Share via

Azure Data Factory Power Query activity fails with “Invalid dataset reference” (auto-generated dataset)

Wen Jie Chin 20 Reputation points
2026-02-27T03:47:12.4966667+00:00

We are experiencing a consistent failure in Azure Data Factory when running pipelines that contain a Power Query activity.

Error:

  • Operation on target <pipeline name> failed: Invalid dataset reference.

Key observations:

  • The dataset name is auto-generated by the Power Query activity (GUID suffix).
  • The pipeline fails almost immediately (~1 second after start).
  • The dataset DOES exist in the factory.
  • Manually cloning / recreating the dataset using the exact same name does NOT resolve the issue.
  • No recent changes were made to pipelines, datasets, or linked services.
  • Other activity types (Copy) work correctly.
  • Issue appears isolated to Power Query activity only.

Environment:

  • Service: Azure Data Factory
  • Activity type: Power Query
  • Region: East US 2
  • Occurs across multiple pipelines

Impact:

  • Power Query activity is currently unusable in production pipelines.

This appears to be a backend regression in how Power Query resolves its internally generated datasets. There are similar reports from other tenants starting around Feb 24, 2026. Has Microsoft identified this as a known issue, and is there an ETA or recommended workaround?

Azure Data Factory
Azure Data Factory
An Azure service for ingesting, preparing, and transforming data at scale.
{count} votes

1 answer

Sort by: Most helpful
  1. Q&A Assist
    2026-02-27T03:47:23.68+00:00
    AI generated content. This question contains an answer created with AI from Q&A Assist. Learn more

    The provided information does not indicate that this specific “Invalid dataset reference” failure for Power Query activities (with auto‑generated datasets) is a documented or active known issue, and no ETA or official workaround is described.

    From the available documentation:

    • Power Query activities in Azure Data Factory use Power Query mash-ups and run on the ADF data flow Spark environment, but there is no mention of an “Invalid dataset reference” error or a regression in how internally generated datasets are resolved.
    • The known issues list for Azure Data Factory does not list a Power Query–specific dataset resolution problem similar to what is described.
    • The troubleshooting guides for pipelines, CI/CD, UX, and connectors do not document this error pattern for Power Query activities.

    Given that:

    • The dataset exists and cloning it does not help.
    • Other activities (such as Copy) work correctly.
    • The behavior is isolated to Power Query and appears to have started recently across tenants.

    The appropriate next step is to open an Azure support request so the product team can confirm whether this is a new backend regression and provide an ETA or hotfix/workaround. The known issues page explicitly advises submitting a support request when an issue is not already listed.


    References:

    0 comments No comments

Your answer

Answers can be marked as 'Accepted' by the question author and 'Recommended' by moderators, which helps users know the answer solved the author's problem.