Lakesense Auth Error When Creating Multiple Knowledge Assistant Agents in Databricks

[snarayan]

I’m encountering an issue related to Lakesense authentication when working with Agent Bricks Knowledge Assistant (KA) agents (I recreated thisAgent Bricks Knowledge Assistant agent), and I haven’t found any documentation on this. Hoping someone here has seen this before or can point me in the right direction.

Steps to Reproduce

1. Create two new KA agent from the Agents tab.
2. Add files, instructions, and prompts.
3. Allow the data ingestion process to complete.
4. Test the model after syncing.

---

Issue

• For every agent after the first one, I consistently receive a Lakesense auth error during testing.
• The error occurs even though the ingestion completes successfully.
• I’ve paused creating more agents because only the first one works as expected.

---

What I’ve Tried

• Verified the setup for each agent.
• Checked public documentation but couldn’t find any reference to Lakesense.

---

Has anyone else faced this issue? Is there a known limitation or configuration step I might be missing for multiple KA agents?

Any guidance or workaround would be greatly appreciated!

#generativeai ##agent

[snarayan · Accepted solution]

Update: The new agents I create are now working as expected. A Databricks architect suggested that the earlier issue might have been due to a backend outage, since customers typically shouldn’t encounter Lakesense errors. At this point, any newly created agents are functioning properly.