解决CON$ ORA-600 kdsgrp1错误

联系:手机/微信(+86 17813235971) QQ(107644445)

标题:解决CON$ ORA-600 kdsgrp1错误

作者:惜分飞©版权所有[未经本人同意,不得以任何形式转载,否则有进一步追究法律责任的权利.]

数据库报ORA 600 kdsgrp1错误
数据库报ORA-00600: internal error code, arguments: [kdsgrp1], [], [], [], [], [], [], [], [], [], [], []错

Thread 1 advanced to log sequence 23861 (LGWR switch)
  Current log# 7 seq# 23861 mem# 0: /oradata/easdb/redo07.log
Tue Nov 15 10:00:42 2016
Errors in file /u01/oracle/diag/rdbms/easdb/easdb/trace/easdb_dw00_3165.trc  (incident=908262):
ORA-00600: internal error code, arguments: [kdsgrp1], [], [], [], [], [], [], [], [], [], [], []
Incident details in: /u01/oracle/diag/rdbms/easdb/easdb/incident/incdir_908262/easdb_dw00_3165_i908262.trc
Tue Nov 15 10:00:55 2016
Use ADRCI or Support Workbench to package the incident.
See Note 411.1 at My Oracle Support for error and packaging details.
Tue Nov 15 10:00:56 2016
Errors in file /u01/oracle/diag/rdbms/easdb/easdb/trace/easdb_dw00_3165.trc  (incident=908263):
ORA-00600: internal error code, arguments: [kdsgrp1], [], [], [], [], [], [], [], [], [], [], []
ORA-06512: at "SYS.KUPW$WORKER", line 1751
ORA-06512: at line 2
Incident details in: /u01/oracle/diag/rdbms/easdb/easdb/incident/incdir_908263/easdb_dw00_3165_i908263.trc
Use ADRCI or Support Workbench to package the incident.
See Note 411.1 at My Oracle Support for error and packaging details.
DW00 terminating with fatal err=600, pid=40, wid=1, job SYSTEM.
Tue Nov 15 10:01:01 2016
Thread 1 advanced to log sequence 23862 (LGWR switch)
  Current log# 2 seq# 23862 mem# 0: /oradata/easdb/redo02.log
Tue Nov 15 10:01:23 2016
Errors in file /u01/oracle/diag/rdbms/easdb/easdb/trace/easdb_dm00_3163.trc  (incident=908254):
ORA-31671: Worker process DW00 had an unhandled exception.
ORA-00600: internal error code, arguments: [kdsgrp1], [], [], [], [], [], [], [], [], [], [], []
ORA-06512: at "SYS.KUPW$WORKER", line 1751
ORA-06512: at line 2
Incident details in: /u01/oracle/diag/rdbms/easdb/easdb/incident/incdir_908254/easdb_dm00_3163_i908254.trc
Use ADRCI or Support Workbench to package the incident.
See Note 411.1 at My Oracle Support for error and packaging details.
Tue Nov 15 10:01:26 2016
Tue Nov 15 10:01:28 2016
Thread 1 advanced to log sequence 23863 (LGWR switch)
  Current log# 4 seq# 23863 mem# 0: /oradata/easdb/redo04.log

trace文件中信息

*** 2016-11-15 10:00:35.977
* kdsgrp1-1: *************************************************
            row 0x004459e6.26 continuation at
            0x004459e6.26 file# 1 block# 285158 slot 38 not found
KDSTABN_GET: 0 ..... ntab: 1
curSlot: 38 ..... nrows: 208
kdsgrp - dump CR block dba=0x004459e6
Block header dump:  0x004459e6
 Object id on Block? Y
 seg/obj: 0x1c  csc: 0x01.c712f743  itc: 3  flg: -  typ: 1 - DATA
     fsl: 0  fnx: 0x0 ver: 0x01
 Itl           Xid                  Uba         Flag  Lck        Scn/Fsc
0x01   0x000b.015.0036d715  0x00c01bba.0fbd.02  C---    0  scn 0x0001.c6b4cb1a
0x02   0x000c.004.00044d36  0x04c0dd93.3eec.33  C---    0  scn 0x0001.c6d2c65b
0x03   0x000d.008.00008eb9  0x04c0777a.10e3.02  --U-    2  fsc 0x0056.c7346f21

确定报错对象和确认异常

SQL> select object_name from dba_objects where object_id=28;
OBJECT_NAME
---------------------------------------------------------
CON$
SQL> ANALYZE TABLE sys.CON$ VALIDATE STRUCTURE CASCADE online;
ANALYZE TABLE sys.CON$ VALIDATE STRUCTURE CASCADE online
*
ERROR at line 1:
ORA-01499: table/index cross reference failure - see trace file
SQL> SET LINES 122
SQL> COL INDEX_OWNER FOR A20
SQL> COL INDEX_NAME FOR A30
SQL> COL TABLE_OWNER FOR A20
SQL> COL COLUMN_NAME FOR A25
SQL> SELECT TABLE_OWNER,INDEX_NAME,COLUMN_NAME,COLUMN_POSITION
2  FROM Dba_Ind_Columns
3  WHERE table_name = upper('&TABLE_NAME') order by TABLE_OWNER,INDEX_OWNER,INDEX_NAME,COLUMN_POSITION;
Enter value for table_name: CON$
old   3:  WHERE table_name = upper('&TABLE_NAME') order by TABLE_OWNER,INDEX_OWNER,INDEX_NAME,COLUMN_POSITION
new   3:  WHERE table_name = upper('CON$') order by TABLE_OWNER,INDEX_OWNER,INDEX_NAME,COLUMN_POSITION
TABLE_OWNER	     INDEX_NAME 		    COLUMN_NAME 	      COLUMN_POSITION
-------------------- ------------------------------ ------------------------- ---------------
SYS		     I_CON1			    OWNER#				    1
SYS		     I_CON1			    NAME				    2
SYS		     I_CON2			    CON#				    1
SQL> select owner#,name from con$
2    minus
3   select /*+ full(t) */owner#,name from con$ t;
no rows selected
SQL> select /*+ full(t) */owner#,name from con$ t
2    minus
3   select owner#,name from con$  ;
no rows selected
SQL> select /*+ full(t) */ con# from con$ t
2    minus
3   select con# from con$ ;
no rows selected
SQL> select con# from con$
2    minus
3   select /*+ full(t) */ con# from con$ t   ;
      CON#
----------
   1037224
   1037225
   1037386
   1037387
   1037388
   ……
   1037846
62 rows selected.

通过上述分析,可以确定是由于CON$和I_CON2数据不一致,而且是index的数据比表中多了62条.针对这样情况,考虑通过重建index来解决.

尝试rebuild index

SQL> alter index I_CON2 rebuild online;
alter index I_CON2 rebuild online
*
ERROR at line 1:
ORA-00701: object necessary for warmstarting database cannot be altered
SQL>
SQL>
SQL>
SQL>
SQL> shutdown immediate;
Database closed.
Database dismounted.
ORACLE instance shut down.
SQL> startup upgrade
ORACLE instance started.
Total System Global Area 2421825536 bytes
Fixed Size                  2215744 bytes
Variable Size            1828716736 bytes
Database Buffers          570425344 bytes
Redo Buffers               20467712 bytes
Database mounted.
Database opened.
SQL> alter index I_CON2 rebuild;
alter index I_CON2 rebuild
*
ERROR at line 1:
ORA-00701: object necessary for warmstarting database cannot be altered

因为是数据库核心index,无法直接rebuild解决,只能通过bootstrap$核心index(I_OBJ1,I_USER1,I_FILE#_BLOCK#,I_IND1,I_TS#,I_CDEF1等)异常恢复—ORA-00701错误解决 方式解决

ORA-600 kdsgrp1

联系:手机/微信(+86 17813235971) QQ(107644445)

标题:ORA-600 kdsgrp1

作者:惜分飞©版权所有[未经本人同意,不得以任何形式转载,否则有进一步追究法律责任的权利.]

在硬件恢复,断电,redo异常等恢复case中ORA-600 [kdsgrp1]是一个比较常见的错误,这里该出来官方关于该错误的解释说明和处理方法

RROR:
  Format: ORA-600 [kdsgrp1]
VERSIONS:
  versions 10.1 and above
DESCRIPTION:
 This error was introduced in 10g with the fix to Bug 2442351, it provides
 for an extra health check on a block, we detected a null row header,
 see Note:2442351.9 for more information.
 Error may be caused by:
 Case 1. A row referenced in an index that does not exist in the table.
 Case 2. An non-existent rowid pointed to by a chained row.
 Trace Examples:
 Case 1. Mismatch between table and index:
====================================================
 Trace file has:
 row 02433566.13 continuation at
 file# 9 block# 210278 slot 20 not found
 The file=9 block=210278 is rdba=0x02433566 which was taken from an index:
 row#3[7549] flag: ------, lock: 0, len=85, data:(6):  02 43 35 66 00 14
 But the slot 20 does not exist in the table block:
 tab 0, row 1, @0x1e62
 tl: 2 fb: --HDFL-- lb: 0x3
 tab 0, row 12, @0x191a
 tl: 2 fb: --HDFL-- lb: 0x1
 tab 0, row 17, @0x1675
 tl: 2 fb: --HDFL-- lb: 0x2
 tab 0, row 21, @0x1459
 tl: 2 fb: --HDFL-- lb: 0x4
 ORA-1499 may be produced by analyze:
 analyze table <table name> validate structure cascade;
 Case 2. A row points to another rowid which does not exist (Chained row does not exist).
============================================================================================
 Trace file has:
 row 1186b11a.ffffffff continuation at
 file# 70 block# 441621 slot 1 not found
 It means that row with rdba 0x1186b11a continues in file# 70 block# 441621 slot 1.
 But the information in file# 70 block# 441621 slot 1 does not exist.  It is:
 tab 0, row 16, @0xd7f    ---> This is the slot with the problem.
 tl: 29 fb: -------- lb: 0x0  cc: 11
 nrid:  0x1186bd15.1      ---> It points to rdba=0x1186bd15 slot 1
(file# 70 block# 441621 slot 1) but that row does not exist in that block.
 For this case ANALYZE TABLE .. VALIDATE STRUCTURE is not detecting this logical corruption
Referece Bug 6858313
Run an export (exp) or Full Table Scan to identify if there is a permanent invalid chained row.
 Workaround for Case 2:
 The row producing the ORA-600 [kdsgrp1] can be skipped by setting the Event 10231
 Note that a testcase has concluded that event 10231 does not skip rows in an Index Organized Table (IOT)
 when there is an invalid nrid as explained in Case 2.  It only works for regular tables.
 Event 43810 skip corrupt block in IOT?s (10.2.0.4)
nor  parameter _index_scan_check_skip_corrupt (11g) work for this case 2 on IOTs either.
FUNCTIONALITY:
  Kernel Data layer Seek/Scan
IMPACT:
  PROCESS FAILURE
  POSSIBLE PHYSICAL CORRUPTION

某集团ebs数据库redo undo丢失导致悲剧

联系:手机/微信(+86 17813235971) QQ(107644445)

标题:某集团ebs数据库redo undo丢失导致悲剧

作者:惜分飞©版权所有[未经本人同意,不得以任何形式转载,否则有进一步追究法律责任的权利.]

某集团的ebs系统因磁盘空间不足把redo和undo存放到raid 0之上,而且该库无任何备份。最终悲剧发生了,raid 0异常导致redo undo全部丢失,数据库无法正常启动(我接手之时数据库已经resetlogs过,但是未成功)

Sun Jul 27 11:31:27 2014
SMON: enabling cache recovery
SMON: enabling tx recovery
Sun Jul 27 11:31:27 2014
Database Characterset is ZHS16GBK
Sun Jul 27 11:31:27 2014
Errors in file /prod/oracle/PROD/db/tech_st/10.2.0/admin/PROD_erpserver/bdump/prod_smon_454754.trc:
ORA-00604: error occurred at recursive SQL level 1
ORA-00376: file 42 cannot be read at this time
ORA-01110: data file 42: '/prod/oracle/PROD/logdata/undo/undo1.dbf'
Sun Jul 27 11:31:27 2014
Errors in file /prod/oracle/PROD/db/tech_st/10.2.0/admin/PROD_erpserver/bdump/prod_smon_454754.trc:
ORA-00604: error occurred at recursive SQL level 1
ORA-00376: file 42 cannot be read at this time
ORA-01110: data file 42: '/prod/oracle/PROD/logdata/undo/undo1.dbf'
Sun Jul 27 11:31:27 2014
Errors in file /prod/oracle/PROD/db/tech_st/10.2.0/admin/PROD_erpserver/bdump/prod_smon_454754.trc:
ORA-00604: error occurred at recursive SQL level 1
ORA-00376: file 42 cannot be read at this time
ORA-01110: data file 42: '/prod/oracle/PROD/logdata/undo/undo1.dbf'
Sun Jul 27 11:31:27 2014
Errors in file /prod/oracle/PROD/db/tech_st/10.2.0/admin/PROD_erpserver/udump/prod_ora_663670.trc:
ORA-00604: error occurred at recursive SQL level 1
ORA-00376: file 41 cannot be read at this time
ORA-01110: data file 41: '/prod/oracle/PROD/logdata/undo/undo2.dbf'
Error 604 happened during db open, shutting down database
USER: terminating instance due to error 604
Instance terminated by USER, pid = 663670
ORA-1092 signalled during: ALTER DATABASE OPEN...

查询相关文件状态发现,undo表空间文件丢失,被offline处理
df_status
因为以前alert日志被清理,通过这里大概猜测是offline丢失的undo文件,然后resetlogs了数据库,现在处理方式为
使用_corrupted_rollback_segments屏蔽回滚段,然后尝试启动数据库

Tue Jul 29 11:40:39 2014
SMON: enabling cache recovery
SMON: enabling tx recovery
Tue Jul 29 11:40:39 2014
Database Characterset is ZHS16GBK
Tue Jul 29 11:40:39 2014
Errors in file /prod/oracle/PROD/db/tech_st/10.2.0/admin/PROD_erpserver/bdump/prod_smon_569378.trc:
ORA-00604: error occurred at recursive SQL level 1
ORA-01555: snapshot too old: rollback segment number  with name "" too small
Tue Jul 29 11:40:39 2014
Errors in file /prod/oracle/PROD/db/tech_st/10.2.0/admin/PROD_erpserver/bdump/prod_smon_569378.trc:
ORA-00604: error occurred at recursive SQL level 1
ORA-01555: snapshot too old: rollback segment number  with name "" too small
Tue Jul 29 11:40:39 2014
Errors in file /prod/oracle/PROD/db/tech_st/10.2.0/admin/PROD_erpserver/bdump/prod_smon_569378.trc:
ORA-00604: error occurred at recursive SQL level 1
ORA-01555: snapshot too old: rollback segment number  with name "" too small
Tue Jul 29 11:40:39 2014
Errors in file /prod/oracle/PROD/db/tech_st/10.2.0/admin/PROD_erpserver/udump/prod_ora_585786.trc:
ORA-00604: error occurred at recursive SQL level 1
ORA-01555: snapshot too old: rollback segment number  with name "" too small
Error 604 happened during db open, shutting down database
USER: terminating instance due to error 604
Instance terminated by USER, pid = 585786
ORA-1092 signalled during: alter database open...

该错误是由于数据库启动需要找到对应的回滚段,但是由于undo异常导致该回滚段无法找到,因此出现该错误,解决方法是通过修改数据scn,让其不找回滚段,从而屏蔽该错误.数据库启动后,删除undo重新创建新undo

Tue Jul 29 15:59:22 2014
drop tablespace undo2 including contents and datafiles
Tue Jul 29 15:59:23 2014
Errors in file /prod/oracle/PROD/db/tech_st/10.2.0/admin/PROD_erpserver/udump/prod_ora_782490.trc:
ORA-01122: database file 41 failed verification check
ORA-01110: data file 41: '/prod/oracle/PROD/logdata/undo/undo2.dbf'
ORA-01565: error in identifying file '/prod/oracle/PROD/logdata/undo/undo2.dbf'
ORA-27037: unable to obtain file status
IBM AIX RISC System/6000 Error: 2: No such file or directory
Additional information: 3
Tue Jul 29 15:59:23 2014
Errors in file /prod/oracle/PROD/db/tech_st/10.2.0/admin/PROD_erpserver/udump/prod_ora_782490.trc:
ORA-01259: unable to delete datafile /prod/oracle/PROD/logdata/undo/undo2.dbf
Tue Jul 29 15:59:23 2014
Errors in file /prod/oracle/PROD/db/tech_st/10.2.0/admin/PROD_erpserver/udump/prod_ora_782490.trc:
ORA-01122: database file 42 failed verification check
ORA-01110: data file 42: '/prod/oracle/PROD/logdata/undo/undo1.dbf'
ORA-01565: error in identifying file '/prod/oracle/PROD/logdata/undo/undo1.dbf'
ORA-27037: unable to obtain file status
IBM AIX RISC System/6000 Error: 2: No such file or directory
Additional information: 3
ORA-01259: unable to delete datafile /prod/oracle/PROD/logdata/undo/undo2.dbf
Tue Jul 29 15:59:23 2014
Errors in file /prod/oracle/PROD/db/tech_st/10.2.0/admin/PROD_erpserver/udump/prod_ora_782490.trc:
ORA-01259: unable to delete datafile /prod/oracle/PROD/logdata/undo/undo1.dbf
Tue Jul 29 15:59:23 2014
Completed: drop tablespace undo2 including contents and datafiles
Tue Jul 29 15:59:56 2014
create undo tablespace undotbs1 datafile '/prod/oracle/PROD/logdata/undo_new01.dbf' size 100M autoextend on next 128M maxsize 30G
Tue Jul 29 15:59:57 2014
Completed: create undo tablespace undotbs1 datafile '/prod/oracle/PROD/logdata/undo_new01.dbf' size 100M autoextend on next 128M maxsize 30G
Tue Jul 29 16:00:03 2014
alter tablespace undotbs1 add datafile '/prod/oracle/PROD/logdata/undo_new02.dbf' size 100M autoextend on next 128M maxsize 30G
Completed: alter tablespace undotbs1 add datafile '/prod/oracle/PROD/logdata/undo_new02.dbf' size 100M autoextend on next 128M maxsize 30G

业务运行过程中,数据库报大量ORA-600 4097,ORA-600 kdsgrp1,ORA-600 kcfrbd_3错误

Tue Jul 29 16:07:03 2014
Errors in file /prod/oracle/PROD/db/tech_st/10.2.0/admin/PROD_erpserver/udump/prod_ora_950484.trc:
ORA-00600: internal error code, arguments: [4097], [], [], [], [], [], [], []
Tue Jul 29 16:07:06 2014
Errors in file /prod/oracle/PROD/db/tech_st/10.2.0/admin/PROD_erpserver/udump/prod_ora_950484.trc:
ORA-00600: internal error code, arguments: [kdsgrp1], [], [], [], [], [], [], []
Tue Jul 29 16:10:06 2014
Errors in file /prod/oracle/PROD/db/tech_st/10.2.0/admin/PROD_erpserver/udump/prod_ora_917702.trc:
ORA-00600: internal error code, arguments: [4097], [], [], [], [], [], [], []
Tue Jul 29 16:10:07 2014
Errors in file /prod/oracle/PROD/db/tech_st/10.2.0/admin/PROD_erpserver/udump/prod_ora_917702.trc:
ORA-00600: internal error code, arguments: [kdsgrp1], [], [], [], [], [], [], []
Tue Jul 29 16:12:45 2014
Errors in file /prod/oracle/PROD/db/tech_st/10.2.0/admin/PROD_erpserver/bdump/prod_m000_880692.trc:
ORA-00600: internal error code, arguments: [4097], [], [], [], [], [], [], []
Tue Jul 29 16:21:23 2014
Errors in file /prod/oracle/PROD/db/tech_st/10.2.0/admin/PROD_erpserver/udump/prod_ora_1040638.trc:
ORA-00600: 内部错误代码, 参数: [kcfrbd_3], [41], [231381], [1], [12800], [12800], [], []
Tue Jul 29 16:21:37 2014
Errors in file /prod/oracle/PROD/db/tech_st/10.2.0/admin/PROD_erpserver/udump/prod_ora_1040638.trc:
ORA-00600: 内部错误代码, 参数: [kcfrbd_3], [41], [231381], [1], [12800], [12800], [], []
Tue Jul 29 16:21:56 2014
Errors in file /prod/oracle/PROD/db/tech_st/10.2.0/admin/PROD_erpserver/udump/prod_ora_1040638.trc:
ORA-00600: 内部错误代码, 参数: [kcfrbd_3], [41], [231381], [1], [12800], [12800], [], []
Tue Jul 29 16:22:18 2014
Errors in file /prod/oracle/PROD/db/tech_st/10.2.0/admin/PROD_erpserver/udump/prod_ora_1040638.trc:
ORA-00600: 内部错误代码, 参数: [kcfrbd_3], [41], [231381], [1], [12800], [12800], [], []
Tue Jul 29 16:22:28 2014
Errors in file /prod/oracle/PROD/db/tech_st/10.2.0/admin/PROD_erpserver/udump/prod_ora_1105950.trc:
ORA-00600: 内部错误代码, 参数: [4097], [], [], [], [], [], [], []
Tue Jul 29 16:22:33 2014
Errors in file /prod/oracle/PROD/db/tech_st/10.2.0/admin/PROD_erpserver/udump/prod_ora_1159232.trc:
ORA-00600: 内部错误代码, 参数: [kcfrbd_3], [42], [61235], [1], [12800], [12800], [], []

出现该错误有几个原因和解决方法:
ORA-600 kdsgrp1 是因为相关坏块引起(tab,index,memory,cr block等),结合日志分析对象异常原因,根据具体情况确定对象然后选择合适处理方案(具体参考NOTE:1332252.1)
ORA-600 4097 由于数据库异常关闭然后open,创建回滚段,可能触发bug导致该问题(虽然说在当前版本修复,但是实际处理我确实按照NOTE:1030620.6解决)
ORA-600 kcfrbd_3 有事务的block被访问之后,根据回滚槽信息定位到相关回滚段,而正好新建的回滚段信息又和以前的名字编号一致,从而反馈出来是数据文件大小不够,从而出现该错误(具体参考NOTE:601798.1)
最终该数据库虽然恢复了,抢救了大量数据,但是对于ebs系统来说,丢失redo和undo数据的损失还是巨大的.再次温馨提示:数据库的redo,undo也很重要,数据库的备份更加重要