Genie Agent content search on a managed volume fails with "The request was invalid"

[mramanindia]

Hi all,

I am trying to enable content search on a Unity Catalog managed volume so a Genie Agent can answer questions over PDFs. The Enable action fails every time with:

"Failed to update content search - The request was invalid."

Environment: AWS, eu-west-2 (London), serverless workspace on Default Storage, pay-as-you-go. This started as a trial and the failure is identical before and after adding a payment method.

What I have verified:

- Preview "Analyze Files in Volumes with Genie Agents" is enabled.

- The volume is a managed volume. Catalog binding is "All workspaces have access" (OPEN). Reproduced on two catalogs, including a freshly created one.

- I have CAN MANAGE on the volume.

- ai_parse_document() runs successfully on the PDFs in the volume.

- Foundation Model APIs work (ai_query succeeds).

- Lakebase is available in the region.

- I can create a Vector Search endpoint via the API and it comes ONLINE.

- Files are 4 PDFs, each under 10 MB.

Without content search the agent does not retrieve the files ("no relevant content from the reports directory"), so document and hybrid questions are not possible.

Questions:

1. Is content search supported on Default Storage / serverless-only workspaces? It is not listed under the Default Storage limitations, but it is the one variable I cannot rule out.

2. Is there a way to get a more detailed error than "The request was invalid"? No public API seems to expose this.

Happy to share the request ID from the browser network log if that helps.

Thanks!

[balajij8]

@mramanindia 

The underlying issue here is likely the workspace cross Geo data processing requirement. In Databricks, content search on managed volumes relies on background Vector Search indexing infrastructure that is not deployed in every single cloud region. For eu-west-2 (London) along with Frankfurt, Ireland, Paris, Central Canada, São Paulo, Tokyo, Seoul, Singapore, Jakarta and Mumbai - Databricks handles the document indexing workload in a secondary geography. When a workspace restricts processing to its local boundary, the service rejects the indexing request and surfaces that generic Failed to update content search - The request was invalid error in Catalog Explorer.

Content search is fully supported on Default Storage and serverless workspaces. Managed volumes residing on Default Storage work without issue for volume content search, so the storage architecture is not the limiting factor. There is currently no public API or workspace-level log stream that exposes more granular validation errors for this error. The action triggers backend infrastructure provisioning within Databricks-managed control planes, which throws the downstream geo-restriction errors before reaching the UI or workspace APIs.

To resolve this, an account administrator needs to permit cross-region processing for Designated Services on the workspace

• Open the Databricks Account Console
• Navigate to Workspaces in the sidebar and select your London workspace.
• Open the Security and compliance tab.
• Disable Enforce data processing within workspace Geography for Designated Services.

Toggling this off grants Databricks permission to run the required document parsing and vector indexing pipelines outside eu-west-2. Once updated at the account level, re-enabling content search on your managed volume should complete successfully, allowing the Genie Agent to retrieve content from the PDFs for hybrid queries. More details here

[data_pulse]

@mramanindia 
You’ve already checked most of the documented prerequisites.

On your first question, I don’t see Default Storage or serverless only workspaces called out as a limitation for Content Search in the current docs, so I suppose that probably isn’t the main blocker.

One thing still worth checking is cross-Geo processing, since Databricks lists that as a requirement for Content Search in some regions.

On the error itself, I couldn’t find a documented public API that exposes a richer provisioning error. If the documented prerequisites are met, the practical next step is a Databricks Support case with the browser request ID and timestamp.

One other thing I noticed while testing this: I was able to attach a managed volume with a small PDF to a Genie Agent and query it successfully in Agent mode even without Content Search enabled on the volume. So the "no relevant content" part may actually be a separate retrieval/config issue from the Content Search enablement failure.
Analyze files in volumes with Genie Agents Reference

Below is the response I got from Genie Agent searching on PDF files. 

[mramanindia]

Thanks both, very helpful. Two updates.

First, a correction on my side: the workspace is in us-east-2, not London, so cross-Geo processing should not have been required for this region.

Second, I checked the workspace Security and compliance tab anyway. Compliance security profile is disabled. "Enforce data processing within workspace Geography for Designated Services" was enabled, so I turned it off. Enable content search still fails right after with the same error (it may need time to propagate, I will retry later).

Details in case they help: the UI call is PATCH /ajax-api/2.1/unity-catalog/volumes/{volume} returning 400. GET on the volume with include_content_search=true returns no content_search object at all. The volume is a managed volume on Default Storage (securable_kind VOLUME_DB_STORAGE), catalog binding is open.

If it still fails after the setting propagates I will raise a support case with the request IDs. Will report back.