The problem has been pretty erratic. It was really slow half an hour ago (up to 90 seconds), and then 15 minutes ago it took off (~1 second responses).

At least they're working on it -- The log shows the system was shut down several times overnight.

I am wondering if the db was out of disk space to do any sorts.

It is being consistently quick as I write this (knock on wood), so they may have figured out the problem. It would be nice if the techies would share the solution.