Azure data factory - ADF - Inline Sink Delta

jmarciogsousa 6 Reputation points
2020-08-16T07:52:16.24+00:00

I'm getting the error for the ADF data flow - Delta Sink on key column, but I did the schema import and the column exists:
"Key column 'undefined' does not exist in mapped columns."

Some clue about the problem, looks like a bug, I know the Delta Sink is on preview
17769-image.png

Azure Data Factory
Azure Data Factory
An Azure service for ingesting, preparing, and transforming data at scale.
11,624 questions
{count} votes

6 answers

Sort by: Most helpful
  1. Jack Ma 161 Reputation points
    2020-10-21T01:48:26.293+00:00

    Hi @jmarciogsousa @Wesley Tremayne ,

    Thanks for reporting this issue! It’s a bug on our UX tool, it complains the key column as “undefined” because the schema of sink is not published so that the UX tool can’t recognize it properly. We will create a work item for it and fix this issue soon.

    Meanwhile as a workaround, you can try:

    1. Uncheck delete/upsert/update and only leave insert for the delta sink -> import the schema from “Mapping” panel then save/publish the data flow.
    2. Once saved/published, click “Refresh” button on the page, navigate to the data flow and set delete/upsert/update + key columns again.
      33903-refresh.jpg
    2 people found this answer helpful.

  2. jmarciogsousa 6 Reputation points
    2020-08-18T12:45:35.727+00:00

    I get it working with the procedure

    1. Mapping
    2. insert
    3. publish
    4. refresh page -> if not works go to 5.
    5. turn off debug mode
    6. delete and key column
    7. publish
    1 person found this answer helpful.

  3. Kiran-MSFT 696 Reputation points Microsoft Employee
    2020-08-17T03:10:36.9+00:00

    Check to make sure you have mapped "prrowid" column in the mapping tab. It is most likely left unmapped.

    0 comments No comments

  4. Balan Murugan 1 Reputation point
    2020-09-25T06:23:25.463+00:00

    Hello Everyone,

    I am facing the same issue. My requirement is to build metadata driven mapping data flow (All Parameterized way).

    I have to read multiple CSV files (All are different files having differing schema) from raw_zone and do SCD 1 or 2 based on the parameterized key columns. I have enabled Azure-DevOps in my ADF. Working on feature branch and I could not publish from feature branch.

    Key Points to Note:

    1. In Source transformation, enabled schema drift and thus no projection.
    2. In Sink transformation, specifying the key columns using byName('Drifted_Column') function. Auto-Mapping is selected in Sink.28233-source.png28283-sink.png28301-sink-settings.png28170-dataflow-script.txt

    Could you please help me for the solution ? @jmarciogsousa

    0 comments No comments

  5. jmarciogsousa 6 Reputation points
    2020-09-25T08:27:30+00:00

    @Balan Murugan , you can try refresh the page if not stop-start debug mode


Your answer

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