[kcbnew_3]

客户数据库alert日志一直在报如下错误

引用

Tue Jul 14 10:40:51 2009
Errors in file /oracle/app/oracle/admin/fyxny/udump/fyxny_ora_16791.trc:
ORA-00600: internal error code, arguments: [kcbnew_3], [1], [], [], [], [], [], []

操作系统及数据库版本为

引用

#uname -a
HP-UX orasvr1 B.11.11 U 9000/800 3641685662 unlimited-user license
SQL> select * from v$version;

BANNER
----------------------------------------------------------------
Oracle9i Enterprise Edition Release 9.2.0.1.0 - 64bit Production
PL/SQL Release 9.2.0.1.0 - Production
CORE    9.2.0.1.0       Production
TNS for HPUX: Version 9.2.0.1.0 - Production
NLSRTL Version 9.2.0.1.0 - Production

据metalink描述，又是一个bug，需升级到9202或者9208

引用

Bug 2414972 ( Unpublished )

Abstract : OERI:[kcbnew_3]/OERI:[kcbgtcr_3] after resize in LOCALLY MANAGED tablespace

Versions affected : 9.2.0.1, 9.0.1.3

Fixed Releases : 9.2.0.2

Details : If a locally-managed tablespace is resized (either automatically or manually)

then a process may subsequently error with and ORA-600 error.

Symptoms :

1. A locally-managed tablespace is resized (either automatically or manually) and some time
shortly after this, a process reports ORA-600:[kcb...]. Usually the tablespace involved
is an undo tablespace and the errors displayed are:

2. 9.2.0.1: ORA-600:[kcbnew] [n]

3. 9.0.1.3: ORA-600:[kcbgtc_3]

Workaround :

Avoid tablespace resizing.

Patch Details :

Check Metalink for Patch 2414972 availability.

引用

  Bug# 2414972   See Note 2414972.8
      OERI:[kcbnew_3]/OERI:[kcbgtcr_3] after resize in LOCALLY MANAGED tablespace
      Fixed: 9.2.0.2

  Bug# 2406802   See Note 2406802.8
      OERI[kcbgtcr_3]  / OERI[kcbcxx_1] after DROP TABLE in locally managed tablespace
      Fixed: 9.2.0.2

metalink原理阐述：

引用

DESCRIPTION:

  A cache buffer holding a database block is in the process of
  being reused.

  The buffer is in state "current" and may be reused only if the object
  is of type temp or undo.

  The consistency check comparing the block class in the buffer header
  with the block class passed to the cache by the caller is failing.

由以上DESCRIPTION可知，数据库重启亦可暂时解决问题，因为客户不能长时间宕机，那就重启一下数据库吧，呵呵