Up until a few days ago I was having some issues with the AWS EC2 server I was using for this site. It was regularly locking up — not just stopping, but getting to the point where I was having to recycle the host altogether.
I was getting frustrated, as one does with things not working.
Earlier this week I had logged in while the server was working and started poking around. Upon trying to start up Emacs on one instance I got an out of memory error. This was the key to solving the problem. It seemed that the somehow my PHP server was no longer serving PHP but rather the main Apache server was taking the load.
As it turns out in-process PHP sucks something fierce. It works but it uses memory like it’s going out of style.
I went back and fixed things. Switched back to FPM serving and threaded instead of forking. Now things are back to where they should be.