An hours-long outage from Amazon Web Services (AWS) caused all manner of global havoc on Tuesday, shutting down government and private websites, robotic vacuums and popular streaming services and idling Amazon facilities as networks and apps went dark and operations ground to a halt.
Entire Amazon fulfillment centers shut down, and reports came in of truck drivers reverting to pen and paper when dropping off or picking up loads, the Verge reported.
At the height of peak season, an untold volume of packages didn’t go out on time from Amazon for several hours Tuesday during the outage, further exacerbating already strained networks and cascading through delivery timelines. Amazon has once again resorted to augmenting its fleets of vans for DSPs with rented vehicles to handle the seasonal crush. The company has invested billions in peak season prep, including opening new facilities, chartering planes and cargo ships and hiring hundreds of thousands of temp workers – some even poached school bus drivers.
Amazon first reported outage issues at 9:37 a.m. PST on Tuesday on its AWS status page, saying multiple APIs in the US-EAST-1 Region were impacted. “This issue is also affecting some of our monitoring and incident response tooling, which is delaying our ability to provide updates. We have identified the root cause and are actively working towards recovery,” the company said. It took until 4:35 p.m. PST for all network issues to be resolved.
According to Business Insider, an internal memo at Amazon mentioned massive server traffic from an “as yet unknown source” as being the root cause of the outage. A denial of service (DOS) attack in that scenario wouldn’t be out of the question.
Amazon has an estimated 41% share of the world’s cloud server market, according to Gartner, with just five companies controlling 80% of the market.