ORA-07445 core dump periodically happen

From: Jing Jing Wu <jjwu_at_webfs.com>
Date: Fri, 03 Aug 2001 16:58:58 -0700
Message-ID: <3B6B3AC2.16326ACF@webfs.com>

Hi:

I am looking for the suggestions from experienced oracle DBA. I have spent many nights on it, but I still couldn't find what's wrong.

Problem: ora-07445 core dumps periodly throwed by dead multi-threaded servers one by one. (every minute per core dump). The errors may occured in peak time or non-peak time. On the application which connected to the database server from other machine, there are many ORA-03113 in the application log file correspondingly.

Platform: ORACLE 8.1.7 on Sun OS 5.7

Here are the error message:

---

Exception signal: 11 (SIGSEGV), code: 1 (Address not mapped to object), addr: 0x20, PC: _memcpy()+404
*** 2001-08-03 08:04:08.340
ksedmp: internal or fatal error
ORA-07445: exception encountered: core dump [_memcpy()+404] [SIGSEGV] [Address not mapped to object] [32] [] [ ]
Current SQL statement for this session:
SELECT PERMISSION FROM FSOBJECT_PERMISSIONS WHERE MEMBERID = :b1 AND MEMBERID IS NOT NULL AND FSOBJECT_I
D = :b2 UNION ALL SELECT PERMISSION FROM FSOBJECT_PERMISSIONS A,GROUP_MEMBERS B WHERE B.MEMBERID = :b1 AN D A.GROUP_ID = B.GROUP_ID AND A.FSOBJECT_ID = :b2 AND A.GROUP_ID IS NOT NULL
----- PL/SQL Call Stack -----
  object line object
  handle number name

9f9f16c4      3849  package body WEBFSDBUID.SHARING_PKG
9f9f16c4      3879  package body WEBFSDBUID.SHARING_PKG
9f85ecc4         3  anonymous block
---------------------------------

I am looking forword for your response.

Sincerely

Jingjing Received on Fri Aug 03 2001 - 18:58:58 CDT