坏块问题

tom_fu

Thu Jul 19 16:34:48  2012
Doing block recovery for file 12 block 60242838
Resuming block recovery (PMON) for file 12 block 60242838
Block recovery from logseq 22058, block 54110 to scn 200619091
Thu Jul 19 16:34:48  2012
Recovery of Online Redo Log: Thread 1 Group 2 Seq 22058 Reading mem 0
  Mem# 0: H:\ORADATA\HBCZHIS\REDO02.LOG
Block recovery completed at rba 22058.54795.16, scn 0.200619092
Thu Jul 19 16:35:28  2012
Corrupt Block Found
         TSN = 13, TSNAME = HISDIGITALDB
         RFN = 1024, BLK = 60242838, RDBA = 60242838
         OBJN = 1, OBJD = -1, OBJECT = , SUBOBJECT =
         SEGMENT OWNER = , SEGMENT TYPE =
Thu Jul 19 16:35:31  2012
Errors in file d:\oracle\product\10.2.0\admin\hbczhis\udump\hbczhis_ora_8648.trc:
ORA-00600: 内部错误代码, 参数: [kddummy_blkchk], [12], [60242838], [18018], [], [], [], []
Thu Jul 19 16:35:32  2012
Doing block recovery for file 12 block 60242838
Resuming block recovery (PMON) for file 12 block 60242838
Block recovery from logseq 22058, block 54999 to scn 200619392
Thu Jul 19 16:35:32  2012
Recovery of Online Redo Log: Thread 1 Group 2 Seq 22058 Reading mem 0
  Mem# 0: H:\ORADATA\HBCZHIS\REDO02.LOG
Block recovery completed at rba 22058.55100.16, scn 0.200619393



SQL> select * from v$version;
BANNER
----------------------------------------------------------------
Oracle Database 10g Enterprise Edition Release 10.2.0.5.0 - Prod
PL/SQL Release 10.2.0.5.0 - Production
CORE    10.2.0.5.0      Production
TNS for 32-bit Windows: Version 10.2.0.5.0 - Production
NLSRTL Version 10.2.0.5.0 - Production


http://www.oracledatabase12g.com/archives/ora-00600-kddummy_blkchk%E9%94%99%E8%AF%AF%E4%B8%80%E4%BE%8B.html

按照上面链接的方式查询数据块60242838，找不到任何对象。
没有down机，而且重启也正常。
之前系统是10204也出现同样的问题，升级到10205后用了不到半天现在又出现这种错误，过了一段时间后说是系统又正常了。补充一下，这个错误是在保存图片入库时出现的。

[ 本帖最后由 tom_fu 于 2012-7-19 20:56 编辑 ]

Maclean Liu(刘相兵

action plan:

请上传完整的TRACE文件 d:\oracle\product\10.2.0\admin\hbczhis\udump\hbczhis_ora_8648.trc

aks5201314

楼主可以用DBV实用程序检测一下datafile 12,看是否有块损坏的报告。

dbrabbit

How to Resolve ORA-00600[kddummy_blkchk] [ID 1342443.1]

Maclean Liu(刘相兵

ORA-600 [kddummy_blkchk] [file#] [block#] [check code]

ORA-00600: 内部错误代码, 参数: [kddummy_blkchk], [12], [60242838], [18018], [], [], [], []

即指向 file# 12 block# 60242838

18018 为check code 检测代码 ，check code说明了出现 kddummy_blkchk 函数 块检测出错的原因

ODM FINDING:

Applies to:
Oracle Server - Enterprise Edition - Version: 10.2.0.5 and later   [Release: 10.2 and later ]
Information in this document applies to any platform.
Symptoms
The error received is:

ORA-01595: error freeing extent (51) of rollback segment (91))
ORA-00607: Internal error occurred while making a change to a data block
ORA-00600: internal error code, arguments: [kddummy_blkchk], [30], [3], [18018], [], [], [], []
Corrupt Block Found
TSN = 1, TSNAME = APPS_UNDOTS1
RFN = 30, BLK = 3, RDBA = 125829123
OBJN = 1, OBJD = -1, OBJECT = , SUBOBJECT =
SEGMENT OWNER = , SEGMENT TYPE =


Changes
This happens after trying to mark an extent as used by dbms_space_admin.tablespace_fix_bitmaps


Cause
This is being analyzed in bug 12349316
At the moment of writing this note, the bug is still under investigation.

Solution
The Workaround to avoid the problem once the corruption is introduced is :

1- Move all the objects from the affected tablespace to a new tablespace, that will workaround the problem

Once the bug be fixed, then apply the patch to avoid this situation to happens again.




建议：

就以上note看，这是10.2.0.5中的一个corruption bug
重建 tablespace ， 并将原tablespace 上的对象迁移到新的表空间上