY')>2014;
expdp system/oracle DIRECTORY=EXPDP DUMPFILE=table2data.dmp TABLES='new.table2' logfile=table2data.log query=\"where to_char\(SUB_DATE,\'YYYY\'\)\>2014\"
/**********Impdp 导入 schemas level:
如果分区表的表空间不存在
impdp system/oracle directory=EXPDP dumpfile=20151228expnew.dmp schemas=new remap_schema=new:demo remap_tablespace=new:demo logfile=20151228expnew.log 会不导入该表，其他表可以导入
如果目标库上分区表的表空间 跟 源数据库上分区表的名称 创建的一致
impdp system/oracle directory=EXPDP dumpfile=20151228expnew.dmp schemas=new remap_schema=new:demo remap_tablespace=new:demo logfile=20151228expnew.log 可以全部导入
将分区表聚合到一个表空间
impdp system/oracle directory=EXPDP dumpfile=20151228expnew.dmp schemas=new remap_schema=new:demo remap_tablespace='new:demo,T:demo,T1:demo,T2:demo,T3:demo,T4:demo' logfile=20151228expnew.log
/**********impdp 导入分区表 table level:
分区表不存在
drop table table2 purge;
impdp system/oracle directory=EXPDP dumpfile=20151228expnew.dmp schemas=new remap_schema=new:demo remap_tablespace=new:demo tables='new.table2' logfile=20151228impnew.table2.log
[oracle@primary ~]$ impdp system/oracle directory=EXPDP dumpfile=20151228expnew.dmp schemas=new remap_schema=new:demo remap_tablespace=new:demo tables=table2 logfile=20151228impnew.table2.log
Import: Release 11.2.0.4.0 - Production on Mon Dec 28 13:02:36 2015
Copyright (c) 1982, 2011, Oracle and/or its affiliates. All rights reserved.
Connected to: Oracle Database 11g Enterprise Edition Release 11.2.0.4.0 - 64bit Production
With the Partitioning, Automatic Storage Management, OLAP, Data Mining
and Real Application Testing options
UDI-00010: multiple job modes requested, schema and tables.
不支持 schemas 和 tables 同时存在
这样解决
源分区表 聚合到一个表空间
impdp system/oracle directory=EXPDP dumpfile=20151228expnew.dmp remap_schema=new:demo remap_tablespace='new:demo,T:demo,T1:demo,T2:demo,T3:demo,T4:demo' tables=new.table2 logfile=20151228impnew.table2.log
说明 不设置参数partition_options 这样聚合之后，demo.table2 依然是一个分区表，只不过所有的数据都在一个表空间中。partition_options 默认是none的。
impdp system/oracle directory=EXPDP dumpfile=20151228expnew.dmp remap_schema=new:demo remap_tablespace='new:demo,T:demo,T1:demo,T2:demo,T3:demo,T4:demo' tables=new.table2 partition_options=merge logfile=20151228impnew.table2.log
说明 加入参数 partition_options=merge 选项后demo.table2 变成了一个普通表。
补充
partition_options=departition 这个参数是将各个分区 变成一个新的表。
impdp system/oracle directory=EXPDP dumpfile=20151228expnew.dmp remap_schema=new:demo remap_tablespace='new:demo,T:demo,T1:demo,T2:demo,T3:demo,T4:demo' tables=new.table2 partition_options=departition logfile=20151228impnew.table2.log
SQL> select table_name from user_tables;
TABLE_NAME
------------------------------
TABLE2_T_TEST_MAX
TABLE2_T_TEST_201510
TABLE2_T_TEST_201503
TABLE2_T_TEST_201412
TABLE2_T_TEST_201406
源分区表 和 目标分区表 相同表空间
1、先创建 表空间
查看分区表空间的对象关系 使用如下语句 在源数据库上查询：
select TABLE_NAME,PARTITION_NAME,TABLESPACE_NAME from USER_TAB_PARTITIONS order by table_name;
注意，分区表所依赖的表空间 并不单单指的是 分区的表空间，还包含 那个母表的表空间。
创建语句
##create tablespace new datafile '+DATA' size 200M; -----本例中，将new 表空间 remap_tablespace 到demo表空间中
create tablespace T datafile '+DATA' size 20M;
create tablespace T1 datafile '+DATA' size 20M;
create tablespace T2 datafile '+DATA' size 20M;
create tablespace T3 datafile '+DATA' size 20M;
create tablespace T4 datafile '+DATA' size 20M;
2、创建分区表(11g可不创建)
SELECT DBMS_METADATA.GET_DDL('TABLE',u.table_name) FROM USER_TABLES u where u.table_name='&name'; 获得DDL创建 需要先创建表空间
或者
从dump文件中获得 biao table2的DDL语句
impdp system/oracle directory=EXPDP dumpfile=20151228expnew.dmp sqlfile=EXPDP:table2_ddl.sql tables=new.table2
3、导入表
impdp system/oracle directory=EXPDP dumpfile=20151228expnew.dmp remap_schema=new:demo remap_tablespace=new:demo tables=new.table2 logfile=20151228impnew.table2_sametbs.log
源分区表 和 目标分区表 不同表空间
1、先创建 表空间
创建语句
##create tablespace new datafile '+DATA' size 200M; -----本例中，将new 表空间 remap_tablespace 到demo表空间中
create tablespace TN datafile '+DATA' size 20M;
create tablespace TN1 datafile '+DATA' size 20M;
create tablespace TN2 datafile '+DATA' size 20M;
create tablespace TN3 datafile '+DATA' size 20M;
create tablespace TN4 datafile '+DATA' size 20M;
2、创建分区表(11g可不创建)
SELECT DBMS_METADATA.GET_DDL('TABLE',u.table_name) FROM USER_TABLES u where u.table_name='&name'; 获得DDL创建 需要先创建表空间
或者
从dump文件中获得 biao table2的DDL语句
impdp system/oracle directory=EXPDP dumpfile=20151228expnew.dmp sqlfile=EXPDP:table2_ddl.sql tables=new.table2
3、导入表
impdp system/oracle directory=EXPDP dumpfile=20151228expnew.dmp remap_schema=new:demo remap_tablespace='new:demo,T:TN,T1:TN1,T2:TN2,T3:TN3,T4:TN4' tables=new.table2 logfile=20151228impnew.table2_sametbs_n.log
如果只是导 入分区的话
impdp system/oracle directory=EXPDP dumpfile=20151228expnew.dmp remap_schema=new:demo remap_tablespace='new:demo,T:TN,T1:TN1,T2:TN2,T3:TN3,T4:TN4' tables='new.table2:T_TEST_201503,new.table2:T_TEST_201412' logfile=20151228impnew.table2_pn.log
impdp system/oracle directory=EXPDP dumpfile=20151228expnew.dmp remap_schema=new:demo remap_tablespace='new:demo,T:TN,T1:TN1,T2:TN2,T3:TN3,T4:TN4' tables='new.table2:T_TEST_%' logfile=20151228impnew.table2_pn.log
impdp system/oracle directory=EXPDP dumpfile=20151228expnew.dmp remap_schema=new:demo remap_tablespace=new:demo tables='new.table2:T_TEST_%' logfile=20151228impnew.table2_sametbs.log
/****************/将原来在一个表空间中的分区表，导入为每个分区一个表空间的表，目标也在一个新的表空间中
原来表table3
SQL> select * from table3;
TABLE_ID SUB_DATE VALUE
---------- ------------ ----------
1 01-APR-14 11
1 01-MAY-14 11
1 01-JUN-14 11
1 01-JUL-14 11
1 01-AUG-14 11
1 01-AUG-14 11
1 01-SEP-14 11
1 01-OCT-14 11
1 01-NOV-14 11
1 01-JUN-15 11
1 01-JUL-15 11
1 01-AUG-15 11
1 01-SEP-15 11
1 01-OCT-15 11
1 01-NOV-15 11
1 01-DEC-15 11
1 01-JUN-16 11
目的 将table3迁移成分区表
方案一 手动先创建 目标分区表（表空间是必须要先创建好的）
1、创建一个分区表 demo.table3 on 目标库
select dbms_metadata.get_ddl('TABLE','TABLE3','DEMO') from dual;
##create tablespace demo datafile '+DATA' size 200M; -----本例中，将 demo 表空间为默认表空间 不需要再次创建
create tablespace T datafile '+DATA' size 20M;
create tablespace T1 datafile '+DATA' size 20M;
create tablespace T2 datafile '+DATA' size 20M;
create tablespace T3 datafile '+DATA' size 20M;
create tablespace T4 datafile '+DATA' size 20M;
conn demo/demo
create table table3 (
TABLE_ID NUMBER(8),
SUB_DATE DATE,
VALUE NUMBER(8)
)
PARTITION BY RANGE (SUB_DATE)
(
PARTITION t_test_201406 VALUES LESS THAN (TO_DATE('2014-06-01 00:00:00','yyyy-mm-ddhh24:mi:ss')) TABLESPACE T1,
PARTITION t_test_201412 VALUES LESS THAN (TO_DATE('2014-12-01 00:00:00','yyyy-mm-ddhh24:mi:ss')) TABLESPACE T2,
PARTITION t_test_201503 VALUES LESS THAN (TO_DATE('2015-03-01 00:00:00','yyyy-mm-ddhh24:mi:ss')) TABLESPACE T3,
PARTITION t_test_201510 VALUES LESS THAN (TO_DATE('2015-10-01 00:00:00','yyyy-mm-ddhh24:mi:ss')) TABLESPACE T4,
PARTITION t_test_max VALUES LESS THAN (MAXVALUE) TABLESPACE T
) ;
----注意 parttion是在 T%表空上的，但是 table3是在DEMO表空间中的。 可以创建新的分区名称 以及重新定义表空间的名称
2、导入
impdp system/oracle directory=EXPDP dumpfile=20151228expnewtab3.dmp remap_schema=new:demo tables='new.table3' table_exists_action=append logfile=20151228impnew.table3.log
[oracle@primary expdp]$ impdp system/oracle directory=EXPDP dumpfile=20151228expnewtab3.dmp remap_schema=new:demo tables='new.table3' table_exists_action=append logfile=20151228impnew.table3.log
Import: Release 11.2.0.4.0 - Production on Mon Dec 28 15:11:02 2015
Copyright (c) 1982, 2011, Oracle and/or its affiliates. All rights reserved.
Connected to: Oracle Database 11g Enterprise Edition Release 11.2.0.4.0 - 64bit Production
With the Partitioning, Automatic Storage Management, OLAP, Data Mining
and Real Application Testing options
Master table "SYSTEM"."SYS_IMPORT_TABLE_01" successfully loaded/unloaded
Starting "SYSTEM"."SYS_IMPORT_TABLE_01": system/******** directory=EXPDP dumpfile=20151228expnewtab3.dmp remap_schema=new:demo tables=new.table3 table_exists_action=append logfile=20151228impnew.table3.log
Processing object type SCHEMA_EXPORT/TABLE/TABLE
Table "DEMO"."TABLE3" exists. Data will be appended to existing table but all dependent metadata will be skipped due to table_exists_action of append
Processing object type SCHEMA_EXPORT/TABLE/TABLE_DATA
. . imported "DEMO"."TABLE3" 6.195 KB 17 rows
Job "SYSTEM"."SYS_IMPORT_TABLE_01" successfully completed at Mon Dec 28 15:11:08 2015 elapsed 0 00:00:05
select * from table3 PARTITION(t_test_201406);
SQL> select * from table3 PARTITION(t_test_201406);
TABLE_ID SUB_DATE VALUE
---------- ------------ ----------
1 01-APR-14 11
1 01-MAY-14 11
3、导入到不同的表名称中
(1)创建表
create table tableN (
TABLE_ID NUMBER(8),
SUB_DATE DATE,
VALUE NUMBER(8)
)
PARTITION BY RANGE (SUB_DATE)
(
PARTITION t_test_1 VALUES LESS THAN (TO_DATE('2014-06-01 00:00:00','yyyy-mm-ddhh24:mi:ss')) TABLESPACE T1,
PARTITION t_test_2 VALUES LESS THAN (TO_DATE('2014-12-01 00:00:00','yyyy-mm-ddhh24:mi:ss')) TABLESPACE T2,
PARTITION t_test_3 VALUES LESS THAN (TO_DATE('2015-03-01 00:00:00','yyyy-mm-ddhh24:mi:ss')) TABLESPACE T3,
PARTITION t_test_4 VALUES LESS THAN (TO_DATE('2015-10-01 00:00:00','yyyy-mm-ddhh24:mi:ss')) TABLESPACE T4,
PARTITION t_test_max VALUES LESS THAN (MAXVALUE) TABLESPACE T
) ;
(2) 导入
impdp system/oracle directory=EXPDP dumpfile=20151228expnewtab3.dmp remap_schema=new:demo tables='new.table3' remap_tablespace='new:demo' remap_table='new.table3:tablen' table_exists_action=append logfile=20151228impnew.table3.log
[oracle@primary expdp]$ impdp system/oracle directory=EXPDP dumpfile=20151228expnewtab3.dmp remap_schema=new:demo tables='new.table3' remap_tablespace='new:demo' remap_table='new.table3:tablen' table_exists_action=append logfile=20151228impnew.table3.log
Import: Release 11.2.0.4.0 - Production on Mon Dec 28 15:33:42 2015
Copyright (c) 1982, 2011, Oracle and/or its affiliates. All rights reserved.
Connected to: Oracle Database 11g Enterprise Edition Release 11.2.0.4.0 - 64bit Production
With the Partitioning, Automatic Storage Management, OLAP, Data Mining
and Real Application Testing options
Master table "SYSTEM"."SYS_IMPORT_TABLE_01" successfully loaded/unloaded
Starting "SYSTEM"."SYS_IMPORT_TABLE_01": system/******** directory=EXPDP dumpfile=20151228expnewtab3.dmp remap_schema=new:demo tables=new.table3 remap_tablespace=new:demo remap_table=new.table3:tablen table_exists_action=append logfile=20151228impnew.table3.log
Processing object type SCHEMA_EXPORT/TABLE/TABLE
Table "DEMO"."TABLEN" exists. Data will be appended to existing table but all dependent metadata will be skipped due to table_exists_action of append
Processing object type SCHEMA_EXPORT/TABLE/TABLE_DATA
. . imported "DEMO"."TABLEN" 6.195 KB 17 rows
Job "SYSTEM"."SYS_IMPORT_TABLE_01" successfully completed at Mon Dec 28 15:33:44 2015 elapsed 0 00:00:02
注意 remap_table='new.table3:tablen' 这里 不要再tablen之前加用户名称，否则会创建 “username.tablen” 这样一个表
select * from tableN PARTITION(t_test_1);
SQL> select * from tableN PARTITION(t_test_1);
TABLE_ID SUB_DATE VALUE
---------- ------------ ----------
1 01-APR-14 11
1 01-MAY-14 11
注意：
由于约束是系统名称，导致在执行REMAP_TABLE导入后，同样的约束在源表上重复建立
/**********impdp 分区表 追加:
1、按分区追加
清理分区：
alter table test truncate partiton a;
ALTER TABLE yourTable DROP PARTITION partionName1;
alter table table1 DROP PARTITION SYS_P41;
增加第一个分区
impdp system/oracle directory=EXPDP dumpfile=20151228expnew.dmp remap_schema=new:demo remap_tablespace='new:demo' tables='new.table1:P1' logfile=20151228impnew.table1_1.log
追加后面的分区
----impdp system/oracle directory=EXPDP dumpfile=20151228expnew.dmp remap_schema=new:demo remap_tablespace='new:demo' tables='new.table1:P1' table_exists_action=append参数 logfile=20151228impnew.table1_1.log
impdp system/oracle directory=EXPDP dumpfile=20151228expnew.dmp remap_schema=new:demo remap_tablespace='new:demo' tables='new.table1:SYS_P41' table_exists_action=append参数 logfile=20151228impnew.table1_SYS_P41.log 切记，table_exists_action的参数 尽量不要使用replace 和 truncate
append 是为表增加数据；
truncate 是截断表，然后为其增加新数据；
replace 是删除已存在表，重新建表并追加数据。
或者
impdp system/oracle directory=EXPDP dumpfile=20151228expnew.dmp remap_schema=new:demo remap_tablespace='new:demo' tables='new.table1:SYS_P%' table_exists_action=append参数 logfile=20151228impnew.table1_SYS_Pn.log
参考：
exp/imp,expdp/impdp 与分区表有关的知识：使用exp -help 查看：
example: imp scott/tiger ignore=y tables=(emp,dept) full=n
or tables=(t1:p1,t1:p2), if t1 is partitioned table
example: exp scott/tiger grants=y tables=(emp,dept,mgr)
or tables=(t1:p1,t1:p2), if t1 is partitioned table
example: expdp scott/tiger dumpfile=scott.dmp directory=dmpdir schemas=scott or tables=(t1:p1,t1:p2), if t1 is partitioned table
example: impdp scott/tiger directory=dmpdir dumpfile=scott.dmp
从转储文件中提取DDL。
如：impdp system/oracle directory=EXPDP dumpfile=20151228expnew.dmp sqlfile=EXPDP:table22.sql schemas=new 就是指将属于 new模式的所有DDL写入位于EXPDP的table22.sql文件中.
Expdp导出分区表的数据文件再Impdp导入到没分区的表的操作方法
Article_Detail是个按时间的分区表:
导出的Bat如下：假设这个表有四个分区，且每个分区的NUM_ROWs至少有20万记录，导出参数tables=Article_Detail 表示每个分区都会导出20万条记录，实则导出的记录为80万。
expdp test/1234@orcl_test query=article_detail:\"where rownum<=200000 order by Article_Detail_ID desc\" tables=Article_Detail directory=TEST_DUMP_DIR dumpfile=Article_Detail_20W.dmp logfile=Article_Detail_20W.log compression=DATA_ONLY content=DATA_ONLY parallel=2
导入：从分区表导出的文件里导入到非分区表，在导入的批处理里要有partition_options参数，partition_options=merge表示将四个分区的数据合并在一起。因为分区表中对应多个表空间，所以要remap_tablespace一下，导入脚本如下：
impdp test_2/1234@orcl_test2 directory=TEST_DUMP_DIR remap_schema=test:test_2 dumpfile=Article_Detail_20W.dmp logfile=Impdp_Article_Detail_20W.log table_exists_action=APPEND data_options=SKIP_CONSTRAINT_ERRORS parallel=2 Exclude=Trigger partition_options=merge
数据库版本：
Oracle 11.2.0.4
当前有两个分区表
TABLE_NAME PARTITION_NAME TABLESPACE_NAME
------------------------------ ------------------------------ ------------------------------
TABLE1 SYS_P57 NEW
TABLE1 SYS_P66NEW
TABLE2 T_TEST_201503 T3
TABLE2 T_TEST_MAX T
TABLE2 T_TEST_201510 T4
TABLE2 T_TEST_201412 T2
TABLE2 T_TEST_201406 T1
21 rows selected.
SQL> l
1* select TABLE_NAME,PARTITION_NAME,TABLESPACE_NAME from USER_TAB_PARTITIONS order by table_name
当前有两个分区表
table1 是创建在一个表空间中的分区表，有多个分区。
table2 是将有多个分区表空间的分区表，多个分区对应多个表空间。
涉及的表空间
new、T、T1、T2、T3、T4
涉及的用户
new
/**********对new用户进行导出，再远端数据库中导入 schema level
create directory EXPDP as '/backup/expdp';
grant read,write on directory EXPDP to system;
导出整个 new schema:
export NLS_LANG=AMERICAN_AMERICA.ZHS16GBK
expdp system/oracle directory=EXPDP schemas=new dumpfile=20151228expnew.dmp logfile=20151228expnew.log
expdp system/oracle directory=EXPDP schemas=new dumpfile=20151228expnewtab3.dmp logfile=20151228expnewtab3.log
. . exported "NEW"."PF_MEMBER_BANKCARD" 0 KB 0 rows
. . exported "NEW"."PF_MEMBER_CORPORATION" 0 KB 0 rows
. . exported "NEW"."PF_MEMBER_LINKINFO" 0 KB 0 rows
. . exported "NEW"."TABLE2":"T_TEST_201503" 0 KB 0 rows
Master table "SYSTEM"."SYS_EXPORT_SCHEMA_01" successfully loaded/unloaded
******************************************************************************
Dump file set for SYSTEM.SYS_EXPORT_SCHEMA_01 is:
/backup/expdp/20151228expnew.dmp
Job "SYSTEM"."SYS_EXPORT_SCHEMA_01" successfully completed at Mon Dec 28 10:25:40 2015 elapsed 0 00:01:21
目标库导入new schema:
create directory EXPDP as '/backup/expdp';
grant read,write on directory EXPDP to system;
export NLS_LANG=AMERICAN_AMERICA.ZHS16GBK
impdp system/oracle directory=EXPDP dumpfile=20151228expnew.dmp schemas=new remap_schema=new:demo remap_tablespace=new:demo logfile=20151228expnew.log
[oracle@primary ~]$ export NLS_LANG=AMERICAN_AMERICA.ZHS16GBK
[oracle@primary ~]$ impdp system/oracle directory=EXPDP dumpfile=20151228expnew.dmp schemas=new remap_schema=new:demo remap_tablespace=new:demo logfile=20151228expnew.log
ORA-39171: Job is experiencing a resumable wait.
ORA-01658: unable to create INITIAL extent for segment in tablespace DEMO
ORA-39083: Object type TABLE:"DEMO"."TABLE2" failed to create with error:
ORA-00959: tablespace 'T1' does not exist
Failing sql is:
CREATE TABLE "DEMO"."TABLE2" ("TABLE_ID" NUMBER(8,0), "SUB_DATE" DATE, "VALUE" NUMBER(8,0)) PCTFREE 10 PCTUSED 40 INITRANS 1 MAXTRANS 255 STORAGE( BUFFER_POOL DEFAULT FLASH_CACHE DEFAULT CELL_FLASH_CACHE DEFAULT) TABLESPACE "DEMO" PARTITION BY RANGE ("SUB_DATE") (PARTITION "T_TEST_201406" VALUES LESS THAN (TO_DATE(' 2014-06-01 00:00:00', 'SYYYY-MM-DD HH24:MI:SS', 'NLS_CALEN
Processing object type SCHEMA_EXPORT/TABLE/TABLE_DATA
. . imported "DEMO"."PF_AGREEMENT" 0 KB 0 rows
. . imported "DEMO"."PF_MEMBER_BANKCARD" 0 KB 0 rows
. . imported "DEMO"."PF_MEMBER_CORPORATION" 0 KB 0 rows
. . imported "DEMO"."PF_MEMBER_LINKINFO" 0 KB 0 rows
Job "SYSTEM"."SYS_IMPORT_SCHEMA_01" completed with 5 error(s) at Mon Dec 28 10:55:08 2015 elapsed 0 00:07:31
因为 没有对应的表空间，所以导入失败。
SQL> select * from table1;
TABLE_ID SUB_DATE VALUE
---------- ------------ ----------
1 01-APR-14 11
1 01-MAY-14 11
1 01-JUN-14 11
1 01-JUL-14 11
SQL> select * from table2;
select * from table2
*
ERROR at line 1:
ORA-00942: table or view does not exist
换个方法试试 将多个表空间的内容 导入到一个表空间。适用于 一个scheman中有多个表，分布在不同的表空间中，最终汇总到一个表空间。
参考：
impdp ctais3/oracle DIRECTORY=DATA_PUMP_DIR dumpfile=ctais2.dmp remap_schema=ctais2:ctais3 remap_tablespace='(EXAMPLE:ZHANGQIAOC,USERS:ZHANGQIAOC)'
impdp system/oracle directory=EXPDP dumpfile=20151228expnew.dmp schemas=new remap_schema=new:demo remap_tablespace='new:demo,T:demo,T1:demo,T2:demo,T3:demo,T4:demo' logfile=20151228expnew.log
drop user demo cascade;
create user demo identified by demo default tablespace demo;
[oracle@primary ~]$ impdp system/oracle directory=EXPDP dumpfile=20151228expnew.dmp schemas=new remap_schema=new:demo remap_tablespace='new:demo,T:demo,T1:demo,T2:demo,T3:demo,T4:demo' logfile=20151228expnew.log
Processing object type SCHEMA_EXPORT/TABLE/TABLE_DATA
. . imported "DEMO"."PF_AGREEMENT" 0 KB 0 rows
. . imported "DEMO"."PF_MEMBER_BANKCARD" 0 KB 0 rows
. . imported "DEMO"."PF_MEMBER_CORPORATION" 0 KB 0 rows
. . imported "DEMO"."PF_MEMBER_LINKINFO" 0 KB 0 rows
. . imported "DEMO"."TABLE2":"T_TEST_201503" 0 KB 0 rows
Processing object type SCHEMA_EXPORT/TABLE/COMMENT
ORA-39082: Object type ALTER_PROCEDURE:"DEMO"."P_BALANCE_ONLINE_EXPORT" created with compilation warnings
ORA-39082: Object type ALTER_PROCEDURE:"DEMO"."P_CARD_ENCRYPT" created with compilation warnings
Job "SYSTEM"."SYS_IMPORT_SCHEMA_01" completed with 3 error(s) at Mon Dec 28 11:02:12 2015 elapsed 0 00:00:18
这么 调整之后，就可以将多个表空间的内容导入到一个表空间中了。
总结：
如果原schema中 包含了分区表，在进行导出的时候，是可以正常导出的。
使用 schemas=new remap_schema=new:demo remap_tablespace='new:demo,T:demo,T1:demo,T2:demo,T3:demo,T4:demo' 可以将一个schema中分布在不同多个表空间中的表 对应到新的schema下的 一个表空间中。
建议，在进行导出之前，最好先查一下，需要导出的schema中是否有分区表：
select table_name,PARTITION_NAME,TABLESPACE_NAME from USER_TAB_PARTITIONS order by table_name;
建议：在进行导入之前，如果确定包含分区表，最好是创建根之前对象的表空间一样名称的表空间。
如果不创建，或者创建对应表空间的话，使用
remap_tablespace='new:demo,T:T_new,T1:T1_new,T2:T2_new,T3:T3_new,T4:T4_new' 方式来对应---创建后依然是分区表
如果加上 partition_options=merge 选项后 变成了一个普通表
/**********分区表的导出 table level：
导出整个表
export NLS_LANG=AMERICAN_AMERICA.ZHS16GBK
expdp system/oracle DIRECTORY=EXPDP DUMPFILE=table2full.dmp TABLES=new.table2 logfile=table2full.log tables参数格式为user.table
导出一个分区
expdp system/oracle DIRECTORY=EXPDP DUMPFILE=table2one.dmp TABLES='new.table2:T_TEST_201510' logfile=table2one.log
导出多个分区
expdp system/oracle DIRECTORY=EXPDP DUMPFILE=table2some.dmp TABLES='new.table2:T_TEST_201510,new.table2:T_TEST_201412,new.table2:T_TEST_201503' logfile=table2some.log
expdp system/oracle DIRECTORY=EXPDP DUMPFILE=table2someo.dmp TABLES='new.table2:T_%' logfile=table2someo.log 使用通配符 跟exp imp的用法一致
导出部分数据
select * from table2 where to_char(SUB_DATE,'YYYY')>2014;
expdp system/oracle DIRECTORY=EXPDP DUMPFILE=table2data.dmp TABLES='new.table2' logfile=table2data.log query=\"where to_char\(SUB_DATE,\'YYYY\'\)\>2014\"
/**********Impdp 导入 schemas level:
如果分区表的表空间不存在
impdp system/oracle directory=EXPDP dumpfile=20151228expnew.dmp schemas=new remap_schema=new:demo remap_tablespace=new:demo logfile=20151228expnew.log 会不导入该表，其他表可以导入
如果目标库上分区表的表空间 跟 源数据库上分区表的名称 创建的一致
impdp system/oracle directory=EXPDP dumpfile=20151228expnew.dmp schemas=new remap_schema=new:demo remap_tablespace=new:demo logfile=20151228expnew.log 可以全部导入
将分区表聚合到一个表空间
impdp system/oracle directory=EXPDP dumpfile=20151228expnew.dmp schemas=new remap_schema=new:demo remap_tablespace='new:demo,T:demo,T1:demo,T2:demo,T3:demo,T4:demo' logfile=20151228expnew.log
/**********impdp 导入分区表 table level:
分区表不存在
drop table table2 purge;
impdp system/oracle directory=EXPDP dumpfile=20151228expnew.dmp schemas=new remap_schema=new:demo remap_tablespace=new:demo tables='new.table2' logfile=20151228impnew.table2.log
[oracle@primary ~]$ impdp system/oracle directory=EXPDP dumpfile=20151228expnew.dmp schemas=new remap_schema=new:demo remap_tablespace=new:demo tables=table2 logfile=20151228impnew.table2.log
Import: Release 11.2.0.4.0 - Production on Mon Dec 28 13:02:36 2015
Copyright (c) 1982, 2011, Oracle and/or its affiliates. All rights reserved.
Connected to: Oracle Database 11g Enterprise Edition Release 11.2.0.4.0 - 64bit Production
With the Partitioning, Automatic Storage Management, OLAP, Data Mining
and Real Application Testing options
UDI-00010: multiple job modes requested, schema and tables.
不支持 schemas 和 tables 同时存在
这样解决
源分区表 聚合到一个表空间
impdp system/oracle directory=EXPDP dumpfile=20151228expnew.dmp remap_schema=new:demo remap_tablespace='new:demo,T:demo,T1:demo,T2:demo,T3:demo,T4:demo' tables=new.table2 logfile=20151228impnew.table2.log
说明 不设置参数partition_options 这样聚合之后，demo.table2 依然是一个分区表，只不过所有的数据都在一个表空间中。partition_options 默认是none的。
impdp system/oracle directory=EXPDP dumpfile=20151228expnew.dmp remap_schema=new:demo remap_tablespace='new:demo,T:demo,T1:demo,T2:demo,T3:demo,T4:demo' tables=new.table2 partition_options=merge logfile=20151228impnew.table2.log
说明 加入参数 partition_options=merge 选项后demo.table2 变成了一个普通表。
补充
partition_options=departition 这个参数是将各个分区 变成一个新的表。
impdp system/oracle directory=EXPDP dumpfile=20151228expnew.dmp remap_schema=new:demo remap_tablespace='new:demo,T:demo,T1:demo,T2:demo,T3:demo,T4:demo' tables=new.table2 partition_options=departition logfile=20151228impnew.table2.log
SQL> select table_name from user_tables;
TABLE_NAME
------------------------------
TABLE2_T_TEST_MAX
TABLE2_T_TEST_201510
TABLE2_T_TEST_201503
TABLE2_T_TEST_201412
TABLE2_T_TEST_201406
源分区表 和 目标分区表 相同表空间
1、先创建 表空间
查看分区表空间的对象关系 使用如下语句 在源数据库上查询：
select TABLE_NAME,PARTITION_NAME,TABLESPACE_NAME from USER_TAB_PARTITIONS order by table_name;
注意，分区表所依赖的表空间 并不单单指的是 分区的表空间，还包含 那个母表的表空间。
创建语句
##create tablespace new datafile '+DATA' size 200M; -----本例中，将new 表空间 remap_tablespace 到demo表空间中
create tablespace T datafile '+DATA' size 20M;
create tablespace T1 datafile '+DATA' size 20M;
create tablespace T2 datafile '+DATA' size 20M;
create tablespace T3 datafile '+DATA' size 20M;
create tablespace T4 datafile '+DATA' size 20M;
2、创建分区表(11g可不创建)
SELECT DBMS_METADATA.GET_DDL('TABLE',u.table_name) FROM USER_TABLES u where u.table_name='&name'; 获得DDL创建 需要先创建表空间
或者
从dump文件中获得 biao table2的DDL语句
impdp system/oracle directory=EXPDP dumpfile=20151228expnew.dmp sqlfile=EXPDP:table2_ddl.sql tables=new.table2
3、导入表
impdp system/oracle directory=EXPDP dumpfile=20151228expnew.dmp remap_schema=new:demo remap_tablespace=new:demo tables=new.table2 logfile=20151228impnew.table2_sametbs.log
源分区表 和 目标分区表 不同表空间
1、先创建 表空间
创建语句
##create tablespace new datafile '+DATA' size 200M; -----本例中，将new 表空间 remap_tablespace 到demo表空间中
create tablespace TN datafile '+DATA' size 20M;
create tablespace TN1 datafile '+DATA' size 20M;
create tablespace TN2 datafile '+DATA' size 20M;
create tablespace TN3 datafile '+DATA' size 20M;
create tablespace TN4 datafile '+DATA' size 20M;
2、创建分区表(11g可不创建)
SELECT DBMS_METADATA.GET_DDL('TABLE',u.table_name) FROM USER_TABLES u where u.table_name='&name'; 获得DDL创建 需要先创建表空间
或者
从dump文件中获得 biao table2的DDL语句
impdp system/oracle directory=EXPDP dumpfile=20151228expnew.dmp sqlfile=EXPDP:table2_ddl.sql tables=new.table2
3、导入表
impdp system/oracle directory=EXPDP dumpfile=20151228expnew.dmp remap_schema=new:demo remap_tablespace='new:demo,T:TN,T1:TN1,T2:TN2,T3:TN3,T4:TN4' tables=new.table2 logfile=20151228impnew.table2_sametbs_n.log
如果只是导 入分区的话
impdp system/oracle directory=EXPDP dumpfile=20151228expnew.dmp remap_schema=new:demo remap_tablespace='new:demo,T:TN,T1:TN1,T2:TN2,T3:TN3,T4:TN4' tables='new.table2:T_TEST_201503,new.table2:T_TEST_201412' logfile=20151228impnew.table2_pn.log
impdp system/oracle directory=EXPDP dumpfile=20151228expnew.dmp remap_schema=new:demo remap_tablespace='new:demo,T:TN,T1:TN1,T2:TN2,T3:TN3,T4:TN4' tables='new.table2:T_TEST_%' logfile=20151228impnew.table2_pn.log
impdp system/oracle directory=EXPDP dumpfile=20151228expnew.dmp remap_schema=new:demo remap_tablespace=new:demo tables='new.table2:T_TEST_%' logfile=20151228impnew.table2_sametbs.log
/****************/将原来在一个表空间中的分区表，导入为每个分区一个表空间的表，目标也在一个新的表空间中
原来表table3
SQL> select * from table3;
TABLE_ID SUB_DATE VALUE
---------- ------------ ----------
1 01-APR-14 11
1 01-MAY-14 11
1 01-JUN-14 11
1 01-JUL-14 11
1 01-AUG-14 11
1 01-AUG-14 11
1 01-SEP-14 11
1 01-OCT-14 11
1 01-NOV-14 11
1 01-JUN-15 11
1 01-JUL-15 11
1 01-AUG-15 11
1 01-SEP-15 11
1 01-OCT-15 11
1 01-NOV-15 11
1 01-DEC-15 11
1 01-JUN-16 11
目的 将table3迁移成分区表
方案一 手动先创建 目标分区表（表空间是必须要先创建好的）
1、创建一个分区表 demo.table3 on 目标库
select dbms_metadata.get_ddl('TABLE','TABLE3','DEMO') from dual;
##create tablespace demo datafile '+DATA' size 200M; -----本例中，将 demo 表空间为默认表空间 不需要再次创建
create tablespace T datafile '+DATA' size 20M;
create tablespace T1 datafile '+DATA' size 20M;
create tablespace T2 datafile '+DATA' size 20M;
create tablespace T3 datafile '+DATA' size 20M;
create tablespace T4 datafile '+DATA' size 20M;
conn demo/demo
create table table3 (
TABLE_ID NUMBER(8),
SUB_DATE DATE,
VALUE NUMBER(8)
)
PARTITION BY RANGE (SUB_DATE)
(
PARTITION t_test_201406 VALUES LESS THAN (TO_DATE('2014-06-01 00:00:00','yyyy-mm-ddhh24:mi:ss')) TABLESPACE T1,
PARTITION t_test_201412 VALUES LESS THAN (TO_DATE('2014-12-01 00:00:00','yyyy-mm-ddhh24:mi:ss')) TABLESPACE T2,
PARTITION t_test_201503 VALUES LESS THAN (TO_DATE('2015-03-01 00:00:00','yyyy-mm-ddhh24:mi:ss')) TABLESPACE T3,
PARTITION t_test_201510 VALUES LESS THAN (TO_DATE('2015-10-01 00:00:00','yyyy-mm-ddhh24:mi:ss')) TABLESPACE T4,
PARTITION t_test_max VALUES LESS THAN (MAXVALUE) TABLESPACE T
) ;
----注意 parttion是在 T%表空上的，但是 table3是在DEMO表空间中的。 可以创建新的分区名称 以及重新定义表空间的名称
2、导入
impdp system/oracle directory=EXPDP dumpfile=20151228expnewtab3.dmp remap_schema=new:demo tables='new.table3' table_exists_action=append logfile=20151228impnew.table3.log
[oracle@primary expdp]$ impdp system/oracle directory=EXPDP dumpfile=20151228expnewtab3.dmp remap_schema=new:demo tables='new.table3' table_exists_action=append logfile=20151228impnew.table3.log
Import: Release 11.2.0.4.0 - Production on Mon Dec 28 15:11:02 2015
Copyright (c) 1982, 2011, Oracle and/or its affiliates. All rights reserved.
Connected to: Oracle Database 11g Enterprise Edition Release 11.2.0.4.0 - 64bit Production
With the Partitioning, Automatic Storage Management, OLAP, Data Mining
and Real Application Testing options
Master table "SYSTEM"."SYS_IMPORT_TABLE_01" successfully loaded/unloaded
Starting "SYSTEM"."SYS_IMPORT_TABLE_01": system/******** directory=EXPDP dumpfile=20151228expnewtab3.dmp remap_schema=new:demo tables=new.table3 table_exists_action=append logfile=20151228impnew.table3.log
Processing object type SCHEMA_EXPORT/TABLE/TABLE
Table "DEMO"."TABLE3" exists. Data will be appended to existing table but all dependent metadata will be skipped due to table_exists_action of append
Processing object type SCHEMA_EXPORT/TABLE/TABLE_DATA
. . imported "DEMO"."TABLE3" 6.195 KB 17 rows
Job "SYSTEM"."SYS_IMPORT_TABLE_01" successfully completed at Mon Dec 28 15:11:08 2015 elapsed 0 00:00:05
select * from table3 PARTITION(t_test_201406);
SQL> select * from table3 PARTITION(t_test_201406);
TABLE_ID SUB_DATE VALUE
---------- ------------ ----------
1 01-APR-14 11
1 01-MAY-14 11
3、导入到不同的表名称中
(1)创建表
create table tableN (
TABLE_ID NUMBER(8),
SUB_DATE DATE,
VALUE NUMBER(8)
)
PARTITION BY RANGE (SUB_DATE)
(
PARTITION t_test_1 VALUES LESS THAN (TO_DATE('2014-06-01 00:00:00','yyyy-mm-ddhh24:mi:ss')) TABLESPACE T1,
PARTITION t_test_2 VALUES LESS THAN (TO_DATE('2014-12-01 00:00:00','yyyy-mm-ddhh24:mi:ss')) TABLESPACE T2,
PARTITION t_test_3 VALUES LESS THAN (TO_DATE('2015-03-01 00:00:00','yyyy-mm-ddhh24:mi:ss')) TABLESPACE T3,
PARTITION t_test_4 VALUES LESS THAN (TO_DATE('2015-10-01 00:00:00','yyyy-mm-ddhh24:mi:ss')) TABLESPACE T4,
PARTITION t_test_max VALUES LESS THAN (MAXVALUE) TABLESPACE T
) ;
(2) 导入
impdp system/oracle directory=EXPDP dumpfile=20151228expnewtab3.dmp remap_schema=new:demo tables='new.table3' remap_tablespace='new:demo' remap_table='new.table3:tablen' table_exists_action=append logfile=20151228impnew.table3.log
[oracle@primary expdp]$ impdp system/oracle directory=EXPDP dumpfile=20151228expnewtab3.dmp remap_schema=new:demo tables='new.table3' remap_tablespace='new:demo' remap_table='new.table3:tablen' table_exists_action=append logfile=20151228impnew.table3.log
Import: Release 11.2.0.4.0 - Production on Mon Dec 28 15:33:42 2015
Copyright (c) 1982, 2011, Oracle and/or its affiliates. All rights reserved.
Connected to: Oracle Database 11g Enterprise Edition Release 11.2.0.4.0 - 64bit Production
With the Partitioning, Automatic Storage Management, OLAP, Data Mining
and Real Application Testing options
Master table "SYSTEM"."SYS_IMPORT_TABLE_01" successfully loaded/unloaded
Starting "SYSTEM"."SYS_IMPORT_TABLE_01": system/******** directory=EXPDP dumpfile=20151228expnewtab3.dmp remap_schema=new:demo tables=new.table3 remap_tablespace=new:demo remap_table=new.table3:tablen table_exists_action=append logfile=20151228impnew.table3.log
Processing object type SCHEMA_EXPORT/TABLE/TABLE
Table "DEMO"."TABLEN" exists. Data will be appended to existing table but all dependent metadata will be skipped due to table_exists_action of append
Processing object type SCHEMA_EXPORT/TABLE/TABLE_DATA
. . imported "DEMO"."TABLEN" 6.195 KB 17 rows
Job "SYSTEM"."SYS_IMPORT_TABLE_01" successfully completed at Mon Dec 28 15:33:44 2015 elapsed 0 00:00:02
注意 remap_table='new.table3:tablen' 这里 不要再tablen之前加用户名称，否则会创建 “username.tablen” 这样一个表
select * from tableN PARTITION(t_test_1);
SQL> select * from tableN PARTITION(t_test_1);
TABLE_ID SUB_DATE VALUE
---------- ------------ ----------
1 01-APR-14 11
1 01-MAY-14 11
注意：
由于约束是系统名称，导致在执行REMAP_TABLE导入后，同样的约束在源表上重复建立
/**********impdp 分区表 追加:
1、按分区追加
清理分区：
alter table test truncate partiton a;
ALTER TABLE yourTable DROP PARTITION partionName1;
alter table table1 DROP PARTITION SYS_P41;
增加第一个分区
impdp system/oracle directory=EXPDP dumpfile=20151228expnew.dmp remap_schema=new:demo remap_tablespace='new:demo' tables='new.table1:P1' logfile=20151228impnew.table1_1.log
追加后面的分区
----impdp system/oracle directory=EXPDP dumpfile=20151228expnew.dmp remap_schema=new:demo remap_tablespace='new:demo' tables='new.table1:P1' table_exists_action=append参数 logfile=20151228impnew.table1_1.log
impdp system/oracle directory=EXPDP dumpfile=20151228expnew.dmp remap_schema=new:demo remap_tablespace='new:demo' tables='new.table1:SYS_P41' table_exists_action=append参数 logfile=20151228impnew.table1_SYS_P41.log 切记，table_exists_action的参数 尽量不要使用replace 和 truncate
append 是为表增加数据；
truncate 是截断表，然后为其增加新数据；
replace 是删除已存在表，重新建表并追加数据。
或者
impdp system/oracle directory=EXPDP dumpfile=20151228expnew.dmp remap_schema=new:demo remap_tablespace='new:demo' tables='new.table1:SYS_P%' table_exists_action=append参数 logfile=20151228impnew.table1_SYS_Pn.log
参考：
exp/imp,expdp/impdp 与分区表有关的知识：使用exp -help 查看：
example: imp scott/tiger ignore=y tables=(emp,dept) full=n
or tables=(t1:p1,t1:p2), if t1 is partitioned table
example: exp scott/tiger grants=y tables=(emp,dept,mgr)
or tables=(t1:p1,t1:p2), if t1 is partitioned table
example: expdp scott/tiger dumpfile=scott.dmp directory=dmpdir schemas=scott or tables=(t1:p1,t1:p2), if t1 is partitioned table
example: impdp scott/tiger directory=dmpdir dumpfile=scott.dmp
从转储文件中提取DDL。
如：impdp system/oracle directory=EXPDP dumpfile=20151228expnew.dmp sqlfile=EXPDP:table22.sql schemas=new 就是指将属于 new模式的所有DDL写入位于EXPDP的table22.sql文件中.
Expdp导出分区表的数据文件再Impdp导入到没分区的表的操作方法
Article_Detail是个按时间的分区表:
导出的Bat如下：假设这个表有四个分区，且每个分区的NUM_ROWs至少有20万记录，导出参数tables=Article_Detail 表示每个分区都会导出20万条记录，实则导出的记录为80万。
expdp test/1234@orcl_test query=article_detail:\"where rownum<=200000 order by Article_Detail_ID desc\" tables=Article_Detail directory=TEST_DUMP_DIR dumpfile=Article_Detail_20W.dmp logfile=Article_Detail_20W.log compression=DATA_ONLY content=DATA_ONLY parallel=2
导入：从分区表导出的文件里导入到非分区表，在导入的批处理里要有partition_options参数，partition_options=merge表示将四个分区的数据合并在一起。因为分区表中对应多个表空间，所以要remap_tablespace一下，导入脚本如下：
impdp test_2/1234@orcl_test2 directory=TEST_DUMP_DIR remap_schema=test:test_2 dumpfile=Article_Detail_20W.dmp logfile=Impdp_Article_Detail_20W.log table_exists_action=APPEND data_options=SKIP_CONSTRAINT_ERRORS parallel=2 Exclude=Trigger partition_options=merge