Forum Discussion

alozovoy's avatar
alozovoy
Advocate II
2 years ago

Dataflow refresh transaction failed with status 22

Executing a dataflow within a Fabric pipeline frequently fails with the following error message:

 

Error code 20302 link directs to this page: Troubleshoot connectors - Azure Data Factory & Azure Synapse | Microsoft Learn

 

The dataflow copies a table from an Excel file on Sharepoint to a fabric lakehouse table. The Excel file typically does not change between each dataflow run. The dataflow runs successfully ~90% of the time, but sometimes fails with the above error message.

 

Has anyone else run into a similar error message?

Dataflows

9 Replies

  • Anonymous's avatar
    Anonymous
    Not applicable
    2 years ago

    Hi alozovoy 

    Apologies for the inconvenience.

    Please reach out to our support team to gain deeper insights and explore potential solutions. It's highly recommended that you reach out to our support team. Their expertise will be invaluable in suggesting the most appropriate approach.

    Please go ahead and raise a support ticket to reach our support team:

    https://support.fabric.microsoft.com/support

    After creating a Support ticket please provide the ticket number as it would help us to track for more information.

     

    Thank you.

    • Anonymous's avatar
      Anonymous
      Not applicable
      2 years ago

      Hi alozovoy 
      We haven’t heard from you on the last response and was just checking back to see if your query got resolved. Otherwise, will respond back with the more details and we will try to help.
      Thanks

      • JeffGray's avatar
        JeffGray
        Advocate I
        2 years ago

        I experienced exactly the same failure this morning, with exactly the same use case. . . Dataflow Gen 2 retrieving data from an Excel file on Sharepoint.  Ran succesfully each hour previous to this one run since the schedule was launched yesterday morning.  Failed at 12:42AM and then has run succesfully every hour since.

    • paul_hasell's avatar
      paul_hasell
      Regular Visitor
      1 year ago

      Hi, this is happening frequently but randomly for a customer, it's not workload or CU dependent, it's making the use of Fabric impossible for production workloads of even the lightest order. We'll likely have to revert to SQL/ADF for reliability despite MS promoting Fabric as an alternative.

  • Anonymous's avatar
    Anonymous
    Not applicable
    1 year ago

    (Appears to be the) Same issue experienced - ingestion from SQL Managed Instance, via On-premise Gateway.

    Will submit a support ticket and hope they can provide more information.

  • Shubham04's avatar
    Shubham04
    Microsoft Employee
    1 year ago

    Hi, 
    I'm facing this exact issue and seems there is no workaround mentioned anywhere. Do we have solution to this yet?

     

     

    Thanks

    Shubham

    • paul_hasell's avatar
      paul_hasell
      Regular Visitor
      1 year ago

      Hi, I raised a support call with MS and they are working on a fix but no ETA on it I'm afraid. The only 'reliable' workaround we have found is to forego the use of Lakehouse tables and use Azure SQL as the source/target instead. We're hoping that when the issue is final resolved we can use Lakehouse as it's a cleaner solution.

  • Max_'s avatar
    Max_
    Helper I
    1 year ago

    Hi All - This might not be the solution for alozovoy  issue, but I came across this post as I had the same issue in one of my pipelines. However, I discovered that the dataflow itself failed. Opening the "Recent runs" overview for the dataflow via the workspace gave a specific error message. With that, I was able to solve the error, so the dataflow and pipeline can be refreshed again.