Now this is nice. I like more of an explanation than "the service was out; now its not" and the incident report gives a bit more detail. Although its a little vague about exactly what happened, you get the jist that they had a bug in their server load sharing algorithm, so when one was taken off-line for maintenance, too many people were redirected to another. The "multiple downstream overload conditions" sounds like a neat cascading system of propagating errors. But maybe I'm a big dork.P.S. The System Is Down is a nice beat for all your lightswitch rave needs.
P.S.S. I totally didn't notice this outtage and I'm on the gmails like all effing day.
