Data-Factory: Snowflake Linked Service KeyVault

Patrick Doerig 31 Reputation points
2020-06-29T12:01:46.277+00:00

Currently the implementation of KeyVault inside the linked service template for Snowflake is not properly done.

  1. Even when keyVault is referenced for the password I am getting the following message:

10860-ms-bug.png

  1. It would be helpful if the the username, warehouse as well as the database can also be stored in keyVault and not only the password as this makes it difficult to work with multiple environments and devops.
Azure Data Factory
Azure Data Factory

An Azure service for ingesting, preparing, and transforming data at scale.


3 answers

Sort by: Newest
  1. KranthiPakala-MSFT 46,607 Reputation points Moderator
    2020-08-10T07:12:26.44+00:00

    Hi @Patrick Doerig ,

    Just wanted to update you that ADF Product team has successfully implemented the feature request you have shared to support AKV for whole connection string in Snowflake Database. Please feel free to use it and let us know if you see any issues.

    16620-image.png

    Thank you.

    Was this answer helpful?


  2. KranthiPakala-MSFT 46,607 Reputation points Moderator
    2020-07-30T04:55:07.267+00:00

    Hi @PatrickDoerig-3768,

    Thanks for the confirmation. When a pipeline is working by trigger, but not by debug, that suggests either: there is a difference between the published version and the version in the UI or there are parameters that depend upon the trigger. Have you created the Snowflake linked service first with hard coded values and then edited the JSON payload with connection string AKV? If that is the case, then, when the Snowflake linked service was created initially with hard coded values (no AKV) then the linked service is directly published to data factory service, and when you have edited the JSON payload of the linked service with AKV details, then the edited linked service is not immediately published to the data factory service, it has to be merged to master and then to be published from master to data factory service (nothing but adf_publish). This could be the case, where there is a mismatch between published version and local brach/UI version.

    And since the linked service haven't been published yet, debug run would take all linked service resources payload fetched from UI, while trigger run would take all published resources in Azure. Debug run always takes the newest resources user edits on UI, after they publish resources, UI would keep consistency with published resource, everything should be fine.

    Are you noticing the same error when you published the linked service and do a debug run? - Could you please confirm on this?


    Thank you

    Was this answer helpful?


  3. Deleted

    This answer has been deleted due to a violation of our Code of Conduct. The answer was manually reported or identified through automated detection before action was taken. Please refer to our Code of Conduct for more information.

    7 deleted comments

    Comments have been turned off. Learn more

Your answer

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