【Oracle】undo损坏，无备份非常规恢复

客户的一个测试环境，主机异常断电，启动后发现undo文件损坏，无法启动，在open阶段报错如下：

Errors in file /u01/app/oracle/diag/rdbms/cdrdb/CDRDB/trace/CDRDB_ora_4109.trc:
ORA-01122: database file 3 failed verification check
ORA-01110: data file 3: '/u01/app/oracle/oradata/CDRDB/undotbs01.dbf'
ORA-01210: data file header is media corrupt
ORA-1122 signalled during: ALTER DATABASE OPEN...

由于是测试环境，没有备份，但是又需要里边的一些数据，所以我尝试使用非常规恢复方法进行了尝试。

先冷备份现有环境！！！！！！！！！！！！！！！！

创建pfile文件：

create pfile from spfile;
在pfile中修改这两个参数

#*.undo_tablespace='UNDOTBS1'
*.undo_management= MANUAL
之后用这个pfile启动：

SYS@CDRDB>startup force pfile='/u01/app/oracle/product/11.2.0/dbhome_1/dbs/initCDRDB.ora';
ORACLE instance started.

Total System Global Area 523108352 bytes
Fixed Size 1337632 bytes
Variable Size 364906208 bytes
Database Buffers 150994944 bytes
Redo Buffers 5869568 bytes
Database mounted.
ORA-01122: database file 3 failed verification check
ORA-01110: data file 3: '/u01/app/oracle/oradata/CDRDB/undotbs01.dbf'
ORA-01210: data file header is media corrupt
仍然报错

之后尝试drop掉这个undo

SYS@CDRDB>alter database datafile 3 offline drop;

Database altered.
之后重新开库

SYS@CDRDB>alter database open;

Database altered.
之后创建新的undo表空间undotbs2

SYS@CDRDB>create undo tablespace undotbs2 datafile '/u01/app/oracle/oradata/CDRDB/undotbs02_01.dbf' size 100M;
create undo tablespace undotbs2 datafile '/u01/app/oracle/oradata/CDRDB/undotbs02_01.dbf' size 100M
*
ERROR at line 1:
ORA-00604: error occurred at recursive SQL level 1
ORA-01552: cannot use system rollback segment for non-system tablespace
'DATA_OL'
ORA-06512: at line 999
ORA-01552: cannot use system rollback segment for non-system tablespace
'DATA_OL'
产生报错如上

比较疑惑创建undo为什么会影响到DATA_OL表空间,所以做了个10046

SYS@CDRDB>oradebug event 10046 trace name context off
Statement processed.
SYS@CDRDB>oradebug tracefile_name
/u01/app/oracle/diag/rdbms/cdrdb/CDRDB/trace/CDRDB_ora_4279.trc
[oracle@centos-1 ~]$ tkprof /u01/app/oracle/diag/rdbms/cdrdb/CDRDB/trace/CDRDB_ora_4279.trc
output = 1.trm
查看1.trm，发现了原因：

"OGG".DDLReplication.dbQueried IS NULL THEN
SELECT database_role,
open_mode
INTO dbRole, dbOpenMode
FROM v$database;

"OGG".DDLReplication.dbQueried := TRUE;
END IF;
IF NOT (
^@ (dbRole = 'PRIMARY' OR dbRole = 'LOGICAL STANDBY')
AND dbOpenMode =
'READ WRITE'
)
THEN
-- do not write any trace even though it
should work as this is standby
"OGG"
.DDLReplication.setCtxInfo(-1,-1,-1,-1,-1);
RETURN; -- do not use
trigger if not read/write and primary/logical_standby
END IF;
EXCEPTION
......略

原来是因为这个库配置过OGG的DDL同步，有DDL产生时会产生insert操作，使用DATA_OL表空间。

原因找到~ 跑脚本关闭该库OGG的DDL配置即可

SYS@CDRDB>@ddl_disable.sql

Trigger altered.
之后再重新创建undotbs2表空间

SYS@CDRDB>create undo tablespace undotbs2 datafile '/u01/app/oracle/oradata/CDRDB/undotbs02_01.dbf' size 1000M;

Tablespace created.
这次很顺利的添加成功。

undotbs2添加完毕，接下来关闭数据库，修改pfile参数

*.undo_tablespace='UNDOTBS2'
*.undo_management=AUTO

重启数据库用pfile启动

SYS@CDRDB>startup pfile='/u01/app/oracle/product/11.2.0/dbhome_1/dbs/initCDRDB.ora';
ORACLE instance started.

Total System Global Area 523108352 bytes
Fixed Size 1337632 bytes
Variable Size 364906208 bytes
Database Buffers 150994944 bytes
Redo Buffers 5869568 bytes
Database mounted.
Database opened.
数据库成功启动，然而在很多业务表查询时依然会报错：

ERROR at line 1:
ORA-00376: file 3 cannot be read at this time
ORA-01110: data file 3: '/u01/app/oracle/oradata/CDRDB/undotbs01.dbf'
依然需要原来的undotbs01.dbf回滚数据库崩溃时未提交的事务

之后我先尝试了设置event 10513，来屏蔽smon的回滚

SYS@CDRDB>alter system set events '10513 trace name context forever, level 2';

System altered.
发现还是会报之前的错误，看来需要自己手工去屏蔽回滚段了。

SYS@CDRDB>SELECT SEGMENT_NAME,STATUS FROM DBA_ROLLBACK_SEGS;

SEGMENT_NAME STATUS
------------------------------ ----------------
SYSTEM ONLINE
_SYSSMU10_4131489474$ NEEDS RECOVERY
_SYSSMU9_1735643689$ NEEDS RECOVERY
_SYSSMU8_3901294357$ NEEDS RECOVERY
_SYSSMU7_3517345427$ NEEDS RECOVERY
_SYSSMU6_2897970769$ NEEDS RECOVERY
_SYSSMU5_538557934$ NEEDS RECOVERY
_SYSSMU4_1003442803$ NEEDS RECOVERY
_SYSSMU3_1204390606$ NEEDS RECOVERY
_SYSSMU2_967517682$ NEEDS RECOVERY
_SYSSMU1_592353410$ NEEDS RECOVERY

SEGMENT_NAME STATUS
------------------------------ ----------------
_SYSSMU30_244658789$ ONLINE
_SYSSMU29_1020880693$ ONLINE
_SYSSMU28_2912622077$ ONLINE
_SYSSMU27_747253598$ ONLINE
_SYSSMU26_560868814$ ONLINE
_SYSSMU25_1357066082$ ONLINE
_SYSSMU24_103440716$ ONLINE
_SYSSMU23_1006903361$ ONLINE
_SYSSMU22_2808190508$ ONLINE
_SYSSMU21_39626587$ ONLINE

21 rows selected.

之后把查询到的NEEDS RECOVERY的表加到以下两个参数中，屏蔽这些回滚段

_offline_rollback_segments/_corrupted_rollback_segments参数
*._offline_rollback_segments=(_SYSSMU10_4131489474$,_SYSSMU9_1735643689$,_SYSSMU8_3901294357$,_SYSSMU7_3517345427$,_SYSSMU6_2897970769$,_SYSSMU5_538557934$,_SYSSMU4_1003442803$,_SYSSMU3_1204390606$,_SYSSMU2_967517682$,_SYSSMU1_592353410$)
*._corrupted_rollback_segments=(_SYSSMU10_4131489474$,_SYSSMU9_1735643689$,_SYSSMU8_3901294357$,_SYSSMU7_3517345427$,_SYSSMU6_2897970769$,_SYSSMU5_538557934$,_SYSSMU4_1003442803$,_SYSSMU3_1204390606$,_SYSSMU2_967517682$,_SYSSMU1_592353410$)

使用这个pfile启动数据库

startup force pfile='/u01/app/oracle/product/11.2.0/dbhome_1/dbs/initCDRDB.ora';

之后对那些测试表都查询了一下，确认可以查询，之后导出帮他们导出数据，一切OK~