Instance hanging / crashing on Windows with ora-7445 access_violation unable_to_read
- From: "Reardon, Bruce (RTABBAY)" <Bruce.Reardon@xxxxxxxxxxxx>
- To: <oracle-l@xxxxxxxxxxxxx>
- Date: Tue, 9 Nov 2010 15:42:59 +1100
We are using 11.1.0.7 Standard Edition One on Windows 2003 32 bit - with
the 3G switch enabled.
In the last 4 days, one of our sites has had 2 downtimes where Oracle
has stopped accepting new connections and existing ones have stopped
working.
Nothing that we are aware of has changed in the OS, Oracle or
application.
The error in the alert log is like
Exception [type: ACCESS_VIOLATION, UNABLE_TO_READ] [ADDR:0x0]
[PC:0x7C82A774, {empty}]
Errors in file d:\oracle\diag\rdbms\abcd\xyz\trace\xyz_ora_3220.trc
(incident=54056):
ORA-07445: exception encountered: core dump [PC:0x7C82A774]
[ACCESS_VIOLATION] [ADDR:0x0] [PC:0x7C82A774] [UNABLE_TO_READ] []
ORA-03135: connection lost contact
The first argument doesn't contain a function name.
On the first occasion, the database stayed working for 57 minutes and
through 3 scheduled archive log swaps.
Then more errors like following and connections now hung:
Exception [type: ACCESS_VIOLATION, UNABLE_TO_READ] [ADDR:0x0]
[PC:0x7C82C873, {empty}]
Errors in file d:\oracle\diag\rdbms\abcd\xyz\trace\xyz_ora_3480.trc
(incident=54064):
ORA-07445: exception encountered: core dump [PC:0x7C82C873]
[ACCESS_VIOLATION] [ADDR:0x0] [PC:0x7C82C873] [UNABLE_TO_READ] []
Incident details in:
d:\oracle\diag\rdbms\abcd\xyz\incident\incdir_54064\xyz_ora_3480_i54064.
trc
Sat Nov 06 07:57:51 2010
Exception [type: ACCESS_VIOLATION, UNABLE_TO_READ] [ADDR:0x0]
[PC:0x7C82C873, {empty}]
Errors in file d:\oracle\diag\rdbms\abcd\xyz\trace\xyz_j000_1596.trc
(incident=54072):
ORA-07445: exception encountered: core dump [PC:0x7C82C873]
[ACCESS_VIOLATION] [ADDR:0x0] [PC:0x7C82C873] [UNABLE_TO_READ] []
Incident details in:
d:\oracle\diag\rdbms\abcd\xyz\incident\incdir_54072\xyz_j000_1596_i54072
.trc
Sat Nov 06 07:59:57 2010
kkjcre1p: unable to spawn jobq slave process
On the second day this happened (2 days after the first) the instance
hung at the time of the first message in the alert log.
This time the message was very similar:
ORA-07445: exception encountered: core dump [PC:0x7C84D0B3]
[ACCESS_VIOLATION] [ADDR:0x2A] [PC:0x7C84D0B3] [UNABLE_TO_READ] []
The actual dump files contain the message
Symbol file d:\oracle\product\11.1.0\db_1\BIN\oracore11.SYM does not
match binary
But I'm not sure about the relevancy of this.
We've logged a SR, but in the meantime thought I'd ask the list if
others have seen errors like this, or for suggestions on what
information to gather next time?
Thanks for your time,
Bruce Reardon
bruce.reardon@xxxxxxxxxxxx
Other related posts:
- » Instance hanging / crashing on Windows with ora-7445 access_violation unable_to_read - Reardon, Bruce (RTABBAY)
