事情起由是在Oracle g手動建庫腳本中看到dbms_backup_restorezerodbid()過程其中作用是修改數據庫的dbid於是想通過該存儲直接在sqlplus中執行修改dbid
修改之前記錄其dbid
引用
SQL> select dbid from v$database;
DBID
修改dbid
引用
SQL> exec dbms_backup_restorezerodbid();
PL/SQL procedure successfully completed
貌似執行成功了但隨後alert日志顯示ckpt進程將數據實例終止
引用
Tue Mar ::
CKPT: terminating instance due to error
Instance terminated by CKPT pid =
Tue Mar ::
再次啟動數據庫報錯
引用
Tue Mar ::
Errors in file /orag/app/admin/ldbra/udump/ldbra_ora_trc:
ORA: data file is not the same file to a background process
ORA signalled during: ALTER DATABASE OPEN
dump Oracle數據文件頭
引用
SQL> ALTER SESSION SET EVENTS immediate trace name file_hdrs level ;
通過跟蹤文件可以看到dbid以被重置為
引用
V STYLE FILE HEADER:
Compatibility Vsn = =xa
Db ID==x Db Name=LDBRA
Activation ID==x
Control Seq==xfba File size==xff
File Number= Blksiz= File Type= DATA
還有一種途徑是通過bbed工具觀察
引用
struct kcvfhhdr bytes @
ub kccfhswv @ x
ub kccfhcvn @ xa
ub kccfhdbi @ x
當然第一反應是重建控制文件看看能不能恢復成功
引用
SQL> alter database backup controlfile to trace;
Database altered
STARTUP NOMOUNT
CREATE CONTROLFILE REUSE DATABASE LDBRA RESETLOGS ARCHIVELOG
MAXLOGFILES
MAXLOGMEMBERS
MAXDATAFILES
MAXINSTANCES
MAXLOGHISTORY
LOGFILE
GROUP /orag/app/oradata/ldbra/redolog SIZE M
GROUP /orag/app/oradata/ldbra/redolog SIZE M
GROUP /orag/app/oradata/ldbra/redolog SIZE M
STANDBY LOGFILE
DATAFILE
/orag/app/oradata/ldbra/systemdbf
/orag/app/oradata/ldbra/undotbsdbf
/orag/app/oradata/ldbra/sysauxdbf
/orag/app/oradata/ldbra/usersdbf
/orag/app/oradata/ldbra/exampledbf
/orag/app/product//db_/dbs/companydbf
/orag/app/product//db_/dbs/streamsdbf
CHARACTER SET ZHSGBK
;
郁悶的是重建控制文件不成功
引用
CREATE CONTROLFILE REUSE DATABASE LDBRA RESETLOGS NOARCHIVELOG
*
ERROR at line :
ORA: CREATE CONTROLFILE failed
ORA: log is inconsistent with other logs
想到還有另外一種語法重建控制文件（重建控制文件之前備份controlfile和online redolog）
引用
Create controlfile reuse set database LDBRA
MAXINSTANCES
MAXLOGHISTORY
MAXLOGFILES
MAXLOGMEMBERS
MAXDATAFILES
Datafile
/orag/app/oradata/ldbra/systemdbf
/orag/app/oradata/ldbra/undotbsdbf
/orag/app/oradata/ldbra/sysauxdbf
/orag/app/oradata/ldbra/usersdbf
/orag/app/oradata/ldbra/exampledbf
/orag/app/product//db_/dbs/ companydbf
/orag/app/product//db_/dbs/streamsdbf
LOGFILE GROUP (/orag/app/oradata/ldbra/redolog) SIZE K
GROUP (/orag/app/oradata/ldbra/redolog) SIZE K
GROUP (/orag/app/oradata/ldbra/redolog) SIZE K RESETLOGS;
似乎重建成功了！但是進行recover的時候報錯了！
引用
SQL> RECOVER DATABASE USING BACKUP CONTROLFILE;
ORA: recovery session canceled due to errors
ORA: internal error code arguments: [kcvhvdf_] [] [] [] [] [] []
[]
可以看到控制文件在重建的過程中進行了dbid重置
引用
SQL> select dbid from v$database;
DBID
問題到這裡似乎失去了頭緒呵呵拷回之前備份的控制文件替換剛建的控制文件因為我采用的是resetlog選項創建控制文件從理論上來講應該是會重置redolog的即重新創建redolog但是目前采用此選項確報ORA錯誤不可思議！後來一想可能是跟數據文件中的dbid為有關於是采用終極修復方法bbed！首先將所有數據文件的dbid用bbed工具重置為其次將fuzzy標記打為x（因為數據庫被ckpt進程異常終止將標記打為x表示數據庫是shutdown immediate關閉）采用上述方法之後控制文件成功創建！
引用
SQL> STARTUP NOMOUNT
CREATE CONTROLFILE REUSE DATABASE LDBRA RESETLOGS ARCHIVELOG
MAXLOGFILES
MAXLOGMEMBERS
MAXDATAFILES
MAXINSTANCES
MAXLOGHISTORY
LOGFILE
ORACLE instance started
Total System Global Area bytes
Fixed Size bytes
Variable Size bytes
Database Buffers bytes
Redo Buffers bytes
GROUP /orag/app/oradata/ldbra/redolog SIZE M
GROUP /orag/app/oradata/ldbra/redolog SIZE M
GROUP /orag/app/oradata/ldbra/redolog SIZE M
STANDBY LOGFILE
DATAFILE
/orag/app/oradata/ldbra/systemdbf
/orag/app/oradata/ldbra/undotbsdbf
/orag/app/oradata/ldbra/sysauxdbf
/orag/app/oradata/ldbra/usersdbf
/orag/app/oradata/ldbra/exampledbf
/orag/app/product//db_/dbs/companydbf
/orag/app/product//db_/dbs/streamsdbf
CHARACTER SET ZHSGBK
;
Control file created
嘗試打開數據庫
SQL> alter database open RESETLOGS;
出現數據庫掛起狀態後台alert日志顯示[]錯誤呵呵看到這個錯誤希望就來了！
引用
SMON: enabling cache recovery
Tue Mar ::
Errors in file /orag/app/admin/ldbra/udump/ldbra_ora_trc:
ORA: internal error code arguments: [] [] [] [] [] [] [] []
Tue Mar ::
Errors in file /orag/app/admin/ldbra/udump/ldbra_ora_trc:
ORA: internal error code arguments: [] [] [] [] [] [] [] []
Tue Mar ::
由於shutdown abort實例不起作用就采用殺Oracle進程刪除共享內存段的做法將掛起的數據庫實力強制abort
殺Oracle核心進程
引用
[orag@test bdump]$ ps ef|grep ora_
orag Feb ? :: ora_pmon_streams
orag Feb ? :: ora_psp_streams
orag Feb ? :: ora_mman_streams
orag Feb ? :: ora_dbw_streams
orag Feb ? :: ora_lgwr_streams
orag Feb ? :: ora_ckpt_streams
orag Feb ? :: ora_smon_streams
orag Feb ? :: ora_reco_streams
orag Feb ? :: ora_cjq_streams
orag Feb ? :: ora_mmon_streams
orag Feb ? :: ora_mmnl_streams
orag Feb ? :: ora_d_streams
orag Feb ? :: ora_s_streams
orag Feb ? :: ora_qmnc_streams
orag Feb ? :: ora_q_streams
orag Feb ? :: ora_q_streams
orag : ? :: ora_j_streams
orag : pts/ :: grep ora_
[orag@test bdump]$ kill
刪除Oracle 共享內存段
引用
[orag@test bdump]$ ipcs
Shared Memory Segments
key shmid owner perms bytes nattch status
xccbc orag
xbb orag
Semaphore Arrays
key semid owner perms nsems
xdec orag
Message Queues
key msqid owner perms usedbytes messages
[root@test ~]# ipcrm m
[root@test ~]# ipcrm s
再次嘗試將實例打開這裡用到了事件
引用
SQL> alter session set events trace name adjust_scn level ;
Session altered
SQL> alter database open;
alter database open
*
ERROR at line :
ORA: file needs media recovery
ORA: data file : /orag/app/oradata/ldbra/systemdbf
SQL> recover database;
Media recovery complete
SQL> alter database open;
Database altered
後續工作就是將tempfile添加到temp表空間中終於恢復成功
引用
SQL> alter tablespace temp add tempfile /orag/app/oradata/ldbra/tempdbf size m reuse;
Tablespace altered
From:http://tw.wingwit.com/Article/program/Oracle/201311/18247.html
