Occasionally our internal network can be unreliable so we end up relying on retries in some requests.
There are no retries within the Authentication classes(specifically BearerAuth) so our pipelines can occasionally fail when encountering a short lived 503 or 504.
Occasionally our internal network can be unreliable so we end up relying on retries in some requests.
There are no retries within the Authentication classes(specifically BearerAuth) so our pipelines can occasionally fail when encountering a short lived 503 or 504.
Would it be alright to add retries for authentication here? I would add with the same set of request codes as laid out here: https://github.com/blackducksoftware/hub-rest-api-python/blob/master/blackduck/Client.py#L34