oradebug释放句柄引发的错误

saup007

1. 环境：
   
2. 11.2.0.2.0 - 64bit Production$ cat /etc/redhat-release Red Hat Enterprise Linux Server release 5.6 (Tikanga)
   
3. 我的操作步骤;
   
4. create undo tablespace undotbs2 DATAFILE  SIZE 1G AUTOEXTEND ON NEXT 50M MAXSIZE UNLIMITED;
   
5. 
   
6. alter system set undo_tablespace=UNDOTBS1 scope=both;
   
7. 
   
8. drop tablespace UNDOTBS1 including contents and datafiles;
   
9. 
   
10. /usr/sbin/lsof | grep undotbs1 | awk '{print $2}'
    
11. 
    
12. sys> oradebug setospid  5989
    
13. sys> oradebug call close 257;
    
14. 
    
15. kill inactive的业务session
    
16. kill -9
    
17. 
    
18. create undo tablespace undotbs1 DATAFILE  SIZE 1G  AUTOEXTEND ON NEXT 50M MAXSIZE UNLIMITED;
    
19. 
    
20. alter system set undo_tablespace=UNDOTBS1 scope=both;
    
21. 
    
22. drop tablespace UNDOTBS2 including contents and datafiles;
    
23. 
    
24. sys> oradebug setospid  5989
    
25. 
    
26. sys> oradebug call close 259;
    
27. 
    
28. kill inactive的业务session
    
29. kill -9
    
30. 
    
31. alter日志报错如下;（我操作的时间为下午1点多）报错时间为下午3点多
    
32. Fri Aug 03 15:18:17 2012
    
33. 
    
34. Errors in file
    
35. /u01/app/oracle/diag/rdbms/sid/SID/trace/SID_q000_5989.trc:
    
36. 
    
37. ORA-01115: IO error reading block from
    
38. file  (block # )
    
39. 
    
40. ORA-01110: data file 3:
    
41. '/u01/oradata/SID/datafile/o1_mf_undotbs1_81pr2654_.dbf'
    
42. 
    
43. ORA-27072: File I/O error
    
44. 
    
45. Linux-x86_64 Error: 9: Bad file descriptor
    
46. 
    
47. Additional information: 4
    
48. 
    
49. Additional information: 96991
    
50. 
    
51. Additional information: -1
    
52. 附trace文件

复制代码

[ 本帖最后由 saup007 于 2012-8-8 18:17 编辑 ]

saup007

1. alter日志算吗
   
2. 
   
3. $ tail -10000 alert_SID.log | grep  -C 5 ORA-27072
   
4. 
   
5. Fri Aug 03 15:18:17 2012
   
6. Errors in file /u01/app/oracle/diag/rdbms/sid/SID/trace/SID_q000_5989.trc:
   
7. ORA-01115: IO error reading block from file  (block # )
   
8. ORA-01110: data file 3: '/u01/oradata/SID/datafile/o1_mf_undotbs1_81pr2654_.dbf'
   
9. ORA-27072: File I/O error
   
10. Linux-x86_64 Error: 9: Bad file descriptor
    
11. Additional information: 4
    
12. Additional information: 96991
    
13. Additional information: -1

复制代码

Maclean Liu(刘相兵

ODM DATA:

Trace file /u01/app/oracle/diag/rdbms/oracle_sid/oracle_sid/trace/oracle_sid_q000_5989.trc
Oracle Database 11g Enterprise Edition Release 11.2.0.2.0 - 64bit Production
With the Partitioning, OLAP, Data Mining and Real Application Testing options
ORACLE_HOME = /u01/app/oracle/product/11.2.0/db_1
System name:        Linux
Node name:        dbserver2
Release:        2.6.18-238.19.1.el5xen
Version:        #1 SMP Fri Jul 15 08:16:59 EDT 2011
Machine:        x86_64
VM name:        Xen Version: 3.1 (PVM)
Instance name: oracle_sid
Redo thread mounted by this instance: 1
Oracle process number: 19
Unix process pid: 5989, image: oracle@dbserver2 (Q000)


*** 2012-08-03 13:19:54.179
*** SESSION ID:(592.11) 2012-08-03 13:19:54.179
*** CLIENT ID:() 2012-08-03 13:19:54.179
*** SERVICE NAME:(SYS$BACKGROUND) 2012-08-03 13:19:54.179
*** MODULE NAME:(Streams) 2012-08-03 13:19:54.179
*** ACTION NAME:(QMON Slave) 2012-08-03 13:19:54.179

Received ORADEBUG command (#1) 'call close 257' from process 'Unix process pid: 32285, image: <none>'

*** 2012-08-03 13:19:54.194
Finished processing ORADEBUG command (#1) 'call close 257'

*** 2012-08-03 13:26:04.138
Received ORADEBUG command (#2) 'call close 259' from process 'Unix process pid: 3904, image: <none>'

*** 2012-08-03 13:26:04.138
Finished processing ORADEBUG command (#2) 'call close 259'
DDE rules only execution for: ORA 1110

*** 2012-08-03 15:18:17.521
----- START Event Driven Actions Dump ----
---- END Event Driven Actions Dump ----
----- START DDE Actions Dump -----
Executing SYNC actions
----- START DDE Action: 'DB_STRUCTURE_INTEGRITY_CHECK' (Async) -----
Successfully dispatched
----- END DDE Action: 'DB_STRUCTURE_INTEGRITY_CHECK' (SUCCESS, 0 csec) -----
Executing ASYNC actions
----- END DDE Actions Dump (total 0 csec) -----
kwqitmmit during kwqitmmit:time mgr IOT proc: Error 1115 in Queue Table SYS.ALERT_QT
kwqicaclcur: Error 1115
             Cursor Session Number : 611
             Cursor Session Serial : 21975
             Cursor Pin     Number : 28681
kwqitmmit during kwqitmmit:end proc: Error 1115 in Queue Table SYS.ALERT_QT
kwqitptm: error 1115 encountered during serving 12580
KSV 1115 error in slave process

*** 2012-08-03 15:18:17.546
ORA-01115: IO error reading block from file  (block # )
ORA-01110: data file 3: '/u01/oradata/oracle_sid/datafile/o1_mf_undotbs1_81pr2654_.dbf'
ORA-27072: File I/O error
Linux-x86_64 Error: 9: Bad file descriptor
Additional information: 4
Additional information: 96991
Additional information: -1
OPIRIP: Uncaught error 447. Error stack:
ORA-00447: fatal error in background process
ORA-01115: IO error reading block from file  (block # )
ORA-01110: data file 3: '/u01/oradata/oracle_sid/datafile/o1_mf_undotbs1_81pr2654_.dbf'
ORA-27072: File I/O error
Linux-x86_64 Error: 9: Bad file descriptor
Additional information: 4
Additional information: 96991
Additional information: -1

Maclean Liu(刘相兵

如果仅仅是 后台TRACE中有 ORA-27072: File I/O error 错误的话 可以忽略

Maclean Liu(刘相兵

为什么要用 oradebug call close 259;？


我在这篇文章里说明了 http://www.askmaclean.com/archiv ... A5%94%E6%BA%83.html


call close 不要用在 产品环境， 会引起的问题

saup007

我也是参考原来我发的帖子做的。

http://t.askmaclean.com/viewthread.php?tid=1105&highlight=