Forum Discussion

BenFord's avatar
BenFord
Frequent Visitor
5 years ago

deployment pipelines compare

I recently created a deployment pipeline from our Production workspace.  I deployed backwards to STG.  I then proceeded to make several changes the datasets for that workspace and republished them.  Since they were named the same, they simply replaced the ones that were in the STG workspace.  Now however when i compare with production, it says the STG datasets are new even they they are named the same.  In fact even when i deploy to Prd it creates a duplicate version of the datasets.  Also, if i publish the dataset to PRD (again with the same name) it replaces it as it should.  What is wrong with my deployment pipeline?  

7 Replies

  • Anonymous's avatar
    Anonymous
    Not applicable

    Hi BenFord ,

     

    Are you saying that when you publish a dataset that has the same name in Power BI Service, it won't replace the exists one?

     

    Best Regards,

    Jay

    • BenFord's avatar
      BenFord
      Frequent Visitor

      No, if i publish the dataset from power bi desktop, it replaces as it should.  the issue is using the deployment pipeline.  deploying from STG to PRD which have datasets that named the same, results in duplicate datasets in the prd workspace.  Note, i use the deployment pipelines for 2 other projects and they work as expected.

      • Nimrod_Shalit's avatar
        Nimrod_Shalit
        Power BI Team

        Is there an auto-sync setup in any of the workspaces in that pipeline, with either OneDrive or SharePoint online?

  • Anonymous's avatar
    Anonymous
    Not applicable

    Hey BenFord 

    I am running into this same issue and was wondering if you ever heard back? I believe my issue could be because we are using parameters to determine which SQL server to use? If it is the BUILD workspace then it uses the BUILD SQL server, if it is in TEDI then it uses our TEDI SQL Server and if it is PROD workspace then it uses our PROD sql server. Could a change in data source cause the issue with duplicated datasets and the compare saying that the two identical reports are different? 

     

    Any Insight would be greatly appreciated!

    Nimrod_Shalit Anonymous 

    Thanks,

    Dillon 

    • LeeBenjamin's avatar
      LeeBenjamin
      Microsoft Employee

      Hi Anonymous ,
      Using parameters to determine different datasource for the report shouldn't lead to diff between the reports.
      There's maybe an issue with the connecting the two reports per the work process of creating this pipeline. In this case it could be helpful removing/delete the item from the target stage and then deploy it from the source again. Please advise that any pipeline deployment's rule in place will be removed and have to be set again.
      Best,

      Lee.