Massive AWS Outage Disrupts Major Apps; DNS Issue in us-east-1 Blamed
A significant Amazon Web Services (AWS) outage impacted a wide range of apps and websites, from Alexa and Venmo to Snapchat, Fortnite, Reddit, Lyft, and Disney+. The incident was tied to DNS resolution issues affecting DynamoDB API endpoints in the US-EAST-1 (N. Virginia) region, one of AWS’s busiest hubs.
- Root cause: DNS resolution problems for regional DynamoDB service endpoints in us-east-1.
- Timeline (ET): Error rates began overnight (around 3:11 AM); DNS issue mitigated by early morning; lingering impacts affected EC2 instance launches into the afternoon; AWS reported full resolution by evening.
- Services hit: Reports included Alexa, Venmo, Lyft, Snapchat, Canva, Fortnite, Reddit, Disney+, Apple Music, Pinterest, Roblox, banks, airlines, and news sites.
- Knock-on effects: Even after DNS stabilized, AWS limited new EC2 launches in us-east-1 to aid recovery and advised avoiding zone-specific deployments during restoration.
AWS noted that most services recovered progressively once the DNS issue was mitigated, but cascading dependencies caused elevated errors across multiple services. One expert described it as “apps being separated from their data,” highlighting how platform-wide DNS issues can temporarily sever access to otherwise intact data stores.
Why it matters
As of mid-2025, AWS holds roughly 30% of global cloud infrastructure share, making outages in us-east-1 highly visible. The event underscores concentration risk and the need for multi-region architectures, graceful degradation, and tested failover plans for critical workloads.
Key links and status pages
Takeaways for teams on AWS
- Design for region-level failures (active-active or warm standby across regions).
- Avoid single-AZ coupling and use health checks that can bypass a failing zone.
- Use DNS and service discovery patterns that can tolerate resolver or endpoint issues.
- Regularly game-day failover procedures and test rate-limit behavior during recoveries.
Even with resilient cloud architectures, widespread outages can still ripple across the internet given shared dependencies. Did this one affect your stack or daily apps?
Discussion: What resiliency changes will you consider after this outage—multi-region, alternative providers, or improved incident comms?