none
Copy activity in Azure Data Factory RRS feed

  • Question

  • Hi,

    In a Copy Data Activity, we are getting following error while the Source and Sink are Azure Data lake Storage types. We try to copy one file path (folders/sub folders with hierarchy structure) to another file path.

    "errorCode": "2200",
        "message": "ErrorCode=UserErrorFormatIsRequired,'Type=Microsoft.DataTransfer.Common.Shared.HybridDeliveryException,Message=Format setting is required for file based store(s) in this scenario.,Source=Microsoft.DataTransfer.ClientLibrary,'",
        "failureType": "UserError",
        "target": "copydata"
    }

    Can anybody help. please

    Srinivasan


    Srinivasan

    Wednesday, November 21, 2018 10:22 AM

All replies

  • Hi Srinivasan,

    The error message is saying you didn't specify the file format in source/sink dataset, would you please clarify the copy scenario in your side? This doc describes the details of the file format in the copy activity. Or you could provide us with the failed activity runID to identify the problem.

    Thanks.

    Wednesday, November 21, 2018 1:53 PM
  • Hi Srinivasan -  I have  a similar issue occurring, as well. I have opened a support ticket with Microsoft 


    118112121004011

    We are using ADF to pull for source and target all file to an Azure data lake.   In the middle of the pipe we are doing the merge to support file compression of the file between the source and target. The issue seems to be occurring during this process since this the only time we are using merge.  Also, this is impacting every pipe we have running and have had pipes running for 5 months.

    { "errorCode": "2200", "message": "ErrorCode=UserErrorFormatRequiredWithCopyBehaviorMergeFiles,'Type=Microsoft.DataTransfer.Common.Shared.HybridDeliveryException,Message=Format setting is required on both source and sink for 'MergeFiles' copy behavior.,Source=Microsoft.DataTransfer.ClientLibrary,'", "failureType": "UserError", "target": "Copy F0911 from E1" }

    Wednesday, November 21, 2018 5:07 PM
  • Hi Mike,

    You can use the ZipDeflate compression type in target dataset to merge the multiple files into single zip file.

    If you don't want the zip file, then the format info is required so that ADF can parse the source file correctly and then serialize into single file. Please specify the format info in the source and sink dataset to make your copy work.


    Wednesday, November 21, 2018 11:59 PM