ORA-00600 (KKDOILSN1)

DG has a 600 error today.
Tue Dec 10 17:04:49 2013
Errors in File/tol/app/oracle/admin/coredb/diag/rdbms/dg_111/coredb/trace/coredb_ora_29573.trc (incident=68329):
ORA-00600: Internal error code, parameters: [KKDOILSN1], [], [], [], [], [], [], [], [], [], [], []
Incident details in:/tol/app/oracle/admin/coredb/diag/rdbms/dg_111/coredb/incident/incdir_68329/coredb_ora_29573 _i68329.trc


Trace file Contents:
========= Dump for incident 67809 (ORA [KKDOILSN1]) ========

2013-12-10 17:05:36.189 ()
Dbkeddefdump (): Starting incident default dumps (flags=0x2, level=3, mask=0x0)
-----current SQL Statement to this session (SQL_ID=FU2VS5GAD3PNQ)-----
SELECT * from User_help_service_record where user_help_service_id=:1 and Status=1 order by update_time ASC


-----Call Stack Traccall entry argument values in hex
Location type point (? means dubious value)
-------------------- -------- -------------------- ----------------------------


See User_help_service_record This table, a bit of a spectrum, because the main library has just changed the table of two fields to CLOB, and rebuilt all the indexes, including the primary key

Two repositories, only one of the queries will be error, and then check the information about this table (bad block, index, etc.) on the repository and find no exception

Due to the development of the more anxious, did not go to specific query the cause of the error, but this error gave me the first reaction is to suspect that it should be caused by the cache
So the first time I flush Buffer_cache and shared_pool.
And then test it and find that it's no longer a mistake.

Record first, another day to continue this error ...