Google App Engine: Post-mortem for February 24th, 2010 outage (groups.google.com)

Original post here: http://news.ycombinator.com/item?id=1168493

A pretty good write-up about what went wrong. I like the mitigation procedure of implementing recovery drills that "include the rare and complicated procedures"... whoever runs that is going to have a wickedly fun job!

This information will be invaluable to the resistance when sky.. google becomes concious