René's URL Explorer Experiment

{"@context":"https://schema.org","@type":"DiscussionForumPosting","headline":"ServiceUnavailable missing in polling retries","articleBody":"\"503 service unavailable\" errors are not being retried when long running operations (e.g. [dataproc create_cluster](https://googleapis.dev/python/dataproc/latest/dataproc_v1/services.html?highlight=create_cluster#google.cloud.dataproc_v1.services.cluster_controller.ClusterControllerClient.create_cluster)) are polled.\r\n\r\nCurrently, the [default retry predicate (retry.py)](https://github.com/googleapis/python-api-core/blob/ed092271ab94c7ef53c40ce8e73ca2c48e97159c/google/api_core/retry.py#L100) includes `google.api_core.exceptions.ServiceUnavailable`, but the [polling retry predicate (polling.py)](https://github.com/googleapis/python-api-core/blob/ed092271ab94c7ef53c40ce8e73ca2c48e97159c/google/api_core/future/polling.py#L32) overrides the former and doesn't include ServiceUnavailable. This doesn't look intentional: if retry.py considers ServiceUnavailable to be transient and retriable, so should polling.py.\r\n\r\nThis is causing issues for customers: e.g. their Cloud Function crashes during polling when they encounter a single `connection reset`.","author":{"url":"https://github.com/hoonji","@type":"Person","name":"hoonji"},"datePublished":"2021-05-10T08:06:40.000Z","interactionStatistic":{"@type":"InteractionCounter","interactionType":"https://schema.org/CommentAction","userInteractionCount":0},"url":"https://github.com/183/python-api-core/issues/183"}