数据文件采用错误方式删除后的解决办法

问题描述：当采用以下步骤删除数据文件后发现v$datafile视图中仍然有该文件记录,且状态recover，意识到该删除步骤有问题，生产环境中存在此种情况，此时作为dba应该怎么办？

a、 使用alter database datafile 5 offline;

b、rm 命令删掉数据文件

实验目标：模拟该异常，并提供解决方案.

实验环境：oracle 11.2.0.4+ 单实例

1、模拟异常

SQL>alter tablespace users add datafile '/u01/app/oracle/oradata/leodb/users02.dbf'size 10m autoextend on;

Tablespacealtered.

SQL>select file#,name from v$datafile

FILE# NAME

------------------------------------------------------------

1/u01/app/oracle/oradata/leodb/system01.dbf

2/u01/app/oracle/oradata/leodb/sysaux01.dbf

3/u01/app/oracle/oradata/leodb/undotbs01.dbf

4/u01/app/oracle/oradata/leodb/users01.dbf

5/u01/app/oracle/oradata/leodb/users02.dbf

SQL>select d.tablespace_name,v.file#,d.file_name,d.bytes/1024/1024,v.status,v.enabled,d.autoextensible,d.increment_byfrom dba_data_files d,v$datafile v where d.file_id=v.file#

TABLESPACE_NAME FILE# FILE_NAME D.BYTES/1024/1024 STATUS ENABLED AUT INCREMENT_BY

------------------------- -------------------------------------------------- ------------------------ ---------- --- ------------

USERS 4/u01/app/oracle/oradata/leodb/users01.dbf 5 ONLINE READ WRITE YES 160

UNDOTBS1 3/u01/app/oracle/oradata/leodb/undotbs01.dbf 70 ONLINE READ WRITE YES 640

SYSAUX 2/u01/app/oracle/oradata/leodb/sysaux01.dbf 490 ONLINE READ WRITE YES 1280

SYSTEM 1/u01/app/oracle/oradata/leodb/system01.dbf 740 SYSTEM READ WRITE YES 1280

USERS 5/u01/app/oracle/oradata/leodb/users02.dbf 10 ONLINE READ WRITE YES 1

现使用错误方法删数据文件users02.dbf

SQL>alter database datafile 5 offline;

Databasealtered.

现在系统层面删除数据文件users02.dbf

[oracle@leoleodb]$ ll

total1531236

-rw-r-----1 oracle asmadmin 9748480 Aug 7 13:54 control01.ctl

-rw-r-----1 oracle asmadmin 52429312 Aug 7 13:23 redo01.log

-rw-r-----1 oracle asmadmin 52429312 Aug 7 13:53 redo02.log

-rw-r-----1 oracle asmadmin 52429312 Aug 7 13:23 redo03.log

-rw-r-----1 oracle asmadmin 534781952 Aug 7 13:54sysaux01.dbf

-rw-r-----1 oracle asmadmin 775954432 Aug 7 13:54system01.dbf

-rw-r-----1 oracle asmadmin 30416896 Aug 7 12:33 temp01.dbf

-rw-r-----1 oracle asmadmin 73408512 Aug 7 13:54 undotbs01.dbf

-rw-r-----1 oracle asmadmin 5251072 Aug 7 13:23 users01.dbf

-rw-r-----1 oracle asmadmin 10493952 Aug 7 13:23 users02.dbf

[oracle@leoleodb]$ rm -rf users02.dbf

[oracle@leoleodb]$ pwd

/u01/app/oracle/oradata/leodb

此时通过以下语句查询发现users02.dbf状态为recover，v$datafile中仍然有该数据文件记录,意识到操作错误.

SQL> selectd.tablespace_name,v.file#,d.file_name,d.bytes/1024/1024,v.status,v.enabled,d.autoextensible,d.increment_byfrom dba_data_files d,v$datafile v where d.file_id=v.file#;

TABLESPACE_NAME FILE# FILE_NAME D.BYTES/1024/1024STATUS ENABLED AUT INCREMENT_BY

------------------------- -------------------------------------------------- ------------------------ ---------- --- ------------

USERS 4/u01/app/oracle/oradata/leodb/users01.dbf 5 ONLINE READ WRITE YES 160

UNDOTBS1 3/u01/app/oracle/oradata/leodb/undotbs01.dbf 70 ONLINE READ WRITE YES 640

SYSAUX 2/u01/app/oracle/oradata/leodb/sysaux01.dbf 510 ONLINE READ WRITE YES 1280

SYSTEM 1/u01/app/oracle/oradata/leodb/system01.dbf 740 SYSTEM READ WRITE YES 1280

USERS 5/u01/app/oracle/oradata/leodb/users02.dbf RECOVERREAD WRITE

SQL>select name from v$datafile;

NAME

--------------------------------------------------

/u01/app/oracle/oradata/leodb/system01.dbf

/u01/app/oracle/oradata/leodb/sysaux01.dbf

/u01/app/oracle/oradata/leodb/undotbs01.dbf

/u01/app/oracle/oradata/leodb/users01.dbf

/u01/app/oracle/oradata/leodb/users02.dbf

说明：成功模拟出异常现象.

2、解决方案

a、添加同路径同名数据文件

b、RECOVER并ONLINE该文件

c、用正确方法进行删除

操作过程：

SQL>alter database create datafile 5 as '/u01/app/oracle/oradata/leodb/users02.dbf'size 10m;

Databasealtered.

[oracle@leoleodb]$ ll

total1531244

-rw-r-----1 oracle asmadmin 9748480 Aug 7 14:24 control01.ctl

-rw-r-----1 oracle asmadmin 52429312 Aug 7 13:23 redo01.log

-rw-r-----1 oracle asmadmin 52429312 Aug 7 14:24 redo02.log

-rw-r-----1 oracle asmadmin 52429312 Aug 7 13:23 redo03.log

-rw-r-----1 oracle asmadmin 534781952 Aug 7 14:24sysaux01.dbf

-rw-r-----1 oracle asmadmin 775954432 Aug 7 14:24system01.dbf

-rw-r-----1 oracle asmadmin 30416896 Aug 7 14:18 temp01.dbf

-rw-r-----1 oracle asmadmin 73408512 Aug 7 14:23 undotbs01.dbf

-rw-r-----1 oracle asmadmin 5251072 Aug 7 13:23 users01.dbf

-rw-r-----1 oracle asmadmin 10493952 Aug 7 14:24 users02.dbf

SQL>recover datafile 5;

Mediarecovery complete.

SQL>alter database datafile '/u01/app/oracle/oradata/leodb/users02.dbf' online;

Databasealtered.

SQL>alter tablespace users drop datafile '/u01/app/oracle/oradata/leodb/users02.dbf';

Tablespacealtered.

SQL>host ls -l /u01/app/oracle/oradata/leodb/users02.dbf

ls:cannot access /u01/app/oracle/oradata/leodb/users02.dbf: No such file ordirectory

SQL>selectd.tablespace_name,v.file#,d.file_name,d.bytes/1024/1024,v.status,v.enabled,d.autoextensible,d.increment_byfrom dba_data_files d,v$datafile v where d.file_id=v.file#;

TABLESPACE_NAME FILE# FILE_NAME D.BYTES/1024/1024 STATUS ENABLED AUT INCREMENT_BY

------------------------- -------------------------------------------------- ------------------------ ---------- --- ------------

USERS 4/u01/app/oracle/oradata/leodb/users01.dbf 5 ONLINE READ WRITE YES 160

UNDOTBS1 3/u01/app/oracle/oradata/leodb/undotbs01.dbf 70 ONLINE READ WRITE YES 640

SYSAUX 2/u01/app/oracle/oradata/leodb/sysaux01.dbf 510 ONLINE READ WRITE YES 1280

SYSTEM 1/u01/app/oracle/oradata/leodb/system01.dbf 740 SYSTEM READ WRITE YES 1280

[oracle@leoleodb]$ ll

total1520996

-rw-r-----1 oracle asmadmin 9748480 Aug 7 14:26 control01.ctl

-rw-r-----1 oracle asmadmin 52429312 Aug 7 13:23 redo01.log

-rw-r-----1 oracle asmadmin 52429312 Aug 7 14:26 redo02.log

-rw-r-----1 oracle asmadmin 52429312 Aug 7 13:23 redo03.log

-rw-r-----1 oracle asmadmin 534781952 Aug 7 14:24sysaux01.dbf

-rw-r-----1 oracle asmadmin 775954432 Aug 7 14:24system01.dbf

-rw-r-----1 oracle asmadmin 30416896 Aug 7 14:18 temp01.dbf

-rw-r-----1 oracle asmadmin 73408512 Aug 7 14:23 undotbs01.dbf

-rw-r-----1 oracle asmadmin 5251072 Aug 7 13:23 users01.dbf

说明：users02.dbf被成功删除，系统层面及v$datafile中已不存在该文件记录