We’re experiencing major database issues again, due to a significant bug somewhere in MySQL.
We’re trying various options to fix the problem, but please bear with us for a little while longer. The problem manifests itself as a sudden (and catastrophic!) slow down, which on the server itself is represented by a load average exceeding 100 on some occasions. Under normal load that number hovers around 0.8 to 1.2… That gives you some idea as to what the numbers represent at least. (100 renders the server almost unuseable).
We’ll let you know how we get on.