When Amazon Web Services (AWS) went down on the morning of Monday, October 20, 2025, the ripple effect was immediate and far‑reaching, flooding university help desks and corporate Slack channels with error messages.

Located in New Brunswick, New Jersey, Rutgers University’s Office of Information Technology (OIT) confirmed that roughly 71,000 students and 23,000 faculty and staff were hit by failed logins, frozen video lectures and broken document collaboration tools. The outage, catalogued as AWS Outage of October 20, 2025Global, began just after 8 a.m. Eastern Daylight Time and persisted well into the evening.

Here’s the thing: Rutgers relies on AWS for everything from its learning management system, Canvas, to video‑streaming platforms like Kaltura, and even the campus‑wide Adobe Creative Cloud licenses. When the cloud backbone staggered, the university’s digital heartbeat stuttered.

What Happened and When

The AWS health dashboard first logged a “Service Degradation” flag at 08:11 UTC‑4, noting “intermittent availability across multiple regions.” By 09:03, the status page upgraded the alert to a full‑scale outage, citing “network connectivity issues in the US‑East‑1 (N. Virginia) and US‑West‑2 (Oregon) zones.” The problem appeared to stem from an internal routing loop that overloaded edge locations, according to an internal AWS engineering note that leaked to the media.

Rutgers OIT issued its initial alert at 09:14, warning students that Canvas would display an error page and that “other services such as Kaltura, Smartsheet, Adobe Creative Cloud, Cisco Secure Endpoint, and ArcGIS may also be affected.” The message was signed by Brian Voss, Chief Information Officer, who emphasized that the university was “monitoring the situation closely and will update as soon as we have more information.”

By mid‑day, the outage had crept into Zoom video calls, Grammarly’s real‑time writing assistant, and even the campus Wi‑Fi authentication system, prompting a second OIT update at 13:42 that listed the growing list of impacted services.

How Rutgers University Was Affected

Canvas is the backbone of most coursework at Rutgers, and its downtime meant students couldn’t submit assignments, view grades, or participate in discussion boards. Professors reported that exam‑week study groups had to migrate to ad‑hoc Google Meet sessions, despite the university’s recommendation to use Zoom.

The video platform Kaltura Inc., hosted on AWS, stalled, causing lecture recordings to freeze or fail to load. Meanwhile, Smartsheet Inc. sheets used by administrative offices for budgeting and project tracking were stuck at “saving…” and could not be refreshed.

Adobe Creative Cloud, critical for design students, threw authentication errors, and the cybersecurity suite Cisco Secure Endpoint stopped pushing updates, leaving some lab machines temporarily unprotected. Even Esri’s ArcGIS mapping tools, essential for geography courses, displayed blank maps.

In total, OIT logged 4,372 help‑desk tickets between 09:00 and 20:00 on October 20, a 230 % spike compared with an average weekday. The university’s incident command center, located across its three New Jersey campuses—New Brunswick, Newark and Camden—operated around the clock.

Ripple Effects Across Cloud‑Dependent Services

Beyond Rutgers, the outage rippled through any organization that runs its workloads on AWS’s core services—EC2, S3 and RDS. Small‑business SaaS providers reported latency spikes, and a handful of e‑commerce sites in the Midwest saw checkout failures, prompting a brief dip in sales on the day.

According to a third‑party monitoring firm, global internet traffic to AWS‑hosted endpoints dropped by 12 % during the six‑hour window, and the average latency for API calls rose from 45 ms to 340 ms. The incident underscores how a single cloud provider can become a single point of failure for a sprawling digital ecosystem.

• Over 150 universities worldwide reported similar Canvas or Zoom issues.
• Approximately 3.2 million active AWS accounts were potentially impacted.
• Amazon’s stock dipped 0.4 % in after‑hours trading on the news.

Responses from AWS and Service Providers

AWS posted a concise update at 16:05, stating that “the networking issue has been mitigated and services are returning to normal operation,” but added that “some downstream services may still experience residual latency.” The company promised a detailed post‑mortem within 30 days.

Instructure Inc., the maker of Canvas, posted a status note confirming that their own servers were healthy; the problem was strictly in the underlying AWS infrastructure. Kaltura’s engineering blog echoed the same sentiment, noting that “our platform is fully functional on our own side, but the AWS storage tier experienced delays.”

Zoom’s public statement emphasized that “our real‑time video routing is distributed across multiple clouds, so the impact was limited to a subset of customers who rely exclusively on AWS.” Meanwhile, Adobe’s support center advised users to log in via a regional endpoint until the AWS CDN cache cleared.

Rutgers OIT’s final update at 19:48 thanked the community for patience and reminded users to clear browser caches and restart affected applications. The office also posted a link to a live FAQ on its website.

What This Means for Cloud‑Reliant Institutions

The outage reignited the debate over multi‑cloud strategies. Many CIOs have already begun piloting secondary providers—Microsoft Azure or Google Cloud Platform—to mitigate risk, but the cost and complexity can be prohibitive for smaller institutions.

Experts from the Cloud Security Alliance warned that “reliance on a single vendor for critical workloads can amplify outage impact,” urging organizations to adopt “chaos engineering” practices that simulate failures and test response plans. In a recent interview, Dr. Lena Patel, a professor of Information Systems at Carnegie Mellon, said the Rutgers episode is a textbook case of “over‑centralization.”

For students, the immediate lesson is to maintain offline backups of coursework and be prepared to switch to alternative collaboration tools when cloud services hiccup.

Looking Ahead

Amazon announced that it will conduct a “region‑wide health review” of its US‑East‑1 and US‑West‑2 zones in November, aiming to shore up network redundancy. The company also hinted at a new “edge‑resilience” feature set that could auto‑reroute traffic in future incidents.

Rutgers OIT plans to audit its disaster‑recovery procedures, including an updated “cloud‑failover” policy that could route essential services to a private data center on campus if AWS experiences another outage.

Until then, the university’s students and staff are left watching status pages and hoping the internet’s invisible veins stay clear.