Experiment warning when legacy artifact storage location is used

Resolve experiment warnings when a legacy artifact storage location is used instead of the MLflow managed location.

Written by Adam Pavlacka

Last published at: May 16th, 2022

Problem

A new icon appears on the MLflow Experiments page with the following open access warning:

MLflow legacy artifact location open access warning.

Cause

MLflow experiment permissions (AWS | Azure | GCP) are enforced on artifacts in MLflow Tracking, enabling you to easily control access to datasets, models, and other files.

In MLflow 1.11 and above, new experiments store artifacts in an MLflow-managed location (dbfs:/databricks/mlflow-tracking/) that enforces experiment access controls.

Certain older experiments use a legacy storage location (dbfs:/databricks/mlflow/) that can be accessed by all users of your workspace.

This warning indicates that your experiment uses a legacy artifact storage location.

Solution

You should always use the MLflow-managed DBFS storage locations when logging artifacts to experiments. This protects against unintended or unauthorized access to your MLflow artifacts.

Was this article helpful?