btlqql opened a new issue, #2172:
URL: https://github.com/apache/rocketmq-dashboard/issues/2172

   ### What happened?
   
   Testing an HTTP LLM configuration can report `Connection successful` even 
when the selected provider is not supported by the server's OpenAI-compatible 
client. Azure and Bedrock currently take this path: validation succeeds, 
`supports()` returns false, and the method falls through to a success result 
without making a request.
   
   ### Expected behavior
   
   The test endpoint should return the same `llm.config.unsupported_provider` 
failure used by the runtime gateway, so the UI does not confirm a configuration 
that cannot serve requests.
   
   ### Reproduction
   
   1. Select the HTTP engine and either the Azure or Bedrock provider.
   2. Fill in the provider-specific required fields.
   3. Test the configuration.
   4. Observe a successful result even though the HTTP gateway rejects that 
provider.


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]

Reply via email to