By: Nick Gambino
You may have noticed some apps or internet tools not working for you on Monday. Well that was no minor hiccup. Amazon Web Services went down and took a huge network of seemingly unrelated pieces of the internet with it.
The AWS crash gave us a tiny glimpse into what would happen if the internet suddenly crashed. We tend to forget that it’s all built on a tapestry owned and controlled by a few tech juggernauts. From the online mom and pop to massive corporations, they are supported by Amazon’s cloud services.
So what happened exactly? Well, there was a disruption in the data server farm in Northern Virginia not 20 minutes from where I’m writing this. More technically, the US-EAST 1 server cluster was having trouble with its DNS (Domain Name System) so applications weren’t able to distribute cloud-based information to the correct web addresses across its vast international network. That disruption started small in the wee hours of Monday and spread through the rest of the day, forcing them to respond like a five-alarm fire.
Except that fire wasn’t put out right away. The outage began at 3am and we didn’t get the “All AWS services returned to normal operations” update until 6:01pm. That’s 15 hours of small business not being able to process payments. That’s 15 hours of passengers on United flights not being able to pull up their reservations. That’s 15 hours of Ring cameras not surveilling.
Of course this is a momentary impact with effects that will probably only ripple out over the next week or so and then it will all go back to normal. But what this has done, and what earlier outages in Microsoft and Google have done, is exposed how fragile it all is.
The internet is not just a bunch of individual websites completely unrelated from each other. And while they do a lot to ensure outages aren’t constant, we have to hope that these mega cloud service platforms are responding to these issues by adding more redundancies and resources to ensure it doesn’t happen again.
