>>sillys+(OP)
> Being down all morning is an impressive recovery time, because they had to provision an EC2 server and transfer all data to it
It looks like the data was already in S3: [1]
The recovery probably would have been a lot faster had they had a fully provisioned EC2 image standing by... which I'd bet they will from now on.
[1] - https://news.ycombinator.com/item?id=32032316