We’ve been having major problems with the interconnect between the main server and the NAS over the past few days. What that means is that some of the software that powers the site sometimes gets transmitted wrong, and then crashes. While the crash is being logged, the broken software is held in memory, and is executed again and again, causing a cascade that, depending on the module, brings down the whole server. As I write this, the machine is (I presume) running a disk check. It should come back up when it’s done (there are a lot of files so it will take a while), but it may continue to be unstable.
Update (6PM EST): Back online, but it may not last. These crashes are unrelated to images and other user files in case anyone was worried about that!
Tomorrow (Monday) we’ll be reconfiguring the server to hopefully completely eliminate this problem. I apologize for the inconvenience and annoyance.