Skip to Content

The AWS outage affects Canvas and other apps

Amazon Web Services (AWS) encountered connectivity issues across multiple platforms like Amazon and Canvas, affecting CWRU students on Oct. 20.
Amazon Web Services (AWS) encountered connectivity issues across multiple platforms like Amazon and Canvas, affecting CWRU students on Oct. 20.
Tyler Sun

On Monday, Oct 20, Amazon Web Services (AWS) experienced problems from around 3:11 a.m. Eastern Time to 6:01 p.m. later that day. In technical terms, the health monitoring system of network load balancers inside AWS’s Elastic Compute Cloud 2 (EC2) internal network in US-East-1 malfunctioned. Simply put, there was a database malfunction in a major cloud computing hub. Amazon reported increased error rates and connectivity issues across multiple services across the United States.

 

AWS is a cloud computing platform that provides technology infrastructure, such as data processing, storage and management to other companies that purchase it. Instead of buying hardware and infrastructure that needs to be maintained, companies can simply rent the services it runs on from AWS. Because AWS already has and operates the infrastructure needed to power Amazon, they are able to also rent their cloud services out to others for a profit.

 

However, AWS sharing its service with these companies also means it powers them. These companies include apps like Zoom, Reddit, Snapchat, Duolingo, Microsoft 365 and more. For some of these companies, like Snapchat and Duolingo, websites and services were completely down during the outage, while some just experienced slowdowns. 

 

Canvas was one of the apps that was completely down, not only for Case Western Reserve University students but for high school and college students across the country. Students couldn’t access educational resources or their classes on Canvas for the day. For some students at other universities, professors cancelled classes that were highly dependent on Canvas or pushed deadlines that passed during the outage.

 

By 3:35 a.m. Pacific Daylight Time on AWS Health Dashboard, AWS reported, “the underlying DNS issue has been fully mitigated, and most AWS Service operations are succeeding normally now. Some requests may be throttled while we work toward full resolution.” However, these issues continued further into the afternoon and even into the evening.

 

The outage received mixed reactions from CWRU students coming back from fall break. Some were happy and hoping it wouldn’t be back up when Wednesday rolled around, while others were worried about upcoming deadlines and midterms. 

 

“I actually thought it was nice because I could relax during the break since I couldn’t do any work,” said Maria Burrus, a first-year student at CWRU.

 

On the other hand, first-year student Zoe Shankland said “I was anxious because I have a midterm tomorrow, and I couldn’t look at the notes that my professor uploaded on Canvas.”

 

However, around 6 p.m., AWS repaired its issues, allowing students to regain access to their courses and assignments and prepare for the upcoming classes.