Bug #71135 [Com]: Random memory corruption with strings

From: Date: Fri, 05 Feb 2016 13:47:39 +0000
Subject: Bug #71135 [Com]: Random memory corruption with strings
References: 1  Groups: php.bugs 
Request: Send a blank email to php-bugs+get-199067@lists.php.net to get a copy of this message
Edit report at https://bugs.php.net/bug.php?id=71135&edit=1

 ID:                 71135
 Comment by:         vojtech dot kurka at gmail dot com
 Reported by:        iquito at gmx dot net
 Summary:            Random memory corruption with strings
 Status:             Assigned
 Type:               Bug
 Package:            opcache
 Operating System:   Debian Jessie
 PHP Version:        7.0.0
 Assigned To:        laruence
 Block user comment: N
 Private report:     N

 New Comment:

I think we're running into something related: We have a static testcase, that calls an API,
that takes a result from a database and uses json_encode() to turn it into JSON and send it back.
The testcase is called every 10 minutes in production to check that API is working. Randomly, that
tescase starts failing on an error (json_encode reports recursion detected). The dataset is still
the same. When we restart php70-php-fpm, everything is working again.

Update to PHP 7.0.3 didn't resolve this issue.

Our environment:
Centos 6, 2.6.32-573.12.1.el6.x86_64
PHP 7.0.3 (remi) with opcache
php70-php-fpm

Opcache config contains only these settings:

zend_extension=opcache.so
opcache.enable=1
opcache.memory_consumption=128
opcache.interned_strings_buffer=8
opcache.max_accelerated_files=4000
opcache.blacklist_filename=/etc/opt/remi/php70/php.d/opcache*.blacklist


I think some fuzzy-testing for this issue would be useful. We are still trying to find a
reproducible testcase, but without any success.

We are running PHP7 and PHP5.5.32 in parallel in production and PHP5 is completely OK, no such
problems observed.


Previous Comments:
------------------------------------------------------------------------
[2016-02-02 21:52:11] iquito at gmx dot net

fast_shutdown was always set to 0 in my configuration.

------------------------------------------------------------------------
[2016-02-02 19:34:06] sroussey at gmail dot com

Try with fast_shutdown set to 0. This cleared up issues we had with string memory corruption.

------------------------------------------------------------------------
[2016-01-18 08:49:22] w1zzard at techpowerup dot com

I can confirm this issue.

Seeing it on PHP 7.0.2 from remi. Just like OP it happens roughly once a day. 

Running inside Docker container.

------------------------------------------------------------------------
[2015-12-27 23:12:18] iquito at gmx dot net

I cannot edit the bug report / change the status, the error message

"You aren't allowed to change a bug to that state."

always appears, although I provide the correct password, and therefore I cannot submit anything with
the "Edit" method or provide "official" feedback. I therefore provided feedback
via Add Comment, which does not seem to count.

------------------------------------------------------------------------
[2015-12-27 04:22:29] php-bugs at lists dot php dot net

No feedback was provided. The bug is being suspended because
we assume that you are no longer experiencing the problem.
If this is not the case and you are able to provide the
information that was requested earlier, please do so and
change the status of the bug back to "Re-Opened". Thank you.

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


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=71135


--
Edit this bug report at https://bugs.php.net/bug.php?id=71135&edit=1


Thread (52 messages)

« previous php.bugs (#199067) next »