Skip to content

Add response server in step executor error log

Shinya Maeda requested to merge log-v2-error-message-and-origin into master

What does this MR do and why?

This MR adds response server in step executor error log. We need visibility on the errors at the intermediate servers https://gitlab.com/groups/gitlab-org/-/epics/15402+.

Closes Log response body and origin at "Received error... (#497472 - closed) and Make sure what error message is going to be sur... (#497494 - closed)

MR acceptance checklist

Please evaluate this MR against the MR acceptance checklist. It helps you analyze changes to reduce risks in quality, performance, reliability, security, and maintainability.

Screenshots or screen recordings

Sample llm.log:

{"severity":"ERROR","time":"2024-10-03T05:35:12.168Z","correlation_id":"77c125370695dc12f00c7b532989e030","status":422,"ai_response_server":"uvicorn","message":"Received error from Duo Chat Agent","class":"Gitlab::Duo::Chat::StepExecutor","ai_event_name":"error_returned","ai_component":"duo_chat"}

In this example, I intentionally raised an error in AI Gateway, so the ai_response_server is uvicorn.

How to set up and validate locally

Numbered steps to set up and validate the change are strongly suggested.

Edited by Shinya Maeda

Merge request reports