startup with ora-00392,ora-312

admin

用rman得备份在异机做恢复，打开数据库得时候报错如下：

1. SQL> alter database open resetlogs;
   
2. alter database open resetlogs
   
3. *
   
4. ERROR at line 1:
   
5. ORA-00392: log 1 of thread 1 is being cleared, operation not allowed
   
6. ORA-00312: online log 1 thread 1: '/<PATH>/group_1.log'
   
7. ORA-00312: online log 1 thread 1: '/<PATH>/group_12.log'

复制代码

原因：

1. The first command "alter database open resetlogs" is abnormally abrupted leaving the redo log status as
   
2. CLEARING/CLEARING_CURRENT in controlfile :
   
3. SQL> select GROUP#,THREAD#,SEQUENCE#,MEMBERS,ARCHIVED,STATUS,FIRST_CHANGE# from v$log order
   
4. by first_change# ;
   
5. GROUP# THREAD# SEQUENCE# MEMBERS ARCHIVED STATUS FIRST_CHANGE#
   
6. ------------- ----------- ----------- -------------- ---------- ---------------- -------------------
   
7. 2 1 0 2 YES CLEARING 12142264718335
   
8. 3 1 0 2 YES CLEARING 12142306351551
   
9. 1 1 0 2 NO CLEARING_CURRENT 12142306351562

复制代码

处理：

1. At first, see if below command helps :
   
2. alter database clear unarchived logfile group 1 ;
   
3. alter database clear unarchived logfile group 2 ;
   
4. alter database clear unarchived logfile group 3 ;
   
5. alter database open resetlogs;
   
6. ELSE,
   
7. Recreate the controflile with the RESETLOGS option.
   
8. 1) Get the controlfile trace from below command :
   
9. SQL> ALTER DATABASE BACKUP CONTROLFILE TO TRACE AS '/tmp/control.sql' resetlogs ;
   
10. 2) Modify the CREATE CONTROLFILE script /tmp/control.sql and ensure that all directories for the online redo logs
    
11. exist and Oracle has permission to write to it
    
12. 3) Create the controlfile in NOMOUNT state :
    
13. SQL> STARTUP FORCE NOMOUNT
    
14. SQL> @/tmp/control.sql
    
15. controlfile created
    
16. 4) Run a fake recovery :
    
17. SQL> RECOVER DATABASE USING BACKUP CONTROLFILE UNTIL CANCEL ;
    
18. Type <CANCEL> when prompted
    
19. 5) Open with RESETLOGS option:
    
20. SQL> ALTER DATABASE OPEN RESETLOGS ;

复制代码