Amazon Web Services (AWS) experienced a significant outage starting late on October 19, 2025, primarily affecting the US-EAST-1 region (Northern Virginia). The root cause was a DNS resolution failure for the DynamoDB service. AWS engineers have mitigated the main DNS issue by early morning October 20, but some service disruptions and increased error rates, especially for launching new EC2 instances, persisted for several hours afterward. AWS indicated that full restoration might take some additional time due to server backlogs and propagation of fixes across the internet. While the DNS problem was resolved around 2:24 AM PDT on October 20, customer impact continued with gradual recovery through the day. Amazon stated its services have returned to normal operations after about 15 hours of outage, with ongoing monitoring and work to clear backlogs and prevent future disruptions. Users were advised to flush DNS caches to resolve any lingering endpoint issues immediately.
In summary, AWS is expected to be fully back up following the October 19-20 outage by the end of October 20, 2025, with core services mostly operational now but some residual delays clearing.
