I am getting a lot or error messages after upgrading the PHP, Apache, MySQL. I think this may be due to some error messages that were referred to just as we released 2.8.3 having to due with PHP5 being stricter than PHP4 at catching certain syntactical errors.
OpenEMR 2.8.2-dev (just before the 2.8.3 release)
mysql-5.0.54
apache-2.2.8
php-5.2.5-r1
mod_ssl/2.2.8
OpenSSL/0.9.8g
Don’t know about that error message, but check over your php.ini. It sounds like you are running out of memory for the scripts - the default setting is not usually enough. Some settings that I commonly use:
You might create a “<?php phpinfo() ?>” script amd make sure the php.ini that you configured is the same one that’s being used. Also browse the output for anything that seems wrong.
This Multi-Processing Module (MPM) implements a hybrid multi-process multi-threaded server. By using threads to serve requests, it is able to serve a large number of requests with less system resources than a process-based server. Yet it retains much of the stability of a process-based server by keeping multiple processes available, each with many threads.
******
The description on the Gentoo Wiki describes this as the greatest thing on "well behaved" web sites. They then describe the web-behaved behavior but no trouble shooting in case you happen to have a "non well behaved site."
Run away threads are a problem for PHP and may be affecting the Calender module (and my database storage of images).
Last I heard, PHP requires the "prefork" MPM. The multi-threading MPMs are not compatible with it because PHP uses libraries that are not thread-safe. So yeah, that might be your problem.
Interestly the Gentoo community feels that PHP is generally thread safe and the default or recommended "USE" variables include "threads" for both the Apache and PHP. Using the "threads "USE" variable causes the MPM-worker module to be installed.
This is mutually exclusive with the mpm_prefork_module. I am pretty confident that the Calender is not thread safe and apparently the code that I wrote for the image - documents database uploading routine is not.
I have removed the "threads" "USE" variables from /etc/portage/package.use and added "mpm-prefork" to the APACHE2_MODULES. I will recompile Apache and report back.
After removing "threads" from the pakage.use and adding "mpm_prefork_module" to the Apache "USE" variables, then recompiling we have not had any more issues. We have been up significantly longer than previously. Initially this looks like this solved this problem.