ID: 26393
User updated by: troy dot tinnes at motorola dot com
Reported By: troy dot tinnes at motorola dot com
-Status: Feedback
+Status: Open
Bug Type: OCI8 related
Operating System: Solaris 2.8
PHP Version: 4.3.4
New Comment:
Ok tried the patch. We went back to 4.3.3 to apply the patch - since
that was the version bug #24531 fixed. Came up with a Seg Fault again -
here's the backtrace:
solaris8:/opt/web/apache_php/apache2/logs# gdb
/opt/web/apache_php/apache2/bin/httpd
GNU gdb 5.3
Copyright 2002 Free Software Foundation, Inc.
GDB is free software, covered by the GNU General Public License, and
you are welcome to change it and/or distribute copies of it under
certain conditions. Type "show copying" to see the conditions. There is
absolutely no warranty for GDB. Type "show warranty" for details. This
GDB was configured as "sparc-sun-solaris2.8"...
(gdb) run -X
Starting program: /opt/web/apache_php/apache2/bin/httpd -X
[New LWP 1]
[New LWP 2]
[New LWP 3]
[New LWP 4]
[New LWP 5]
Program received signal SIGSEGV, Segmentation fault.
0xff0331f0 in strlen () from /usr/lib/libc.so.1
(gdb) bt
#0 0xff0331f0 in strlen () from /usr/lib/libc.so.1
#1 0xfed8b538 in _oci_close_session (session=0x0)
at /opt/web/apache_php/source/php-4.3.3/ext/oci8/oci8.c:2443
#2 0xfee52358 in list_entry_destructor (ptr=0x3ea3d0)
at /opt/web/apache_php/source/php-4.3.3/Zend/zend_list.c:183
#3 0xfee50ce8 in zend_hash_apply_deleter (ht=0xfeec29f4, p=0x3e1178)
at /opt/web/apache_php/source/php-4.3.3/Zend/zend_hash.c:608
#4 0xfee50d80 in zend_hash_graceful_reverse_destroy (ht=0xfeec29f4)
at /opt/web/apache_php/source/php-4.3.3/Zend/zend_hash.c:674
#5 0xfee42684 in shutdown_executor ()
at
/opt/web/apache_php/source/php-4.3.3/Zend/zend_execute_API.c:213
#6 0xfee4ba28 in zend_deactivate ()
at /opt/web/apache_php/source/php-4.3.3/Zend/zend.c:665
#7 0xfee1d520 in php_request_shutdown (dummy=0x0)
at /opt/web/apache_php/source/php-4.3.3/main/main.c:998
#8 0xfee5e5e4 in php_handler (r=0x17b790)
at
/opt/web/apache_php/source/php-4.3.3/sapi/apache2handler/sapi_apache2.c:550
#9 0x0004a37c in ap_run_handler (r=0x17b790) at config.c:195 #10
0x0004a950 in ap_invoke_handler (r=0x17b790) at config.c:401 #11
0x00037d60 in ap_process_request (r=0x17b790) at http_request.c:288 #12
0x00033274 in ap_process_http_connection (c=0x1e0a20) at
http_core.c:293 #13 0x00055740 in ap_run_process_connection
(c=0x1e0a20) at connection.c:85 #14 0x00048b78 in child_main
(child_num_arg=1960368) at prefork.c:694 #15 0x00048d70 in make_child
(s=0x9bb80, slot=0) at prefork.c:734 #16 0x00048dfc in startup_children
(number_to_start=4) at prefork.c:806 #17 0x0004962c in ap_mpm_run
(_pconf=0x1, plog=0xd1558, s=0x9bb80) ---Type <return> to continue, or
q <return> to quit---
at prefork.c:1022
#18 0x0004f23c in main (argc=2, argv=0xffbefd1c) at main.c:660
Previous Comments:
------------------------------------------------------------------------
[2003-11-25 15:19:50] [EMAIL PROTECTED]
Did you try the patch or not?
------------------------------------------------------------------------
[2003-11-25 14:59:08] troy dot tinnes at motorola dot com
I assume the resolution in this case is to switch to persistent
connections? Here's why I'm hesitant to use persistent connections with
my application:
1) the application is not transaction based, its simple inquiry/lookup.
A person would do potentially 2-3 quick look-ups per day. The
persistent connection would potentially cost more than the
connect/tear-down cost for 1 user to access the system for a few
minutes.
2) depending on the request, I sometimes have data from 2 or 3
different database instances represented on one web page output. ie. 1
database has a ISO country code HK entered to indicate the currency
code - the master currencies are located in a data warehouse
elsewhere. Establishing persistent connections for each user for all 3
databases may be a bit excessive?
------------------------------------------------------------------------
[2003-11-25 14:18:22] [EMAIL PROTECTED]
Oops..I meant the Open reports.. :)
See bug #24531 which has some patch in it.
Try the patch..
------------------------------------------------------------------------
[2003-11-25 09:02:28] troy dot tinnes at motorola dot com
I went through all 1300 of them. The few that looked similar ended up
with no answer. ie. You asked them to do a backtrace and there was no
response.
------------------------------------------------------------------------
[2003-11-25 03:09:32] [EMAIL PROTECTED]
Please check the other OCI8 related bug reports, I'm not sure but most
likely one or two of them are reporting this exact same problem.
------------------------------------------------------------------------
The remainder of the comments for this report are too long. To view
the rest of the comments, please view the bug report online at
http://bugs.php.net/26393
--
Edit this bug report at http://bugs.php.net/?id=26393&edit=1
