The graph below shows the overall failure rate, excluding 4xx client-side warnings.

Failures here include 5xx server-side errors, network errors in which no response is returned, slow errors in which a response is received after an exceptionally long period, and redirect errors in which a 3xx redirect HTTP status code is returned.

As expected, the overall failure rate is nearly identical for all clouds. Deprecated Yahoo! APIs caused the February 2019 peak, the December 2019-February 2020 peak by service issues with Cisco Spark, and the July-August 2020 one by Amazon. As these APIs were returning 5xx HTTP status codes to calls that usually passed, these can be taken as failures within the provider's control. Any 5xx indicates an issue with the API outside user control and is considered a failure.