artsnla

On October 20, 2025, a major outage affected Amazon’s cloud service, Amazon Web Services (AWS). This outage created service interruptions on a global scale, impacting companies including Spotify, Zoom, and Canvas. Here is a recap of the AWS outage:

When did this outage start?

The AWS outage started at around 3 a.m. ET. (~1:28 MDT), after Canvas identified an issue “caused by an operational issue affecting multiple services at Amazon Web Services”. This outage affected companies including T-Mobile, Venmo, Google, PlayStation, Zoom, Fortnite, and Starbucks. It also affected social media, streaming apps like Spotify, food delivery services, and remote learning apps like Canvas.

How many apps and websites were affected?

The AWS outage affected millions of people worldwide. In fact, users reported problems with popular apps and websites like Netflix, Spotify, WhatsApp, and Zoom. Amazon’s Alexa, Kindle, Prime Video, and Ring products were also affected by this global AWS outage. Canvas experienced global outages that prevented students and teachers from accessing coursework and class dashboards. On October 20, 2025, at 3 a.m. ET, Canvas was officially down for maintenance. After the AWS outage occurred, Canvas users received this notification: “We are working on recovering as soon as possible. You can follow the status of the issue at AWS Health Dashboard and Instructure Status.”

What caused this global AWS outage?

Specifically, Amazon attributed this outage to problems with its Domain Name System (DNS). Like a smartphone charger, DNS is crucial for loading websites and apps. According to Reuters, AWS said this issue originated from within the “EC2 internal network”. The EC2, Amazon’s “Elastic Compute Cloud” service, provides cloud capacity within AWS. Amazon’s cloud service, Amazon Web Services (AWS), supports power operations for government departments, schools, and the media. According to Ars Technica, this global AWS outage might cost billions of dollars because it disrupted over 28 AWS services.

What steps is AWS taking to monitor outages and work towards improving service continuity on the cloud?

According to PBS, AWS’s longest outage occurred in 2021, delaying services from airline reservations to video streaming services. (In 2023, there was another short AWS outage.) On October 20, 2025, at 3 A.M. ET, the third AWS outage occurred. This outage was resolved at around 4 P.M. ET. AWS implements preventative measures during operations meetings. By holding operations meetings weekly, AWS is constantly working on maintaining service continuity and monitoring incidents.

AWS posted this statement: “Although the likelihood of such incidents is very low, AWS is prepared to manage large-scale events that affect our infrastructure and services. AWS becomes aware of incidents or degradations in service based on continuous monitoring through metrics and alarms, high-severity tickets, customer reports, and the 24x7x365 service and technical support hotlines.”