AWS Outage Crashes Amazon, Prime Video, Fortnite, Perplexity and More

Date of Data Posted: 2025‑10‑20

What You Need to Be Aware Of

  • Millions of websites, including major e‑commerce, streaming, gaming and AI services, experienced downtime.
  • The outage originated in the US‑EAST‑1 region and was linked to a DNS resolution issue affecting the DynamoDB API endpoint.
  • Services such as Amazon.com, Prime Video, Fortnite login, Perplexity AI chat, Canva design tools, Roblox, Hulu, Robinhood, Grammarly, and others were impacted.

How It Might Effect You

  • Operational Impact: Any business relying on AWS services in US‑EAST‑1 may see latency spikes or service failures, potentially disrupting customer experience or internal processes.
  • Security Exposure: Prolonged outages can delay patch deployment, log collection, and incident response, increasing the window for attackers to exploit unpatched vulnerabilities.
  • Compliance & Legal Risks: Service disruptions that affect data availability could breach SLAs, regulatory obligations (e.g., GDPR, PCI‑DSS), or contractual commitments to customers.

Mitigation Steps

  1. Immediate Actions – Short‑Term:
    • Verify the health of your AWS resources via the AWS Health Dashboard and monitor CloudWatch metrics for latency or error spikes.
    • If you experience downtime, switch traffic to a standby region (e.g., US‑WEST‑1) if you have multi‑region architecture.
    • Communicate outage status to stakeholders using your incident management system.
  2. Long‑Term Measures – Strategic:
    • Design for geographic redundancy: replicate critical workloads across multiple regions and enable Route 53 failover routing.
    • Implement automated health checks that trigger DNS failover or traffic shifting when thresholds are breached.
    • Conduct a post‑mortem review to identify single points of failure in your architecture.
    • Update incident playbooks to include procedures for third‑party outages, ensuring clear escalation paths and communication channels.

Sources
– BleepingComputer – 2025‑10‑20
– AWS Health Dashboard update on the outage – 2025‑10‑20