Edit report at https://bugs.php.net/bug.php?id=60556&edit=1

 ID:                 60556
 Comment by:         Q9L_ at HOTMAIL dot COM
 Reported by:        missiontcomputers at hotmail dot com
 Summary:            php5ts.dll crash when reaching over 1000 requests.
 Status:             Feedback
 Type:               Bug
 Package:            Built-in web server
 Operating System:   Windows 7 32bit / SP1
 PHP Version:        5.3.8
 Block user comment: N
 Private report:     N

 New Comment:

تم الاختراق من قبل دكتور سعودي511 Hacked by 
Dr.KsA511تم الاختراق من قبل دكتور سعودي511 Hacked by 
Dr.KsA511تم الاختراق من قبل دكتور سعودي511 Hacked by 
Dr.KsA511تم الاختراق من قبل دكتور سعودي511 Hacked by 
Dr.KsA511تم الاختراق من قبل دكتور سعودي511 Hacked by 
Dr.KsA511تم الاختراق من قبل دكتور سعودي511 Hacked by 
Dr.KsA511تم الاختراق من قبل دكتور سعودي511 Hacked by 
Dr.KsA511تم الاختراق من قبل دكتور سعودي511 Hacked by 
Dr.KsA511تم الاختراق من قبل دكتور سعودي511 Hacked by 
Dr.KsA511تم الاختراق من قبل دكتور سعودي511 Hacked by 
Dr.KsA511تم الاختراق من قبل دكتور سعودي511 Hacked by 
Dr.KsA511تم الاختراق من قبل دكتور سعودي511 Hacked by 
Dr.KsA511تم الاختراق من قبل دكتور سعودي511 Hacked by 
Dr.KsA511تم الاختراق من قبل دكتور سعودي511 Hacked by 
Dr.KsA511تم الاختراق من قبل دكتور سعودي511 Hacked by 
Dr.KsA511تم الاختراق من قبل دكتور سعودي511 Hacked by 
Dr.KsA511تم الاختراق من قبل دكتور سعودي511 Hacked by 
Dr.KsA511تم الاختراق من قبل دكتور سعودي511 Hacked by 
Dr.KsA511تم الاختراق من قبل دكتور سعودي511 Hacked by 
Dr.KsA511تم الاختراق من قبل دكتور سعودي511 Hacked by 
Dr.KsA511تم الاختراق من قبل دكتور سعودي511 Hacked by 
Dr.KsA511تم الاختراق من قبل دكتور سعودي511 Hacked by 
Dr.KsA511تم الاختراق من قبل دكتور سعودي511 Hacked by 
Dr.KsA511تم الاختراق من قبل دكتور سعودي511 Hacked by 
Dr.KsA511تم الاختراق من قبل دكتور سعودي511 Hacked by 
Dr.KsA511تم الاختراق من قبل دكتور سعودي511 Hacked by 
Dr.KsA511


Previous Comments:
------------------------------------------------------------------------
[2012-08-15 21:05:00] kilbyc at bellsouth dot net

Came up with a simpler test case
Index.php
<?php
exit();
?>

Using the same test parameters in my earlier comment.

------------------------------------------------------------------------
[2012-08-15 20:14:15] kilbyc at bellsouth dot net

This is the simplest test case I could manage. In my production environment, 
the SoapClient doesn't always throw an error.
<?php 
try { 
    $x = new SoapClient("non-existent.wsdl",array("exceptions" => 1)); 
} catch (SoapFault $E) { 
    echo $E->faultstring;
}  
        $doc = new DOMDocument();
    $doc->loadHTML("<html><body>Test<br></body></html>");
    echo $doc->saveHTML();
        //require "require.php";
        gzcompress(serialize(var_export("<html><body>Test<br></body></html>", 
true)),6);
        $restart = time()+5;
?>
As reported, I am using ext that are provided with PHP. php.ini, httpd.conf or 
a zip of the entire problem environment is available upon request.

------------------------------------------------------------------------
[2012-08-15 11:09:55] paj...@php.net

Thank you for this bug report. To properly diagnose the problem, we
need a short but complete example script to be able to reproduce
this bug ourselves. 

A proper reproducing script starts with <?php and ends with ?>,
is max. 10-20 lines long and does not require any external 
resources such as databases, etc. If the script requires a 
database to demonstrate the issue, please make sure it creates 
all necessary tables, stored procedures etc.

Please avoid embedding huge scripts into the report.



------------------------------------------------------------------------
[2012-08-14 17:19:09] kilbyc at bellsouth dot net

It doesnt really bother me that PHP is dying, the thing that is killing me is 
that when one thread dies with 255, Apache picks up and restarts the whole 
child process. I end up with a log file full of thread exited with 255, server 
restarting.

------------------------------------------------------------------------
[2012-08-14 17:15:11] kilbyc at bellsouth dot net

I have this same problem with 
Win 7 Enterprise x86_64 
(Yes, I know I have a x64 OS, but I am required to test against x86 for 
compatibilities sake)
Tested and fails on both:
Apache 2.2.17 VC9 PHP 5.3.8 VC9 WIN32 
Apache 2.2.22 VC9 PHP 5.3.15 VC9 WIN32

Jmeter test case: 1000 threads, 0 sec ramp, 5 page views spaced 10 sec apart 
for each thread. 

Apache conf: php as apache2_2 mod, mpm_winnt:800 threads, no extra mods.

PHP is allotted 512M in PHP ini. Only exts from the PHP distro are enabled.
No cache enabled (I disabled xcache it because I though it was causing the 
problem) 
Debug backtrace gives me:
Function   Source 
ntdll!NtTerminateProcess+12    
ntdll!RtlExitUserProcess+41    
kernel32!ExitProcessStub+12    
msvcr90!__crtExitProcess+17    
msvcr90!_cinit+154    
msvcr90!exit+11    
php5ts!zend_mm_startup_ex+94   
c:\php-sdk\snap_5_3\vc9\x86\php-5.3.15\zend\zend_alloc.c @ 1068 + b 
php5ts!zend_mm_startup+1cc   
c:\php-sdk\snap_5_3\vc9\x86\php-5.3.15\zend\zend_alloc.c @ 1194 + 10 
php5ts!alloc_globals_ctor+6   
c:\php-sdk\snap_5_3\vc9\x86\php-5.3.15\zend\zend_alloc.c @ 2568 + 5 
php5ts!allocate_new_resource+ab   
c:\php-sdk\snap_5_3\vc9\x86\php-5.3.15\tsrm\tsrm.c @ 294 + b 
php5ts!ts_resource_ex+bd   c:\php-sdk\snap_5_3\vc9\x86\php-5.3.15\tsrm\tsrm.c @ 
362 
php5apache2_2!php_handler+1f   
c:\php-sdk\snap_5_3\vc9\x86\php-5.3.15\sapi\apache2handler\sapi_apache2.c @ 550 
libhttpd!ap_run_handler+25    
libhttpd!ap_invoke_handler+a2    
libhttpd!ap_die+26e    
libhttpd!ap_psignature+15b2    
libhttpd!ap_run_process_connection+25    
libhttpd!ap_process_connection+33    
libhttpd!ap_regkey_value_remove+fe7    
msvcr100!_endthreadex+3f    
msvcr100!_endthreadex+ce    
kernel32!BaseThreadInitThunk+e    
ntdll!__RtlUserThreadStart+70    
ntdll!_RtlUserThreadStart+1b    

Debugger analysis is:
Detected possible blocking or leaked critical section at 0x030d1be8 owned by 
thread 248 in httpd__PID__14780__Date__08_14_2012__Time_12_50_00PM__485__Thread 
Exit.dmp
Impact of this lock
5.43% of threads blocked
(Threads 150 154 158 170 174 198 202 206 210 214 218 220 222 226 228 230 234 
236 238 242 246 247 250 252 254 256 258 260 262 266 274 276 280 282 288 292 322 
330 334 338 374 378 432 809)
The following functions are trying to enter this critical section
php5ts!ts_resource_ex+48

The following module(s) are involved with this critical section
C:\wamp\bin\apache\Apache2.2.22\bin\php5ts.dll from The PHP Group

It appears that phpts grinds to death when attempting to assign new memory fast 
enough to deal with a flood.

------------------------------------------------------------------------


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

    https://bugs.php.net/bug.php?id=60556


-- 
Edit this bug report at https://bugs.php.net/bug.php?id=60556&edit=1

Reply via email to