Workspace Role

[sudip2607]

Hello beutiful people out there.. Please help me on below one simple question please.

Which workspace role is required to modify Spark pool settings for a workspace?
A. Viewer B. Contributor C. Admin D. Member

My personal experience is 'Contributor'. I have an user, who has given contributor role to a workspace. He can make changes to the pool and save the setting. 

But as per this --> https://learn.microsoft.com/en-us/fabric/data-engineering/workspace-admin-settings, it says user needs to have admin role. Got little confused. Even ChatGPT is saying minimum role required is Admin. Am I doing something wrong? 

[v-prasare]

Hi @sudip2607,

We would like to follow up to see if the solution provided by the super user resolved your issue. Please let us know if you need any further assistance.

Thanks,

Prashanth Are

MS Fabric community support
If our super user response resolved your issue, please mark it as "Accept as solution" and click "Yes" if you found it helpful.

[sudip2607]

I am good now. Thanks. 

[sudip2607]

Hi @v-prasare 

I would wait for little more to accept response as a Solution. Please accept my apologies 

[v-prasare]

@sudip2607, is there any progress on above mentioned ask?

Thanks,

Prashanth Are

MS Fabric community support

[v-prasare]

Hi @sudip2607 ,

We would like to follow up to see if the solution provided by the super user resolved your issue. Please let us know if you need any further assistance.

@burakkaragoz , @Srisakthi, @suparnababu8  thanks for your prompt response.

Thanks,

Prashanth Are

MS Fabric community support
If our super user response resolved your issue, please mark it as "Accept as solution" and click "Yes" if you found it helpful.

[burakkaragoz · Accepted solution]

Hi @sudip2607 ,

Great question — this is one of those areas where behavior and documentation can feel a bit out of sync.

You're right that the official documentation states that modifying Spark pool settings requires Admin role in the workspace. You can find that here:
🔗Microsoft Learn 

However, in practice, Contributor role can sometimes modify Spark pool settings if the workspace permissions are loosely scoped or inherited from broader Fabric-level permissions. This might explain why you were able to make changes with Contributor access.

That said, relying on Contributor for this task isn't guaranteed or recommended. Microsoft enforces role-based access control (RBAC) strictly in some environments, especially in enterprise tenants. So while it might work now, it could break later if policies are tightened.

To be safe and compliant with best practices:

• Use Admin role for Spark pool configuration.
• Use Contributor for pipeline, notebook, and dataset development.

Also, here's another helpful doc on Fabric roles and permissions:
🔗 Microsoft Learn 

Let me know if you want to test this behavior in a controlled setup or need help checking your workspace's permission inheritance.

If my response resolved your query, kindly mark it as the Accepted Solution to assist others. Additionally, I would be grateful for a 'Kudos' if you found my response helpful.

This response was assisted by AI for translation and formatting purposes.

[sudip2607]

Hi @burakkaragoz 

Thanks for your explanation. I'm convinced but wanted to learn more on this, how the loosly coupled setting works. By the way, your given links are not working for some reason. 

[burakkaragoz]

Hi @sudip2607 ,

You're right about the links - sorry about that! Let me explain the "loosely coupled" permissions thing because it's actually pretty interesting how Fabric handles this.

What I meant by "loosely scoped permissions":

In some Fabric environments, workspace permissions can inherit from higher-level settings or get influenced by:

1. Tenant-level admin rights - If someone has broader Fabric admin permissions at the tenant level, those can sometimes override workspace-level restrictions
2. Capacity admin permissions - Users with capacity admin rights might be able to modify Spark settings even with Contributor workspace role
3. Legacy Power BI admin inheritance - Since Fabric evolved from Power BI, some permission behaviors carry over in unexpected ways

Here's what's probably happening in your case: Your Contributor user likely has one of these broader permissions without you realizing it. Check if they have:

• Any Fabric admin roles at the tenant level
• Capacity admin access on the capacity your workspace uses
• Power BI admin permissions that are carrying over

To test this properly: Create a test user with ONLY Contributor workspace access (no other admin rights anywhere) and see if they can still modify Spark settings. My guess is they won't be able to.

The documentation is correct about needing Admin role - but Fabric's permission inheritance can create these edge cases where it seems like Contributor works.

Does your Contributor user have any other admin rights you can think of?

[Srisakthi]

Hi @sudip2607 ,

A User with Contributor role cannot update spark pool settings. But can modify spark porperties in environment.

Regards,

Srisakthi

[sudip2607]

Hi @Srisakthi 

I had the same understanding but surprisingly have checked and experienced that with the Contributor role user can update spark pool settings. Thanks for your response by the way. 

[suparnababu8]

Hello @sudip2607 

Which workspace role is required to modify Spark pool settings for a workspace?

Answer : Admin

I replicated your scenario by granting Contributor access to the workspace. 

Later, I opened Contributor role workspace and went to edit and clicked on edit icon.

I increased the number of nodes to 16.

then when I clicked on save. it show the below

Now again If i click on Save, I got a notification 'Spark settings update failed'

The complete error mentioned in below image.

So, MS documentation is correct. Hope you are clear now.

Thank you!

Did I answer your question? Mark my post as a solution!

Proud to be a Super User!

[sudip2607]

I checked again, and now I can't do it. You're right to showcase the replicated scenario result.

However, I feel @burakkaragoz explained why I got successful last time. Hope you wouldn't mind accepting that response as a solution. Thanks for your time and assistance, much appreciated. 

[sudip2607]

Hi @suparnababu8 

Thanks for your detailed steps. Surprisingly when I clicked on 'Save' button after changing the node size as 'Contributor', that got saved. Let me recreate the scenario, this time would definitely attach screenshot.