diff --git a/sdk/python/kubeflow/training/api/training_client.py b/sdk/python/kubeflow/training/api/training_client.py index c3ee98b267..782863586b 100644 --- a/sdk/python/kubeflow/training/api/training_client.py +++ b/sdk/python/kubeflow/training/api/training_client.py @@ -871,7 +871,8 @@ def wait_for_job_conditions( Raises: ValueError: Invalid input parameters. TimeoutError: Timeout to get Job. - RuntimeError: Failed to get Job or Job reaches unexpected Failed condition. + RuntimeError: Failed to get Job, or Job reaches Failed condition and + Failed is not in `expected_conditions` set. """ namespace = namespace or self.namespace @@ -904,9 +905,9 @@ def wait_for_job_conditions( if callback: callback(job) - # Raise an exception if Job is Failed and Failed is not expected condition. + # Raise an exception if Job is Failed and Failed is not the expected condition. if ( - constants.JOB_CONDITION_FAILED not in conditions + constants.JOB_CONDITION_FAILED not in expected_conditions and utils.has_condition(conditions, constants.JOB_CONDITION_FAILED) ): raise RuntimeError(