Add llm log events to docs

What does this MR do and why?

It documents every event logged to llm.log file and updates documentation on logging to reflect recent changes in this MR

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

Screenshots are required for UI changes, and strongly recommended for all other merge requests.

Before	After

How to set up and validate locally

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

assigned to @mksionek

added development guidelines docsimprovement documentation maintenancerefactor typemaintenance labels

	1 Warning
	This merge request is quite big (734 lines changed), please consider splitting it into multiple merge requests.

	1 Message
	This MR contains docs in the /doc/development directory, but any Maintainer (other than the author) can merge. You do not need tech writer review.

If needed, you can retry the danger-review job that generated this comment.

Generated by Danger

added docs-only label

added 1 commit

• e70a2caa - Format classes names nicely

Compare with previous version

added 1 commit

• a3c33d46 - Update logs documentation

Compare with previous version

marked this merge request as ready

changed title from Draft: Mk add llm log events to docs to Add llm log events to docs

changed the description

changed milestone to %17.5

requested review from @sselhorn

mentioned in epic gitlab-org#15150 (closed)

resolved all threads

resolved all threads

resolved all threads

added 1 commit

• 1bef8992 - Apply 3 suggestion(s) to 1 file(s)

Compare with previous version

Hi @mksionek You are testing the limits of what our docs site is capable of! Take a look at the rendered content: https://main-ee-167304.docs.gitlab-review.app/ee/development/ai_features/logged_events.html

I'm on my laptop, so my screen is a bit narrower, but I see things overwriting each other and going off the right side of the screen.

Could we separate the information out by one or more criteria and have multiple tables instead? I just gave the table to Claude and asked for advice, and here's what it said. Does this make any sense? I'm open to ideas, I think we just need a few fewer columns.

Event Logging Tables

Table 1: Event Basic Information

Event message	description	class	ai_event_name	level
Returning from Service due to validation	user not permitted to perform action	Llm::BaseService	permission_denied	info
Enqueuing CompletionWorker	scheduling completion worker in sidekiq	Llm::BaseService	worker_enqueued	info
aborting: missing resource	If there is no resource for slash command	Llm::ChatService	missing_resource	info
Performing CompletionService	performing completion	Llm::Internal::CompletionService	completion_service_performed	info
Answer from LLM response	Get answer from response	Gitlab::Llm::Chain::Answer	answer_received	info
Final answer	Get final answer from response	Gitlab::Llm::Chain::Answer	final_answer_received	info
Default final answer	Default final answer: I'm sorry, I couldn't respond in time. Please try a more specific request or enter /clear to start a new chat.	Gitlab::Llm::Chain::Answer	default_final_answer_received	info
Error message/ "Error"	when answering with an error	Gitlab::Llm::Chain::Answer	error_returned	error

Table 2: Logging Details

Event message	Logged with monolith logs	Logged with sidekiq logs	Part of the system	Expanded logging?
Returning from Service due to validation	yes	no	abstraction_layer	no
Enqueuing CompletionWorker	yes	no	abstraction_layer	yes
aborting: missing resource	yes	no	duo_chat	no
Performing CompletionService	no	yes	abstraction_layer	yes
Answer from LLM response	no	yes	duo_chat	yes
Final answer	no	yes	duo_chat	yes
Default final answer	no	yes	duo_chat	yes
Error message/ "Error"	no	yes	duo_chat	no

Table 3: Additional Event Information

Event message	arguments	error_code
Returning from Service due to validation	none	-
Enqueuing CompletionWorker	user_id: message.user.id, resource_id: message.resource&.id, resource_class: message.resource&.class&.name, request_id: message.request_id, action_name: message.ai_action, options: job_options	-
aborting: missing resource	none	-
Performing CompletionService	user_id: prompt_message.user.to_gid, resource_id: prompt_message.resource&.to_gid, action_name: prompt_message.ai_action, request_id: prompt_message.request_id, client_subscription_id: prompt_message.client_subscription_id, completion_service_name: completion_class_name	-
Answer from LLM response	llm_answer_content: content	-
Final answer	llm_answer_content: content	-
Default final answer	-	"A6000"
Error message/ "Error"	error: content, error_code: error_code, source: source	-