Flaky MXNext e2e test

[tenzen-y]

        TRAINING_CLIENT.create_mxjob(mxjob, job_namespace)
        logging.info(f"List of created {constants.MXJOB_KIND}s")
        logging.info(TRAINING_CLIENT.list_mxjobs(job_namespace))
    
>       verify_job_e2e(
            TRAINING_CLIENT,
            JOB_NAME,
            job_namespace,
            constants.MXJOB_KIND,
            CONTAINER_NAME,
        )

sdk/python/test/e2e/test_e2e_mxjob.py:152: 
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ 
sdk/python/test/e2e/utils.py:36: in verify_job_e2e
    client.wait_for_job_conditions(name, namespace, job_kind, timeout=timeout)
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ 

full log

self = <kubeflow.training.api.training_client.TrainingClient object at 0x7fe7543da910>
name = 'mxjob-mnist-ci-test', namespace = 'default', job_kind = 'MXJob'
expected_conditions = {'Succeeded'}, timeout = 600, polling_interval = 15
callback = None, apiserver_timeout = 120

    def wait_for_job_conditions(
        self,
        name: str,
        namespace: str = utils.get_default_target_namespace(),
        job_kind: str = constants.TFJOB_KIND,
        expected_conditions: Set = {constants.JOB_CONDITION_SUCCEEDED},
        timeout: int = 600,
        polling_interval: int = 15,
        callback: Callable = None,
        apiserver_timeout: int = constants.DEFAULT_TIMEOUT,
    ):
        """Wait until Training Job reaches any of the specified conditions.
        By default it waits for the Succeeded condition.
    
        Args:
            name: Name for the Job.
            namespace: Namespace for the Job.
            job_kind: Kind for the Training job to wait for conditions.
                It should be one of these: `TFJob, PyTorchJob, MXJob, XGBoostJob, MPIJob, or PaddleJob`.
            expected_conditions: Set of expected conditions. It must be subset of this:
                `{"Created", "Running", "Restarting", "Succeeded", "Failed"}`
            timeout: How many seconds to wait until Job reaches one of
                the expected conditions.
            polling_interval: The polling interval in seconds to get Job status.
            callback: Optional callback function that is invoked after Job
                status is polled. This function takes a single argument which
                is current Job object.
            apiserver_timeout: Optional, Kubernetes API server timeout in seconds
                to execute the request.
    
        Returns:
            object: Training Job object of type `KubeflowOrgV1TFJob, KubeflowOrgV1PyTorchJob,
            KubeflowOrgV1MXJob, KubeflowOrgV1XGBoostJob, KubeflowOrgV1MPIJob, or
            KubeflowOrgV1PaddleJob` which is reached required condition.
    
        Raises:
            ValueError: Expected conditions are invalid or Job kind is invalid
            TimeoutError: Timeout to get Job.
            RuntimeError: Failed to get Job.
        """
    
        if not expected_conditions.issubset(constants.JOB_CONDITIONS):
            raise ValueError(
                f"Expected conditions: {expected_conditions} must be subset of {constants.JOB_CONDITIONS}"
            )
        for _ in range(round(timeout / polling_interval)):
    
            # We should get Job only once per cycle and check the statuses.
            job = utils.get_job(
                custom_api=self.custom_api,
                api_client=self.api_client,
                name=name,
                namespace=namespace,
                job_model=constants.JOB_KINDS[job_kind]["model"],
                job_kind=job_kind,
                job_plural=constants.JOB_KINDS[job_kind]["plural"],
                timeout=apiserver_timeout,
            )
            conditions = self.get_job_conditions(
                name, namespace, job_kind, job, timeout
            )
            if len(conditions) > 0:
                status_logger(
                    name, conditions[-1].type, conditions[-1].last_transition_time,
                )
            # Execute callback function.
            if callback:
                callback(job)
    
            # Raise an exception if Job is Failed and Failed is not expected condition.
            if (
                constants.JOB_CONDITION_FAILED not in conditions
                and utils.has_condition(conditions, constants.JOB_CONDITION_FAILED)
            ):
>               raise RuntimeError(
                    f"{job_kind} {namespace}/{name} is Failed. "
                    f"{job_kind} conditions: {job.status.conditions}"
                )
E               RuntimeError: MXJob default/mxjob-mnist-ci-test is Failed. MXJob conditions: [{'last_transition_time': datetime.datetime(2023, 5, 31, 12, 26, 48, tzinfo=tzlocal()),
E                'last_update_time': datetime.datetime(2023, 5, 31, 12, 26, 48, tzinfo=tzlocal()),
E                'message': 'MXJob mxjob-mnist-ci-test is created.',
E                'reason': 'MXJobCreated',
E                'status': 'True',
E                'type': 'Created'}, {'last_transition_time': datetime.datetime(2023, 5, 31, 12, 27, 2, tzinfo=tzlocal()),
E                'last_update_time': datetime.datetime(2023, 5, 31, 12, 27, 2, tzinfo=tzlocal()),
E                'message': 'MXJob mxjob-mnist-ci-test is running.',
E                'reason': 'MXJobRunning',
E                'status': 'False',
E                'type': 'Running'}, {'last_transition_time': datetime.datetime(2023, 5, 31, 12, 28, 1, tzinfo=tzlocal()),
E                'last_update_time': datetime.datetime(2023, 5, 31, 12, 28, 1, tzinfo=tzlocal()),
E                'message': 'mxjob mxjob-mnist-ci-test is failed because 1 Worker replica(s) '
E                           'failed.',
E                'reason': 'MXJobFailed',
E                'status': 'True',
E                'type': 'Failed'}]

sdk/python/kubeflow/training/api/training_client.py:378: RuntimeError

[tenzen-y]

/kind kind/e2e-test-failure

[google-oss-prow]

@tenzen-y: The label(s) kind/kind/e2e-test-failure cannot be applied, because the repository doesn't have them.

In response to this:

/kind kind/e2e-test-failure

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository.

[tenzen-y]

/kind e2e-test-failure

[github-actions]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.

[tenzen-y]

/remove-lifecycle stale

[tenzen-y]

/remove-lifecycle stale

[tenzen-y]

There is a incident with Github webhook: https://www.githubstatus.com/

[github-actions]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.

[tenzen-y]

/lifecycle frozen

[tenzen-y]

MXJob was removed from the training-operator.
/close

[google-oss-prow]

@tenzen-y: Closing this issue.

In response to this:

MXJob was removed from the training-operator.
/close

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository.