Multiple header rows in output file


  • In the last 5 days, there appears to have been a change to Azure Data Lake Analytics causing a header row to be written twice to an output file in USQL.  The header row is written at the beginning, and also appears in the middle of the input file.

    Is there a setting that may address this issue?

    This is causing ADF copy jobs to fail since ADF copy cannot skip invalid rows.

    This is the code outputting the file.  It is run from within a USQL stored procedure, scheduled as an hourly slice ADF job.

    OUTPUT @out
      TO @outputlocation
      USING Outputters.Tsv(outputHeader : true, quoting : false);

    Andrew Sears

    Thursday, December 8, 2016 8:11 PM

All replies

  • Surprising. Are you sure this second header is not the input header line that was parsed as a line of data?

    Can you reproduce? If you think it's a bug, please create a support ticket (directly from the Azure portal).

    Monday, December 12, 2016 6:44 PM
  • Thanks, fixed by support team.

    Andrew Sears

    Tuesday, December 13, 2016 2:13 AM
  • What do you mean with fixed by support team? Was it indeed a regression? Or some other reason? Thanks!

    Michael Rys

    Wednesday, December 14, 2016 8:43 AM