https://www.msn.com/en-us/technology/cloud-computing/smart-beds-flipped-out-during-the-aws-outage-and-so-did-their-sleepy-owners/ar-AA1OYol8

Smart beds flipped out during the AWS outage, and so did their sleepy owners

Banks failed to transfer their clients’ money. Gamers lost their streaks. Peloton riders struggled to log into classes.

For a few hours early Monday, a service outage at Amazon Web Services triggered a rippling cascade of error messages and system malfunctions across the internet, revealing just how much daily life functions on the cloud-based servers of one provider.

But perhaps the rudest awakening was felt by those sleeping on Eight Sleep’s high-spec internet-enabled mattresses.As the outage caused cloud servers to fail, people reported being awoken in discomfort by beds that locked in an upright incline, became unbearably warm, blinked flashing lights or even sounded a wake-up alarm.

“That is not the experience we want to provide and I want to apologize for it,” said Matteo Franceschetti, the chief executive of Eight Sleep, which also manufactures and sells internet-enabled mattress covers and pillow covers. In a social media post, he explained that the malfunctions were the result of the AWS outageand said engineers were racing to build an outage-proof modein the event of a future outage.

Eight Sleep did not immediately respond to a request for further comment early Wednesday. As of Monday evening, Franceschetti said all devices were functioning although some were still experiencing “data processing delays.”

The company’s internet-enabled mattresses allow users to adjust their bed’s temperature between 55 and 110 degrees, elevate their body into different positions, as well as activate immersive “soundscapes” and vibrational alarms. The most advanced product bundles retail for over $5,000, in addition to a yearly subscription costing $199 to $399 that is required to enable the temperature controls.

The malfunctions also highlighted some of the risks in making so many aspects of everyday life dependent on cloud-based technologies, which Alan Woodward, a professor of computing at England’s University of Surrey, said will always be prone to occasional failure.

“Complexity always causes issues,” he said in a phone interview Wednesday. “The more complex you make the functionality, the more difficult it is to understand what will happen when some parts of that whole chain of computing fails.”

Amazon said that the outage was caused by a DNS issue, referring to the Domain Name System — effectively the global online directory that automatically translates domain names into IP addresses. (Amazon founder Jeff Bezos owns The Washington Post.)

“The servers basically disappeared off the internet. So as far as the devices were concerned, there was nothing for them to talk to,” Woodward said. “If they don’t know what to do, they will go into some failure mode.”

“Computers are relatively dumb. They will do what you tell them, but if you don’t tell them what to do, they just run around in a panic,” he said.