Between 4:05 PM and 4:39 PM PST, we experienced increased error rates and latencies for the EC2 APIs, and elevated launch failures for new EC2 instances in the US-WEST-2 Region. The issue has been resolved and the service is operating normally.