包含了MVIEW的表领域，在进行导出，表领域改名，再导入后，MVIEW会消失不见。

测试环境12.1.0.2

=================
步骤1：数据的准备

[oracle@db12102 admin]$ sqlplus / as sysdba

SQL*Plus: Release 12.1.0.2.0 Production on 木 8月 17 13:01:48 2017
...

SQL> alter session set container=pdb1;
SQL> create tablespace userdata1 datafile '/home/oracle/data5.dbf' size 20M;

SQL> grant dba to user001 identified by user001;
SQL> alter user user001 default tablespace userdata1;

SQL> conn user001/user001@mypdb1

SQL> create table journal_tab001(ss_pk integer primary key, base_date date, cancel_date date,app_date date);
SQL> create materialized view log on journal_tab001;

SQL> insert into journal_tab001 values(1,sysdate,sysdate,sysdate);
SQL> insert into journal_tab001 values(2,sysdate,sysdate,sysdate);
SQL> insert into journal_tab001 values(3,sysdate,sysdate,sysdate);
SQL> commit;

创建 MVIEW：

SQL> create materialized view journal_tab001_MDATE_MV  TABLESPACE USERDATA1
       AS SELECT  ss_pk,
                  MAX (GREATEST (base_date, NVL (cancel_date,TO_DATE ('00010101', 'yyyymmdd')),app_date) ) basedate
          FROM journal_tab001 GROUP BY ss_pk;  2

SQL> select * from journal_tab001_MDATE_MV;

SS_PK          BASEDATE
-------------- --------
            1 17-08-17
            2 17-08-17
            3 17-08-17

=================
步骤2：数据的导出

SQL> create or replace directory DATAPUMP_DIR_D001 as '/home/oracle';

SQL> show con_name;

CON_NAME
------------------------------
PDB1
SQL>

以表领域 模式导出：

[oracle@db12102 ~]$ expdp "'"sys/oracle@mypdb1 as sysdba"'" tablespaces=USERDATA1 directory=DATAPUMP_DIR_D001 dumpfile=userdata1.dmp logfile=userdata1_export.log

Export: Release 12.1.0.2.0 - Production on 木 8月 17 13:39:02 2017

Copyright (c) 1982, 2014, Oracle and/or its affiliates.  All rights reserved.

Connect to: Oracle Database 12c Enterprise Edition Release 12.1.0.2.0 - 64bit Production
With the Partitioning, OLAP, Advanced Analytics and Real Application Testing options
"SYS". "SYS_EXPORT_TABLESPACE_01" is running: "sys / ******** @ mypdb1 AS SYSDBA" tablespaces = USERDATA1 directory = DATAPUMP_DIR_D001 dumpfile = userdata1.dmp logfile = userdata1_export.log
Estimating using BLOCKS method ...
Processing of object type TABLE_EXPORT / TABLE / TABLE_DATA
Total estimates using the BLOCKS method: 192 KB
Processing of object type TABLE_EXPORT / TABLE / TABLE
Processing of object type TABLE_EXPORT / TABLE / COMMENT
Processing of object type TABLE_EXPORT / TABLE / INDEX / INDEX
Processing of object type TABLE_EXPORT / TABLE / INDEX / FUNCTIONAL_INDEX / INDEX
Processing of object type TABLE_EXPORT / TABLE / CONSTRAINT / CONSTRAINT
Processing of object type TABLE_EXPORT / TABLE / INDEX / STATISTICS / INDEX_STATISTICS
Processing of object type TABLE_EXPORT / TABLE / INDEX / STATISTICS / FUNCTIONAL_INDEX / INDEX_STATISTICS
Processing of object type TABLE_EXPORT / TABLE / STATISTICS / TABLE_STATISTICS
Processing of object type TABLE_EXPORT / TABLE / STATISTICS / MARKER
Processing of object type TABLE_EXPORT / TABLE / POST_TABLE_ACTION
. "user001". "journal_tab001" 6.445 KB 3 rows have been exported
. "user001". "journal_tab001_MDATE_MV" 5.984 KB 3 rows have been exported
. "user001". "MLOG $ _journal_tab001" 7.273 KB 3 rows have been exported
Master table "SYS". "SYS_EXPORT_TABLESPACE_ 01" has been successfully loaded / unloaded
************************************************** ****************************
The dump file set to SYS.SYS_EXPORT_TABLESPACE_01 is as follows:
 /home/oracle/userdata1.dmp
Job "SYS". "SYS_EXPORT_TABLESPACE_01" completed successfully with tree August 17 13:39:55 2017 elapsed 0 00: 00: 50

[oracle@db12102 ~]$

===

SQL> set pages 0
SQL> set longchunksize 3000
SQL> set long 2000000000

SQL> select dbms_metadata.get_ddl('MATERIALIZED_VIEW','journal_tab001_MDATE_MV','user001') from dual;

CREATE MATERIALIZED VIEW "user001"."journal_tab001_MDATE_MV" ("ss_pk", "BASEDATE")
 ORGANIZATION HEAP PCTFREE 10 PCTUSED 40 INITRANS 1 MAXTRANS 255
NOCOMPRESS LOGGING
 STORAGE(INITIAL 65536 NEXT 1048576 MINEXTENTS 1 MAXEXTENTS 2147483645
 PCTINCREASE 0 FREELISTS 1 FREELIST GROUPS 1
 BUFFER_POOL DEFAULT FLASH_CACHE DEFAULT CELL_FLASH_CACHE DEFAULT)
 TABLESPACE "USERDATA1"
 BUILD IMMEDIATE
 USING INDEX PCTFREE 10 INITRANS 2 MAXTRANS 255
 STORAGE(INITIAL 65536 NEXT 1048576 MINEXTENTS 1 MAXEXTENTS 2147483645
 PCTINCREASE 0 FREELISTS 1 FREELIST GROUPS 1
 BUFFER_POOL DEFAULT FLASH_CACHE DEFAULT CELL_FLASH_CACHE DEFAULT)
 TABLESPACE "USERDATA1"
 REFRESH FORCE ON DEMAND
 USING DEFAULT LOCAL ROLLBACK SEGMENT
 USING ENFORCED CONSTRAINTS DISABLE QUERY REWRITE
 AS SELECT  ss_pk, MAX (GREATEST (base_date, NVL (cancel_date,TO_
DATE ('00010101', 'yyyymmdd')),app_date) ) basedate FROM journal_tab001 GROUP BY ss_pk

SQL>

SQL> alter tablespace USERDATA1 rename to USEROLD; <<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<   为了导出作准备，原有表领域改名

SQL> select dbms_metadata.get_ddl('MATERIALIZED_VIEW','journal_tab001_MDATE_MV','user001') from dual;

CREATE MATERIALIZED VIEW "user001"."journal_tab001_MDATE_MV" ("ss_pk", "
BASEDATE")
 ORGANIZATION HEAP PCTFREE 10 PCTUSED 40 INITRANS 1 MAXTRANS 255
NOCOMPRESS LOGGING
 STORAGE(INITIAL 65536 NEXT 1048576 MINEXTENTS 1 MAXEXTENTS 2147483645
 PCTINCREASE 0 FREELISTS 1 FREELIST GROUPS 1
 BUFFER_POOL DEFAULT FLASH_CACHE DEFAULT CELL_FLASH_CACHE DEFAULT)
 TABLESPACE "USEROLD"
 BUILD IMMEDIATE
 USING INDEX PCTFREE 10 INITRANS 2 MAXTRANS 255
 STORAGE(INITIAL 65536 NEXT 1048576 MINEXTENTS 1 MAXEXTENTS 2147483645
 PCTINCREASE 0 FREELISTS 1 FREELIST GROUPS 1
 BUFFER_POOL DEFAULT FLASH_CACHE DEFAULT CELL_FLASH_CACHE DEFAULT)
 TABLESPACE "USEROLD" ★
 REFRESH FORCE ON DEMAND
 USING DEFAULT LOCAL ROLLBACK SEGMENT
 USING ENFORCED CONSTRAINTS DISABLE QUERY REWRITE
 AS SELECT  ss_pk, MAX (GREATEST (base_date, NVL (cancel_date,TO_
DATE ('00010101', 'yyyymmdd')),app_date) ) basedate FROM journal_tab001 GROUP BY ss_pk

SQL>

SQL> create tablespace USERDATA1 LOGGING DATAFILE '/home/oracle/data6.dbf' SIZE 20M; <<<<<<<<<<<<<<<<<<<<<<<<<,创建同名的表领域

SQL> select object_id,object_name,object_type, status from dba_objects where owner='user001' order by object_id;  [***1]

91961 journal_tab001                    TABLE                          VALID
    91962 SYS_C0010023                   　 INDEX                          VALID
    91963 MLOG$_journal_tab001              TABLE                          VALID
    91964 RUPD$_journal_tab001              TABLE                          VALID
    91965 I_MLOG$_journal_tab001            INDEX                          VALID
    91966 journal_tab001_MDATE_MV           TABLE                          VALID
    91967 I_SNAP$_journal_tab001_MAXDAT     INDEX                          VALID
    91968 journal_tab001_MDATE_MV           MATERIALIZED VIEW              VALID

SQL>

=================
步骤3：数据的导入

[oracle @ db12102 ~] $ impdp "'" sys / oracle @ mypdb1 as sysdba "'" tablespaces = USERDATA1 directory = DATAPUMP_DIR_D001 dumpfile = userdata1.dmp logfile = userdata1_import.log table_exists_action = replace

Import: Release 12.1.0.2.0 - Production on Tree August 17 13:45:58 2017

Copyright (c) 1982, 2014, Oracle and / or its affiliates. All rights reserved.

Connect to: Oracle Database 12c Enterprise Edition Release 12.1.0.2.0 - 64bit Production
With the Partitioning, OLAP, Advanced Analytics and Real Application Testing options
The master table "SYS". "SYS_IMPORT_TABLESPACE_ 02" was successfully loaded / unloaded
"SYS". "SYS_IMPORT_TABLESPACE_02" is running: "sys / ******** @ mypdb1 AS SYSDBA" tablespaces = USERDATA1 directory = DATAPUMP_DIR_D001 dumpfile = userdata1.dmp logfile = userdata1_import.log table_exists_action = replace
Processing of object type TABLE_EXPORT / TABLE / TABLE
Processing of object type TABLE_EXPORT / TABLE / TABLE_DATA
. "user001". "journal_tab001" 6.445 KB 3 rows imported
. "user001". "journal_tab001_MDATE_MV" 5.984 KB 3 rows imported
. "user001". "MLOG $ _journal_tab001" 7.273 KB 3 rows imported
Processing of object type TABLE_EXPORT / TABLE / COMMENT
Processing of object type TABLE_EXPORT / TABLE / INDEX / INDEX
Processing of object type TABLE_EXPORT / TABLE / INDEX / FUNCTIONAL_INDEX / INDEX
Processing of object type TABLE_EXPORT / TABLE / CONSTRAINT / CONSTRAINT
Processing of object type TABLE_EXPORT / TABLE / INDEX / STATISTICS / INDEX_STATISTICS
Processing of object type TABLE_EXPORT / TABLE / INDEX / STATISTICS / FUNCTIONAL_INDEX / INDEX_STATISTICS
Processing of object type TABLE_EXPORT / TABLE / STATISTICS / TABLE_STATISTICS
Processing of object type TABLE_EXPORT / TABLE / STATISTICS / MARKER
Processing of object type TABLE_EXPORT / TABLE / POST_TABLE_ACTION
Job "SYS". "SYS_IMPORT_TABLESPACE_02" completed successfully with tree August 17 13:46:18 2017 elapsed 0 00: 00: 17

SQL> select object_id,object_name,object_type, status from dba_objects where owner='user001' order by object_id;   [***2]

92014 journal_tab001                    TABLE                          VALID
    92015 MLOG$_journal_tab001              TABLE                          VALID
    92016 journal_tab001_MDATE_MV           TABLE                          VALID
    92017 I_MLOG$_journal_tab001            INDEX                          VALID
    92018 I_SNAP$_journal_tab001_MAXDAT     INDEX                          VALID
    92019 SYS_C0010034                      INDEX                          VALID

对比上面的 [***1] 和 [**2], 可以发现，缺少了几个内部表。journal_tab001_MDATE_MV 也不见了。

SQL>
SQL> select dbms_metadata.get_ddl ('MATERIALIZED_VIEW', 'journal_tab001_MDATE_MV', 'user001') from dual;
ERROR:
ORA-31603: Object "journal_tab001_MDATE_MV" (type MATERIALIZED_VIEW) not found in schema "user001". ORA-06512:
"SYS.DBMS_METADATA", line 6069
ORA-06512: "SYS.DBMS_METADATA", line 8666
ORA-06512: Row 1

no record selected.

事实上，即便是如下的方式来进行，导入后也会发现 mview 不见了。
   Export、MVIEW删除、表领域改名、再新建一个同名表领域、Import (过程略)

此现象，是否是 Bug， 是不明确的。 在如下几种环境中都会发生：

　 11.2.0.4
　 12.1.0.2 Non-CDB
　 12.1.0.2 CDB/PDB
   12.2.0.1 Non-CDB
   12.2.0.1 CDB/PDB

对策是 ：

通过 Schema 模式来进行导出/导入。

[Oracle]包含了MVIEW的表领域，在进行导出，表领域改名，再导入后，MVIEW会消失不见。的更多相关文章

  1. Oracle - 数据库的实例、表空间、用户、表之间关系

    完整的Oracle数据库通常由两部分组成:Oracle数据库和数据库实例. 1) 数据库是一系列物理文件的集合(数据文件,控制文件,联机日志,参数文件等): 2) Oracle数据库实例则是一组Ora ...

  2. 【转】Oracle - 数据库的实例、表空间、用户、表之间关系

    [转]Oracle - 数据库的实例.表空间.用户.表之间关系 完整的Oracle数据库通常由两部分组成:Oracle数据库和数据库实例. 1) 数据库是一系列物理文件的集合(数据文件,控制文件,联机 ...

  3. oracle高级查询（实例基于scott用户四张表）

    oracle高级查询(实例基于scott用户四张表) 分组查询 多表查询 子查询 综合实例 ====================================================== ...

  4. oracle下的数据库实例、表空间、用户及其表的区分

    完整的Oracle数据库通常由两部分组成:Oracle数据库和数据库实例. 1) 数据库是一系列物理文件的集合(数据文件,控制文件,联机日志,参数文件等): 2) Oracle数据库实例则是一组Ora ...

  5. [转]Oracle - 数据库的实例、表空间、用户、表之间关系

    本文转自:http://www.cnblogs.com/adforce/p/3312252.html 完整的Oracle数据库通常由两部分组成:Oracle数据库和数据库实例. 1) 数据库是一系列物 ...

  6. Oracle 优化器_访问数据的方法_单表

    Oracle 在选择执行计划的时候,优化器要决定用什么方法去访问存储在数据文件中的数据.我们从数据文件中查询到相关记录,有两种方法可以实现:1.直接访问表记录所在位置.2.访问索引,拿到索引中对应的r ...

  7. Oracle大表改为分区表及表空间切换方案

    Oracle大表改为分区表及表空间切换方案 一.            背景 由于之前数据库表和索引放在一个表空间导致表空间数据文件增长太快,文件数量即将达到Oracle表空间的限制,需要对表(没有分 ...

  8. Oracle基础维护02-表、主键、索引、表结构维护手册

    目录 一.项目新建表.主键.索引注意事项 二.举例说明建表.主建.索引的操作方法 2.1 设定需求如下 2.1.1 查询数据库有哪些表空间 2.1.2 本文档假设数据库有这两个业务用户的表空间 2.2 ...

  9. oracle的minus返回第一个表中有、第二个表中没有的数据

    oracle的minus返回第一个表中有.第二个表中没有的数据 CREATE TABLE hovertree_union_1 ( id INT, val ) ); CREATE TABLE hover ...

随机推荐

  1. 测试思想-测试设计 史上最详细测试用例设计实践总结 Part2

    史上最详细测试用例设计实践总结 by:授客 QQ:1033553122 -------------------------接 Part1-------------------------- 方法:这里 ...

  2. python中strip()方法学习笔记

    Python strip() 方法用于移除字符串头尾指定的字符(默认为空格). 当使用strip('xxx'),只要字符串头尾有"xxx"中的一个,就会去掉,而不是符合字符串''x ...

  3. sql server递归

    with cte as ( select belongsAgent from [QPProxyDB].[dbo].[BS_ProxyInfo] where ProxyID = @ProxyID uni ...

  4. knockoutjs关于ko.bindingHandlers的updata订阅

    ko.bindingHandlers是先执行init进行初始化数据的绑定(如果需要执行updata进行数据更新可以不用初始化); init: function(element, valueAccess ...

  5. SQL易错总结1

    SQL易错总结1 进阶 select语句.select * 查询所有不规范,写出要查的属性.distinct慎用,性能消耗很大 like 模糊查询 ,空值判断是 is null 单行函数:lower( ...

  6. python+mongodb+flask的基本使用

    最近在做一个设备管理系统的后端,需要用python结合mongodb来实现,查了一下flask框架是比较合适的,自己摸索了好久一步步慢慢实现基本功能. 在程序开始之前请确保mongodb服务是开启的, ...

  7. February 10th, 2018 Week 6th Saturday

    It is not enough to have a good mind. The main thing is to use it well. 头脑聪明还不够,重要的是好好运用. From Rene ...

  8. GUI_菜单练习

    package com.mywindow.test; import java.awt.event.ActionEvent; import java.awt.event.ActionListener; ...

  9. (9)Python循环结构

  10. 月球美容计划之最小生成树（MST）

    寒假学的两个算法,普里姆,克鲁斯卡尔最终弄明确了.能够发总结了 先说说普里姆,它的本质就是贪心.先从随意一个点開始,找到最短边,然后不断更新更新len数组,然后再选取最短边并标记经过的点,直到全部的点 ...