Troubleshooting of corrupted logfile files (ORA-00327)

1, database log error ORA-00327, pay attention to the garbled here

1. Errors in file/Oracle/app/oracle/diag/rdbms/tyolap/tyolap3/trace/tyolap3_lgwr_22388.trc:
2. ORA-00327: Too Many I 31 (too M too <80> 3 ): <8c> <8d> WTI malong 102400 yundun 1048576
3. ORA-00312: Memory <80> Memory I 31 memory M memory <80> 3: '+ DATA/tyolap/onlinelog/redo31a'
4. Errors in file/oracle/app/oracle/diag/rdbms/tyolap/tyolap3/trace/tyolap3_lgwr_22388.trc:
5. ORA-00321: Too Many I 31 (too M too <80> 3 ): <9f> o Lu Wu Yi Mei Yu I <99> n <98> Yu Mao ^
6. ORA-00312: Memory <80> Memory I 31 memory M memory <80> 3: '+ DATA/tyolap/onlinelog/redo31a'

Check the error details. The redo file + DATA/tyolap/onlinelog/redo31a is damaged.

1. [Oracle @ dtydb5 scripts] $ oerr ora 327
2. 00327,000 00, "log % s of thread % s, physical size % s less than needed % s"
3. // * Cause: A log file has shrunk insize. This is likely
4. // To have been caused by operator or operating system error.
5. // * Action: Restore the log file from backup. If backup isnot available,
6. // Drop this log and re-create. If the database was shut down
7. // Cleanly, no further action shocould be required; otherwise
8. // Incomplete recovery may be required.
10. [Oracle @ dtydb5 scripts] $ oerr ora 312
11. 00312,000 00, "online log % s thread % s: '% S '"
12. // * Cause: This message reports the filename for details of another message.
13. // * Action: Other messages will accompany this message. See
14. // Associated messages for the appropriate actionto take.
16. [Oracle @ dtydb5 scripts] $ oerr ora 321
17. 00321,000 00, "log % s of thread % s, cannot update log file header"
18. // * Cause: Cannot write to the log file.
19. // * Action: Restore the access to the file.

2. Check the trc file. The specific cause is not clear, but the log file is damaged.

DE: Problem Key 'ora 313 'was flood controlled (0x1) (no incident)
ORA-00313 :?????? 31 (???? 3 )?????
ORA-00313 :?????? 31 (???? 3 )?????
DDE rules only execution for: ORA 312
----- START Event Driven Actions Dump ----
---- END Event Driven Actions Dump ----
----- Start dde Actions Dump -----
Executing SYNC actions
----- Start dde Action: 'db _ STRUCTURE_INTEGRITY_CHECK '(Async )-----
DDE Action 'db _ STRUCTURE_INTEGRITY_CHECK 'was flood controlled
----- End dde Action: 'db _ STRUCTURE_INTEGRITY_CHECK '(flood controlled, 5 csec )-----
Executing ASYNC actions
----- End dde Actions Dump (total 0 csec )-----

3. Check the status of the redo file. The status is INVALID, the other is stale, and the system is still running normally.

1. 31 3 575 2 536,870,912 YES ACTIVE
3. 32 3 576 2 536,870,912 NOCURRENT
5. GROUP # status type member is _
6. -----------------------------------------------------------------------------
7. 31 stale online + DATA/tyolap/onlinelog/redo31a NO
8. ONLINE + DATA/tyolap/onlinelog/redo31b NO
10. 32 invalid online + DATA/tyolap/onlinelog/redo32a NO
11. ONLINE + DATA/tyolap/onlinelog/redo32b NO
13. STATUS VARCHAR2 (7) Status of the log member:
15. INVALID-File is inaccessible
17. STALE-File's contents are incomplete
19. DELETED-File isno longer used
21. Null-File isin use

4. log files are damaged. The simplest way is to recreate the log files online.

1. Sys @ TYOLAP3 (dtydb5)> alterdatabasedrop logfile member '+ DATA/tyolap/onlinelog/redo31a ';
3. The database has been changed.
6. Sys @ TYOLAP3 (dtydb5)> alterdatabaseadd logfile member '+ DATA/tyolap/onlinelog/redo31a 'togroup 31;
8. The database has been changed.
10. Sys @ TYOLAP3 (dtydb5)> alter system switch logfile;
12. System Log Recovery
14. GROUP # status type member is _
15. -----------------------------------------------------------------------------
16. 31 ONLINE + DATA/tyolap/onlinelog/redo31a NO
17. ONLINE + DATA/tyolap/onlinelog/redo31b NO
19. 32 ONLINE + DATA/tyolap/onlinelog/redo32a NO
20. ONLINE + DATA/tyolap/onlinelog/redo32b NO

5. The system returns to normal, but the cause of Log File Corruption is not clear yet.