For added fun, their EC2 console is down. I got this for a while:
<html><body><b>Http/1.1 Service Unavailable</b></body> </html>
... then an empty console saying "loading" for the last 20 minutes. Then recently it upgraded to saying "Request limit exceeded." in place of the loading message (because hey, I'd refreshed the page four times over the course of 20 minutes).
On the upside, their status page shows all green lights.
They have standardized icons to represent various levels of issues (orange = perf issue, red = disruption). But they don't even use them. Instead they add [i] to the green icons to indicate perf issues (Amazon Elastic Compute Cloud - N. Virginia) and disruptions of service (Amazon Relational Database Service - N. Virginia).
Maybe this status page is controlled by marketing bozos who want to pretend the situation is not so bad.
Does Amazon explain anywhere what regions, availability zones or whatever have single points of failure for each of their services? I guess that's supposed to be what an "availability zone" is but somehow that doesn't quite seem to capture it. It's pretty hard to build reliable apps without knowing where the single points of failure are in the underlying infrastructure.
On the upside, their status page shows all green lights.