Forum Discussion

Irwan's avatar
Irwan
Super User
2 months ago
Solved

Need Advice: Dataflow Gen1 vs Semantic Model

hello experts,   need an advice here. in my current project, i tried to do exact same transformation step in both dataflow gen1 and semantic model. however, the result show that the refresh i...
Need Help
  • MFelix's avatar
    MFelix
    2 months ago

    Hi Irwan ,

     

    Without additional information about the error in question on the Dataflow is difficult to pin point the problem. As mentioned by v-echaithra  the way both artifact work differ so the problem may be caused by different things.

     

    If you can get the information of the error that would be helpfull.

  • v-echaithra's avatar
    v-echaithra
    2 months ago

    Hi Irwan ,

    Based on the error message (`Container exited unexpectedly with code 0x0000DEAD` and `MashupValueException`), this is often related to the Dataflow Gen1 mashup engine running into a resource or processing limitation.

    A few things you could try:

    Disable transformation steps one at a time to identify whether a specific step is causing the failure.
    Review any Merge, Group By, Pivot/Unpivot, or custom column operations, as these can be memory-intensive in Dataflow Gen1.
    Check whether query folding is being broken early in the query.
    Filter the source data to a smaller subset and test the refresh. If it succeeds, the issue may be related to data volume.
    If possible, move the transformation to a Semantic Model, Dataflow Gen2, or the source system, as these can handle some workloads more efficiently.

    Since the same logic succeeds in the Semantic Model, the transformation itself appears valid, which points more toward a Dataflow Gen1 execution limitation than a query logic issue.

    Could you share the applied steps or M code? That may help identify the specific operation triggering the container exit.

    Thank you.

v-echaithra's avatar
v-echaithra
Community Support
2 months ago

Hi  Irwan ,

From the screenshots, I can see that the Semantic Model refresh completed successfully while the Dataflow Gen1 refresh failed.

Although the transformation steps may be identical, Dataflow Gen1 and Semantic Models use different execution paths and resource handling, so a query can succeed in one and fail in the other.

Could you share the detailed error message from the Dataflow Gen1 refresh history? That would help identify whether the failure is related to a specific transformation, connector limitation, or resource constraint.

Thank you.