系统AIX5.3 64bit，ora9.2.0.5
最近alert每天都出现这样的错误。
Errors in file /app/prod/proddb/9.2.0/admin/PROD1_ibmup/udump/prod1_ora_4890704.trc:
ORA-00600: 内部错误代码，参数: [4454], [], [], [], [], [], [], []        
ORA-06512: 在"APPS.CUX_RECEIPT_STORAGE_MIDDLE_PKG", line 335         
在metalink也没找到。兄弟们给点建议。

trc文件我给出一部分吧，太大了10M。
Dump file /app/prod/proddb/9.2.0/admin/PROD1_ibmup/udump/prod1_ora_4890704.trc
Oracle9i Enterprise Edition Release 9.2.0.5.0 - 64bit Production
With the Partitioning, Real Application Clusters, OLAP and Oracle Data Mining options
JServer Release 9.2.0.5.0 - Production
ORACLE_HOME = /app/prod/proddb/9.2.0
System name:        AIX
Node name:        ibmup
Release:        3
Version:        5
Machine:        00C5D96D4C00
Instance name: PROD1
Redo thread mounted by this instance: 1
Oracle process number: 58
Unix process pid: 4890704, image: oracle@ibmup (TNS V1-V3)

*** SESSION ID144.53254) 2008-07-04 01:51:01.671
*** 2008-07-04 01:51:01.671
ksedmp: internal or fatal error
ORA-00600: 内部错误代码，参数: [4454], [], [], [], [], [], [], []
Current SQL statement for this session:
SELECT TO_CHAR(ADD_MONTHS(TO_DATE(:B1, 'yyyy-mm-dd'), -1), 'MON-yy') FROM DUAL
----- PL/SQL Call Stack -----
  object      line  object
  handle    number  name
7000002e98788f0       335  package body APPS.CUX_RECEIPT_STORAGE_MIDDLE_PKG
7000002e98788f0       173  package body APPS.CUX_RECEIPT_STORAGE_MIDDLE_PKG
7000002d5fea248         1  anonymous block
----- Call Stack Trace -----

有人看，没人理啊，顶

SELECT TO_CHAR(ADD_MONTHS(TO_DATE(:B1, 'yyyy-mm-dd'), -1), 'MON-yy') FROM DUAL
把:B1用个值代入,执行这语句看看出问题吗

没有问题啊

我这边长时间运行的job有可能出这个错误，重新启动job就好了~

Subject:  ORA-600 [4454]
  Doc ID:  Note:138836.1 Type:  REFERENCE
  Last Revision Date:  15-OCT-2004 Status:  PUBLISHED


Note: For additional ORA-600 related information please read Note 146580.1

PURPOSE:            
  This article discusses the internal error "ORA-600 [4454]", what
  it means and possible actions. The information here is only applicable
  to the versions listed and is provided only for guidance.

ERROR:              
  ORA-600 [4454]

VERSIONS:           
  Versions 7.0 to 9.2

DESCRIPTION:

  This exception is signalled because the savepoint number is not what was
  expected.

  The current transaction savepoint is less than the beginning savepoint of
  the transaction.

FUNCTIONALITY:      
  TRANSACTION CONTROL SAVEPOINT

IMPACT:            
  PROCESS FAILURE
  NON CORRUPTIVE - No underlying data corruption.

SUGGESTIONS:

  Break the job into smaller chunks and reconnect for each part of the job
  to reset the savepoint number.

  For repeatable occurrences, please submit the trace files and alert.log
  to Oracle Support Services for further analysis.

  Known Issues:
  Bug# 1402161   See Note 1402161.8
      OERI:4411/OERI:4454 on long running job
      Fixed: 8.1.7.3, 9.0.1.3, 9.2.0.1

job错误应该生成例如 orcl_j000_...trc的文件吧。我这个是 orcl_ora_...可能还不一样