An AWS Employee Typo Broke the Internet
Amazon Web Services (AWS) has revealed more details about the S3 cloud storage service outage that affected many websites earlier this week. It seems that an employee made a typing error when taking some servers offline for debugging. “Unfortunately, one of the inputs to the command was entered incorrectly and a larger set of servers was removed than intended,” Amazon said. “The servers that were inadvertently removed supported two other S3 subsystems.”
Re-booting those systems took a long time, meaning that websites relying on S3 experienced an outage that lasted several hours. According to SimilarTech, 148,13 websites rely on Amazon S3, including Quora, GitHub, Indeed.com, StackOverflow and, of course, Amazon itself. Somewhat ironically, the dashboard that reports on the status of Amazon cloud computing services relies on S3, so it also went down during the outage.