AWS Outage And Top Cloud Disruptions Of 2025 That Brought Web To A Standstill

In 2025, the interconnectedness of cloud infrastructure exposed vulnerabilities, as failures in core components like DNS, configuration systems, and authentication triggered widespread outages affecting millions of users and thousands of businesses.

These incidents underscored risks in concentrated cloud ecosystems. They highlighted the need for enhanced redundancy, phased rollouts, and multi-region architectures to mitigate failures.

Here are five major cloud outages that disrupted the digital world in 2025.

A widespread outage disrupted several popular online gaming platforms on Dec. 25, preventing thousands of players from logging in or playing. The hardest-hit game was ARC Raiders, with nearly 35,000 reports of server connection errors and network timeouts. Other affected titles included Fortnite, Rocket League, and various Epic Games properties. The outage, which was initially attributed to Amazon Web Services (with the company denying such claims), was reportedly later traced to an authentication failure in Epic Online Services.

Cloudflare experienced multiple incidents, including a Nov. 18 bot-management bug that caused a 3-6 hour global outage hitting Spotify, ChatGPT, and X. There was a shorter Dec. 5 firewall-related disruption affecting LinkedIn and Zoom — both stemming from configuration errors.

The year’s most severe outage hit AWS on Oct. 20, when a DNS automation bug in the US-East-1 region corrupted DynamoDB records, causing failures across services like Lambda, EC2, and API Gateway. Lasting nearly 15 hours, it generated over 17 million Downdetector reports and impacted platforms including Snapchat, Netflix, Roblox, and banking apps.

Just days later, on Oct. 29, Microsoft Azure suffered an eight-hour global disruption from an inadvertent configuration change in Azure Front Door, leading to DNS issues, latencies, and timeouts. This affected Microsoft 365, Xbox Live, Minecraft, Copilot, and enterprises in aviation, telecom, and retail.

In June, Google Cloud faced a multi-hour outage after a faulty quota policy update with blank fields crashed its Service Control system, resulting in widespread 503 errors. It disrupted Google Cloud Platform services like Compute Engine and Cloud Storage, while rippling to external platforms such as Spotify, Discord, Snapchat, and Cloudflare-dependent sites.