ORA-07445 : 出現異常錯誤: 核心轉儲

ORA-07445: exception encountered: core dump [kghalp()+58] [SIGSEGV] [Address not mapped to object] [0x000000068] [] []

DUMP文件

*** SERVICE NAME:(orcl) 2012-07-12 14:18:55.016-
 

*** SESSION ID:(258.58198) 2012-07-12 14:18:55.016

Exception signal: 11 (SIGSEGV), code: 1 (Address not mapped to object), addr: 0x68, PC: [0x3c175e2, kghalp()+58]

*** 2012-07-12 14:18:55.030

ksedmp: internal or fatal error

ORA-07445: exception encountered: core dump [kghalp()+58] [SIGSEGV] [Address not mapped to object] [0x000000068] [] []

Current SQL statement for this session:

INSERT INTO MAILSENDER.GARBAGEDISTRIBUTING SELECT :B6 ,DOMAIN,MAX(CASE WHEN FLODER='ad' THEN COUNTNUM ELSE 0 END ) AS AD, MAX(CASE W

HEN FLODER='Garbage' THEN COUNTNUM ELSE 0 END) AS GARBAGE,:B3 ,:B5 ,:B4 FROM ( SELECT DOMAIN, FLODER, COUNT(MESSAGEINFOID) AS COUNTN

UM FROM( SELECT MBL.DOMAIN,T.MESSAGEINFOID,CASE WHEN REGEXP_SUBSTR(T.FLODER,'(Junk)|(QQ閭�錃g璁㈤槄)|(鍨冨溇閭�錃g)|(璁㈤槄閭�錃g)') IS NOT NULL

 THEN 'Garbage' WHEN REGEXP_SUBSTR(T.FLODER,'(騫垮憡閭�錃g)') IS NOT NULL THEN 'ad' ELSE '' END AS FLODER FROM MESSAGEINFO T,MAILBOXLIST

 MBL,TABLE(SPLITSTR(:B1 ,',')) TBEMAIL WHERE MBL.MAILBOXID=T.MAILBOXID AND TBEMAIL.COLUMN_VALUE=MBL.EMAIL AND T.ADDDATE BETWEEN :B3

AND :B2 AND (SELECT COUNT(*) FROM MESSAGEINFO M WHERE T.MAILID=M.MAILID AND T.MESSAGEINFOID<M.MESSAGEINFOID)<1) GROUP BY DOMAIN,FLOD

ER ) GROUP BY DOMAIN

----- PL/SQL Call Stack -----

 object     line object

 handle   number name

0x25b104658       23 package body MONITOR.PACKAGE_GARBAGEDISTRIBUTING

0x238b77b30        1 anonymous block

----- Call Stack Trace -----

calling             call    entry               argument values in hex    

location            type    point               (? means dubious value)   

-------------------- -------- -------------------- ----------------------------

Cannot find symbol

Cannot find symbol

Cannot find symbol

ksedst()+31         call    ksedst1()           000000001 ? 000000001 ?

                                                  000000000 ? 000000000 ?

                                                  000000000 ? 000000001 ?

ksedmp()+610        call    ksedst()            000000001 ? 000000001 ?

                                                  000000000 ? 000000000 ?

                                                  000000000 ? 000000001 ?

ssexhd()+630        call    ksedmp()            000000003 ? 000000001 ?

                                                  000000000 ? 000000000 ?

                                                  000000000 ? 000000001 ?

__restore_rt()+0    call    ssexhd()            00000000B ? 2B6124A48D70 ?

                                                  2B6124A48C40 ? 000000000 ?

                                                  000000000 ? 000000001 ?

<(nil)>             call    __restore_rt()      00000000B ? 2B6124A48D70 ?

                                                  2B6124A48C40 ? 000000000 ?

                                                  000000000 ? 000000001 ?

我們的ORA-07445，根據DUMP文件分析，多半是因為亂碼問題照成的，然后修改了服務器的字符集。（即添加export NLS_LANG=AMERICAN_AMERICA.ZHS16GBK）。然后跟進我們這個包MONITOR.PACKAGE_GARBAGEDISTRIBUTING。一切OK了。

總結：亂碼問題也可以導致ora-07445問題。遇到問題沉著冷靜，細心才可以把問題解決好。

附表：

以下為網上收集的ORA-07445問題

select owner,object_name,object_type,debuginfo
rom all_probe_objects
where object_name='TRI_XXX' and object_type like 'TRIGGER';

OWNER        OBJECT_NAME    OBJECT_TYPE     D
------------ -------------- ---------------     -
COPERATOR    TRI_XXX        TRIGGER             T

1 rows selected.

  天哪，真是DEBUG狀態！
   一個因為在DEBUG狀態，且是基于DELETE的TRIGGER觸發了這個Oracle的BUG！
    ......
   現在要么改這個TRIGGER的狀態，要么打補丁（9.2.0.6）。第一種選擇是明智而快捷且安裝的。重新編譯，檢查不是DEBUG狀態。應用啟動，正常運行

服務器租用/服務器托管中國五強！虛擬主機域名注冊頂級提供商！15年品質保障！--億恩科技[ENKJ.COM]