Oracle内部错误:ORA-00600[25012]一例

chinaab

　　
SQL> select count(*) from WWork;
　　

　　
COUNT(*)
　　
----------
　　
116114
　　

　　
select count(*) from WWork where Work_WorkID = 100;
　　
select count(*) from WWork where Work_WorkID = 100
　　
*
　　
ERROR at line 1:
　　
ORA-00600: internal error code, arguments: [25012], [15], [8], [], [], [], [],
　　
[]
　　


　　
Tablespace>　　

　　
analyze table livelink.WWORK validate structure cascade;
　　

　　
analyze table livelink.WWORK validate structure cascade
　　
*
　　
ERROR at line 1:
　　
ORA-00600: internal error code, arguments: [25012], [15], [8], [], [], [], [],
　　
[]
　　

　　
analyze index livelink.WWORK_PRIMARY validate structure;
　　
analyze index livelink.WWORK_PRIMARY validate structure
　　
*
　　
ERROR at line 1:
　　
ORA-08100: index is not valid - see trace file for diagnostics
　　

　　
I'm wondering if the issue could be resolved recreating the index.
　　

　　
alter index WWORK_PRIMARY rebuild online noparallel;
　　

　　
On the alert log file, ORA-00600 began at:
　　
Sun Jun 5 23:29:10 2011
　　
Errors in file /u001/app/oracle/admin/motpcom/udump/motpcom_ora_26554.trc:
　　
ORA-00600: internal error code, arguments: [ktbair1], [1], [6], [], [], [], [], []
　　
Mon Jun 6 00:35:51 2011
　　
Thread 1 advanced to log sequence 303722
　　
Current log# 19 seq# 303722 mem# 0: /u002/oradata/motpcom/redo19.log
　　
Mon Jun 6 00:35:51 2011
　　
And then below error often raised in alert log:
　　
Mon Jun 6 02:45:00 2011
　　
Errors in file /u001/app/oracle/admin/motpcom/udump/motpcom_ora_28348.trc:
　　
ORA-00600: internal error code, arguments: [25012], [15], [8], [], [], [], [], []
　　
Mon Jun 6 02:50:01 2011
　　
I also found other errors:
　　
Mon Jun 6 05:00:01 2011
　　
ORA-01555 caused by SQL statement below (Query Duration=18448 sec, SCN: 0x09f0.52bb91c8):
　　
Mon Jun 6 05:00:01 2011
　　
SELECT COUNT(d.dataid)
　　
FROM allemployees a, networks_dataids d, kuaf k, dversdata v
　　
WHERE k.name=a.user_id(+) AND d.userid=k.id AND a.user_id is null
　　
Mon Jun 6 05:00:13 2011
　　
Errors in file /u001/app/oracle/admin/motpcom/udump/motpcom_ora_28348.trc:
　　
ORA-00600: internal error code, arguments: [25012], [15], [8], [], [], [], [], []
　　
Mon Jun 6 05:05:13 2011
　　
And
　　
Tue Jun 7 02:02:14 2011
　　
Errors in file /u001/app/oracle/admin/motpcom/udump/motpcom_ora_25842.trc:
　　
ORA-07445: exception encountered: core dump [00000001011B8AD8] [SIGSEGV] [Address not mapped to object] [0x000000A88] [] []
　　
Tue Jun 7 02:03:11 2011
　　
And
　　
Tue Jun 7 02:41:35 2011
　　
Errors in file /u001/app/oracle/admin/motpcom/udump/motpcom_ora_27325.trc:
　　
ORA-00600: internal error code, arguments: [2662], [2544], [1396224900], [28954], [3445424704], [786437], [], []
　　
Tue Jun 7 02:41:35 2011
　　
Errors in file /u001/app/oracle/admin/motpcom/udump/motpcom_ora_27325.trc:
　　
ORA-00603: ORACLE server session terminated by fatal error
　　
ORA-00600: internal error code, arguments: [2662], [2544], [1396224900], [28954], [3445424704], [786437], [], []
　　
Tue Jun 7 02:41:59 2011
　　

　　
SQL> select count(*) FROM WWork where Work_WorkID=100;
　　
select count(*) FROM WWork where Work_WorkID=100
　　
*
　　
ERROR at line 1:
　　
ORA-00600: internal error code, arguments: [25012], [15], [8], [], [], [], [],
　　
[]
　　

　　


　　
SQL>>　　


　　
Index>　　

　　
SQL> select count(*) FROM WWork where Work_WorkID=100;
　　

　　
COUNT(*)
　　
----------
　　
0
　　

　　
Index corruptions do not always mean that there is a Bug.

　　
Indexes corrupt when the transaction>　　
In majority of the cases, this happens when a table has a large amount of DDL or DML processing as in a OLAP

　　
processing. The index buffer cache is over written with an incorrect transaction>　　
The error could be a ORA-00600 [25027] error, an ORA-8102/8103, or another ORA-00600 error.

　　
Possibly, adding another index may resolve the issue of>　　

　　
In addition, the index space must be reviewed to determine whether there is enough space.
　　
As in the first example above, if there isn't space, then the index can be overwritten and an error can appear.
　　
I suggest reading My Oracle Knowledge Note 33343.1 "How to Find Out How Much Space an Index is Using" which provides select

　　
statements to show the actual usage of blocks within an index. This gives an>
　　
allows the DBA to adjust next extent>　　

　　
In addition, you can start the Index Tuning Wizard from Enterprise Manager in order to get advice on the indexes within the database.
　　

　　
We would like to emphasize that the best method to resolve a corrupt index is to drop and recreate it and not use a rebuild.
　　
If this index corrupts again, then we suggest that it be dropped and recreated.
　　

　　

　　
ORA-600 [25012] "Relative to Absolute File Number Conversion Error"
　　

　　


　　
Note: For additional ORA-600>　　

　　
PURPOSE:
　　This article discusses the internal error "ORA-600 [25012]", 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 [25012] [a] [c]
　　

　　
VERSIONS:
　　versions 8.0 and above
　　

　　
DESCRIPTION:
　　

　　We are trying to generate the absolute file number given a tablespace

　　number and>　　or the file number is zero.
　　

　　
ARGUMENTS:
　　Arg [a] Tablespace Number

　　Arg >
　　Arg [c] Absolute file number (This arg is present is more recent>　　

　　
FUNCTIONALITY:
　　KERNEL FILE MANAGEMENT TABLESPACE COMPONENT
　　

　　
IMPACT:
　　POSSIBLE PHYSICAL CORRUPTION
　　

　　
SUGGESTIONS:
　　

　　The possibility of physical corruption exists.
　　

　　Obtain the trace files and alert.log for this error and log a Service Request
　　with Oracle Support Services for diagnosis.
　　


　　If the Arg >　　that can cause this error.
　　

　　The following query list fake indexes :
　　

　　select a.*,b.flags from dba_objects a, sys.ind$ b
　　where a.object_id = b.obj#
　　and bitand(b.flags,4096)=4096;
　　

　　Known Issues:
　　

　　

　　

　　

　　
Known Bugs

　　
You can restrict the list below to issues likely to affect one of the following versions by clicking the>　　

　　

　　NBBugFixedDescription
　　5653641 11.2.0.1 Corrupt dictionary from DROP TABLESPACE containing _offline_rollback_segments
　　* 8198906 10.2.0.5, 11.2.0.1 OERI [kddummy_blkchk] / OERI [5467] for an aborted transaction of allocating extents
　　4925342 9.2.0.8, 10.2.0.3, 11.1.0.6 OERI [25027] / OERI [25012] on IOT analyze estimate statistics
　　3258674 9.2.0.7, 10.1.0.4, 10.2.0.1 QMN process may dump or raise an internal error during DML
　　4186885 10.2.0.1 Partition numbers for IOT index/overflow segments are not synchronized
　　3751874 9.2.0.6, 10.1.0.4, 10.2.0.1 OERI[25012] can occur when _old_connect_by_enabled set to true
　　3430832 9.2.0.6, 10.1.0.3, 10.2.0.1 OERI[25012] in DML after ONLINE create index against PARTITIONED table
　　3915343 9.2.0.7, 10.1.0.5, 10.2.0.1 OERI [25012] on COMMIT with refresh ON-COMMIT Materialized view over cluster
　　4305391 9.2.0.7, 10.1.0.5, 10.2.0.1 OERI[25012] with kditalp can occur during a temporary LOB operation
　　P* 6047085 Linux x64-64: SGA corruption / crash following any ORA-7445
　　2526334 9.2.0.3, 10.1.0.2 OERI:25012 from AND EQUAL with B-Tree and DOMAIN index
　　2531519 9.2.0.3, 10.1.0.2 OERI:[25012] from parallel direct load of bitmap managed segments
　　3150268 9.2.0.5, 10.1.0.2 OERI[kcbgcur_1] / OERI:25012 deleting rows from PARENT table with LIST subpartitions
　　3070856 9.2.0.5, 10.1.0.2 OERI:12700 / 25012 / ktrexc_1 on transported tablespace with SMU
　　3771508 9.2.0.7 OERI[kcbgtcr_1] / OERI[25012] selecting from V$AQ in a shared server
　　3900237 9.2.0.7 OERI[25012] changing length of indexed column for a global temporary table
　　1834530 8.1.7.4, 9.0.1.4, 9.2.0.1 OERI:25012 / wrong results after EXCHANGE
　　PARTITION with indexes with different FREELIST /FREELIST GROUPS
　　1698789 9.2.0.1 Wrong results, ORA-1410, ORA-8103, OERI:25012 on SELECT of UNSCOPED REF with ROWID
　　2189615 8.1.7.4, 9.0.1.4, 9.2.0.1 OERI:25012 / OERI:504 [cache buffers chains] selecting from certain V$ tables
　　1784708 8.1.7.3, 9.0.1.1, 9.2.0.1 OERI:KCBGTCR_1/OERI:25012 accessing LONG / LONG RAW in a HASH CLUSTER
　　1872985 9.0.1.2, 9.2.0.1 Dump / OERI:25012 from BITMAP CONVERSION of INDEX on GLOBAL TEMPORARY TABLE
　　1968815 9.0.1.2, 9.2.0.1 OERI:25012 possible from SMON following DROP TABLESPACE command
　　2287815 9.0.1.4, 9.2.0.1 OERI:KTSPISCHNT / OERI:25012 after exchange PARTITION with bitmap managed segments
　　2184731 8.1.7.4, 9.2.0.1 OERI:25012 possible from index prefetch
　　1837529 8.1.7.3, 9.0.1.1, 9.2.0.1 OERI:KFTR2BZ_1/OERI:25012 from CREATE sub-partitioned local INDEX ONLINE
　　2214167 9.0.1.4, 9.2.0.1 OERI:25012 / wrong results possible after TRUNCATE of bitmap managed index

　　2212389 9.0.1.4, 9.2.0.1 OERI:25012 / Cursor frame memory corruption when cursor with BINDS aged out />　　1788648 8.1.7.3, 9.0.1.1, 9.2.0.1 OERI:25012 [2147483647] possible selecting from certain V$ views
　　1527982 8.1.7.2, 9.0.1.0 OERI:25012 / Bitmap indextable mismatch after UPDATE of PARTITION KEY moves rows

　　1949273 8.1.7.3, 9.0.1.0 OERI:25012 / ORA-1555 / ORA-22922 accessing LOB data after>　　1264970 8.1.7.1, 9.0.1.0 OERI:25012 / OERI:6050 possible coalescing index with freelists
　　1396571 8.0.6.3, 8.1.6.3, 8.1.7.1, 9.0.1.0 OERI:25012 possible importing with TRANSPORT_TABLESPACE=Y
　　1678963 8.1.7.2, 9.0.1.0 OERI:25012 / OERI:4142 possible on TRUNCATE of a table with a CLOB column
　　1297674 8.0.6.2, 8.1.6.3, 8.1.7.0 OERI:25012 Analyzing partitioned table with ESTIMATE
　　1138239 8.1.6.2, 8.1.7.0 OERI-25012 possible on direct read from plugged in (transportable) tablespace
　　1228658 8.1.6.2, 8.1.7.0 Create INDEX on SNAPSHOT/MV can produce corrupt index (OERI:13004 / OERI:25012 / ORA-1499)
　　1108002 8.1.7.0 OERI:25012 when DBMS_STATS tries to gather stats for a GLOBAL TEMPORARY table
　　986928 8.1.7.0 DBMS_SPACE.UNUSED_SPACE reports OERI:25012 for TEMPORARY tables
　　1312233 8.1.6.3, 8.1.7.0 OERI:25012 / OERI:KCBGCUR_1 possible on PQ STAR query with PARTITIONED fact table
　　718499 8.0.6.1, 8.1.5.0 OERI:25012 possible on OPS parallel create index
　　677243 8.0.6.0 OERI:kdddgbX on DELETE / UPDATE / SELECT for UPDATE with an invalid ROWID
　　679817 8.0.5.1 OERI:25012 from view containing 2+ LOB columns
　　595698 8.0.4.3, 8.0.5.0 DELETE on table with many child constraints may dump.
　　

　　'*' against a bug indicates that an alert exists for that issue.
　　'+' indicates a particularly notable bug.
　　'P' indicates a port specific bug.
　　"OERI:nnnnn" is used as shorthand for ORA-600 [nnnnn].
　　

　　
Symptom(s)
　　
~~~~~~~~~~
　　

　　
ORA-600 [25012] errors with same first argument and different
　　
second arguments (second arguments changes during every SELECT Run)
　　

　　
Eg.:
　　

　　
ORA-600 [25012], [5], [1023], [], [], [], []
　　

　　

　　
The first argument corresponds to a tablespace and second argument
　　
corresponds to a file number.
　　

　　
The tablespace points to an index tablespace
　　

　　

　　
Cause
　　
~~~~~
　　

　　
Bug:2184731
　　

　　

　　
Fix
　　
~~~~
　　

　　
Bug:2184731 is fixed in 8.1.7.4 and 9.2
　　

　　
The workaround is to set _db_file_noncontig_mblock_read_count=1.
　　

　　

　　
References
　　
~~~~~~~~~~
　　

　　
Note:100073.1  ORA-600 [25012] "Relative to Absolute File Number
　　Conversion Error"
　　

　　
Bug:2184731    ORA-600 [25012] WHEN SELECT FROM A TABLE USING AN INDEX