enlarge the maximum waiting period to 4s (1s before)
use a larger base interval of 1s (100ms before)
Related Issue
150
Motivation and Context
This would better tune calls retries. We do not want in general to retry too much, and we want more time between each of the calls to wait for the called system to recover. Note that even in case of a retry after header present in the response, the maximum waiting period will be used in place of the time specified in the response to place the next call when lower:
if (e.retryAfter() != null) {
interval = e.retryAfter().getTime() - currentTimeMillis();
if (interval > maxPeriod) {
interval = maxPeriod;
}
which is ok to protect the service for hanging up for too long. 4s seems like a reasonable time to wait before retrying and eventually fail, also, the maximal waiting period that can be calculated with the given parameters by the exponential backoff strategy fits the maximum as well.
Description
Configure Retries in Feign utility:
Related Issue
150
Motivation and Context
This would better tune calls retries. We do not want in general to retry too much, and we want more time between each of the calls to wait for the called system to recover. Note that even in case of a retry after header present in the response, the maximum waiting period will be used in place of the time specified in the response to place the next call when lower:
which is ok to protect the service for hanging up for too long. 4s seems like a reasonable time to wait before retrying and eventually fail, also, the maximal waiting period that can be calculated with the given parameters by the exponential backoff strategy fits the maximum as well.
Another solution would be the approach proposed here: https://github.com/adobe/aio-lib-java/pull/149
How Has This Been Tested?
Screenshots (if appropriate):
Types of changes
Checklist: