Oracle9i R2 Startup NoMount Fails with ORA-07445
From: <japplewhite_at_austinisd.org>
Date: Mon, 16 Jul 2012 11:46:49 -0500
Message-ID: <OFBB1F38DF.FE3FD5F8-ON86257A3D.005A85DF-86257A3D.005C2671_at_austinisd.org>
HPUX PA-RISC 11.11 64bit with Oracle9iR2 (9.2.0.8) 64bit. While bouncing an existing DB the other day, I got the ORA-07445 error and then the DB wouldn't even startup nomount. The MOS 7445 lookup tool doesn't recognize the first argument and I can't find anything on MOS or by Googling that helps Bouncing the HPUX server didn't help. Even with a totally clean re-install of Oracle9iR2 I get the same error now. Here's what happens:
SQL> startup nomount
... long pause
ORA-03113: end-of-file on communication channel SQL> In the Alert log I see:
ORA-07445: exception encountered: core dump [kqlnfy()+1336] [SIGFPE] [HP Conditional trap] [0x40000000019AA0F3] [] []
Jack C. Applewhite - Database Administrator Austin I.S.D. - MIS Department
512.414.9250 (wk)
Date: Mon, 16 Jul 2012 11:46:49 -0500
Message-ID: <OFBB1F38DF.FE3FD5F8-ON86257A3D.005A85DF-86257A3D.005C2671_at_austinisd.org>
HPUX PA-RISC 11.11 64bit with Oracle9iR2 (9.2.0.8) 64bit. While bouncing an existing DB the other day, I got the ORA-07445 error and then the DB wouldn't even startup nomount. The MOS 7445 lookup tool doesn't recognize the first argument and I can't find anything on MOS or by Googling that helps Bouncing the HPUX server didn't help. Even with a totally clean re-install of Oracle9iR2 I get the same error now. Here's what happens:
SQL> startup nomount
... long pause
ORA-03113: end-of-file on communication channel SQL> In the Alert log I see:
ORA-07445: exception encountered: core dump [kqlnfy()+1336] [SIGFPE] [HP Conditional trap] [0x40000000019AA0F3] [] []
In the udump trace file referenced in the Alert Log: Exception signal: 8 (SIGFPE), code: 13 (HP Conditional trap), addr: 0x40000000019aa0f3, PC: [0x40000000019aa0f0, kqlnfy()+1336]
Of course Oracle9i is de-supported, but, due to the App Vendor and our own HR/Finance group's foot-dragging, we have to be on 9i, so upgrading to 10g isn't an option. This database and others like it on 9.2.0.8 have been functioning just fine for years on several different HPUX servers. They get bounced all the time and I've never seen anything like this. We also have 10gR2 and 11gR2 DBs on Linux.
Anybody got any clues about this?
Thanks.
Jack C. Applewhite - Database Administrator Austin I.S.D. - MIS Department
512.414.9250 (wk)
--
http://www.freelists.org/webpage/oracle-l
Received on Mon Jul 16 2012 - 11:46:49 CDT