oracle 数据泵导入导出 expdp/impdp

一、导出/入前检查

查看用户默认表空间

select USERNAME,DEFAULT_TABLESPACE,TEMPORARY_TABLESPACE from dba_users where USERNAME='USER_NAME';

查看表空间对应数据文件

select tablespace_name,file_name from dba_data_files where tablespace_name='TBSNAME';

查询目录实际路径

select OWNER,DIRECTORY_PATH from dba_directories where DIRECTORY_NAME='DATA_PUMP_DIR';

目录属主授权

grant read,write on directory DATA_PUMP_DIR to system;

或者system用户登录创建目录

alter user system identified by "xxxxx";
create or replace directory DATA_PUMP as '/data/tmp';

二、导出数据

1. 常用选项

排除指定表 EXCLUDE=
导出指定表 tables=
导出表名带关键字的表 INCLUDE=TABLE:\"LIKE \'%BAK%\'\"
只导表结构 CONTENT=METADATA_ONLY
只导表数据 CONTENT=data_only
输出日志 logfile=expdp1026.log
指定并行度 PARALLEL=4
压缩 compression=all （测试压缩比约为8:1，但对导出性能有影响）
指定版本（高版本导至低版本） version=12.1.0.2.0

2. 导出例子

导出指定用户所有数据

expdp system DIRECTORY=DATA_PUMP schemas=USER_NAME dumpfile=USER_NAME1025.dmp logfile=USER_NAME1025.log

导出指定用户所有数据（排除 IM_HIS 和 IM_OR_HIS表）

expdp system@mydb schemas=user_name dumpfile=user_name.dump DIRECTORY=DATA_PUMP_DIR EXCLUDE=TABLE:\"IN\(\'IM_HIS\',\'IM_OR_HIS\'\)\"

导出用户指定表

expdp system DIRECTORY=DATA_PUMP tables=USER_NAME.t1,USER_NAME.t2,USER_NAME.t3 dumpfile=USER_NAME1206.dmp

导出用户下表名带指定关键字的表

expdp system DIRECTORY=DATA_PUMP schemas=USER_NAME INCLUDE=TABLE:\"LIKE \'%BAK%\'\" dumpfile=USER_NAME1206.dmp logfile=USER_NAME1206.log

高版本导入12.1.0.2版本数据

expdp system@orcl_dev schemas=user1,user2 dumpfile=orcl.dump version=12.1.0.2.0 DIRECTORY=DATA_PUMP_DIR;

只导出用户元数据，且不包含统计信息

expdp system directory=jy schemas=scott content=metadata_only exclude=statistics dumpfile=scott_meta.dmp logfile=scott_meta.log

三、导入数据

0. 注意事项

注意先检查目标表空间是否存在
检查目标磁盘空间及表空间大小
导入时注意检查DB等待事件及alert log，出现异常时及时处理

1. 常用选项

导入新用户下 REMAP_SCHEMA=old:new
导入新表空间下 remap_tablespace=old:new
导入时重命名表 REMAP_TABLE=scott.T_OLD:T_NEW
只导表结构 CONTENT=METADATA_ONLY
指定并行度 PARALLEL=4
目标库与源库相同 transform=oid:n
将dmp文件转为sql语句（不真正导入）sqlfile=xxx.sql
指定表已存在时动作 table_exists_action

table_exists_action的值如下：

不指定：表已存在会报错
skip：跳过已存在表并处理下一个对象（不安全，谨慎使用）
append：为表追加数据（不安全，谨慎使用）
truncate：截断已存在表然后插入新数据（不安全，谨慎使用）
replace：删除已存在表，重建并插入数据（不安全，谨慎使用）

2. 导入例子

导入前准备（可选）

create tablespace tbs datafile size 2g autoextend on next 100m maxsize 30g;create role role_user;
grant CREATE JOB,CREATE MATERIALIZED VIEW,CREATE PUBLIC DATABASE LINK,CREATE PUBLIC SYNONYM,CREATE VIEW,ALTER SESSION to role_user;create user USER_NAME identified by "xxxx";
ALTER USER USER_NAME QUOTA UNLIMITED ON tbs;
grant role_user,connect,resource to USER_NAME;
alter user USER_NAME default tablespace tbs;--drop tablespace tbs including contents and datafiles;
--drop user USER_NAME cascade;

导入指定用户数据（需要与源库相同数据/temp表空间存在）

impdp system DIRECTORY=DATA_PUMP schemas=USER_NAME dumpfile=USER_NAME1025.dmp logfile=USER_NAME1025.log

导入指定用户部分表

impdp system DIRECTORY=DATA_PUMP schemas=USER_NAME tables=USER_NAME.t1,USER_NAME.t2,USER_NAME.t3 dumpfile=USER_NAME1206.dmp logfile=USER_NAME1026.log

导入到其他用户及表空间

impdp system DIRECTORY=DATA_PUMP  schemas=USER_NAME  dumpfile=USER_NAME1206.dmp remap_schema=USER_NAME:userbak remap_tablespace=old:new logfile=USER_NAME1026.log

导入时重命名表

impdp system/oracle directory=dump dumpfile=expdp_T_A.dmp logfile= t_a.log REMAP_TABLE=scott.T_A:EMPS

目标库与源库相同

type默认不能导入到相同的数据库中，因为type的创建SQL中包含源库oid信息，导入时会引起oid冲突，进而导致相关表不能创建。可以使用impdp的transform参数重新生成原有dmp文件中的oid映射，transform=oid:n含义为不加载oid信息而重新生成。

impdp scottback/scottback@wilson directory=my_dir dumpfile=scott.dmp logfile=resimp2.log remap_schema=scott:scottback transform=oid:n

将dmp文件转为sql语句（不真正导入）

impdp scott/tiger directory=dir_name dumpfile=myfile sqlfile=xyz.sql

四、expdp/impdp管理

交互式命令列表

Activity	Command Used
Exit interactive-command mode.	CONTINUE_CLIENT
Stop the import client session, but leave the current job running.	EXIT_CLIENT
Display a summary of available commands.	HELP
Detach all currently attached client sessions and terminate the current job.	KILL_JOB
Increase or decrease the number of active worker processes for the current job. This command is valid only in Oracle Database Enterprise Edition.	PARALLEL
Restart a stopped job to which you are attached.	START_JOB
Display detailed status for the current job.	STATUS
Stop the current job.	STOP_JOB

查看dump job信息

select * from dba_datapump_jobs;

连接到现有dump job（查看dump进度）

--设置attach为job_name即可查看进度。
expdp system/oracle attach=SYS_EXPORT_SCHEMA_01
--或者
impdp system/oracle attach=SYS_EXPORT_SCHEMA_01

输出大致如下

[oracle@ocm1 ~]$ expdp system/oracle attach=SYS_EXPORT_SCHEMA_01...
Job: SYS_EXPORT_SCHEMA_01Owner: SYSTEM                         Operation: EXPORT                         Creator Privs: TRUE                           GUID: 482CA277B7350F88E053479DA8C00EFBStart Time: Friday, 10 February, 2017 20:57:05Mode: SCHEMA                         Instance: PROD1Max Parallelism: 1EXPORT Job Parameters:Parameter Name      Parameter Value:CLIENT_COMMAND        system/******** schemas=eoda directory=ext dumpfile=eoda.dmp State: EXECUTING                      Bytes Processed: 2,113,260,264Percent Done: 65Current Parallelism: 1Job Error Count: 0Dump File: /home/oracle/ext/eoda.dmpbytes written: 2,113,449,984Worker 1 Status:Process Name: DW00State: EXECUTING                      Object Schema: EODAObject Name: NEW_TABLE  <---------Object Type: SCHEMA_EXPORT/TABLE/TABLE_DATA  <---------Completed Objects: 3Total Objects: 52Completed Rows: 1,338,029 <---------Worker Parallelism: 1

暂停dump job

执行ctrl+c
# 或者
expdp system/oracle attach=SYS_EXPORT_SCHEMA_02
Export> STOP_JOB

开始暂停的dump job

Export> START_JOB
Export> CONTINUE_CLIENT# 重新启动暂停的dump job并跳过当前导入/出的对象
Export> START_JOB=SKIP_CURRENT=YES

终止dump job

expdp system/oracle attach=SYS_EXPORT_SCHEMA_02
Export> stop_job=immediate
# 或者
KILL_JOB

参考

Commands Available in Import's Interactive-Command Mode

https://www.cnblogs.com/huacw/p/3888807.html

https://blog.csdn.net/gxc1222/article/details/79272766/

Kill, cancel, resume or restart datapump expdp and impdp jobs (ORA-31626, ORA-31633, ORA-06512, ORA-00955) - Oracle 48

Oracle数据泵(Data Dump)错误汇集 - 潇湘隐者 - 博客园

convert expdp dmp file to SQL DDL statements

如何清除 DBA_DATAPUMP_JOBS 视图中的异常数据泵作业?(文档 ID 1626201.1)