Amazon Finds the Cause of Its Outage: A Typo

Amazon.com (NYSE:AMZN) on Thursday blamed human error for an outage at its cloud-services unit that caused widespread disruption to internet traffic across the U.S. earlier this week.

In a post on its website, Amazon said the outage started with a typo at Amazon's North Virginia data centers Tuesday.

An employee trying to speed up the company's S3 cloud-storage billing system tried to take a few servers offline. The employee mistyped the command, however, affecting more servers than intended, which led to a cascade of failures that ultimately knocked out S3 and other Amazon services. It also took longer than expected to restart certain services, Amazon said.

Amazon said it is adding safeguards to prevent server capacity from falling too quickly or below a minimum level.

Amazon Web Services, the largest global seller of cloud infrastructure, has more than a million users. The hourslong outage Tuesday disabled and slowed apps and websites from a wide section of U.S. companies, including Quora Inc., Slack Technologies Inc. and Medium.com Inc.

The AWS outage cost companies in the S&P 500 index $150 million, according to Cyence Inc., a startup that specializes in estimating cyberrisks. Apica Inc., a website-monitoring company, said 54 of the internet's top 100 retailers saw website performance slow by 20% or more.

Robert McMillan contributed to this article.

Write to Laura Stevens at laura.stevens@wsj.com

(END) Dow Jones Newswires

03-02-17 1650ET