Skip to content

az ml create update does not return output #28476

@yavorivanov-cw

Description

@yavorivanov-cw

Describe the bug

Our pipelines stopped working because the az ml job create returns only warnings and no other output. The job is created and running, but we need the output to get the name of the job.
Command is implemented in a Azure DevOps pipeline.

Related command

az ml job create -f train_job.yml \
-g $(azure.resourceGroup) -w $(azureml.WorkspaceName) \
--set trial.environment=azureml:$(amlEnvironmentName)@latest \
--set experiment_name=$(experimentName) \
--set display_name=$(displayName) \
--set compute=azureml:$(azureml.cpuComputeName) --query name -o tsv

Errors

WARNING: Failed to deserialize response: Value 'Null' passed is not in set ['spot', 'basic', 'standard', 'premium']

Issue script & Debug output

DEBUG: cli.azext_mlv2.manual.custom.utils: Traceback (most recent call last):
File "/opt/az/azcliextensions/ml/azext_mlv2/manual/custom/utils.py", line 62, in _dump_entity_with_warnings
return entity._to_dict() # type: ignore
^^^^^^^^^^^^^^^^^
File "/opt/az/azcliextensions/ml/marshmallow/schema.py", line 517, in _serialize
value = field_obj.serialize(attr_name, obj, accessor=self.get_attribute)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/opt/az/azcliextensions/ml/marshmallow/fields.py", line 340, in serialize
return self._serialize(value, attr, obj, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/opt/az/azcliextensions/ml/azext_mlv2/manual/vendored_curated_sdk/azure/ai/ml/_schema/core/fields.py", line 73, in _serialize
raise ValidationError(f"Value {value!r} passed is not in set {self.allowed_values}")
marshmallow.exceptions.ValidationError: Value 'Null' passed is not in set ['spot', 'basic', 'standard', 'premium']

DEBUG: cli.knack.cli: Event: CommandInvoker.OnTransformResult [<function _resource_group_transform at 0x7f70ac9b6e80>, <function _x509_from_base64_to_hex_transform at 0x7f70ac9b6f20>]
DEBUG: cli.knack.cli: Event: CommandInvoker.OnFilterResult []
DEBUG: cli.knack.cli: Event: Cli.SuccessfulExecute []
DEBUG: cli.knack.cli: Event: Cli.PostExecute [<function AzCliLogging.deinit_cmd_metadata_logging at 0x7f70ac971bc0>]
INFO: az_command_data_logger: exit code: 0
INFO: cli.main: Command ran in 5.133 seconds (init: 0.184, invoke: 4.949)
INFO: cli.azure.cli.core.decorators: Suppress exception:
Traceback (most recent call last):
File "/opt/az/lib/python3.11/site-packages/azure/cli/main.py", line 62, in
raise ex
File "/opt/az/lib/python3.11/site-packages/azure/cli/main.py", line 55, in
sys.exit(exit_code)
SystemExit: 0

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
File "/opt/az/lib/python3.11/site-packages/azure/cli/core/decorators.py", line 79, in _wrapped_func
return func(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^
File "/opt/az/lib/python3.11/site-packages/azure/cli/core/telemetry.py", line 532, in _get_secrets_warning_config
show_secrets_warning = _get_config().getboolean('clients', 'show_secrets_warning', fallback=None)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/opt/az/lib/python3.11/site-packages/knack/config.py", line 147, in getboolean
raise ValueError('Not a boolean: {}'.format(val))
ValueError: Not a boolean: None

INFO: telemetry.main: Begin splitting cli events and extra events, total events: 1
INFO: telemetry.client: Accumulated 0 events. Flush the clients.
INFO: telemetry.main: Finish splitting cli events and extra events, cli events: 1
INFO: telemetry.save: Save telemetry record of length 3470 in cache
INFO: telemetry.main: Begin creating telemetry upload process.
INFO: telemetry.process: Creating upload process: "/opt/az/bin/python3 /opt/az/lib/python3.11/site-packages/azure/cli/telemetry/init.py /home/vsts/work/_temp/.azclitask"
INFO: telemetry.process: Return from creating process
INFO: telemetry.main: Finish creating telemetry upload process.

Expected behavior

No warnings and get the output/response from the command.

Environment Summary

azure-cli 2.57.0

core 2.57.0
telemetry 1.1.0

Extensions:
azure-devops 1.0.0
ml 2.23.0

Dependencies:
msal 1.26.0
azure-mgmt-resource 23.1.0b2

Additional context

While using the --debug option we can see a response with the information about the job, but most likely because of the warning we cannot get it as normal output of the command. Also --only-show-errors does not help. A workaround that works is to use ml version 2.22.0.

Metadata

Metadata

Assignees

Labels

Auto-AssignAuto assign by botMachine Learningaz mlService AttentionThis issue is responsible by Azure service team.act-codegen-extensibility-squadbugThis issue requires a change to an existing behavior in the product in order to be resolved.customer-reportedIssues that are reported by GitHub users external to the Azure organization.

Type

No type

Projects

No projects

Milestone

No milestone

Relationships

None yet

Development

No branches or pull requests

Issue actions