An outage in Amazon's US-EAST-1 region caused widespread chaos, taking websites and services offline globally, including in Europe.
The problems began with increased error rates and latencies for multiple services, and Amazon's techies identified DNS as a potential root cause, specifically the resolution of the DynamoDB API endpoint.
After all, cloud operations are supposed to have some built-in resiliency, right?
The outage raised difficult questions about the dependence on one cloud provider and one location.
Author's summary: AWS outage causes global chaos due to service dependencies.