Azure AI Foundry UI is unable to fetch knowledgebases in Azure AI Foundry

Question

Azure AI Foundry UI is unable to fetch knowledgebases in Azure AI Foundry

Dhanaji More 1 Microsoft Employee

Getting this error message in the UI.

Error loading knowledge basesFailed to fetch knowledge bases for connection mycontososearchxnpcrk.

Note - I have granted Search Index Data Reader, Search index data contributor and Search Service contributor to foundry project managed identitiy as well as to managed identity of foundry resource itself.

Still no success. Any ideas what I am missing here ?

Aditya N 2,550 Reputation points Microsoft External Staff Moderator

2026-04-10T01:30:07.07+00:00

Hello @Dhanaji More

Thank you for reaching out Microsoft Q&A. We're looking into your query, we'll get back to you shortly.
Dhanaji More 1 Reputation point Microsoft Employee

2026-04-10T17:42:53.2666667+00:00
This is what fixed my issue. I was able to make it work. Please make sure of these two settings:

Azure AI Search service: Enable both API Keys and RBAC authN

Storage and AI Search Service: Create role assignments for the system-assigned managed identities at the project level. Make sure the managed identity of azure ai foundry project has all the correct RBACs on search service.
Khanh Nguyen 0 Reputation points

2026-04-12T02:35:21.58+00:00

For Storage and AI Search Service, check for these 2 roles.
Anshika Varshney 9,900 Reputation points Microsoft External Staff Moderator

2026-04-13T08:39:29.2533333+00:00

Hi Dhanaji More,

Yes, this behavior is usually related to how authentication and role-based access is configured between Azure AI Foundry and Azure AI Search.

When using Knowledge Base in Foundry, the project managed identity is used to access the Search service. So, it is important to ensure that the managed identity of the Azure AI Foundry project has the required RBAC roles assigned on the Azure AI Search resource such as:

Search Index Data Contributor Search Service Contributor [learn.microsoft.com]

Also, as you mentioned, enabling both API key authentication and RBAC authentication on the Azure AI Search service helps avoid access issues when the Knowledge Base tries to retrieve data using identity-based access.

For Knowledge Base usage, permissions to query and manage objects on Azure AI Search are required, and Microsoft recommends role-based access configuration for this setup. [learn.microsoft.com]

Thanks for sharing the fix. Since I’ve converted comment into an answer, could you please take a moment to mark it as Accepted? This helps others in the community with the same question find the solution more easily.

Thank you!

1 answer

Your answer

Aditya N 2,550 Reputation points Microsoft External Staff Moderator

2026-04-10T01:30:07.07+00:00

Hello @Dhanaji More

Thank you for reaching out Microsoft Q&A. We're looking into your query, we'll get back to you shortly.
Dhanaji More 1 Reputation point Microsoft Employee

2026-04-10T17:42:53.2666667+00:00

This is what fixed my issue. I was able to make it work. Please make sure of these two settings:

Azure AI Search service: Enable both API Keys and RBAC authN

Storage and AI Search Service: Create role assignments for the system-assigned managed identities at the project level. Make sure the managed identity of azure ai foundry project has all the correct RBACs on search service.
Khanh Nguyen 0 Reputation points

2026-04-12T02:35:21.58+00:00

For Storage and AI Search Service, check for these 2 roles.
Anshika Varshney 9,900 Reputation points Microsoft External Staff Moderator

2026-04-13T08:39:29.2533333+00:00

Hi Dhanaji More,

Yes, this behavior is usually related to how authentication and role-based access is configured between Azure AI Foundry and Azure AI Search.

When using Knowledge Base in Foundry, the project managed identity is used to access the Search service. So, it is important to ensure that the managed identity of the Azure AI Foundry project has the required RBAC roles assigned on the Azure AI Search resource such as:

Search Index Data Contributor Search Service Contributor [learn.microsoft.com]

Also, as you mentioned, enabling both API key authentication and RBAC authentication on the Azure AI Search service helps avoid access issues when the Knowledge Base tries to retrieve data using identity-based access.

For Knowledge Base usage, permissions to query and manage objects on Azure AI Search are required, and Microsoft recommends role-based access configuration for this setup. [learn.microsoft.com]

Thanks for sharing the fix. Since I’ve converted comment into an answer, could you please take a moment to mark it as Accepted? This helps others in the community with the same question find the solution more easily.

Thank you!

Answer 1

Hi @Dhanaji More,

The Error loading knowledge bases Failed to fetch knowledge bases for connection usually means Azure AI Foundry is not able to connect properly with the Azure AI Search service which is used behind the knowledge base.

Below are a few things you can check.

First, please make sure that your Foundry Project Managed Identity has the required access on the Azure AI Search resource.

As per Microsoft documentation, the managed identity should have:

Search Service Contributor role to create and manage the knowledge base Search Index Data Reader role to run queries on the knowledge base [learn.microsoft.com]

Next, please verify that the knowledge base actually exists inside your Azure AI Search service and has not been renamed or removed. If the semantic configuration or index name changes, Foundry may not be able to retrieve it correctly.

Also check the network settings on your Azure AI Search resource.

If the Search service is configured with private endpoint or selected networks only, then Foundry may not be able to reach it unless proper DNS and firewall rules are configured. This can silently cause retrieval failures even when role assignments look correct.

In addition, please confirm that:

The Azure AI Search service is in Basic tier or higher if you are using managed identity access.

Finally, revalidate the connection from the Knowledge tab inside your Foundry project to ensure the correct Search service is linked. Sometimes RBAC or credential changes can cause the agent to skip the search tool due to connectivity or authentication issues.

Hope this helps. Let us know if the issue still continues after checking the above.

Thankyou!

Share via

Azure AI Foundry UI is unable to fetch knowledgebases in Azure AI Foundry

1 answer

Your answer