Retry strategy
Properties can be specified in the service definition which determine how to retry an API call should it fail.
The primary purpose of defining retry parameters is when an API is rate-limited, whereby the API only permits a certain number of calls to be made in a given time period.
Retry is controlled by these parameters:
- Number of retries - the maximum number of times that an API is called if it returns a failure status.
- Initial delay - the initial time interval that the process waits before retrying the API call.
- Delay multiplier - a decimal value that is multiplied by the delay value to extend the delay after each failing call. By default this is set to 1.0, which means that the delay between each call is the same. If set to a value larger than 1.0, then the delay between failing calls will be progressively larger. For example, if the multiplier is set to 2.0 and the initial delay time is 1000 ms (1 second), then the first interval will be 1 second, the second interval will be 2 seconds, the third interval will be 4 seconds, and so on.