Bug has resurfaced. Tested with php5-200509271430. Reproduce 
code is the same as before.

[Tue Sep 27 15:15:31 2005] [notice] child pid 15934 exit 
signal Segmentation fault (11)

Here is the backtrace:

(gdb) run -X
Starting program: /usr/local/apache2/bin/httpd -X
Reading symbols for shared libraries .+++ done
Reading symbols for shared libraries .... done

Program received signal EXC_BAD_ACCESS, Could not access 
memory.
Reason: KERN_INVALID_ADDRESS at address: 0x7269748d
0x020a1504 in dispatch_param_event (stmt=0x49ac08, 
event_type=PDO_PARAM_EVT_FETCH_POST) at /Users/stewey/
Desktop/php5-200509271430/ext/pdo/pdo_stmt.c:133
133             if (!stmt->methods->param_hook) {
(gdb) bt
#0  0x020a1504 in dispatch_param_event (stmt=0x49ac08, 
event_type=PDO_PARAM_EVT_FETCH_POST) at /Users/stewey/
Desktop/php5-200509271430/ext/pdo/pdo_stmt.c:133
#1  0x020a26b4 in do_fetch_common (stmt=0x49ac08, 
ori=PDO_FETCH_ORI_NEXT, offset=0, do_bind=1) at /Users/
stewey/Desktop/php5-200509271430/ext/pdo/pdo_stmt.c:542
#2  0x020a36d8 in do_fetch (stmt=0x49ac08, do_bind=1, 
return_value=0x49fb78, how=PDO_FETCH_ASSOC, 
ori=PDO_FETCH_ORI_NEXT, offset=0, return_all=0x0) at /Users/
stewey/Desktop/php5-200509271430/ext/pdo/pdo_stmt.c:754
#3  0x020a56e8 in zif_PDOStatement_fetchAll (ht=1, 
return_value=0x49bba8, return_value_ptr=0x0, 
this_ptr=0x49a7e8, return_value_used=1) at /Users/stewey/
Desktop/php5-200509271430/ext/pdo/pdo_stmt.c:1348
#4  0x0232daac in zend_do_fcall_common_helper_SPEC 
(execute_data=0xbfffe578) at /Users/stewey/Desktop/
php5-200509271430/Zend/zend_vm_execute.h:187
#5  0x0232eca4 in ZEND_DO_FCALL_BY_NAME_SPEC_HANDLER 
(execute_data=0xbfffe578) at /Users/stewey/Desktop/
php5-200509271430/Zend/zend_vm_execute.h:310
#6  0x0232d4e8 in execute (op_array=0x499678) at /Users/
stewey/Desktop/php5-200509271430/Zend/zend_vm_execute.h:88
#7  0x022ed6b0 in zend_execute_scripts (type=8, retval=0x0, 
file_count=3) at /Users/stewey/Desktop/php5-200509271430/
Zend/zend.c:1087
#8  0x02282420 in php_execute_script 
(primary_file=0xbfffef44) at /Users/stewey/Desktop/
php5-200509271430/main/main.c:1677
#9  0x023a2840 in php_handler (r=0x18bde50) at /Users/
stewey/Desktop/php5-200509271430/sapi/apache2handler/
sapi_apache2.c:568
#10 0x0003daf4 in ap_run_handler (r=0x18bde50) at config.c:
152
#11 0x0003e0bc in ap_invoke_handler (r=0x18bde50) at 
config.c:364
#12 0x0001203c in ap_process_request (r=0x18bde50) at 
http_request.c:249
#13 0x0000cc14 in ap_process_http_connection (c=0x18b7f28) 
at http_core.c:251
#14 0x00044898 in ap_run_process_connection (c=0x18b7f28) at 
connection.c:43
#15 0x000298f8 in child_main (child_num_arg=4828168) at 
prefork.c:610
#16 0x000299dc in make_child (s=0x180ad58, slot=0) at 
prefork.c:650
#17 0x00029b14 in startup_children (number_to_start=5) at 
prefork.c:722
#18 0x00029fe0 in ap_mpm_run (_pconf=0x1806418, plog=0x5, 
s=0x180ad58) at prefork.c:941
#19 0x0002baf4 in main (argc=2, argv=0xbffffb20) at main.c:
618

How did you compile/configure PHP?

$ ./configure --with-apxs2=/usr/local/apache2/bin/apxs --
with-mysql=/usr/local/mysql --enable-pdo --prefix=/usr/
local/php5 --with-xmlrpc --with-pdo-mysql --enable-debug

$ make

$ make install

Hope that helps!

Can't reproduce.
Please show your CREATE TABLE command.
Also, are you able to replicate it using PHP CLI ?

It's actually a 'longtext'. I'm very sorry for the 
misinformation. To the obvious question: yes, this makes the 
difference. If it's 'text', no problem, longtext = segfault. 
Again, I'm sorry.

Yes, I can repro via the CLI. Would you like a backtrace of 
that?


CREATE TABLE `article` (
  `articleId` int(10) unsigned NOT NULL auto_increment,
  `parentId` int(10) unsigned NOT NULL default '0',
  `templateId` int(10) unsigned NOT NULL default '0',
  `title` varchar(100) NOT NULL default '',
  `subTitle` varchar(255) default NULL,
  `text` longtext NOT NULL,
  `creationTime` timestamp NULL default NULL,
  `modificationTime` timestamp NULL default 
CURRENT_TIMESTAMP on update CURRENT_TIMESTAMP,
  `state` enum('Online','Offline') NOT NULL default 
'Online',
  `audience` enum('Private','Public') NOT NULL default 
'Private',
  `childSortOrder` enum
('Title','Date','Popularity','Arbitrary') NOT NULL default 
'Title',
  `sortWeight` int(11) NOT NULL default '0',
  PRIMARY KEY  (`articleId`),
  KEY `parentId` (`parentId`,`templateId`),
  FULLTEXT KEY `title` (`title`,`subTitle`,`text`)
) ENGINE=MyISAM DEFAULT CHARSET=latin1 AUTO_INCREMENT=1 ;

Still no luck in reproducing it.
Please do the following in gdb (when you got the backtrace):
gdb> p *stmt
gdb> p *stmt->methods
and paste results here.

Hope this helps...


(gdb) p *stmt
$1 = {
  ce = 0x712f6461, 
  properties = 0x74615f69, 
  in_get = 0, 
  in_set = 1, 
  methods = 0x72697479, 
  driver_data = 0x2f696e64, 
  executed = 0, 
  supports_placeholders = 3, 
  _reserved = 91762280, 
  column_count = 1953328163, 
  columns = 0x646f776e, 
  database_object_handle = {
    value = {
      lval = 1735549284, 
      dval = 2.0473549222889057e+190, 
      str = {
        val = 0x67726164 <Address 0x67726164 out of bounds>, 
        len = 1696742984
      }, 
      ht = 0x67726164, 
      obj = {
        handle = 1735549284, 
        handlers = 0x65223e48
      }
    }, 
    refcount = 1870078052, 
    type = 111 'o', 
    is_ref = 32 ' '
  }, 
  dbh = 0x646f776e, 
  bound_params = 0x67726164, 
  bound_param_map = 0x65206120, 
  bound_columns = 0x63686563, 
  row_count = 1802530665, 
  query_string = 0x6e742073 <Address 0x6e742073 out of 
bounds>, 
  query_stringlen = 1864397928, 
  active_query_string = 0x61742061 <Address 0x61742061 out 
of bounds>, 
  active_query_stringlen = 1847616865, 
  error_code = "rlier ", 
  lazy_object_ref = {
    value = {
      lval = 1920361842, 
      dval = 2.3894187096828918e+243, 
      str = {
        val = 0x72766572 <Address 0x72766572 out of bounds>, 
        len = 537725472
      }, 
      ht = 0x72766572, 
      obj = {
        handle = 1920361842, 
        handlers = 0x200d0a20
      }
    }, 
    refcount = 538976355, 
    type = 97 'a', 
    is_ref = 110 'n'
  }, 
  refcount = 1700881440, 
  default_fetch_type = 1769226044, 
  fetch = {
    column = 794902048, 
    cls = {
      ce = 0x2f613e20, 
      ctor_args = 0xd0a2020, 
      retval_ptr = 0x3c6c693e, 
      fci = {
        size = 1012998248, 
        function_table = 0x7265663d, 
        function_name = 0x22687474, 
        symbol_table = 0x703a2f2f, 
        retval_ptr_ptr = 0x636f6d70, 
        param_count = 1970562418, 
        params = 0x2f646570, 
        object_pp = 0x6f742f69, 
        no_separation = 110 'n'
      }, 
      fcc = {
        initialized = 110 'n', 
        function_handler = 0x74732f66, 
        calling_scope = 0x61712f64, 
        object_pp = 0x6174615f
      }
    }, 
    func = {
      function = 0x2f613e20, 
      fetch_args = 0xd0a2020, 
      object = 0x3c6c693e, 
      fci = {
        size = 1012998248, 
        function_table = 0x7265663d, 
        function_name = 0x22687474, 
        symbol_table = 0x703a2f2f, 
        retval_ptr_ptr = 0x636f6d70, 
        param_count = 1970562418, 
        params = 0x2f646570, 
        object_pp = 0x6f742f69, 
        no_separation = 110 'n'
      }, 
      fcc = {
        initialized = 110 'n', 
        function_handler = 0x74732f66, 
        calling_scope = 0x61712f64, 
        object_pp = 0x6174615f
      }, 
      values = 0x696e7465
    }, 
    into = 0x2f613e20
  }, 
  named_rewrite_template = 0x67726974 <Address 0x67726974 
out of bounds>
}
(gdb) p *stmt->methods
Cannot access memory at address 0x72697479

And what does valgrind say about it?
valgrind --tool=memcheck --leak-check=yes --num-callers=30 php /path/to/your/script.php

No valgrind on my system :(  A brief look around suggests it 
isn't available (won't build?) on OS X.  If you know 
otherwise, or if there's an alternative tool for OS X, please 
let me know.

Are you certain that you are using the latest pdo and pdo_mysql extensions from the CVS?

Nope; I'm using: php5-200509271430. I'll test again with the 
latest snap and report back.

Problem persists with php5-200509290230.

This bug has been fixed in CVS.

Snapshots of the sources are packaged every three hours; this change
will be in the next snapshot. You can grab the snapshot at
http://snaps.php.net/.
 
Thank you for the report, and for helping us make PHP better.