topic Re: Do errors get logged as part of the artifact tracking in MLflow? Or is there a way to log errors in general? in Data Engineering https://community.databricks.com/t5/data-engineering/do-errors-get-logged-as-part-of-the-artifact-tracking-in-mlflow/m-p/22004#M15026 <P>For the tracking server? Yes, it does produce logs which you could see if running the tracking server as a standalone service. They are not exposed from the hosted tracking server in Databricks. However there typically aren't errors or logs of interest in the tracking server; errors of interest would occur (and be logged) from the MLflow client that is accessing it, for example, in a notebook cell.</P><P></P><P>For model serving, errors are likewise logged, but would be of interest and are exposed in the hosted MLflow serving service. You would see them in the Serving tab of a registered model. Or if deploying to Azure ML or SageMaker, these logs would likewise appear in the console of those services.</P> Tue, 22 Jun 2021 09:50:53 GMT sean_owen 2021-06-22T09:50:53Z Do errors get logged as part of the artifact tracking in MLflow? Or is there a way to log errors in general? https://community.databricks.com/t5/data-engineering/do-errors-get-logged-as-part-of-the-artifact-tracking-in-mlflow/m-p/22003#M15025 Mon, 21 Jun 2021 21:15:25 GMT https://community.databricks.com/t5/data-engineering/do-errors-get-logged-as-part-of-the-artifact-tracking-in-mlflow/m-p/22003#M15025 Anonymous 2021-06-21T21:15:25Z Re: Do errors get logged as part of the artifact tracking in MLflow? Or is there a way to log errors in general? https://community.databricks.com/t5/data-engineering/do-errors-get-logged-as-part-of-the-artifact-tracking-in-mlflow/m-p/22004#M15026 <P>For the tracking server? Yes, it does produce logs which you could see if running the tracking server as a standalone service. They are not exposed from the hosted tracking server in Databricks. However there typically aren't errors or logs of interest in the tracking server; errors of interest would occur (and be logged) from the MLflow client that is accessing it, for example, in a notebook cell.</P><P></P><P>For model serving, errors are likewise logged, but would be of interest and are exposed in the hosted MLflow serving service. You would see them in the Serving tab of a registered model. Or if deploying to Azure ML or SageMaker, these logs would likewise appear in the console of those services.</P> Tue, 22 Jun 2021 09:50:53 GMT https://community.databricks.com/t5/data-engineering/do-errors-get-logged-as-part-of-the-artifact-tracking-in-mlflow/m-p/22004#M15026 sean_owen 2021-06-22T09:50:53Z