Description of changes:
HTTP errors detected as aborted attempts are retried very quickly (2 ms), without exponential backoff. When those errors are caused by connection issues that are transient, but still take more than 10 ms to resolve, calls will fail. With this change, we would transition to using the regular retry loop (with exponential backoff) after the retry-fast attempts are exhausted.
By submitting this pull request, I confirm that my contribution is made under the terms of the Apache 2.0 license.
Issue #, if available:
Description of changes: HTTP errors detected as aborted attempts are retried very quickly (2 ms), without exponential backoff. When those errors are caused by connection issues that are transient, but still take more than 10 ms to resolve, calls will fail. With this change, we would transition to using the regular retry loop (with exponential backoff) after the retry-fast attempts are exhausted.
By submitting this pull request, I confirm that my contribution is made under the terms of the Apache 2.0 license.