查询dba_data_files报ORA-600 ktfbhget-4

oracle321

查询dba_data_files出现下面报错：

1. SQL> select count(*) from dba_data_files;
   
2. select count(*) from dba_data_files
   
3. *
   
4. ERROR at line 1:
   
5. ORA-00600: internal error code, arguments: [ktfbhget-4], [0], [5], [], [], [], [], []

复制代码

逐一查询dba_data_files视图相关表，定位x$ktfbhc查询时异常:

1. SQL> select count(*) from x$ktfbhc;
   
2. select count(*) from x$ktfbhc
   
3. *
   
4. ERROR at line 1:
   
5. ORA-00600: internal error code, arguments: [ktfbhget-4], [0], [5], [], [], [], [], []
   

复制代码

其它故障现象：
1. 当sys用户创建表时，也会出现该问题：

1. SQL> create table iii as select * from dba_objects;
   
2. create table iii as select * from dba_objects
   
3.                                   *
   
4. ERROR at line 1:
   
5. ORA-00600: internal error code, arguments: [ktfbhget-4], [0], [5], [], [], [],
   
6. [], []
   

复制代码

如果在表放在其它表空间则正常：

1. 
   
2. SQL>  create table iii tablespace users as select * from dba_objects;
   
3. Table created.
   

复制代码

2. system表空间有临时段未清除

1. 
   
2. SQL>  select * from dba_segments where segment_type='TEMPORARY';
   
3. 
   
4. OWNER                          SEGMENT_NAME                                                                      PARTITION_NAME                 SEGMENT_TYPE       TABLESPACE_NAME
   
5. ------------------------------ --------------------------------------------------------------------------------- ------------------------------ ------------------ ------------------------------
   
6. HEADER_FILE HEADER_BLOCK      BYTES     BLOCKS    EXTENTS INITIAL_EXTENT NEXT_EXTENT MIN_EXTENTS MAX_EXTENTS PCT_INCREASE  FREELISTS FREELIST_GROUPS RELATIVE_FNO BUFFER_
   
7. ----------- ------------ ---------- ---------- ---------- -------------- ----------- ----------- ----------- ------------ ---------- --------------- ------------ -------
   
8. SYS                            1.29873                                                                                                          TEMPORARY          SYSTEM
   
9.           1        29873    1048576        128         16          65536                       1  2147483645                       1               1            1 DEFAULT
   

复制代码

已尝试操作：
1. 禁用smon的临时段清理功能，故障仍然存在

1. 
   
2. NAME TYPE VALUE
   
3. 
   
4. --------------------------------------------------------------------------------
   
5. -----------
   
6. --------------------------------------------------------------------------------
   
7. event string 10061 trace name context forever, level 10
   

复制代码

2. 手工删除临时段
alter session set events ‘immediate trace name DROP_SEGMENTS level 1′;
命令执行正常，但后台仍有600报错

3.重启数据后，该临时段仍然存在。

4. 试图运行hcheck脚本，但hcheck3.sql脚本执行中途退出，检查alert日志，同样出现600

ORA-600 ktfbhget-4错误是否与临时段无法清除有关？
system表空间的临时段是否能手工删除？
谢谢大家！


补充：
1.  数据库版本：
10.2.0.4    测试环境

2. 告警日志信息

1. 
   
2. ......
   
3. Sun May  1 22:09:08 2011
   
4. Completed: ALTER DATABASE OPEN
   
5. Sun May  1 22:15:23 2011
   
6. Errors in file /home/oracle/admin/davidloc/bdump/davidloc_smon_4622.trc:
   
7. ORA-00600: internal error code, arguments: [ktfbhget-4], [0], [5], [], [], [], [], []
   
8. Sun May  1 22:15:24 2011
   
9. Non-fatal internal error happenned while SMON was doing logging scn->time mapping.
   
10. SMON encountered 1 out of maximum 100 non-fatal internal errors.
    
11. Sun May  1 22:16:13 2011
    
12. Errors in file /home/oracle/admin/davidloc/udump/davidloc_ora_4677.trc:
    
13. ORA-00600: internal error code, arguments: [ktfbhget-4], [0], [5], [], [], [], [], []
    
14. Sun May  1 22:16:15 2011
    
15. Errors in file /home/oracle/admin/davidloc/udump/davidloc_ora_4677.trc:
    
16. ORA-00603: ORACLE server session terminated by fatal error
    
17. ORA-00600: internal error code, arguments: [ktfbhget-4], [0], [5], [], [], [], [], []
    
18. Sun May  1 22:17:52 2011
    
19. Errors in file /home/oracle/admin/davidloc/udump/davidloc_ora_4718.trc:
    
20. ORA-00600: internal error code, arguments: [ktfbhget-4], [0], [5], [], [], [], [], []
    
21. Sun May  1 22:17:52 2011
    
22. Errors in file /home/oracle/admin/davidloc/udump/davidloc_ora_4718.trc:
    
23. ORA-00603: ORACLE server session terminated by fatal error
    
24. ORA-00600: internal error code, arguments: [ktfbhget-4], [0], [5], [], [], [], [], []
    
25. Sun May  1 22:18:48 2011
    
26. Errors in file /home/oracle/admin/davidloc/udump/davidloc_ora_4778.trc:
    
27. ORA-00600: internal error code, arguments: [ktfbhget-4], [0], [5], [], [], [], [], []
    
28. Sun May  1 22:18:48 2011
    
29. Errors in file /home/oracle/admin/davidloc/udump/davidloc_ora_4778.trc:
    
30. ORA-00603: ORACLE server session terminated by fatal error
    
31. ORA-00600: internal error code, arguments: [ktfbhget-4], [0], [5], [], [], [], [], []
    
32. Sun May  1 22:21:17 2011
    
33. Errors in file /home/oracle/admin/davidloc/udump/davidloc_ora_4813.trc:
    
34. ORA-00600: internal error code, arguments: [ktfbhget-4], [0], [5], [], [], [], [], []
    
35. Sun May  1 22:21:18 2011
    
36. Errors in file /home/oracle/admin/davidloc/udump/davidloc_ora_4813.trc:
    
37. ORA-00603: ORACLE server session terminated by fatal error
    
38. ORA-00600: internal error code, arguments: [ktfbhget-4], [0], [5], [], [], [], [], []
    
39. Sun May  1 22:21:33 2011
    
40. Errors in file /home/oracle/admin/davidloc/bdump/davidloc_smon_4622.trc:
    
41. ORA-00600: internal error code, arguments: [ktfbhget-4], [0], [5], [], [], [], [], []
    
42. Sun May  1 22:21:33 2011
    
43. Non-fatal internal error happenned while SMON was doing logging scn->time mapping.
    
44. SMON encountered 2 out of maximum 100 non-fatal internal errors.
    

复制代码

测试环境的主机时间没做调整。

第一次发贴，谢谢刘大提醒。

Maclean Liu(刘相兵

至少数据库版本写一下吧？

Maclean Liu(刘相兵

/home/oracle/admin/davidloc/bdump/davidloc_smon_4622.trc
/home/oracle/admin/davidloc/udump/davidloc_ora_4677.trc

上传看一下

oracle321

已经补充trc文件
冒似写东西到system表空间均会报错。。。。

Maclean Liu(刘相兵

dbv file=system.dbf

1. 1) $ rman target / nocatalog
   
2. 
   
3. 2) RMAN> run {
   
4. allocate channel d1 type disk;
   
5. allocate channel d2 type disk;
   
6. allocate channel d3 type disk;
   
7. allocate channel d4 type disk;
   
8. backup validate check logical database;
   
9. }
   
10. 
    
11. 3) select * from V$DATABASE_BLOCK_CORRUPTION ;
    
12. 
    
13. REM www.oracledatabase12g.com & www.askmaclean.com
    
14. 
    
15. 4) If V$DATABASE_BLOCK_CORRUPTION contains rows please run this query to
    
16. find the objects that contains the corrupted blocks:
    
17. 
    
18. 
    
19. SELECT e.owner,
    
20.        e.segment_type,
    
21.        e.segment_name,
    
22.        e.partition_name,
    
23.        c.file#,
    
24.        greatest(e.block_id, c.block#) corr_start_block#,
    
25.        least(e.block_id + e.blocks - 1, c.block# + c.blocks - 1) corr_end_block#,
    
26.        least(e.block_id + e.blocks - 1, c.block# + c.blocks - 1) -
    
27.        greatest(e.block_id, c.block#) + 1 blocks_corrupted,
    
28.        null description
    
29.   FROM dba_extents e, v$database_block_corruption c
    
30. WHERE e.file_id = c.file#
    
31.    AND e.block_id <= c.block# + c.blocks - 1
    
32.    AND e.block_id + e.blocks - 1 >= c.block#
    
33. UNION
    
34. SELECT s.owner,
    
35.        s.segment_type,
    
36.        s.segment_name,
    
37.        s.partition_name,
    
38.        c.file#,
    
39.        header_block corr_start_block#,
    
40.        header_block corr_end_block#,
    
41.        1 blocks_corrupted,
    
42.        'Segment Header' description
    
43.   FROM dba_segments s, v$database_block_corruption c
    
44. WHERE s.header_file = c.file#
    
45.    AND s.header_block between c.block# and c.block# + c.blocks - 1
    
46. UNION
    
47. SELECT null owner,
    
48.        null segment_type,
    
49.        null segment_name,
    
50.        null partition_name,
    
51.        c.file#,
    
52.        greatest(f.block_id, c.block#) corr_start_block#,
    
53.        least(f.block_id + f.blocks - 1, c.block# + c.blocks - 1) corr_end_block#,
    
54.        least(f.block_id + f.blocks - 1, c.block# + c.blocks - 1) -
    
55.        greatest(f.block_id, c.block#) + 1 blocks_corrupted,
    
56.        'Free Block' description
    
57.   FROM dba_free_space f, v$database_block_corruption c
    
58. WHERE f.file_id = c.file#
    
59.    AND f.block_id <= c.block# + c.blocks - 1
    
60.    AND f.block_id + f.blocks - 1 >= c.block#
    
61. order by file#, corr_start_block#;
    
62. 
    
63. 
    
64. SELECT tablespace_name, segment_type, owner, segment_name
    
65. FROM dba_extents
    
66. WHERE file_id = &fileid
    
67. and &blockid between block_id AND block_id + blocks - 1;
    

复制代码

oracle321

刘大，按你的步骤逐一执行，收集输出如下：

1. dbv检查结果

1. 
   
2. [oracle@win0105rac1 ~]$ dbv file=/oradata/davidloc/davidloc/system01.dbf
   
3. 
   
4. DBVERIFY: Release 10.2.0.4.0 - Production on Mon May 2 04:14:29 2011
   
5. 
   
6. Copyright (c) 1982, 2007, Oracle.  All rights reserved.
   
7. 
   
8. DBVERIFY - Verification starting : FILE = /oradata/davidloc/davidloc/system01.dbf
   
9. 
   
10. 
    
11. DBVERIFY - Verification complete
    
12. 
    
13. Total Pages Examined         : 38400
    
14. Total Pages Processed (Data) : 11706
    
15. Total Pages Failing   (Data) : 0
    
16. Total Pages Processed (Index): 3348
    
17. Total Pages Failing   (Index): 0
    
18. Total Pages Processed (Other): 1784
    
19. Total Pages Processed (Seg)  : 0
    
20. Total Pages Failing   (Seg)  : 0
    
21. Total Pages Empty            : 21562
    
22. Total Pages Marked Corrupt   : 0
    
23. Total Pages Influx           : 0
    
24. Highest block SCN            : 452340 (0.452340)
    
25. [oracle@win0105rac1 ~]$ dbv file=/oradata/davidloc/davidloc/system02.dbf
    
26. 
    
27. DBVERIFY: Release 10.2.0.4.0 - Production on Mon May 2 04:14:48 2011
    
28. 
    
29. Copyright (c) 1982, 2007, Oracle.  All rights reserved.
    
30. 
    
31. DBVERIFY - Verification starting : FILE = /oradata/davidloc/davidloc/system02.dbf
    
32. 
    
33. 
    
34. DBVERIFY - Verification complete
    
35. 
    
36. Total Pages Examined         : 6400
    
37. Total Pages Processed (Data) : 0
    
38. Total Pages Failing   (Data) : 0
    
39. Total Pages Processed (Index): 0
    
40. Total Pages Failing   (Index): 0
    
41. Total Pages Processed (Other): 1
    
42. Total Pages Processed (Seg)  : 0
    
43. Total Pages Failing   (Seg)  : 0
    
44. Total Pages Empty            : 6399
    
45. Total Pages Marked Corrupt   : 0
    
46. Total Pages Influx           : 0
    
47. Highest block SCN            : 0 (0.0)
    

复制代码

2. rman检查

1. [oracle@win0105rac1 ~]$ rman target / nocatalog
   
2. 
   
3. Recovery Manager: Release 10.2.0.4.0 - Production on Mon May 2 04:22:12 2011
   
4. 
   
5. Copyright (c) 1982, 2007, Oracle.  All rights reserved.
   
6. 
   
7. connected to target database: DAVIDLOC (DBID=764783080)
   
8. using target database control file instead of recovery catalog
   
9. 
   
10. RMAN> run {
    
11. 2> allocate channel d1 type disk;
    
12. 3> allocate channel d2 type disk;
    
13. 4> allocate channel d3 type disk;
    
14. 5> allocate channel d4 type disk;
    
15. 6> backup validate check logical database;
    
16. 7> release channel d1;
    
17. 8> release channel d2;
    
18. 9> release channel d3;
    
19. 10> release channel d4;
    
20. 11> }
    
21. 
    
22. allocated channel: d1
    
23. channel d1: sid=145 devtype=DISK
    
24. 
    
25. allocated channel: d2
    
26. channel d2: sid=144 devtype=DISK
    
27. 
    
28. allocated channel: d3
    
29. channel d3: sid=143 devtype=DISK
    
30. 
    
31. allocated channel: d4
    
32. channel d4: sid=142 devtype=DISK
    
33. 
    
34. Starting backup at 02-MAY-11
    
35. channel d1: starting full datafile backupset
    
36. channel d1: specifying datafile(s) in backupset
    
37. input datafile fno=00004 name=/oradata/davidloc/davidloc/users01.dbf
    
38. channel d2: starting full datafile backupset
    
39. channel d2: specifying datafile(s) in backupset
    
40. input datafile fno=00001 name=/oradata/davidloc/davidloc/system01.dbf
    
41. input datafile fno=00005 name=/oradata/davidloc/davidloc/system02.dbf
    
42. channel d3: starting full datafile backupset
    
43. channel d3: specifying datafile(s) in backupset
    
44. input datafile fno=00002 name=/oradata/davidloc/davidloc/undotbs101.dbf
    
45. input datafile fno=00003 name=/oradata/davidloc/davidloc/sysaux01.dbf
    
46. channel d4: starting full datafile backupset
    
47. channel d4: specifying datafile(s) in backupset
    
48. including current control file in backupset
    
49. channel d4: backup set complete, elapsed time: 00:00:08
    
50. channel d4: starting full datafile backupset
    
51. channel d4: specifying datafile(s) in backupset
    
52. including current SPFILE in backupset
    
53. channel d4: backup set complete, elapsed time: 00:00:07
    
54. channel d3: backup set complete, elapsed time: 00:00:20
    
55. channel d1: backup set complete, elapsed time: 00:00:35
    
56. channel d2: backup set complete, elapsed time: 00:00:35
    
57. Finished backup at 02-MAY-11
    
58. 
    
59. released channel: d1
    
60. 
    
61. released channel: d2
    
62. 
    
63. released channel: d3
    
64. 
    
65. released channel: d4
    
66. 
    
67. RMAN> exit
    
68. 
    
69. 
    
70. Recovery Manager complete.

复制代码

3. v$database_block_corruption视图查询

1. 
   
2. SQL>  select * from V$DATABASE_BLOCK_CORRUPTION
   
3.   2  ;
   
4. 
   
5. no rows selected

复制代码

4. 步骤3无结果返回，步骤4未执行。

数据库冒似没有坏块。。。

Maclean Liu(刘相兵

X$KTFBHC
         [k]ernel [t]ablespace [f]ile [b]itmapped
           [h]eader [c]ontrol

重启一下是否问题消失？

怀疑是 文件头的 bitmap 有问题了， 考虑 导出数据 并重建数据库吧

oracle321

恩，已经让他们导出数据了，准备重建。

重启过，还是一样的问题。