I run Red Hat Enterprise Linux 5 server. If I were to install PHP-FPM, which patches the PHP package, it would be replaced when PHP is updated on the next yum update. I have spawn-cgi installed, and can run it only from the command line as such:

/usr/local/bin/spawn-fcgi -a 127.0.0.1 -p 9001 -C 5 -u apache -g apache -f "/usr/bin/php-cgi /maxconns=1024 /maxreqs=1024" -P /var/run/fastcgi-php.pid

When I try running fastcgi from an init script I get the following message when I go to any page on my blog:

The webpage at https://my_IP_address_omitted/init/start_stop.cgi?file=%2Fetc%2Frc%2Ed%2Finit%2Ed%2Fphp%2Dfastcgi&name=php%2Dfastcgi&start=1&back=edit%5Faction%2Ecgi%3F0%2Bphp%2Dfastcgi might be temporarily down or it may have moved permanently to a new web address.

It doesn't matter if I user the nginx user and group.

Apache can run PHP reliably on heavy loads, and it gets updated automatically on a yum update, so for me it works, and it automatically restarts if the sever does, because the init script works.

I've read that under heavy load Apache can handle PHP much more reliably than fastcgi as well, which is what I'm after.

Apache running as a backend in the configuration I cited above uses as much or less memory and CPU cycles as fastcgi, so the only issues for me are reliability, and the ability to have an init script that works.

I am also concerned about SEO: http://wordpress.org/extend/plugins/nginx-compatibility/

My config handles different files in different ways as you can see. I have several high traffic blogs, and they have very specific needs.