James in php.windows (Sat, 14 Apr 2012 10:29:17 -0700):
>I need one more extension for PHP 5.4, Imagick the non-thread safe
>version for IIS. I tried the one from PHP 5.3, but it doesn't show up
>under phpinfo();
php_imagemagick does not compile anymore under PHP 5.4, so there is
little chance
"Keith Davis" in php.windows (Sat, 5 May 2012 09:41:39 -0500):
>Strange, this guy says he has it working (but apparently with some
>relative directory problem):
>
>http://stackoverflow.com/questions/9954764/imagick-php-5-4-extension-does-not-work-with-relative-paths-windows
Apparently, he has got
"Keith Davis" in php.windows (Sat, 5 May 2012 09:41:39 -0500):
>Strange, this guy says he has it working (but apparently with some
>relative directory problem):
>
>http://stackoverflow.com/questions/9954764/imagick-php-5-4-extension-doe
>s-not-work-with-relative-paths-windows
>
>We will need it to
"Keith Davis" in php.windows (Sat, 5 May 2012 09:41:39 -0500):
>Strange, this guy says he has it working (but apparently with some
>relative directory problem):
>
>http://stackoverflow.com/questions/9954764/imagick-php-5-4-extension-does-not-work-with-relative-paths-windows
>
>We will need it to (w
"Keith Davis" in php.windows (Mon, 7 May 2012 11:09:35 -0500):
>I tried to compile Wincache under 5.4.0 and got all kinds of errors. I'm
>not very experienced at compiling PHP extensions, I admit, but I did get
>it to compile under 5.3.x.
I used the 1.3.0 beta sources:
http://pecl.php.net/package/
"Keith Davis" in php.windows (Mon, 7 May 2012 18:39:09 -0500):
>Wait. Did you compile 1.3.0? We were never told that this was compiled
>and we're actively working with MS to get this done.
>
>http://forums.iis.net/t/1187707.aspx
As you can see at http://x32.elijst.nl/phpinfo.php5f I compiled the
W
directly instead of using random binaries?
>
>On Tue, May 8, 2012 at 8:59 AM, Jan Ehrhardt wrote:
>> "Keith Davis" in php.windows (Mon, 7 May 2012 18:39:09 -0500):
>>>Wait. Did you compile 1.3.0? We were never told that this was compiled
>>>and we're a
Pierre Joye in php.windows (Tue, 8 May 2012 09:28:37 +0200):
>why don't you ask the authors directly instead of using random binaries?
It seems you are (maybe without knowing it) one of those authors. The
1.3.0RC1 of the PECL wincache sources contained the line:
ADD_EXTENSION_DEP('memcache', 'ses
Ferenc Kovacs in php.windows (Tue, 8 May 2012 18:36:08 +0200):
>> ADD_EXTENSION_DEP('memcache', 'session');
>that was introduced in
>http://svn.php.net/viewvc/pecl/wincache/trunk/config.w32?r1=324182&r2=324181&pathrev=324182
OK. I can see you are quite more experienced with svn than I am ;-)
Jan
Momchil Bozhinov in php.windows (Wed, 3 Apr 2013 16:13:48 +0300):
>I have this huge problem, hope you can help me with.
>
>I compile my own PHP with an additional PECL
>(php_win32service)
>
>Unfortunately, the ext was not working properly. The windows
>service wouldnt start.
>
>So I downloaded the
Tomasz Krawczyk in php.windows (Sun, 30 Jun 2013 15:46:23 +0200):
>I don't have VC 2012 yet.
>I see there are deps available to download for VC9. I know support for
>Windows XP and 2003 was dropped since PHP 5.5. So I can't run PHP 5.5 on
>XP/2003. But I don't know if machine used for building ca
Jan Ehrhardt in php.windows (Mon, 01 Jul 2013 03:13:26 +0200):
>Tomasz Krawczyk in php.windows (Sun, 30 Jun 2013 15:46:23 +0200):
>>I don't have VC 2012 yet.
>>I see there are deps available to download for VC9. I know support for
>>Windows XP and 2003 was dropped sin
I am having troubles configuring multiple PHP-versions, all running
under Apache 2.4.4 with their own OPcache. My feeling at the moment is
that under Windows you can only have one PHP instance with OPcache, but
I would love to be contradicted.
See this issue at github for details:
https://github.c
"Anatol Belski" in php.windows (Mon, 8 Jul 2013 18:54:30 +0200):
>You can try to run from another user account, then the mmap filename will
>change.
When I saw the name ZendOPcache.MemoryBase@Administrator I already was
thinking about such a solution. That must be possible with mote than one
insta
"Anatol Belski" in php.windows (Mon, 8 Jul 2013 18:54:30 +0200):
>You can try to run from another user account, then the mmap filename will
>change.
Just curious: how is this done on Linux? As far as I know both PHP's on
my Centos boxes run as the same user (the one as mod_php, the other as
mod_fc
Richard Quadling in php.windows (Tue, 9 Jul 2013 06:44:58 +0100):
>On 9 July 2013 06:21, Jan Ehrhardt wrote:
>
>> Just curious: how is this done on Linux? As far as I know both PHP's on
>> my Centos boxes run as the same user (the one as mod_php, the other as
>&g
Hi Pierre,
Pierre Joye in php.windows (Tue, 9 Jul 2013 07:46:48 +0200):
>impersonation should solve this problem. I've to check if another
>function(s) has to be used to get the impersonated user.
Please check it out.
I suppose I will not be the only one interested in running multiple
PHP's with
Jan Ehrhardt in php.windows (Tue, 09 Jul 2013 07:16:49 +0200):
>opcache.mmap_base
>
>Could this be a way out? It looks like this setting overrides the
>setting in ZendOPcache.MemoryBase. What I want is that the OPcache of
>PHP 5.5 does not share the memory with PHP 5.3. Can tha
Jan Ehrhardt in php.windows (Sun, 07 Jul 2013 22:41:12 +0200):
>I am having troubles configuring multiple PHP-versions, all running
>under Apache 2.4.4 with their own OPcache. My feeling at the moment is
>that under Windows you can only have one PHP instance with OPcache, but
>I woul
Hi Pierre,
Pierre Joye in php.windows (Tue, 9 Jul 2013 19:06:04 +0200):
>On Tue, Jul 9, 2013 at 4:23 PM, Jan Ehrhardt wrote:
>> Hopping back and forth between this list and github. I seem to have a
>> really ugly workaround now. I recompiled the x64 php_opcache.dll after
Hi Pierre,
Pierre Joye in php.windows (Tue, 9 Jul 2013 22:39:17 +0200):
>On Jul 9, 2013 7:54 PM, "Jan Ehrhardt" wrote:
>> For what purpose should I use two php.ini's?
>
>For having two different settings for x86 and x64. Which is what your patch
>does. You al
Jan Ehrhardt in php.windows (Tue, 09 Jul 2013 23:11:36 +0200):
>Pierre Joye in php.windows (Tue, 9 Jul 2013 22:39:17 +0200):
>>I do not understand why you build everything from scratch. Providing the
>>missing ext should be enough. Also we work with everyone to be sure that we
&g
Hi Pierre,
Getting back on topic.
Jan Ehrhardt in php.windows (Tue, 09 Jul 2013 23:11:36 +0200):
>Steffen from Apachelounge said something interesting:
>"Opcache is creating in his temp dir "
>http://www.apachelounge.com/viewtopic.php?t=5461
I followed Steffen's ide
Hi Keith,
Long tme, no hear.
Keith Davis in php.windows (Sat, 20 Jul 2013 07:14:39 -0500):
>We are trying to move to 5.5, but are stopped by this bug which we reported:
>
>https://bugs.php.net/bug.php?id=65247
>
>Description:
>
>If Opcache is enabled for CLI, php.exe crashes on shutdo
Keith Davis in php.windows (Sat, 20 Jul 2013 07:14:39 -0500):
>We are trying to move to 5.5, but are stopped by this bug which we reported:
BTW: my NTS builds of 5.5 also contain a php_wincache.dll, compiled from
the most recent sources.
Jan
--
PHP Windows Mailing List (http://www.php.net/)
To
Keith Davis in php.windows (Sat, 20 Jul 2013 07:14:39 -0500):
>We are trying to move to 5.5, but are stopped by this bug which we reported:
>
>https://bugs.php.net/bug.php?id=65247
A tip: there is yet another way to get backtraces. I once used it to
track segfaults in httpd.exe, but there is a cha
Keith Davis in php.windows (Mon, 22 Jul 2013 08:04:44 -0500):
>Seems to be the same problem with this method, no debugging symbols found.
>I've added the debug pack to my c:\PHP directory (with the extension
>.pdb files in \ext.)
I had the same problem. I am not sure if the GNU debugger looks at
Keith Davis in php.windows (Mon, 22 Jul 2013 08:04:44 -0500):
>Program received signal SIGTRAP, Trace/breakpoint trap.
>0x7756e696 in ntdll!KiRaiseUserExceptionDispatcher ()
> from C:\Windows\system32\ntdll.dll
Your bug report was talking about
C:\Windows\SysWOW64\ntdll.dll
https://bugs.php.net/
Keith Davis in php.windows (Mon, 22 Jul 2013 10:36:32 -0500):
>"Maybe the developers of OPcache can do something with this information,
>although it is deep inside Windows. Is it a Windows Server? Which
>edition? 2008 R2 or higher?"
>
>How do I contact them?
https://github.com/zendtech/ZendOptimiz
Keith Davis in php.windows (Mon, 22 Jul 2013 12:49:59 -0500):
>A, geez. It looks like it's caused by Wincache. I disable and enable
>that extension and the crashing stops and starts.
>
>However, again, I've tested Wincache thoroughly with Opcache in IIS and
>have no issues. It's only on the com
Keith Davis in php.windows (Mon, 22 Jul 2013 12:49:59 -0500):
>A, geez. It looks like it's caused by Wincache. I disable and
>enable that extension and the crashing stops and starts.
>
>However, again, I've tested Wincache thoroughly with Opcache in
>IIS and have no issues. It's only on the com
Keith Davis in php.windows (Tue, 23 Jul 2013 06:43:06 -0500):
>It's only with:
>wincache.enablecli=1 & opcache.enable_cli=1
OK. And those two do not go along when php is running from th
ecommandline. They produce errors while closing the script.
There is a nasty workaround: leave PHP before the e
Eric Stenson in php.windows (Thu, 25 Jul 2013 18:35:25 +):
>I've filed bug https://bugs.php.net/bug.php?id=65338 to track this
>issue. The root cause of the AV is when more than one module hooks the
>interned strings, a "wrong" value will be left in
>CG(interned_strings_start), which the Zend
Keith Davis in php.windows (Fri, 26 Jul 2013 13:52:53 -0500):
>Jan, this version does not work (crashed.)
Crashed at startup or on closing? What is the date of php_wincache.dll?
Should be something like 2013-07-26 17:49. Filesize: 131584.
Maybe dropbox did not synchronize yet...
>I'm showing Fil
Eric Stenson in php.windows (Thu, 25 Jul 2013 18:35:25 +):
>I've filed bug https://bugs.php.net/bug.php?id=65338 to track this
>issue. The root cause of the AV is when more than one module hooks the
>interned strings, a "wrong" value will be left in
>CG(interned_strings_start), which the Zend
Keith Davis in php.windows (Fri, 26 Jul 2013 16:24:26 -0500):
>Ok, I copied both php5.dll and php_wincache.dll from the ones you
>compiled, Jan, with Eric's patch and it works!
New compilation without Eric's patch, but with Dmitry's patch of today:
https://dl.dropboxusercontent.com/u/8954372/php-5
Jan Ehrhardt in php.windows (Tue, 30 Jul 2013 18:15:50 +0200):
>Keith Davis in php.windows (Fri, 26 Jul 2013 16:24:26 -0500):
>>Ok, I copied both php5.dll and php_wincache.dll from the ones you
>>compiled, Jan, with Eric's patch and it works!
>
>New compilation with
Keith Davis in php.windows (Tue, 30 Jul 2013 15:57:05 -0500):
>But I thought from the previous fix, what was updated was php5.dll. So,
>to be clear, I want to use a base (revert my php5.dll file to base)
>5.5.1 build with the new php_wincache.dll and php_opcache.dll files?
Yes. Dmitry took another
Eric Stenson in php.windows (Thu, 25 Jul 2013 18:35:25 +):
>I've filed bug https://bugs.php.net/bug.php?id=65338 to track this
>issue. The root cause of the AV is when more than one module hooks the
>interned strings, a "wrong" value will be left in
>CG(interned_strings_start), which the Zend
Tomasz Krawczyk in php.windows (Wed, 21 Aug 2013 08:55:01 +0200):
>Another question. Are you planing to make x64 version? PHP 5.5 is ready
>for x64 and a lot of people is interested in.
Just try it:
http://www.apachelounge.com/viewtopic.php?t=5520
The NTS x64 version contains php_wincache.dll. I
Pierre Joye in php.windows (Fri, 23 Aug 2013 09:01:55 +0200):
>In the long run yes, or maybe. Right now the last release is totally
>unusable. For various reasons, some fixable and other will be hardly
>fixable without changes in the engine.
I know there are some issues:
https://github.com/zendtec
Keith Davis in php.windows (Tue, 27 Aug 2013 16:00:21 -0500):
>I have fixed this 2 different ways. The first is to run
>opcache_reset() from localhost
Is not this related to issue
https://github.com/zendtech/ZendOptimizerPlus/issues/118
It might not be Windows-specific at all.
Jan
--
PHP Windo
"Anatol Belski" in php.windows (Thu, 26 Dec 2013 19:37:21 +0100):
>There was a number of commits to opcache last week, I also merged them
>into str_size_and_int64 branch. So the code base should be comparable. I
>think the best way to handle it yet is using bugs.php.net, especially as
>you mean the
Jan Ehrhardt in php.windows (Thu, 26 Dec 2013 21:24:39 +0100):
>Fatal error: Undefined class constant 'MYSQL_ATTR_USE_BUFFERED_QUERY' in
>includes\database\mysql\database.inc on line 46
False alarm, this had something to do with the extension_dir settings. I
have php 5.7 x86 T
"Anatol Belski" in php.windows (Thu, 26 Dec 2013 22:35:06 +0100):
>On Thu, December 26, 2013 20:47, Jan Ehrhardt wrote:
>> I will try to get a BT with GDB. Probably that wil mean I have to load
>> PHP 5.7 as mod_php in a VC11 Apache, because GDB does not catch errors
>&
Jan Ehrhardt in php.windows (Thu, 26 Dec 2013 20:47:08 +0100):
>I will try to get a BT with GDB. Probably that wil mean I have to load
>PHP 5.7 as mod_php in a VC11 Apache, because GDB does not catch errors
>of a php-cgi that is loaded from within Apache.
>
>I will install Apache 3.
Jan Ehrhardt in php.windows (Thu, 26 Dec 2013 22:25:25 +0100):
>Program received signal SIGSEGV, Segmentation fault.
>[Switching to Thread 872.0xb44]
>0x6fdfdf9f in zend_mm_set_custom_handlers ()
> from D:\phpdev\php57.x86\php5ts.dll
>(gdb) bt
>#0 0x6fdfdf9f in zend_mm_s
"Anatol Belski" in php.windows (Thu, 26 Dec 2013 23:10:41 +0100):
>Yeah, there is a page
>https://bugs.php.net/bugs-generating-backtrace-win32.php , the debug tool
>is in current version 2.0 now, but the screenshots are compatible.
I am now getting this result with a x64 debug build:
WARNING - De
Jan Ehrhardt in php.windows (Thu, 02 Jan 2014 17:16:33 +0100):
>This made it easy to switch the second PHP version and test when the
>500-error occurs. A bit to my surprise it only happens for the NTS
>versions of PHP 5.4 and PHP 5.5. The NTS version of PHP 5.3 runs without
>probl
Jan Ehrhardt in php.windows (Sun, 05 Jan 2014 17:25:31 +0100):
>OK, got a ZwTerminateProcess.dmp by tracking the php-cgi.exe PID on
>memory leaks and the like. No debug symbols yet, so I will run another
>analysis.
I cannot get DebugDiag to find the correct symbols. And I do not know
Jan Ehrhardt in php.windows (Thu, 26 Dec 2013 00:58:34 +0100):
>Ontopic again: the bad news is that I am now getting segfaults with the
>experimental builds in Drupal7. I do not know what caused the change
>(might be the upgrade of some Drupal modules). Anyway: opcache in the
>experim
Jan Ehrhardt in php.windows (Sun, 12 Jan 2014 23:09:13 +0100):
>Dmitry Stogov in php.windows (Mon, 30 Dec 2013 11:23:19 +0400):
>>I would very appreciate, if you provide more or less compact reproducable
>>case.
>
>OK, I tracked it down to the module that causes the inter
Dmitry Stogov in php.windows (Mon, 30 Dec 2013 11:23:19 +0400):
>I would very appreciate, if you provide more or less compact reproducable
>case.
OK, I tracked it down to the module that causes the internal server
error 500: Superfish.
https://drupal.org/project/superfish
Take any Drupal7 site, u
53 matches
Mail list logo
