It happens a lot, when you have so much infrastructure and redundancy you think it is too big to fail. Then you lose S3 in US-East1 and break everything.

https://www.theregister.com/2017/03/01/aws_s3_outage/