This is the part that I find fascinating: 7:48 AM - Internal monitoring graphs first begin to show that traffic has problems in our primary datacenter 9:

This is the part that I find fascinating: 7:48 AM - Internal monitoring graphs first begin to show that traffic has problems in our primary datacenter 9:35 AM - An engineer with familiarity with the unplanned failover procedure is reached And the blog entry keeps sounding like there was only one engineer making decisions for the first two hours.