Amazon Web Services Outage Disrupts Global Internet Services
Just the facts

Amazon Web Services Outage Disrupts Global Internet Services

Summary

A 15-hour AWS outage on October 20, 2025, caused by a DNS failure in the US-EAST-1 region, led to widespread service disruptions and significant financial losses worldwide.

On October 20, 2025, Amazon Web Services (AWS) experienced a 15-hour outage originating from its US-EAST-1 region in Northern Virginia. The disruption began at 12:11 AM ET due to a Domain Name System (DNS) resolution failure affecting DynamoDB, AWS's core database service. This failure prevented applications from locating necessary servers, leading to a cascading effect that impacted numerous services globally.

The outage affected over 3,500 companies across more than 60 countries, with major platforms like Snapchat, AWS, and Roblox among the most reported services impacted. According to network intelligence company Ookla, this event was "among the largest internet outages on record for Downdetector." The financial repercussions were substantial, with estimates suggesting global losses exceeding $1 billion due to halted online payments, delayed logistics, and interrupted operations.

AWS engineers identified the root cause as a software bug in the DynamoDB DNS management system, specifically within the DNS Enactor component. This bug led to the deletion of all IP addresses for the regional endpoint, necessitating manual intervention to restore services. The failure also strained Amazon's EC2 services in the US-EAST-1 region, causing delays in network state propagation and further service disruptions.

In response, Amazon has disabled the DynamoDB DNS Planner and DNS Enactor automation worldwide while working to fix the issue and implement safeguards against similar failures. The incident underscores the fragility of centralized cloud infrastructure and highlights the need for businesses to adopt multi-region or multi-cloud strategies to mitigate such risks.

The AWS outage serves as a cautionary tale for the digital economy, emphasizing the importance of designing systems with resilience and redundancy to prevent widespread disruptions from single points of failure.

Fact-checking

Fact-check the facts of the article using external sources and databases.

!
Partly Confirmed

On October 20, 2025, Amazon Web Services (AWS) experienced a 15-hour outage originating from its US-EAST-1 region in Northern Virginia.

✓
Confirmed

The disruption began at 12:11 AM ET due to a Domain Name System (DNS) resolution failure affecting DynamoDB, AWS's core database service.

!
Partly Confirmed

The outage affected over 3,500 companies across more than 60 countries, with major platforms like Snapchat, AWS, and Roblox among the most reported services impacted.

!
Unverified

According to network intelligence company Ookla, this event was 'among the largest internet outages on record for Downdetector.'

!
Unverified

The financial repercussions were substantial, with estimates suggesting global losses exceeding $1 billion due to halted online payments, delayed logistics, and interrupted operations.

!
Unverified

AWS engineers identified the root cause as a software bug in the DynamoDB DNS management system, specifically within the DNS Enactor component.

!
Unverified

In response, Amazon has disabled the DynamoDB DNS Planner and DNS Enactor automation worldwide while working to fix the issue and implement safeguards against similar failures.

✓
Confirmed

The incident underscores the fragility of centralized cloud infrastructure and highlights the need for businesses to adopt multi-region or multi-cloud strategies to mitigate such risks.

Keep reading — for free

Create a free account to follow the news. No card required.