【数据泵】EXPDP导出表结构

【数据泵】EXPDP导出表结构(真实案例)

BLOG文档结构图

因工作需要现需要把一个生产库下的元数据（表定义，索引定义，函数定义，包定义，存储过程）导出到测试库上，本来以为很简单的，可是做的过程发现很多的问题，现记录如下，希望有同样需要的朋友不要再走弯路了。

导读

各位技术爱好者，看完本文后，你可以掌握如下的技能，也可以学到一些其它你所不知道的知识，~O(∩_∩)O~：

① EXPDP和IMPDP如何导出导入元数据，包括表定义，索引定义，函数定义，包定义，存储过程（重点）

② 表的初始化定义参数initial，及如何批量修改该参数

③ 如何导出DMP文件中的DDL语句（重点）

④ 10g和11g默认情况下有哪些用户及其作用

⑤ linux中的批量替换sed命令

⑥ sqlldr和spool命令

本文如有错误或不完善的地方请大家多多指正，ITPUB留言或QQ皆可，您的批评指正是我写作的最大动力。

实验环境介绍

源库：10.2.0.1 AIX

目标库：11.2.0.3 RHEL6.5

执行导出工作
1. 确定需要导出的用户

oracle安装好后有很多的系统默认用户，比如sys和system，对于这2个用户里的元数据我们就没有必要再重新导出嘛，不然导入的时候还提示错误，看着实在不好。

官网信息：

All databases created by the Database Configuration Assistant (DBCA) include the SYS, SYSTEM, SYSMAN, and DBSNMP database accounts. In addition, Oracle Database provides several other administrative accounts. Before using these accounts, you must unlock them and reset their passwords.

11g默认用户比较多：

User Name	Description	See Also
ANONYMOUS	Enables HTTP access to Oracle XML DB.	Oracle XML DB Developer's Guide
APEX_030200	The account owns the Application Express schema and metadata.	Oracle Application Express Application Builder User's Guide
APEX_PUBLIC_USER	The minimally privileged account used for Application Express configuration with Oracle HTTP Server and mod_plsql.	Oracle Application Express Application Builder User's Guide
APPQOSSYS	Used for storing and managing all data and metadata required by Oracle Quality of Service Management.	None
BI	The account that owns the Business Intelligence schema included in the Oracle Sample Schemas. It is available only if you loaded the sample schemas.	Oracle Database Sample Schemas
CTXSYS	The Oracle Text account.	Oracle Text Reference
DBSNMP	The account used by the Management Agent component of Oracle Enterprise Manager to monitor and manage the database.	Oracle Enterprise Manager Grid Control Installation and Basic Configuration
DIP	The account used by the Directory Integration Platform (DIP) to synchronize the changes in Oracle Internet Directory with the applications in the database.	None
DVSYS	There are two roles associated with this account. The Database Vault owner role manages the Database Vault roles and configurations. The Database Vault Account Manager is used to manage database user accounts.	Oracle Database Vault Administrator's Guide

	Note: Part of Oracle Database Vault user interface text is stored in database tables in the DVSYS schema. By default, only the English language is loaded into these tables. You can use Oracle Database Vault Configuration Assistant to add more languages to Oracle Database Vault. For the necessary steps, see Appendix C in Oracle Database Vault Administrator's Guide
EXFSYS	The account owns the Expression Filter schema.	None
FLOWS_FILES	The account owns the Application Express uploaded files.	Oracle Application Express Application Builder User's Guide
HR	The account that owns the Human Resources schema included in the Oracle Sample Schemas. It is available only if you loaded the sample schemas.	Oracle Database Sample Schemas
IX	The account that owns the Information Transport schema included in the Oracle Sample Schemas. It is available only if you loaded the sample schemas.	Oracle Database Sample Schemas
LBACSYS	The Oracle Label Security administrator account.	Oracle Label Security Administrator's Guide
MDDATA	The schema used by Oracle Spatial for storing geocoder and router data.	Oracle Spatial Developer's Guide
MDSYS	The Oracle Spatial and Oracle Multimedia Locator administrator account.	Oracle Spatial Developer's Guide
MGMT_VIEW	An account used by Oracle Enterprise Manager Database Control.	None
OE	The account that owns the Order Entry schema included in the Oracle Sample Schemas. It is available only if you loaded the sample schemas.	Oracle Database Sample Schemas
ORDPLUGINS	The Oracle Multimedia user. Plug-ins supplied by Oracle and third-party plug-ins are installed in this schema.	Oracle Multimedia Reference
ORDSYS	The Oracle Multimedia administrator account.	Oracle Multimedia Reference
ORDDATA	This account contains the Oracle Multimedia DICOM data model.	Oracle Multimedia DICOM Developer's Guide
OUTLN	The account that supports plan stability. Plan stability enables you to maintain the same execution plans for the same SQL statements. OUTLN acts as a role to centrally manage metadata associated with stored outlines.	Oracle Database Concepts
ORACLE_OCM	This account contains the instrumentation for configuration collection used by the Oracle Configuration Manager.	Oracle Configuration Manager Installation and Administration Guide
OWBSYS	The account used by Oracle Warehouse Builder as its default repository. You must unlock this account after installing the Oracle Database and before launching the Warehouse Builder Repository Assistant.	Oracle Warehouse Builder Installation and Administration Guide
OWBSYS_AUDIT	This account is used by the Warehouse Builder Control Center Agent to access the heterogeneous execution audit tables in the OWBSYS schema.	Oracle Warehouse Builder Installation and Administration Guide
PM	The account that owns the Product Media schema included in the Oracle Sample Schemas. It is available only if you loaded the sample schemas.	Oracle Database Sample Schemas
SCOTT	An account used by Oracle sample programs and examples.	Oracle Database Administrator's Guide
SH	The account that owns the Sales History schema included in the Oracle Sample Schemas. It is available only if you loaded the sample schemas during an Enterprise Edition installation.	Oracle Database Administrator's Guide
SI_INFORMTN_SCHEMA	The account that stores the information views for the SQL/MM Still Image Standard.	Oracle Multimedia Reference
SPATIAL_CSW_ADMIN_USR	The Catalog Services for the Web (CSW) account. It is used by the Oracle Spatial CSW cache manager to load all record type metadata, and record instances from the database into the main memory for the record types that are cached.	Oracle Spatial Developer's Guide
SPATIAL_WFS_ADMIN_USR	The Web Feature Service (WFS) account. It is used by the Oracle Spatial WFS cache manager to load all feature-type metadata, and feature instances from the database into main memory for the feature types that are cached.	Oracle Spatial Developer's Guide
SYS	The account used to perform database administration tasks.	Oracle Database Administrator's Guide
SYSMAN	The account used to perform Oracle Enterprise Manager database administration tasks.	Oracle Enterprise Manager Grid Control Installation and Basic Configuration
SYSTEM	Another account used to perform database administration tasks.	Oracle Database Administrator's Guide
WMSYS	The account used to store the metadata information for Oracle Workspace Manager.	Oracle Database Workspace Manager Developer's Guide
XDB	The account used for storing Oracle XML DB data and metadata.	Oracle XML DB Developer's Guide

10g下比较少：

CTXSYS	CTXSYS	The Oracle Text account	Oracle Text Reference
DBSNMP	DBSNMP	The account used by the Management Agent component of Oracle Enterprise Manager to monitor and manage the database	Oracle Enterprise Manager Grid Control Installation and Basic Configuration
LBACSYS	LBACSYS	The Oracle Label Security administrator account	Oracle Label Security Administrator's Guide
MDDATA	MDDATA	The schema used by Oracle Spatial for storing Geocoder and router data	Oracle Spatial User's Guide and Reference
MDSYS	MDSYS	The Oracle Spatial and Oracle interMedia Locator administrator account	Oracle Spatial User's Guide and Reference
DMSYS	DMSYS	The Oracle Data Mining account.	Oracle Data Mining Administrator's Guide

			Oracle Data Mining Concepts
OLAPSYS	MANAGER	The account used to create OLAP metadata structures. It owns the OLAP Catalog (CWMLite).	Oracle OLAP Application Developer's Guide
ORDPLUGINS	ORDPLUGINS	The Oracle interMedia user. Plug-ins supplied by Oracle and third party format plug-ins are installed in this schema.	Oracle interMedia User's Guide
ORDSYS	ORDSYS	The Oracle interMedia administrator account	Oracle interMedia User's Guide
OUTLN	OUTLN	The account that supports plan stability. Plan stability enables you to maintain the same execution plans for the same SQL statements. OUTLN acts as a role to centrally manage metadata associated with stored outlines.	Oracle Database Performance Tuning Guide
SI_INFORMTN_SCHEMA	SI_INFORMTN_SCHEMA	The account that stores the information views for the SQL/MM Still Image Standard	Oracle interMedia User's Guide
SYS	CHANGE_ON_INSTALL	The account used to perform database administration tasks	Oracle Database Administrator's Guide
SYSMAN	CHANGE_ON_INSTALL	The account used to perform Oracle Enterprise Manager database administration tasks. Note that SYS and SYSTEM can also perform these tasks.	Oracle Enterprise Manager Grid Control Installation and Basic Configuration
SYSTEM	MANAGER	Another account used to perform database administration tasks.	Oracle Database Administrator's Guide

确定需要导出的用户在哪些表空间，及其表初始化时占用的表空间大小

SELECT D.tablespace_name, SUM(D.initial_extent)/1024/1024 initial_extent

FROM DBA_SEGMENTS D

WHERE D.owner IN

(SELECT a.username

FROM DBA_USERS A

WHERE A.account_status = 'OPEN'

AND A.username NOT IN

('SYS', 'SYSTEM', 'MGMT_VIEW', 'SYSMAN', 'DBSNMP'))

GROUP BY D.tablespace_name

ORDER BY initial_extent desc ;

SELECT SUM(D.initial_extent)/1024/1024 initial_extent

FROM DBA_SEGMENTS D

WHERE D.owner IN

(SELECT a.username

FROM DBA_USERS A

WHERE A.account_status = 'OPEN'

AND A.username NOT IN

('SYS', 'SYSTEM', 'MGMT_VIEW', 'SYSMAN', 'DBSNMP'))

;

由此可以知道，创建这些元数据大约需要49G的空间，如果涉及到数据的话，还需要判断数据占用空间，这里一定需要判断这个，不然执行导入的时候会因为表空间不足而不能导入，我第一次导入的时候就是因为这里没有判断导致花费了很长的时间，我一直扩展表空间，但是就是就表空间不足的错误(ORA-01659),想想建表不会花这么大的空间的吧，最后查看了表的定义才知道，原来表初始化的时候就很大，这个问题后边还需要再处理一下的，不然测试库没法导入，当然存储够的话就另当别论了。

ORA-39171: Job is experiencing a resumable wait.

ORA-01659: unable to allocate MINEXTENTS beyond 4 in tablespace DWII_SOR_F_01

确定需要导出的用户中有哪些无效的对象、及总共需要导出的对象数量

这一步也很重要，决定着最终导出结果的正确性验证。

SELECT d.OWNER,count(1)

FROM dba_objects d

WHERE d.OWNER in (SELECT a.username

FROM DBA_USERS A

WHERE A.account_status = 'OPEN'

AND A.username NOT IN

('SYS', 'SYSTEM', 'MGMT_VIEW', 'SYSMAN', 'DBSNMP'))

group by d.OWNER;

SELECT d.OWNER,

d.status,count(1)

FROM dba_objects d

WHERE d.OWNER in (SELECT a.username

FROM DBA_USERS A

WHERE A.account_status = 'OPEN'

AND A.username NOT IN

('SYS', 'SYSTEM', 'MGMT_VIEW', 'SYSMAN', 'DBSNMP'))

group by d.OWNER, d.status;

SELECT d.OWNER,

d.OBJECT_NAME,

d.OBJECT_TYPE,

d.status

FROM dba_objects d

WHERE d.status = 'INVALID'

and d.owner in (SELECT a.username

FROM DBA_USERS A

WHERE A.account_status = 'OPEN'

AND A.username NOT IN

('SYS', 'SYSTEM', 'MGMT_VIEW', 'SYSMAN', 'DBSNMP'));

expdp数据泵利用content=metadata_only导出元数据

导出命令，注意这里不导出数据只导出定义我们采用content=metadata_only来处理：

expdp lhr/lhr directory=DATA_PUMP_DIR dumpfile=lhrsql20150515.dmp logfile=lhrsql20150515.log content=metadata_only schemas=TEST,SQCHECK,DWUSER,DPA,CNYDM,ONL1,LHR,TEST1,FXDM,DWII_ETL,DWUSER1,SOR,DW_ETL,NRDM,NRDM_ETL,FXDM_ETL,LCM2,CNY_ETL

由于是事后写文档，所以这里只贴出导出元数据的日志：

;;;

Export: Release 10.2.0.1.0 - 64bit Production on Friday, 15 May, 2015 13:05:54

;;;

Connected to: Oracle Database 10g Enterprise Edition Release 10.2.0.1.0 - 64bit Production

With the Partitioning, OLAP and Data Mining options

Starting "LHR"."SYS_EXPORT_SCHEMA_01": lhr/******** directory=DATA_PUMP_DIR dumpfile=lhrsql20150515.dmp logfile=lhrsql20150515.log content=metadata_only schemas=TEST,SQCHECK,DWUSER,DPA,CNYDM,ONL1,LHR,TEST1,FXDM,DWII_ETL,DWUSER1,SOR,DW_ETL,NRDM,NRDM_ETL,FXDM_ETL,LCM2,CNY_ETL

Processing object type SCHEMA_EXPORT/USER

Processing object type SCHEMA_EXPORT/SYSTEM_GRANT

Processing object type SCHEMA_EXPORT/ROLE_GRANT

Processing object type SCHEMA_EXPORT/DEFAULT_ROLE

Processing object type SCHEMA_EXPORT/PRE_SCHEMA/PROCACT_SCHEMA

Processing object type SCHEMA_EXPORT/DB_LINK

Processing object type SCHEMA_EXPORT/SEQUENCE/SEQUENCE

Processing object type SCHEMA_EXPORT/TABLE/TABLE

Processing object type SCHEMA_EXPORT/TABLE/GRANT/OWNER_GRANT/OBJECT_GRANT

Processing object type SCHEMA_EXPORT/TABLE/INDEX/INDEX

Processing object type SCHEMA_EXPORT/TABLE/CONSTRAINT/CONSTRAINT

Processing object type SCHEMA_EXPORT/TABLE/INDEX/STATISTICS/INDEX_STATISTICS

Processing object type SCHEMA_EXPORT/TABLE/COMMENT

Processing object type SCHEMA_EXPORT/PACKAGE/PACKAGE_SPEC

Processing object type SCHEMA_EXPORT/FUNCTION/FUNCTION

Processing object type SCHEMA_EXPORT/FUNCTION/GRANT/OWNER_GRANT/OBJECT_GRANT

Processing object type SCHEMA_EXPORT/PROCEDURE/PROCEDURE

Processing object type SCHEMA_EXPORT/PROCEDURE/GRANT/OWNER_GRANT/OBJECT_GRANT

Processing object type SCHEMA_EXPORT/PACKAGE/COMPILE_PACKAGE/PACKAGE_SPEC/ALTER_PACKAGE_SPEC

Processing object type SCHEMA_EXPORT/FUNCTION/ALTER_FUNCTION

Processing object type SCHEMA_EXPORT/PROCEDURE/ALTER_PROCEDURE

Processing object type SCHEMA_EXPORT/VIEW/VIEW

Processing object type SCHEMA_EXPORT/VIEW/GRANT/OWNER_GRANT/OBJECT_GRANT

Processing object type SCHEMA_EXPORT/VIEW/COMMENT

Processing object type SCHEMA_EXPORT/TABLE/CONSTRAINT/REF_CONSTRAINT

Processing object type SCHEMA_EXPORT/TABLE/STATISTICS/TABLE_STATISTICS

Master table "LHR"."SYS_EXPORT_SCHEMA_01" successfully loaded/unloaded

******************************************************************************

Dump file set for LHR.SYS_EXPORT_SCHEMA_01 is:

/oracle/product/10.2.0/db_1/rdbms/log/lhrsql20150515.dmp

Job "LHR"."SYS_EXPORT_SCHEMA_01" successfully completed at 13:09:49

接下来就是把导出来的文件利用ftp工具或scp传递到目标端，我这里就直接传到我的虚拟机上吧。

执行导入工作
1. 先建立相应的表空间

根据上边的工作我们知道需要在测试库上建立以下的表空间，我这里都设置的是20M，实际情况下应该根据上边查询出来的结果来设置相应的大小，表空间应用包含用户的默认表空间及用户下对象所在的表空间：

create tablespace DWII_CNY_BK_F_01 datafile '+DATA' size 20M;

create tablespace DWII_DPA_F_01 datafile '+DATA' size 20M;

create tablespace DWII_DPA_I_01 datafile '+DATA' size 20M;

create tablespace DWII_DPA_S_01 datafile '+DATA' size 20M;

create tablespace DWII_SOR_F_01 datafile '+DATA' size 20M;

create tablespace DWII_SOR_I_01 datafile '+FRA' size 20M;

create tablespace DW_USER datafile '+FRA' size 20M;

create tablespace SQCHECK datafile '+FRA' size 20M;

create tablespace SD_CNY_D_01 datafile '+FRA' size 20M;

create tablespace SD_CNY_F_01 datafile '+FRA' size 20M;

create tablespace SD_DPA_D_01 datafile '+FRA' size 20M;

create tablespace SD_DPA_F_01 datafile '+FRA' size 20M;

create tablespace SD_SORT_T_01 datafile '+FRA' size 20M;

create tablespace DWII_FXDM_F_01 datafile '+FRA' size 20M;

create tablespace SD_SOR_T_01 datafile '+FRA' size 20M;

如果空间不够，我们可以追加数据文件：alter tablespace DWII_DPA_F_01 add datafile '+FRA' size 50M;

执行导入语句

一般情况下，如果存储够的话，我们把相应的表空间设置大一点之后这里直接执行导入语句就可以了，但是我是在本机的虚拟机里执行的，由前边的情况我们可以知道大约需要49G的空间，这个显然不太合适，哪该怎么办呢？我能想到的办法只有如下2种，如果大家还有好的办法可以给我留言。

① 在源库上修改表的定义后然后再执行导出命令

② 从已经导出来的dmp文件中抽取其中的DDL语句，然后将DDL语句导入到数据库中，update掉其中的STORAGE(INITIAL 参数后再将语句导出到sql文本中执行sql语句，这样可以解决表的定义问题。

③ 从已经导出来的dmp文件中抽取其中的DDL语句，然后利用linux的sed批量替换功能替换掉不正确的参数。

显然，第一种比较方便，也比较快，但是不实用，由于是生产库，参数不能随便修改，我们就采用第二或第三种办法，也可以多演示一种impdp的用法。

导出DMP文件中的DDL语句

我们在impdp的导入命令中添加sqlfile参数后执行导入并不会真正将数据导入到数据库，而会抽取出dmp文件中的DDL语句，如下：

[oracle@rhel6_lhr dpdump]$ impdp lhr/lhr directory=DATA_PUMP_DIR dumpfile=lhrsql20150515.dmp logfile=imp_exptest.log sqlfile=lhrsql20150515.sql

Import: Release 11.2.0.3.0 - Production on Fri May 15 15:08:03 2015

Connected to: Oracle Database 11g Enterprise Edition Release 11.2.0.3.0 - 64bit Production

With the Partitioning, Automatic Storage Management, OLAP, Data Mining

and Real Application Testing options

Master table "LHR"."SYS_SQL_FILE_FULL_01" successfully loaded/unloaded

Starting "LHR"."SYS_SQL_FILE_FULL_01": lhr/******** directory=DATA_PUMP_DIR dumpfile=lhrsql20150515.dmp logfile=imp_exptest.log sqlfile=lhrsql20150515.sql

Processing object type SCHEMA_EXPORT/USER

Processing object type SCHEMA_EXPORT/SYSTEM_GRANT

Processing object type SCHEMA_EXPORT/ROLE_GRANT

Processing object type SCHEMA_EXPORT/DEFAULT_ROLE

Processing object type SCHEMA_EXPORT/PRE_SCHEMA/PROCACT_SCHEMA

Processing object type SCHEMA_EXPORT/DB_LINK

Processing object type SCHEMA_EXPORT/SEQUENCE/SEQUENCE

Processing object type SCHEMA_EXPORT/TABLE/TABLE

Processing object type SCHEMA_EXPORT/TABLE/GRANT/OWNER_GRANT/OBJECT_GRANT

Processing object type SCHEMA_EXPORT/TABLE/INDEX/INDEX

Processing object type SCHEMA_EXPORT/TABLE/CONSTRAINT/CONSTRAINT

Processing object type SCHEMA_EXPORT/TABLE/INDEX/STATISTICS/INDEX_STATISTICS

Processing object type SCHEMA_EXPORT/TABLE/COMMENT

Processing object type SCHEMA_EXPORT/PACKAGE/PACKAGE_SPEC

Processing object type SCHEMA_EXPORT/FUNCTION/FUNCTION

Processing object type SCHEMA_EXPORT/FUNCTION/GRANT/OWNER_GRANT/OBJECT_GRANT

Processing object type SCHEMA_EXPORT/PROCEDURE/PROCEDURE

Processing object type SCHEMA_EXPORT/PROCEDURE/GRANT/OWNER_GRANT/OBJECT_GRANT

Processing object type SCHEMA_EXPORT/PACKAGE/COMPILE_PACKAGE/PACKAGE_SPEC/ALTER_PACKAGE_SPEC

Processing object type SCHEMA_EXPORT/FUNCTION/ALTER_FUNCTION

Processing object type SCHEMA_EXPORT/PROCEDURE/ALTER_PROCEDURE

Processing object type SCHEMA_EXPORT/VIEW/VIEW

Processing object type SCHEMA_EXPORT/VIEW/GRANT/OWNER_GRANT/OBJECT_GRANT

Processing object type SCHEMA_EXPORT/VIEW/COMMENT

Processing object type SCHEMA_EXPORT/TABLE/CONSTRAINT/REF_CONSTRAINT

Processing object type SCHEMA_EXPORT/TABLE/STATISTICS/TABLE_STATISTICS

Job "LHR"."SYS_SQL_FILE_FULL_01" successfully completed at 15:21:50

[oracle@rhel6_lhr dpdump]$ ll lhrsql20150515.sql

-rw-r--r-- 1 oracle asmadmin 65707967 May 15 15:21 lhrsql20150515.sql

[oracle@rhel6_lhr dpdump]$ more lhrsql20150515.sql

-- CONNECT LHR

ALTER SESSION SET EVENTS '10150 TRACE NAME CONTEXT FOREVER, LEVEL 1';

ALTER SESSION SET EVENTS '10904 TRACE NAME CONTEXT FOREVER, LEVEL 1';

ALTER SESSION SET EVENTS '25475 TRACE NAME CONTEXT FOREVER, LEVEL 1';

ALTER SESSION SET EVENTS '10407 TRACE NAME CONTEXT FOREVER, LEVEL 1';

ALTER SESSION SET EVENTS '10851 TRACE NAME CONTEXT FOREVER, LEVEL 1';

ALTER SESSION SET EVENTS '22830 TRACE NAME CONTEXT FOREVER, LEVEL 192 ';

-- new object type path: SCHEMA_EXPORT/USER

-- CONNECT SYSTEM

CREATE USER "CNY_ETL" IDENTIFIED BY VALUES '4686A1050F638F44'

DEFAULT TABLESPACE "DW_USER"

TEMPORARY TABLESPACE "TEMP";

CREATE USER "LCM2" IDENTIFIED BY VALUES '48BCFDF435352212'

DEFAULT TABLESPACE "DWII_SOR_F_01"

TEMPORARY TABLESPACE "TEMP";

CREATE USER "FXDM_ETL" IDENTIFIED BY VALUES 'EA010AEA839BFA14'

DEFAULT TABLESPACE "DW_USER"

TEMPORARY TABLESPACE "TEMP";

CREATE USER "NRDM_ETL" IDENTIFIED BY VALUES '54A4A046AEE8B31E'

DEFAULT TABLESPACE "DW_USER"

TEMPORARY TABLESPACE "TEMP";

CREATE USER "NRDM" IDENTIFIED BY VALUES '1AE3DF7368DF560D'

DEFAULT TABLESPACE "SD_CNY_F_01"

TEMPORARY TABLESPACE "TEMP";

CREATE USER "DW_ETL" IDENTIFIED BY VALUES '91635F9C0744E7EC'

DEFAULT TABLESPACE "DW_USER"

TEMPORARY TABLESPACE "TEMP";

CREATE USER "SOR" IDENTIFIED BY VALUES 'BA3A6C912E6BFF14'

DEFAULT TABLESPACE "DWII_SOR_F_01"

TEMPORARY TABLESPACE "TEMP";

。。。。。。。。。。。。。。。。。。。。。。。。。。。。。。。。。。。。。。。。。。。。。。。。。。。。。。省略

[oracle@rhel6_lhr dpdump]$ tail -n 50 lhrsql20150515.sql

c := 'SPOT_EXCHNG_RATE_SRC';

EXECUTE IMMEDIATE stmt USING 'C',t,p,sp,c,s,

2,.5,2,156,0,2.65784513562818E+35,2.65784513872303E+35,5,0,nv,nv,

TO_DATE('2015-05-14 22:00:18',df),'33303031','33303032',nv,2,nv;

c := 'SPOT_EXCHNG_RATE_SRC_NM';

EXECUTE IMMEDIATE stmt USING 'C',t,p,sp,c,s,

2,.5,2,156,0,1.18758942587854E+36,1.20817519861257E+36,12,0,nv,nv,

TO_DATE('2015-05-14 22:00:18',df),'E4B8ADE997B4E4BBB7','E8AFA2E4BBB7E59D87E580BC',nv,2,nv;

c := 'FLEG_SWAP_PNT';

EXECUTE IMMEDIATE stmt USING 'C',t,p,sp,c,s,

14,.0714285714285714,14,156,0,-64.01,16.5,4,0,nv,nv,

TO_DATE('2015-05-14 22:00:18',df),'3E256466','C11133',nv,2,nv;

c := 'SRC_SYS_LBL';

EXECUTE IMMEDIATE stmt USING 'C',t,p,sp,c,s,

1,1,1,156,0,3.44097282552972E+35,3.44097282552972E+35,5,0,nv,nv,

TO_DATE('2015-05-14 22:00:18',df),'42454E4D','42454E4D',nv,2,nv;

END;

/

DECLARE

c varchar2(60);

nv varchar2(1);

df varchar2(21) := 'YYYY-MM-DD:HH24:MI:SS';

s varchar2(60) := 'DPA';

t varchar2(60) := 'BNCHMK_OPTN_DLT_PARAM_F';

p varchar2(1);

sp varchar2(1);

stmt varchar2(300) := 'INSERT INTO "SYS"."IMPDP_STATS" (type,version,c1,c2,c3,c4,c5,n1,n2,n3,n4,n5,n6,n7,n8,n9,n10,n11,d1,r1,r2,ch1,flags,cl1) VALUES (:1,6,:2,:3,:4,:5,:6,:7,:8,:9,:10,:11,:12,:13,:14,:15,:16,:17,:18,:19,:20,:21,:22,:23)';

BEGIN

NULL;

c := 'CRT_TMST';

EXECUTE IMMEDIATE stmt USING 'C',t,p,sp,c,s,

1,1,1,156,0,2457157.57244213,2457157.57244213,11,0,nv,nv,

TO_DATE('2015-05-14 22:00:18',df),'7873050E0E2D14','7873050E0E2D14',nv,2,nv;

c := 'PPLN_WKDT';

EXECUTE IMMEDIATE stmt USING 'C',t,p,sp,c,s,

1,1,1,156,0,20150422,20150422,6,0,nv,nv,

TO_DATE('2015-05-14 22:00:18',df),'C415100517','C415100517',nv,2,nv;

c := 'PPLN_TMST';

EXECUTE IMMEDIATE stmt USING 'C',t,p,sp,c,s,

1,1,1,156,0,2457157.57244213,2457157.57244213,11,0,nv,nv,

TO_DATE('2015-05-14 22:00:18',df),'7873050E0E2D14','7873050E0E2D14',nv,2,nv;

DBMS_STATS.IMPORT_TABLE_STATS('"DPA"','"BNCHMK_OPTN_DLT_PARAM_F"',NULL,'"IMPDP_STATS"',NULL,NULL,'"SYS"');

DELETE FROM "SYS"."IMPDP_STATS";

END;

/

[oracle@rhel6_lhr dpdump]$

可以看到生成的SQL语句开始部分是ddl的建用户语句，最后是导入表的统计信息部分。

处理导出来的ddl文件

方法一，利用sed命令批量替换（推荐）

这一步其实如果linux的sed命令熟悉的话可以直接替换掉sql语句中的相关行，我今天也是临时网上搜了下，经过大量实验，然后居然成功了，如下：

[oracle@rhel6_lhr dpdump]$ sed 's/^ STORAGE(INITIAL.*/STORAGE(INITIAL 65536 NEXT 1048576 MINEXTENTS 1 MAXEXTENTS 2147483645/g' lhrsql20150515.sql > lhrsql20150515.sql_bk2

[oracle@rhel6_lhr dpdump]$ more lhrsql20150515.sql_bk2 | grep "STORAGE(INITIAL"

STORAGE(INITIAL 65536 NEXT 1048576 MINEXTENTS 1 MAXEXTENTS 2147483645

[oracle@rhel6_lhr dpdump]$ more lhrsql20150515.sql_bk2 | grep "STORAGE(INITIAL 131072"

[oracle@rhel6_lhr dpdump]$ more lhrsql20150515.sql | grep "STORAGE(INITIAL 131072"

STORAGE(INITIAL 131072 NEXT 1048576 MINEXTENTS 1 MAXEXTENTS 2147483645

。。。。。。。。。。。。。

可以看到替换已经成功，如果不熟悉该命令，那么先熟悉该命令就麻烦了。

方法二：sqlldr导入到数据库，利用update来更新

都是搞数据库的，这样做虽然麻烦点，但是绝对不会出错，而且也是一种技巧，如果有的文本很大又很难处理的话我们就可以导入到数据库中，然后处理。

首先建表：

create table imp_sql_lhr (id number ,text varchar2(4000)) ;

sqlldr的控制文件内容:sqlldr_table.ctl ：

UNRECOVERABLE

load data

LENGTH CHARACTER

infile 'lhrsql20150515.sql'

APPEND imp_sql_lhr

trailing nullcols

(

id SEQUENCE(1,1),

text char(4000) "TRIM(:text)"

)

sqlldr命令：

sqlldr lhr/lhr control=sqlldr_table.ctl log=a.log parallel=y readsize=4194304 streamsize=10485760 date_cache=5000 direct=true

导入到数据库后，我们就可以非常方便的来处理表中的数据了，如下：

SELECT * FROM imp_sql_lhr a where a.text like '%STORAGE(INITIAL%' and a.text not like '%STORAGE(INITIAL 65536%' ;

执行更新语句：

UPDATE imp_sql_lhr t

SET t.text = 'STORAGE(INITIAL 65536 NEXT 1048576 MINEXTENTS 1 MAXEXTENTS 2147483645'

WHERE t.text LIKE '%STORAGE(INITIAL%'

AND t.text NOT LIKE '%STORAGE(INITIAL 65536 %';

最后利用spool来导出到sql文本中：

set echo on

set trimspool on

set trimout on

set linesize 4000

set pagesize 0

set sqlblanklines on

set feedback off

set serveroutput off

set term off

set echo off

spool lhrsql20150515.sql_bk2

SELECT t.text from imp_sql_lhr t order by id;

spool off

执行处理好的sql语句

该步骤比较简单，就是把处理好的ddl语句提前执行一下，让数据库中包含相应的对象，这样再执行impdp导入的时候就不会再创建这些表了。

需要注意的是：DDL语句中创建存过、函数、包的语句中是不包含对象所属的schema的，这样的话如果那个用户执行sql脚本的话，这些对象就创建在那个用户下了，这个显然是错误的，不是我们期望的，那么如何处理这个问题呢，想了想，很简单的嘛，我们sql脚本执行完毕后，再执行一次impdp的命令就可以把这些对象重建，然后把错误的存过删除就可以了。

SELECT t.* from imp_sql_lhr t where t.text like 'CREATE PROCEDURE%' order by id;

[oracle@rhel6_lhr dpdump]$ sqlplus lhr/lhr

SQL*Plus: Release 11.2.0.3.0 Production on Fri May 15 16:32:08 2015

Connected to:

Oracle Database 11g Enterprise Edition Release 11.2.0.3.0 - 64bit Production

With the Partitioning, Automatic Storage Management, OLAP, Data Mining

and Real Application Testing options

16:32:09 SQL> set echo off;

16:32:21 SQL> set serveroutput off;

16:32:30 SQL> set timing on;

16:32:40 SQL> set time on;

16:32:45 SQL> set timing off;

16:32:50 SQL> set time off;

SQL>

SQL> @lhrsql20150515.sql_bk2;

Session altered.

。。。。。。。。。。。。。。。。。。。。。。。。。。。省略

SQL> exit

Disconnected from Oracle Database 11g Enterprise Edition Release 11.2.0.3.0 - 64bit Production

With the Partitioning, Automatic Storage Management, OLAP, Data Mining

and Real Application Testing options

impdp继续导入

[oracle@rhel6_lhr dpdump]$ impdp lhr/lhr directory=DATA_PUMP_DIR dumpfile=lhrsql20150515.dmp logfile=lhrsql20150515_imp.log parallel=4;

Import: Release 11.2.0.3.0 - Production on Fri May 15 19:05:29 2015

Connected to: Oracle Database 11g Enterprise Edition Release 11.2.0.3.0 - 64bit Production

With the Partitioning, Automatic Storage Management, OLAP, Data Mining

and Real Application Testing options

Master table "LHR"."SYS_IMPORT_FULL_02" successfully loaded/unloaded

Starting "LHR"."SYS_IMPORT_FULL_02": lhr/******** directory=DATA_PUMP_DIR dumpfile=lhrsql20150515.dmp logfile=lhrsql20150515_imp.log parallel=4

Processing object type SCHEMA_EXPORT/USER

ORA-31684: Object type USER:"CNY_ETL" already exists

ORA-31684: Object type USER:"LCM2" already exists

ORA-31684: Object type USER:"FXDM_ETL" already exists

ORA-31684: Object type USER:"NRDM_ETL" already exists

ORA-31684: Object type USER:"NRDM" already exists

ORA-31684: Object type USER:"DW_ETL" already exists

ORA-31684: Object type USER:"SOR" already exists

ORA-31684: Object type USER:"DWUSER1" already exists

ORA-31684: Object type USER:"DWII_ETL" already exists

ORA-31684: Object type USER:"FXDM" already exists

ORA-31684: Object type USER:"TEST1" already exists

ORA-31684: Object type USER:"LHR" already exists

ORA-31684: Object type USER:"ONL1" already exists

ORA-31684: Object type USER:"CNYDM" already exists

ORA-31684: Object type USER:"DPA" already exists

ORA-31684: Object type USER:"DWUSER" already exists

ORA-31684: Object type USER:"SQCHECK" already exists

ORA-31684: Object type USER:"TEST" already exists

Processing object type SCHEMA_EXPORT/SYSTEM_GRANT

Processing object type SCHEMA_EXPORT/ROLE_GRANT

Processing object type SCHEMA_EXPORT/DEFAULT_ROLE

Processing object type SCHEMA_EXPORT/PRE_SCHEMA/PROCACT_SCHEMA

Processing object type SCHEMA_EXPORT/DB_LINK

ORA-31684: Object type DB_LINK:"SOR"."COG_DB" already exists

ORA-31684: Object type DB_LINK:"DPA"."COG_DB" already exists

Processing object type SCHEMA_EXPORT/SEQUENCE/SEQUENCE

ORA-31684: Object type SEQUENCE:"SOR"."SEQ_BDS_CL_IP_RL_ID" already exists

ORA-31684: Object type SEQUENCE:"SOR"."SEQ_GOLD_RATE_INFO" already exists

ORA-31684: Object type SEQUENCE:"SOR"."SEQ_MBR_API_ELGBLTY" already exists

ORA-31684: Object type SEQUENCE:"SOR"."SEQ_MBR_MKT_ELGBLTY" already exists

ORA-31684: Object type SEQUENCE:"SOR"."SEQ_MBR_MMKNG_ROLE" already exists

ORA-31684: Object type SEQUENCE:"SOR"."SEQ_ORG_CD_HSTRY" already exists

ORA-31684: Object type SEQUENCE:"SOR"."SQ_DW_IP_IP_RL" already exists

ORA-31684: Object type SEQUENCE:"SOR"."SQ_MKT_CLNDR_ID" already exists

ORA-31684: Object type SEQUENCE:"FXDM"."SEQ_GOLD_MBR_MMKNG_ROLE_D" already exists

ORA-31684: Object type SEQUENCE:"FXDM"."SEQ_GOLD_MINAMNT_CONFIG" already exists

ORA-31684: Object type SEQUENCE:"DPA"."SEQ_BDS_MBR_D_ID" already exists

ORA-31684: Object type SEQUENCE:"DPA"."SEQ_DL_MKT_INFO" already exists

ORA-31684: Object type SEQUENCE:"DPA"."SEQ_GOLD_RATE_INFO" already exists

ORA-31684: Object type SEQUENCE:"DPA"."SEQ_MBR_AUTH_BY_API_D" already exists

ORA-31684: Object type SEQUENCE:"DPA"."SEQ_MBR_MMKNG_ROLE_D" already exists

ORA-31684: Object type SEQUENCE:"DPA"."SEQ_ORG_D" already exists

ORA-31684: Object type SEQUENCE:"DPA"."SQ_DW_BEST_QUOTE" already exists

ORA-31684: Object type SEQUENCE:"DPA"."SQ_DW_BOND_MID_QUOTE_ID" already exists

ORA-31684: Object type SEQUENCE:"DPA"."SQ_DW_MEMBER_D_ID" already exists

Processing object type SCHEMA_EXPORT/TABLE/TABLE

ORA-39151: Table "DPA"."FX_DL_MKT_INFO" exists. All dependent metadata and data will be skipped due to table_exists_action of skip

ORA-39151: Table "DPA"."FX_FWD_CPI_QT" exists. All dependent metadata and data will be skipped due to table_exists_action of skip

ORA-39151: Table "DPA"."FX_FWD_DL_BY_SF_CP" exists. All dependent metadata and data will be skipped due to table_exists_action of skip

ORA-39151: Table "DPA"."FX_FWD_QT_BY_DIR" exists. All dependent metadata and data will be skipped due to table_exists_action of skip

ORA-39151: Table "DPA"."FX_SPOT_CPI_QT" exists. All dependent metadata and data will be skipped due to table_exists_action of skip

ORA-39151: Table "DPA"."FX_SPOT_DL_BY_SF_CP" exists. All dependent metadata and data will be skipped due to table_exists_action of skip

ORA-39151: Table "DPA"."FX_SPOT_QT_BY_DIR" exists. All dependent metadata and data will be skipped due to table_exists_action of skip

ORA-39151: Table "DPA"."FX_SWAP_CPI_QT" exists. All dependent metadata and data will be skipped due to table_exists_action of skip

ORA-39151: Table "DPA"."FX_SWAP_DL_BY_SF_CP" exists. All dependent metadata and data will be skipped due to table_exists_action of skip

ORA-39151: Table "DPA"."FX_SWAP_DL_LEG_MKT_INFO" exists. All dependent metadata and data will be skipped due to table_exists_action of skip

ORA-39151: Table "DPA"."FX_SWAP_QT_BY_DIR" exists. All dependent metadata and data will be skipped due to table_exists_action of skip

ORA-39151: Table "DPA"."BOND_TURNOVER_RATE_F" exists. All dependent metadata and data will be skipped due to table_exists_action of skip

ORA-39151: Table "DPA"."MONI_TORT_CRCLTN_AMNT_F" exists. All dependent metadata and data will be skipped due to table_exists_action of skip

ORA-39151: Table "DPA"."MONI_TURNOVER_RATE_F" exists. All dependent metadata and data will be skipped due to table_exists_action of skip

ORA-39151: Table "SOR"."BST_QT_HIST" exists. All dependent metadata and data will be skipped due to table_exists_action of skip

ORA-39151: Table "SOR"."BST_QT" exists. All dependent metadata and data will be skipped due to table_exists_action of skip

。。。。。。。。。。。。。。。。。。。。。。。。。省略

ORA-39112: Dependent object type TABLE_STATISTICS skipped, base object type TABLE:"SOR"."QUE_BOND_DTL" creation failed

ORA-39112: Dependent object type TABLE_STATISTICS skipped, base object type TABLE:"SOR"."QUE_FX_FWD_SWAP_DTL" creation failed

ORA-39112: Dependent object type TABLE_STATISTICS skipped, base object type TABLE:"SOR"."QUE_CCS_DTL" creation failed

ORA-39112: Dependent object type TABLE_STATISTICS skipped, base object type TABLE:"SOR"."QUE_CREPO_DTL" creation failed

ORA-39112: Dependent object type TABLE_STATISTICS skipped, base object type TABLE:"SOR"."QUE_FCO_DTL" creation failed

ORA-39112: Dependent object type TABLE_STATISTICS skipped, base object type TABLE:"SOR"."QUE_FRA_DTL" creation failed

ORA-39112: Dependent object type TABLE_STATISTICS skipped, base object type TABLE:"SOR"."QUE_IBO_DTL" creation failed

ORA-39112: Dependent object type TABLE_STATISTICS skipped, base object type TABLE:"SOR"."QUE_IRS_DTL" creation failed

ORA-39112: Dependent object type TABLE_STATISTICS skipped, base object type TABLE:"SOR"."QUE_RIBD_DTL" creation failed

ORA-39112: Dependent object type TABLE_STATISTICS skipped, base object type TABLE:"SOR"."FX_BM_FXO_DL_DAY_HIST" creation failed

Job "LHR"."SYS_IMPORT_FULL_02" completed with 2165 error(s) at 19:16:19

[oracle@rhel6_lhr dpdump]$

删除错误用户下的包、存过和函数

执行如下的脚本来删除相应的错误对象：

BEGIN

FOR CUR IN (SELECT 'DROP ' || D.OBJECT_TYPE || ' ' || D.OBJECT_NAME SQLT

FROM dba_objects d

WHERE d.OWNER = 'LHR'

AND d.CREATED >=

to_date('2015-05-15 09:25:27',

'YYYY-MM-DD HH24:MI:SS')

AND D.OBJECT_TYPE IN ('FUNCTION', 'PROCEDURE', 'PACKAGE')) LOOP

EXECUTE IMMEDIATE CUR.SQLT;

END LOOP;

END;

数据校验

执行如下脚本和源库作比较，查看数据是否完整。

SELECT d.OWNER,count(1)

FROM dba_objects d

WHERE d.OWNER in (SELECT a.username

FROM DBA_USERS A

WHERE A.account_status = 'OPEN'

AND A.username NOT IN

('SYS', 'SYSTEM', 'MGMT_VIEW', 'SYSMAN', 'DBSNMP'))

group by d.OWNER;

SELECT d.OWNER,

d.status,count(1)

FROM dba_objects d

WHERE d.OWNER in (SELECT a.username

FROM DBA_USERS A

WHERE A.account_status = 'OPEN'

AND A.username NOT IN

('SYS', 'SYSTEM', 'MGMT_VIEW', 'SYSMAN', 'DBSNMP'))

group by d.OWNER, d.status;

SELECT d.OWNER,

d.OBJECT_NAME,

d.OBJECT_TYPE,

d.status

FROM dba_objects d

WHERE d.status = 'INVALID'

and d.owner in (SELECT a.username

FROM DBA_USERS A

WHERE A.account_status = 'OPEN'

AND A.username NOT IN

('SYS', 'SYSTEM', 'MGMT_VIEW', 'SYSMAN', 'DBSNMP'));

总结

到此所有的处理算是基本完毕，过程很简单，但是不同的场景处理方式有很多种，我们应该学会灵活变通，核心即expdp和impdp但是需要做很多的处理。

aboout me

...........................................................................................................................................................................................

本文作者：小麦苗，只专注于数据库的技术，更注重技术的运用

ITPUB BLOG：http://blog.itpub.net/26736162

本文地址：http://blog.itpub.net/26736162/viewspace-1657828/

本文pdf版：http://yunpan.cn/QCwUAI9bn7g7w 提取码：af2d

QQ：642808185 若加QQ请注明你所正在读的文章标题

创作时间地点：2015-05-15 10:00~ 2015-05-16 19:00 于唐镇金唐公寓宿舍

...........................................................................................................................................................................................

相关阅读:
代理模式
spring aop
mybatis从入门到精通(其他章节)
mybatis从入门到精通（第6章）
Java中Compareable和Comparator两种比较器的区别
Java的equals方法的使用技巧
Dubbo的配置过程，实现原理及架构详解
什么是IPFS?IPFS与区块链有什么关系
leetCode242 有效的字母异位词
需要多个参数输入时-----------------考虑使用变种的Builder模式

原文地址：https://www.cnblogs.com/lhrbest/p/4508832.html

【数据泵】EXPDP导出表结构

导读

实验环境介绍

执行导出工作

确定需要导出的用户

确定需要导出的用户在哪些表空间，及其表初始化时占用的表空间大小

确定需要导出的用户中有哪些无效的对象、及总共需要导出的对象数量

expdp数据泵利用content=metadata_only导出元数据

执行导入工作

先建立相应的表空间

执行导入语句

导出DMP文件中的DDL语句

处理导出来的ddl文件

方法一，利用sed命令批量替换 （推荐）

方法二：sqlldr导入到数据库，利用update来更新

执行处理好的sql语句

impdp继续导入

删除错误用户下的包、存过和函数

数据校验

总结

aboout me

方法一，利用sed命令批量替换（推荐）