Deployment Pipelines - What is compared?

[Mogry]

Hi Community,

In one of my deployment pipelines exists a datatet in the test stage which is also deployed to the prod stage but it is marked as different. When the dataset (and report) is copied to another stage, how it is possible that they are different? I read about the dataset properties which are not copied during deployment in the documentation. I checked them and they are the same (except for the data source, because this is changed by the dataset rule in the pipeline). Another dataset in the same pipeline is not marked as different which means the datasourece can be different and the dataset is not marked as different. When I delete this dataset, publish it again and then deployed to prod it is not different. Until somewhen it is different again.

Basically the question is: Which metadata is compared by the deployment pipeline? And if I am facing a bug?

Best,

Mogry

[Nimrod_Shalit]

If you still get the 'Diff' label after deployment of the dataset, and it suddenly goes away later, without making any particular operation, then this sounds like a bug.

Please open a support ticket.

[v-janeyg-msft]

Hi, @Mogry 

You can reference the document:Get started with deployment pipelines - Power BI | Microsoft Docs

Mentioned above:

The comparison indicator has two states:

• Green indicator – The metadata for each content item in both stages, is the same.

• Orange indicator - Appears if one of these conditions is met:

  • Some of the content items in each stage, were changed or updated (have different metadata).
  • There is a difference in the number of items between the stages.

If it doesn’t solve your problem, please feel free to ask me.

Best Regards

Janey Guo

If this post helps, then please consider Accept it as the solution to help the other members find it more quickly.

[Mogry]

Hi Janey,

I am aware of the documentation. My question was more in detail: Which metadata is compared by the deployment pipeline?

I am trying to understand why a dataset is sometimes "different", even it gets deployed to production state. This process should make the two dataset same and not different. Am I correct?

[v-janeyg-msft]

Hi, @Mogry 

Dataset metadata contains tables, relationships, columns, measures.

'Different' is normal, each stage can add some data.

Best Regards

Janey Guo

If this post helps, then please consider Accept it as the solution to help the other members find it more quickly.

[Mogry]

Thanks for the answer.

But when I deploy the dataset from one stage to another, they should not be different and the indicator should be green, correct?

[Nimrod_Shalit]

@Mogry ,

In more details- we mark items as 'different' only when there's a meatdata change that we are updating during a deployment. Any of the metatdata that @v-janeyg-msft mentioned above will trigger the 'diff'. But if you just share content with someone, it will not trigger the 'diff', since we don't copy permissions.

If you have set rules on that dataset, it might be a bit more tricky. if you have set a rule but did not apply him (by deploying again), then you might get a 'diff'. is that your case? if so, can you re-deploy to apply the rule and see if the 'diff' hase been removed?

[Mogry]

Thanks for the answers so far.

I am using AAS as source and set up a dataset rule from test to prod stage to change the data source. When I hit the deploy to production button the report will be not different, but the dataset. This not happens for all datasets, only for one of three in this pipeline. A few hours later it is not different anymore. I just wanted to understand why this is happening sometimes, so I can explain it better to one of my customers.