| | | Amazon's US-East-1 datacenter suffered increased error rates today. (CRN)
This is in the same sense as "Jeffrey Epstein didn't increase his own error rates". AWS US-East-1 burned down, fell over, and sank into the swamp.
With it went a few little sites like Netflix and Disney+ and, oh, Amazon itself. Tens of thousands of Adele fans waiting in virtual line for pre-release tickets were not impressed when the whole thing was rescheduled, and they weren't notified by email because email services relying on AWS were also down. (Fortune)
Coinbase, Binance, dYdX (which is a cool name for a derivatives trading platform I must admit) and blockchain gateway Infura all suffered outages, which makes you wonder just how decentralized this decentralized finance thingy really is. (Be In Crypto)
Alexa wouldn't talk to you, Kindles couldn't kindle, Roombas couldn't roomb, and Ring video doorbells couldn't burn down your house. (The Verge)
Within Amazon, not a creature was stirring, not even a wireless mouse. (Business Insider)
AWS is not just a product, it's Amazon's own computer system. With that down, the storefront went down, and all the inventory and logistics services went with it. Delivery drivers were paid and sent home because there was nothing they could do - there's no paper trail for this, not at Amazon's scale.
Instacart, Venmo, CashApp, Roku, McDonald's app and automated kiosks, Tinder, and Delta Air Lines all went off the air. (USA Today)
We only have one direct AWS dependency at work, and that's at US-West-2 a thousand miles away, but it went down for six hours anyway.
We have quite a few indirect dependencies, though, so my day was spent cleaning up after all of those services failed.
There is no cloud, there's just someone else's computer, and it's down.
ace.mu.nu |
|
