When human errors can’t be counteracted
The recent outage of services in the Amazon cloud was caused by a human error. Which? Error in typing (typographical) of a command by a collaborator. The same led to the Amazon Web Services (AWS) servers, also known as S3, stop working. It should be noted that such servers offer data hosting on the Internet to thousands of web pages and applications for mobile devices. And because of that mistake, services from companies like Slack, Quora, Giphy, Venmo, and even part of Apple stopped working. Netflix, Spotify, Pinterest and Buzzfeed were also partially affected.
According to Amazon, the failure occurred while performing maintenance on servers. However, the problem occurred when instead of turning off only a part of the system, a command was introduced that precipitated a chain blackout. When trying to correct the first error, it was decided to restart the system. But they did not expect that many of the fallen servants had never been turned off.
The price paid by the giant of Internet shopping, Amazon
The rebooting and commissioning of the servers took about five hours. They represent an eternity for the services that were affected by the fall of the servers being Amazon one of the biggest clouds.
Cyence, a company specializing in risk management, said that the mistake could have cost the giant about $ 150 million. An important issue that we have to mention was the specialized publication Business Insider where he explained a few months ago that the cloud computing market presents characteristics of a natural monopoly: the bigger a network, the more efficient it is at the operational and financial level. But the other side is that it is more fragile … And this we have seen reflected in the fall of the servers of Amazon Web Services. Same as I drag with thousands of companies and applications.