What causes EntityNotFound?

An incorrect, outdated, or miscopied resource object ID (GUID) used in the API request path or body. The referenced Fabric item (lakehouse, warehouse, dataset, pipeline, etc.) was deleted after the ID was captured. The resource exists in a different workspace than the one specified in the API call. The calling identity lacks access to the resource, causing it to surface as not found rather than forbidden

How do I fix EntityNotFound?

Step 1: Inspect the relatedResource field in the error response body — it identifies the specific missing entity type and ID, narrowing down exactly which resource reference is broken.. Step 2: Navigate to the Fabric portal and confirm the referenced item still exists in the expected workspace; if it has been deleted, restore it from the recycle bin or recreate it.. Step 3: Retrieve the correct current resource ID by listing items of the relevant type via the appropriate Fabric API (e.g., GET /v1/workspaces/{workspaceId}/lakehouses) and update all configurations using the stale ID.. Step 4: Verify that the calling service principal or user account has at least Viewer-level access to the specific item, not just to the workspace.. Step 5: If the resource was recently created, allow a short propagation delay and retry — newly provisioned Fabric items can occasionally return EntityNotFound for a brief period after creation.

High severityconfigurationMicrosoft Fabric →

Power BI Refresh Error:
EntityNotFound

What does this error mean?

A Fabric REST API call returned a 404 Not Found response with error code EntityNotFound, meaning the specific resource (dataset, lakehouse, pipeline, report, etc.) referenced in the request does not exist or is inaccessible. The missing entity is identified in the relatedResource field of the error response.

Common causes

1An incorrect, outdated, or miscopied resource object ID (GUID) used in the API request path or body
2The referenced Fabric item (lakehouse, warehouse, dataset, pipeline, etc.) was deleted after the ID was captured
3The resource exists in a different workspace than the one specified in the API call
4The calling identity lacks access to the resource, causing it to surface as not found rather than forbidden

How to fix it

1Step 1: Inspect the relatedResource field in the error response body — it identifies the specific missing entity type and ID, narrowing down exactly which resource reference is broken.
2Step 2: Navigate to the Fabric portal and confirm the referenced item still exists in the expected workspace; if it has been deleted, restore it from the recycle bin or recreate it.
3Step 3: Retrieve the correct current resource ID by listing items of the relevant type via the appropriate Fabric API (e.g., GET /v1/workspaces/{workspaceId}/lakehouses) and update all configurations using the stale ID.
4Step 4: Verify that the calling service principal or user account has at least Viewer-level access to the specific item, not just to the workspace.
5Step 5: If the resource was recently created, allow a short propagation delay and retry — newly provisioned Fabric items can occasionally return EntityNotFound for a brief period after creation.

Frequently asked questions

How do I know which specific item is missing when I get EntityNotFound?

The Fabric API includes a relatedResource field in the error response body that specifies the resource type and ID that could not be found. Always log the full error response body — not just the HTTP status code — to capture this detail.

Can EntityNotFound appear even if the resource exists, and if so why?

Yes, in two scenarios: first, if the calling identity lacks access to the resource (Fabric returns 404 rather than 403 to avoid revealing the resource exists); and second, very briefly after a resource is first created due to propagation latency. In both cases, confirming access permissions and retrying after a short delay are the right first steps.