[BUG] start_run() modifies tags dictionary

[matheusMoreno]

Willingness to contribute

The MLflow Community encourages bug fix contributions. Would you or another member of your organization be willing to contribute a fix for this bug to the MLflow code base?

• Yes. I can contribute a fix for this bug independently.
• Yes. I would be willing to contribute a fix for this bug with guidance from the MLflow community.
• No. I cannot contribute a bug fix at this time.

System information

• Have I written custom code (as opposed to using a stock example script provided in MLflow): No
• OS Platform and Distribution (e.g., Linux Ubuntu 16.04): Linux Ubuntu 20.04.3 LTS
• MLflow installed from (source or binary): binary (pip)
• MLflow version (run mlflow --version): 1.22.0
• Python version: 3.9.5
• npm version, if running the dev UI: -
• Exact command to reproduce: -

Describe the problem

If you pass a dictionary variable to the start_run() function, it may modify the dictionary. On child runs, the mlflow.parentRunId tag is set and updated on the variable, making it persistent for later (non-nested) runs.

Code to reproduce issue

import mlflow

GLOBAL_TAGS = {'custom': 'tag'}

with mlflow.start_run(tags=GLOBAL_TAGS) as run:
    run_id = run.info.run_id
    assert run.data.tags.get('mlflow.parentRunId') is None      # Should be None
    with mlflow.start_run(tags=GLOBAL_TAGS, nested=True) as child_run:
        assert child_run.data.tags.get('mlflow.parentRunId') == run_id

with mlflow.start_run(tags=GLOBAL_TAGS) as run:
    assert run.data.tags.get('mlflow.parentRunId') is None      # Should be None

This code raises the following error on my machine:

Traceback (most recent call last):
  File "/home/matheus/mlflow_bug/bug.py", line 14, in <module>
    assert run.data.tags.get('mlflow.parentRunId') is run_id
AssertionError

Other info / logs

I believe the issue can be fixed by changing the following lines:

mlflow/mlflow/tracking/fluent.py

Lines 287 to 289 in 0fa849a

	tags = context_registry.resolve_tags(user_specified_tags)
	active_run_obj = client.create_run(experiment_id=exp_id_for_run, tags=tags)

tags is a parameter of the start_run() function; it should not be overwritten. By changing the variable name to something else, the problem will hopefully be solved. I will investigate further and open a PR if I'm able to fix the issue.

What component(s), interfaces, languages, and integrations does this bug affect?

Components

• area/artifacts: Artifact stores and artifact logging
• area/build: Build and test infrastructure for MLflow
• area/docs: MLflow documentation pages
• area/examples: Example code
• area/model-registry: Model Registry service, APIs, and the fluent client calls for Model Registry
• area/models: MLmodel format, model serialization/deserialization, flavors
• area/projects: MLproject format, project running backends
• area/scoring: MLflow Model server, model deployment tools, Spark UDFs
• area/server-infra: MLflow Tracking server backend
• area/tracking: Tracking Service, tracking client APIs, autologging

Interface

• area/uiux: Front-end, user experience, plotting, JavaScript, JavaScript dev server
• area/docker: Docker use across MLflow's components, such as MLflow Projects and MLflow Models
• area/sqlalchemy: Use of SQLAlchemy in the Tracking Service or Model Registry
• area/windows: Windows support

Language

• language/r: R APIs and clients
• language/java: Java APIs and clients
• language/new: Proposals for new client languages

Integrations

• integrations/azure: Azure and Azure ML integrations
• integrations/sagemaker: SageMaker integrations
• integrations/databricks: Databricks integrations