Its nice to see them being so transparent about what happened and how they plan on fixing it in the future. They're obviously working on anticipating problems

Its nice to see them being so transparent about what happened and how they plan on fixing it in the future. They're obviously working on anticipating problems in the future, but what I wonder about is things like this, where they thought they were covered. How does one go about finding these failure points on systems that span multiple locations? I hope they followup with lessons learned on their quest to improve reliability.