And we’re moved to a shiny new (virtual) server, so with any luck this sort of thing won’t happen again.
Update: It was late enough last night by the time it was fixed that I just wanted to post the notice and go to bed, but if you’re interested in more detail, feel free to read on now.
Speed Force runs on an NginX web server so it can handle more traffic than an Apache web server. But that makes it a slightly odd environment because (1) WordPress is built mainly for Apache and (2) DreamHost’s NginX setup doesn’t have as much support for PHP (the programming language WordPress runs on) as their Apache setup does.
Over the weekend I turned off another website on the same server. And PHP stopped running. Completely. Not sure if those were actually connected or just coincidence. I spent several hours chatting with tech support while I tried to fix what I could as a user, and they tried to fix what they could as an admin, and nothing went anywhere. Finally the tech I was chatting with put in a support ticket for followup. Every time I checked in, I got the same response: a specialist was looking at it.
Monday evening, after getting the same response again, I decided to test something else that I could do: Launching an entirely new NginX server and seeing if PHP would run. It did. So I moved everything over. I had to do some cleanup, but once that was done the site was up and running again! I moved everything else over (literally pressing one button for each site and waiting a couple of minutes) and cancelled the old server.
For the most part DreamHost has been stable, and their support team has been responsive. Like I said, I started with several hours on chat with the first rep who tried all kinds of things to track down the problem. I think this was just a perfect storm of a non-standard setup running into a weird situation.