AWS introduces DNS emergency system after US-East outages

Amazon has introduced a new feature for its DNS service, Route 53, which is intended to guarantee companies access to critical DNS management functions even during outages in the US-East region. The feature, called Accelerated Recovery, promises a Recovery Time Objective (RTO) of 60 minutes in the event of disruptions in the region, which is important for AWS, Northern Virginia.

The move is not without reason: the US-East-1 region has been considered a bottleneck in the AWS infrastructure for years. As recently as October 2025, there were severe outages that affected numerous services worldwide, including Signal, the Epic Games Launcher, and Amazon's own streaming services. The full error report revealed a complex cascade of problems triggered by faulty DNS entries in the DynamoDB management system.

AWS justifies the new feature with customer requests for "additional DNS resilience capabilities." Particularly regulated industries such as banks, FinTech companies, and SaaS providers need the assurance that they can make DNS changes even in the event of unexpected regional outages. This allows them to quickly deploy standby resources or reroute traffic.

Technically, accelerated recovery provides access to essential Route 53 API operations during an outage, including ChangeResourceRecordSets, GetChange, ListHostedZones, and ListResourceRecordSets. Customers can continue to use their existing API endpoints without having to modify applications or scripts.

Easy activation, But Criticism Remains

The feature is activated via the AWS Management Console, command line, or infrastructure-as-code tools such as CloudFormation and CDK. After selecting a hosted zone, a new tab labeled "Accelerated Recovery" appears, via which the function can be activated with a click. According to AWS, setup takes a few minutes, and there are no additional costs.

However, the promised 60-minute RTO remains problematic: during this time, outages can cause significant disruptions – a point that DNS expert Carsten Strotmann already criticized after the October outages. In his commentary, he clarified that DNS itself is not the problem, but rather faulty data that was written into DNS by AWS systems.

The mere existence of this new feature underscores the long-standing issue with the US-East region. Already in 2022, the analyst firm Gartner warned that the region represented a weak point in the AWS cloud and impaired crisis management capabilities. AWS itself argued in 2024 in an interview with The Register that the region was no less reliable than others but was more heavily burdened due to its colossal size.

The new feature is now available for all public Route 53 hosted zones. Details on the background and setup can be found in the AWS Blog. However, whether the 60-minute guarantee is sufficient to meet the regulatory requirements of highly regulated industries remains to be seen in practice.

(fo)