Known Issues and Limitations

Section Contents

Known Issues and Limitations#

A list of currently known issues and limitations can be found below. This page is actively maintained and updated with each release of the ibm-watsonx-data-integration SDK for Python.

Known Issues#

  • Users may encounter HTTP 429 errors (‘too many requests’) when making calls from the SDK.

  • Changes to batch flow runtime parameters may not appear in the UI. If you are strictly using the SDK, this will not be a problem.

  • When retrieving subflows that contain subflow stages, you cannot modify that subflow’s DAG. You can work around this by downloading the flow and using the PythonGenerator to create the flow.

  • When using PythonGenerator on a batch flow that has runtime settings or parameters changed at the flow level, the flow will be generated correctly, but the runtime settings and parameters will not.

  • BatchFlow objects with unsupported stage types can result in the flow not being serialized into a Python object.

  • The SDK incorrectly allows connecting outputs to source-only stages in streaming flows. These stages should only act as data sources and cannot accept inputs from other stages, but the SDK does not enforce this restriction.

Limitations#

  • The SDK only partially supports DataStage (batch processing) functionality at this point. DataStage components will be added in future releases.

  • Most SDK models use the name attribute for display names and searching. However, Stages inconsistently uses stage_name instead of name for this purpose.

  • Streaming python generator does not support parameter sets.

Section Contents