ORA-07445: exception encountered: core dump [ptg_tx()+40]

Hi:

After migrating our oracle db from 8.1.6.0.0 to 8.1.7.0.0 on a Sun/Solaris box, we have encount some new core dump problem (interMedia text stuff related).

We have a perl script which has a loop

start loop

   # do something
   select refid from core_ref where contains (abstract, 'xyz')>0    # do something else
end loop;

The whole program usually run for 2-3 days. Sometime during the loop execution, the oracle would create a core dump and the perl program exit. We have never had this problem before. I also tried on another newly-created 8.1.7.0.0 db and the program seems to be running fine.

The table definition is:

SQL> desc core_ref;

 Name                                      Null?    Type
 ----------------------------------------- --------
----------------------------
 REFID                                     NOT NULL NUMBER
 TITLE                                              VARCHAR2(1600)
 JOURNALID                                          NUMBER
 YEAR                                               NUMBER
 ABSTRACT                                           VARCHAR2(4000)

And the trace file is like:

• 2002-04-15 14:26:41.292
• SESSION ID:(25.9933) 2002-04-15 14:26:41.281 Exception signal: 11 (SIGSEGV), code: 1 (Address not mapped to object), addr: 0x 8, PC: ptg_tx()+40
• 2002-04-15 14:26:41.949 ksedmp: internal or fatal error ORA-07445: exception encountered: core dump [ptg_tx()+40] [SIGSEGV] [Address not mapped to object] [8] [] [] Current SQL statement for this session: select refid from core_ref where contains (abstract, 'lsm1')>0 ----- Call Stack Trace ----- calling call entry argument values in hex location type point (? means dubious value) -------------------- -------- --------------------

  ---

  ksedmp()+168 CALL ksedst()+0 540 ? 0 ? FFBE9300 ? FFBE8DA4 ? FFBE8D88 ? 0 ? ssexhd()+380 CALL ksedmp()+0 3 ? 0 ? 1 ? FFBE9A20 ? 1 ? 7 ? sigacthandler()+40 PTR_CALL 00000000 B ? FFBEA278 ? 1A28000 ? 1A28000 ? 0 ? 0 ? ptg_tx()+40 PTR_CALL 00000000 B ? FFBEA278 ? FFBE9FC0 ? 0 ? 0 ? A2EE3665 ? pkrgsm()+1324 CALL ptg_tx()+0 FFBEA78C ? 16B788C ? ....

We tried to put "Alter system flush shared_pool" in the loop and it does not seem to help. We are trying to disconnect-reconnect to see if it helps.

Has anyone experienced this problem before?

My question are: Do you know what might cause this error? And is there a workaround?

TIA. Guang Received on Tue Apr 16 2002 - 12:15:27 CDT