Getting error "{"code":"BadRequest","message":null,"target":"pipeline..." when trying to debug pipeline

Anonymous
2021-03-08T02:37:14.197+00:00

Have confirmed through reproducing a couple of times. I have 3 data factories, dev/uat/prod. I develop something on DEV, then use Azure devops to run through and release the ARM templates onto UAT and prod. This all works OK.

However I then change the prod KeyVault Linked Service's default value to something else, using the UI. This is used in other Linked Services to obtain login information.

After changing the linked service "Default value" parameter only and publishing, I then get this error trying to debug a pipeline:

{"code":"BadRequest","message":null,"target":"pipeline//runid/XXX Run ID XXX","details":null,"error":null} (note I removed the GUID run ID in the code)

Seems like that should work without issue, might be a bug?

EDIT:

I have tried a simple pipeline which sets some variables, which works fine. However any pipeline that requires pass through to Key Vault doesn't seem to work. This was all working last week, so something seems to have been screwed up.

Azure Data Factory
Azure Data Factory
An Azure service for ingesting, preparing, and transforming data at scale.
9,643 questions
{count} votes

Accepted answer
  1. Saurabh Sharma 23,751 Reputation points Microsoft Employee
    2021-03-17T05:07:19.303+00:00

    @Anonymous Products team has identified this as a product bug and they have created an workitem internally and working on to provide a fix for the same. Meanwhile, you can use the workaround you have found to circumvent this issue. Thank you so much for reporting this. Appreciate your patience on the same.

    ----------

    Please do not forget to "Accept the answer" wherever the information provided helps you to help others in the community.

    3 people found this answer helpful.

5 additional answers

Sort by: Most helpful
  1. Trystan Mortimer 31 Reputation points
    2022-08-09T05:19:35.9+00:00

    I had this error when I had some trailing whitespace on an activity name.

    5 people found this answer helpful.

  2. Anonymous
    2021-03-10T23:47:50.143+00:00

    Right I think I found the problem.

    Background - I have been using the process described above for months. Developing on DEV version of data factory linked to GIT. These then get pushed through to PROD. I haven't changed any security settings or parameter settings on the linked services for a long time because they are generic.

    However, on the 3rd of March, after committing a change, data factory has added curly braces to the BaseURL parameter being passed through as a parameter to data factory. I can see this in the Repo commits, that it was changed. However I have not made this change, somehow it was added by the UI on publish. So something is up with the UI on linked services, it's adding curly braces to KeyVault parameters. The only way you can change it is to go into the Linked Service code of the key vault and change it in the JSON file. Subsequently other things using the KeyVault linked service will appear as [object Object] instead.

    3 people found this answer helpful.

  3. Grégoire Thibaud 6 Reputation points
    2021-04-12T14:19:20.55+00:00

    I had the same issue on my Data Factory. In my case, a Copy Data activity had generated automatically User properties (cf. attached image).
    By simply deleting these properties, I was able to run pipeline properly.

    87001-capture-decran-2021-04-12-a-161729.png

    1 person found this answer helpful.

  4. Mike McLean (Shell) 1 Reputation point
    2021-04-07T16:08:40.707+00:00

    This can also happen if a pipeline parameter has a dash ("-") in the name