sysaux大面积坏块的例子

某网友的问题，一个11.2.0.3的windows数据库，出现异常，数据库最后经过多人多次抢救之后能够open了，但是查询v$instance,v$database,v$datafile等任何试图都报错，报如下的600错误：

CJQ0 started with pid=31, OS id=5860
 Errors in file E:\APP\ADMINISTRATOR\diag\rdbms\orcl\orcl\trace\orcl_cjq0_5860.trc (incident=121571):
 ORA-00600: internal error code, arguments: [kdsgrp1], [], [], [], [], [], [], [], [], [], [], []
 Use ADRCI or Support Workbench to package the incident.
 See Note 411.1 at My Oracle Support for error and packaging details.
 Errors in file E:\APP\ADMINISTRATOR\diag\rdbms\orcl\orcl\trace\orcl_cjq0_5860.trc (incident=121572):
 ORA-00600: internal error code, arguments: [600], [ORA-00600: internal error code, arguments: [kdsgrp1], [], [], [], [], [], [], [], [], [], [], []
 ], [], [], [], [], [], [], [], [], [], []
 Incident details in: E:\APP\ADMINISTRATOR\diag\rdbms\orcl\orcl\incident\incdir_121572\orcl_cjq0_5860_i121572.trc
 Use ADRCI or Support Workbench to package the incident.
 See Note 411.1 at My Oracle Support for error and packaging details.
 Tue Sep 02 23:58:02 2014
 Errors in file E:\APP\ADMINISTRATOR\diag\rdbms\orcl\orcl\trace\orcl_ora_4188.trc (incident=119396):
 ORA-00600: 内部错误代码, 参数: [kdsgrp1], [], [], [], [], [], [], [], [], [], [], []

这个错误其实是非常容易解决的，本质上就是Index的问题。当然他这里的问题并不止这一个，还有一个问题就是sysaux01.dbf文件居然存在1280个数据坏块，如下：

DBVERIFY: Release 11.2.0.3.0 - Production on 星期三 9月 3 10:21:23 2014
 
Copyright (c) 1982, 2011, Oracle and/or its affiliates. All rights reserved.
 
DBVERIFY - 开始验证: FILE = E:\DATA\ORADATA\ORCL\SYSAUX01.DBF
 页 138241 标记为损坏
 Corrupt block relative dba: 0x00821c01 (file 2, block 138241)
 Completely zero block found during dbv: 

页 138242 标记为损坏
 Corrupt block relative dba: 0x00821c02 (file 2, block 138242)
 Completely zero block found during dbv: 

.....省略部分内容
 
页 139520 标记为损坏
 Corrupt block relative dba: 0x00822100 (file 2, block 139520)
 Completely zero block found during dbv: 

DBVERIFY - 验证完成
 
检查的页总数: 139520
 处理的页总数 (数据): 39125
 失败的页总数 (数据): 0
 处理的页总数 (索引): 36727
 失败的页总数 (索引): 0
 处理的总页数 (Lob) : 9390
 失败的总页数 (Lob) : 0
 处理的页总数 (其他): 27440
 处理的总页数 (段) : 0
 失败的总页数 (段) : 0
 空的页总数: 25558
 标记为损坏的总页数: 1280
 流入的页总数: 0
 加密的总页数 : 0
 最高块 SCN : 37590532 (0.37590532)

一般来讲出现大面积坏块的可能性是比较小的，而且这里的坏块数量也是很奇怪的。我通过bbed简单看楼下，发现后面的1280个坏块都是连续的，而且都是空块（非正常空块）。如下：

BBED> set file 2 block 138241
        FILE#           2
        BLOCK#          138241

BBED> map
 File: /home/ora10g/SYSAUX01.DBF (2)
 Block: 138241                                Dba:0x00821c01
------------------------------------------------------------
BBED-00400: invalid blocktype (00)


BBED> d /v count 200
 File: /home/ora10g/SYSAUX01.DBF (2)
 Block: 138241  Offsets:    0 to  199  Dba:0x00821c01
-------------------------------------------------------
 00000000 00000000 00000000 00000000 l ................
 00000000 00000000 00000000 00000000 l ................
 00000000 00000000 00000000 00000000 l ................
 00000000 00000000 00000000 00000000 l ................
 00000000 00000000 00000000 00000000 l ................
 00000000 00000000 00000000 00000000 l ................
 00000000 00000000 00000000 00000000 l ................
 00000000 00000000 00000000 00000000 l ................
 00000000 00000000 00000000 00000000 l ................
 00000000 00000000 00000000 00000000 l ................
 00000000 00000000 00000000 00000000 l ................
 00000000 00000000 00000000 00000000 l ................
 00000000 00000000                   l ........

 <16 bytes per line>
BBED> set file 2 block 138247
        FILE#           2
        BLOCK#          138247

BBED> d /v count 200
 File: /home/ora10g/SYSAUX01.DBF (2)
 Block: 138247  Offsets:    0 to  199  Dba:0x00821c07
-------------------------------------------------------
 00000000 00000000 00000000 00000000 l ................
 00000000 00000000 00000000 00000000 l ................
 00000000 00000000 00000000 00000000 l ................
 00000000 00000000 00000000 00000000 l ................
 00000000 00000000 00000000 00000000 l ................
 00000000 00000000 00000000 00000000 l ................
 00000000 00000000 00000000 00000000 l ................
 00000000 00000000 00000000 00000000 l ................
 00000000 00000000 00000000 00000000 l ................
 00000000 00000000 00000000 00000000 l ................
 00000000 00000000 00000000 00000000 l ................
 00000000 00000000 00000000 00000000 l ................
 00000000 00000000                   l ........

 <16 bytes per line>

对于这个sysaux的坏文件，如果我们要进行处理，其实并不难，可以dd截取前面的138240 个block，然后同时修改数据文件头的kccfhbsz（单位为block），以及位于system数据文件中的
file$基表中的对应数据文件的大小信息即可。

当然，做完这些之后，你需要重建controlfile，来使信息保持一致。