臨時(shí)表空間恢復(fù)

1.建立臨時(shí)表空間

SQL> create temporary tablespace temp123 tempfile '/u01/app/oracle/oradata/orcl/temp123.dbf' size 20M;

2.更改為默認(rèn)

SQL> alter database default temporary tablespace temp123;

3.SQL> drop tablespace temp;(刪除)

==========================================================

日志文件,控制文件恢復(fù)

1.刪除2個(gè)日志文件

[oracle@sq orcl]$ rm -rf redo01.log redo03.log?

刪除所有控制文件

[oracle@sq orcl]$ rm -rf *.ctl

2.

SQL> startup force(報(bào)錯(cuò))

3.

CREATE CONTROLFILE REUSE DATABASE "ORCL" RESETLOGS ?ARCHIVELOG

? ? MAXLOGFILES 16

? ? MAXLOGMEMBERS 3

? ? MAXDATAFILES 100

? ? MAXINSTANCES 8

? ? MAXLOGHISTORY 292

LOGFILE

? GROUP 1 '/u01/app/oracle/oradata/orcl/redo01.log' ?SIZE 50M,

? GROUP 2 '/u01/app/oracle/oradata/orcl/redo02.log' ?SIZE 50M,

? GROUP 3 '/u01/app/oracle/oradata/orcl/redo03.log' ?SIZE 50M

-- STANDBY LOGFILE

DATAFILE

? '/u01/app/oracle/oradata/orcl/system01.dbf',

? '/u01/app/oracle/oradata/orcl/undotbs01.dbf',

? '/u01/app/oracle/oradata/orcl/sysaux01.dbf',

? '/u01/app/oracle/oradata/orcl/users01.dbf',

? '/u01/app/oracle/oradata/orcl/aaa.dbf'

CHARACTER SET WE8ISO8859P1

;

4.SQL> select open_mode from v$database;

5.SQL> alter database open;(報(bào)錯(cuò))

ORA-01589: must use RESETLOGS or NORESETLOGS option for database open

SQL> alter database open resetlogs;(報(bào)錯(cuò))

SQL> recover database; (報(bào)錯(cuò))

查看恢復(fù)目錄

SQL> show parameter DB_RECOVERY_FILE_DEST

SQL> recover database using BACKUP CONTROLFILE;

(using backup controlfile. ?恢復(fù)就不會受“當(dāng)前controlfile所紀(jì)錄的SCN”的限制) 成功

提示中輸入: ?/u01/app/oracle/oradata/orcl/redo02.log (當(dāng)前就一個(gè)日志文件)

SQL> alter database open resetlogs;

==========================================

部分?jǐn)?shù)據(jù)庫恢復(fù)

1.SQL> alter tablespace users begin backup;

? [oracle@sq orcl]$ cp users01.dbf /home/oracle/db_bak/

2.SQL> alter tablespace users end backup;?

??

3.SQL> create table tt as select * from dba_objects;

? SQL> select count(*) from tt;

? SQL> commit;

4.[oracle@sq orcl]$ rm -rf users01.dbf (模擬損壞)

5.SQL> alter system flush buffer_cache;

SQL> conn sys/123456 as sysdba

6.SQL> startup force;(報(bào)錯(cuò) 缺少4號文件)

7.SQL> alter database datafile 4 offline;

? [oracle@sq orcl]$ cp /home/oracle/db_bak/users01.dbf .

8.SQL> recover datafile 4;

SQL> alter database datafile 4 online;

9.SQL> startup force;

============================================

完全數(shù)據(jù)庫恢復(fù)

步驟跟上面一樣

recover database

============================================

歸檔模式數(shù)據(jù)文件恢復(fù)

1.SQL> create tablespace tbs_recover datafile '/oracle/app/oradata/TEST/tbs_recover.dbf' size 10M;

2.

SQL> create table tab_recover tablespace tbs_recover as select * from dba_objects;

SQL> commit;

3.SQL> alter system switch logfile;

SQL> /

4.[oracle@sq123 TEST]$ rm -rf tbs_recover.dbf?

5.SQL> alter system flush buffer_cache;

6.SQL> conn sys/123456 as sysdba

? SQL> select * from tab_recover;(報(bào)錯(cuò))

7.SQL> alter database datafile 5 offline;

8.SQL> alter database create datafile '/oracle/app/oradata/TEST/tbs_recover.dbf';

(創(chuàng)建出表空間)

9.SQL> alter database datafile 5 online;(報(bào)錯(cuò))

SQL> recover datafile 5;

AUTO

10.SQL> alter database datafile 5 online;(成功)

? ?SQL> select * from tab_recover;(數(shù)據(jù)回來了)

=============================================

不完全數(shù)據(jù)庫恢復(fù)

日志挖掘技術(shù)確定誤操作時(shí)間與scn

查看當(dāng)前時(shí)間

alter database begin backup;

cp *.dbf /home/oracle/db_bak

alter database end backup;

SQL> select to_char(sysdate,'yyyy-mm-dd hh24:mi:ss') from dual; ?2012-04-24 07:20:34

SQL> drop table tab_recover;

1.日志挖掘技術(shù),確定誤刪除表的時(shí)間

[oracle@sq123 ~]$ mkdir logminer

存儲目錄

SQL> alter system set utl_file_dir='/home/oracle/logminer' scope=spfile;

SQL> shutdown immediate

SQL> startup

2.創(chuàng)建存儲數(shù)據(jù)字典文件的文件

SQL> execute dbms_logmnr_d.build('shwdict.ora','/home/oracle/logminer');

cd /home/oracle/logminer/

3.

SQL> execute dbms_logmnr.add_logfile('/oracel/app/oradtaa/TEST/',dbms_logmnr.new);

4.挖掘數(shù)據(jù)

SQL> exec dbms_logmnr.start_logmnr (dictfilename=>'/home/oracle/logminer/shwdict.ora');

5.

SQL> alter session set nls_date_format='yyyy-mm-dd hh24:mi:ss';

SQL> select SCN,TIMESTAMP,SQL_REDO from v$LOGMNR_CONTENTS where SQL_REDO like '%drop%';

結(jié)果:

? ? 716922 2012-03-12 06:12:38

drop table tab_recover AS "BIN$uv7sdpUIi1TgQAB/AQByfw==$0" ;

恢復(fù)數(shù)據(jù)

6.

SQL> shutdown abort;

SQL> startup mount

[oracle@sq123 db_bak]$ cp *.dbf /oracle/app/oradata/TEST/

SQL> alter session set nls_date_format='yyyy-mm-dd hh24:mi:ss';

7.

SQL> recover database until time '2012-04-24 07:20:34';

SQL> alter database open resetlogs;

8.

SQL> conn sys/123456 as sysdba

SQL> select count(*) from tab_recover; (數(shù)據(jù)恢復(fù)出來了)

=============================================

日志文件恢復(fù)

1.[oracle@sq orcl]$ rm *.log

? SQL> startup force(報(bào)錯(cuò))

2.SQL> alter database open resetlogs;(報(bào)錯(cuò))

3.SQL> recover database until cancel;

(執(zhí)行RECOVER命令，在提示輸入歸檔日志時(shí)，應(yīng)用這些歸檔日志提示輸入歸檔日志序列號時(shí)，不輸入這個(gè)文件的名稱或者不接受建議的名稱，而是輸入CANCEL)

(恢復(fù)不能成功,因?yàn)槲覀儧]有歸檔日志了)

4.添加隱藏參數(shù)

SQL> alter system set "_allow_resetlogs_corruption"=true scope=spfile;

(強(qiáng)制啟動數(shù)據(jù)庫，設(shè)置此參數(shù)之后，在數(shù)據(jù)庫Open過程中，Oracle會跳過某些一致性檢查，從而使數(shù)據(jù)庫可能跳過不一致狀態(tài)，Open打開)

5.SQL> startup force

? SQL> alter database open resetlogs;

? (可以啟動,但是剛才沒有同步到硬盤的數(shù)據(jù)丟失了)

--后續(xù)操作

SQL> create pfile from spfile

然后vi pfile文件?

刪除 *._allow_resetlogs_corruption

SQL> create spfile from pfile;

SQL> startup force

===============================================

轉(zhuǎn)載于:https://blog.51cto.com/chenjisong/1737391

總結(jié)

以上是生活随笔為你收集整理的数据库备份 恢复的全部內(nèi)容，希望文章能夠幫你解決所遇到的問題。

如果覺得生活随笔網(wǎng)站內(nèi)容還不錯(cuò)，歡迎將生活随笔推薦給好友。