I just replace serialized PHP with IGBinary .
I follow this steps:
cd /tmp
git clone https://github.com/phadej/igbinary
cd igbinary
phpize
./configure CFLAGS="-O2 -g" –-enable-igbinary –-with-php-config=/usr/bin/php-config
make
make test
make install
But in the step : make test
I get stuck with:
ERROR: Cannot run tests without CLI sapi.
and this is probably is what is causing this other error:
in
http://censocanino.com/log.php
here I use a code to see if the session starts correctly.
and that brought a couple of errors:
Warning: session_start(): Cannot find serialization handler 'igbinary' - session startup failed in /sites/censocanino.com/public/log.php on line 6
and
The current "session.save_path" is "/var/lib/php5".
Session file name: "sess_".
That error prove why phpmyadmin does't work.
http://censocanino.com/phpmyadmin/
What shows here is :
Cannot start session without errors, please check errors given in your PHP and/or webserver log file and configure your PHP installation properly. Also ensure that cookies are enabled in your browser.
all that brought me to the IGBinary Error.
thanks
SOLVED
You know that time when you haver 12 hors investigating why the code is not working and suddly you decide to type everything again .
well that happened to me , it seems was a typo erro.
may be was in --enable-igbinary that the -- were different characters.
but thanks anyway
Related
I am getting a blank screen and nothing in the error logs for a php program I am trying to get up and running (it happens to be yourls). What do I need to do to get error logging enabled or to otherwise get an indicate of why the page is not loading (I know there are multiple postings on this but none of the fixes suggested in this posts seems to generate any output and don't seem to work for this specific implementation)?
I'm running php from this install: php-8.0.0-nts-Win32-vs16-x64.zip
I'm running Apache HTTPD from this install: httpd-2.4.46-o111i-x64-vc15.zip
Turns out this was a behavior of the yourls software I was trying to run. It had a flag that turned error logging/messaging off AND it was not compatable with php 8. I switched to php 7 and switched to the php.ini-development config (by renaming it to php.ini) and I am now seeing error messages (the driver for the database is missing).
I'm following https://xdebug.org/docs/install guideline
When I run sudo make test In my MAC 2 PHP versions are installed PHP5 & PHP7 as project requires.
PHP : /usr/local/php5/bin/php
PHP_SAPI : cli
PHP_VERSION : 7.2.7
It gives below error
=====================================================================
EXPECTED FAILED TEST SUMMARY
---------------------------------------------------------------------
Test for bug #1530: Code coverage incorrect for last code line in a loop [tests/bug01530.phpt] XFAIL REASON: PHP bug #76046: PHP generates "FE_FREE" opcode on the wrong line.
=====================================================================
You may have found a problem in PHP.
This report can be automatically sent to the PHP QA team at
http://qa.php.net/reports and http://news.php.net/php.qa.reports
This gives us a better understanding of PHP's behavior.
If you don't want to send the report immediately you can choose
option "s" to save it. You can then email it to qa-reports#lists.php.net later.
Do you want to send this report now? [Yns]:
I tried 2-3 times but same issue occurs.
the reason for the test failure is obvious:
XFAIL REASON: PHP bug #76046: PHP generates "FE_FREE" opcode on the wrong line.
you would need to patch zend_compile.c in the PHP source code (or wait for a fixed version). without patching zend_compile.c test-coverage results might be inaccurate - but nevertheless, common debugging should work. make test wouldn't check for it, when xdebug wouldn't partially rely on it (the title of the test even explicitly states why it checks for the bug). here's the diff, which adds CG(zend_lineno) = ast->lineno;. this "fixes the issue 100%" and not only it's symptoms:
index f1dd49a..9c0893b 100644 (file)
--- a/Zend/zend_compile.c
+++ b/Zend/zend_compile.c
## -4807,6 +4807,7 ## void zend_compile_foreach(zend_ast *ast) /* {{{ */
zend_end_loop(opnum_fetch, &reset_node);
+ CG(zend_lineno) = ast->lineno;
opline = zend_emit_op(NULL, ZEND_FE_FREE, &reset_node, NULL);
}
/* }}} */
this bug affects PHP 7.0, 7.1, 7.2 - PHP 5.x was at least not reported.
since this is not a xdebug bug, installing even with the failed test shouldn't make anything worse, than it already is. to install xdebug, it's not sudo make test but make && sudo make install (only make install requires sudo).
I have been trying to figure out whats going on here... I have looked all over and nothing seems like a solid solution. If anyone knows a link or information on how to fix this it would be greatly appreciated.
Ubuntu 14.04 lts
Microsoft Azure Server
Apache
Drupal site with APC enabled (everything looks good on the drupal side) I just keep getting these annoying emails about the below issue.
I added smtp and mailutils to my server, and now i am getting these errors from what looks like apache / php...
--
Cron [ -x /usr/lib/php5/maxlifetime ] && [ -x /usr/lib/php5/sessionclean ] && [ -d /var/lib/php5 ] && /usr/lib/php5/sessionclean /var/lib/php5 $(/usr/lib/php5/maxlifetime)
PHP Warning: PHP Startup: Unable to load dynamic library '/usr/lib/php5/20121212/apc.so' - /usr/lib/php5/20121212/apc.so: cannot open shared object file: No such file or directory in Unknown on line 0
Thanks for the help,
-Frank
These are being generated by the /etc/cron.d/php5 cronjob. If you want the errors generated here to stop being emailed, you can set mailto="" on the line before the command in that file. This is probably not your desired solution here, but it is useful to 1) know where you can find where the error is coming from and 2) know how to control its output.
The command generating these errors is likely maxlifetime. You can confirm this by trying to run /usr/lib/php5/maxlifetime and seeing if the same error comes up.
Your apc.so file might be somewhere else. Try "locate apc.so" to find out where it is. If it doesn't exist, I would recommend reinstalling the extension so that it does exist. If it does exist, your /etc/php5/apache2/php.ini file might be having trouble finding it. A couple solutions I read from other threads had people either creating a symbolic link to to the file (a bit sloppy, but it works) or noticing that "extension_dir=" in php.ini was set to something other than "./" so that it was looking in the wrong place (or the "extension=apc.so" entry itself contained some of the path and had a typo or something).
In my case, my php.ini file was trying to load the extension msql.so which didn't exist on my system. I don't know why it would do this, but I commented out the line and the error messages went away. In most cases, though, I would assume the extension was deliberately installed so you would want to go the route of tracking it down or fixing it.
Good luck.
I have spent several hours searching and attempting things but nothing seems to make a difference.
We have a VPS running nginx + php-fpm that only has one Joomla! website and also a subdirectory for iDevAffiliate.
Everything was working fine until I did a yum update - updated everything that was due and then updated iDevAffiliate from 7.4 to 8. Now I'm not sure which action is the culprit but at this point on iDevAffiliate is not working - but they say it's server related.
Users are able to login and use a fairly big Joomla! website without any problems - this makes me doubt the server is really at fault.
When attempting to login to the front-end or admin page of iDevAffiliate - this error at the bottom of the page
Warning: Unknown: Failed to write session data (user). Please verify that the current setting of session.save_path is correct (/tmp) in Unknown on line 0
Rebooted the server a few times as that's suppose to clean
the directory.
Manually tried clearing the directory (wasn't many files
since the reboot).
Confirmed and tried re-applying CHMOD 1777 (was already set) - also tried CHMOD u+w
Confirmed session.save_path in php config is set to "/tmp" (no trailing spaces)
/tmp is running off / when I checked with "df -k /tmp"
What am I missing? I find it really odd that only iDevAffiliate is causing problems, coincidentally right after I upgraded it.
You help would be much appreciated.
Edit: Here's some DF info I've been asked about. http://imgur.com/VRef01d
We found the problem was actually with a bug in suhosin and have resolved it by disabling suhosin. What a weird one.
Apparently, suhosin isn't fully supported for php 5.4
I'm (very) new to databases. I've designing a site, and was happy that I got a handle on the errors in my error log. Recently my error log keeps logging in "PHP Warning: ...unable to load...suhosin.so"
Obviously a new entry is made whenever I query something. Unfortunately, google searches tell me 2 things: (1)the error isn't really a big thing (but I was proud that my error log stopped growing!) and (2) suhosin.so is an advanced topic (and security patch) for mysql.
The advanced part has me a bit freaked out as to how I can "fix" it...
Please help.
It says extension=suhosin.so somewhere in your php.ini. You can either remove or comment out this line to disable the protection, or install the current suhosin extension from suhosin.org. Simply
grep -r suhosin.so /etc/php5
(The path may be different on your system.)
Suhosin is an advanced protection system for PHP. (not mysql)
Try this...
(1) Install suhosin extension from suhosin.org if it's not installed yet
(2) Un-comment (by removing #) extension=suhosin.so from your php.ini file
(3) Restart your web-server (Apache?)
This should fix your suhosin.so error.