麻豆传媒

BSOD, the blue screen of death, has become the universal metaphor for an unscheduled operational collapse. When it happens to your Windows machine, you know you're in for a day that is only slightly more preferable than a root canal. When it happens systemically in the cloud… that's something else entirely.

Welcome to the biggest lesson in business continuity and resiliency we've experienced (at least since two weeks ago, when 15,000 car dealers couldn't write sales contracts or lease documents for more than a week).

What happened? It seems that Crowdstrike pushed an inadequately-tested security update, which resulted in BSODs for tens of thousands of Microsoft servers across the Azure cloud.

Funnily enough (if there is anything funny about this), the fix is a reboot – which may or may not work, just like on your own computer – but there's a catch: you may need to reboot up to 15 times. If that doesn't work, you have to manually remove a configuration file (which is way more time consuming).

There is absolutely nothing you or I can personally do to prevent stuff like this, but it does give us some things to think about. First, this was an honest mistake (albeit a big one). What if it wasn't? What if a nation-state launched a coordinated attack aimed at taking down transportation, banking, and healthcare in the U.S. (or globally)? What would we do?

These questions lead to the next, more important question: Are we prepared? The answer, obviously, is no. What do we need to do to prepare? We are completely dependent on cloud infrastructure to run our digital lives, and with the exponential growth of AI platforms, we’re about to become even more dependent. How should we prepare for the inevitable attack that will take some or all of it away?

As always your thoughts and comments are both welcome and encouraged. Just reply to this email. -s

Today's Most Interesting Stories