NVIDIA / NVIDIA/nvcf

vanity-gateway: tooManyRequestsMessage is dropped on the LLM Gateway path

Open Beginner friendly
#1,690 0 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
Go
Stars
218
Forks
72
Avg merge
1d 12h
Merged PRs (30d)
427

Description

appendTooManyRequestsMessage recognizes two 429 body shapes: an OpenAI error object and an RFC 7807 problem document. Any other body is returned unchanged.

The LLM Gateway returns its 429 through echo's default error handler, which renders {"message":"rate limit exceeded"}. That matches neither shape, so an operator-configured tooManyRequestsMessage on a model with functionType: LLM never reaches the caller, with no log line or metric indicating the drop.

Models served by the invocation service are unaffected: that upstream returns an RFC 7807 document, which is recognized.

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Trace appendTooManyRequestsMessage and the LLM Gateway's 429 error-handling path, starting with the code that handles the echo-rendered {"message":"rate limit exceeded"} body. Confirm the configured tooManyRequestsMessage reaches the caller for functionType: LLM without changing the unaffected invocation-service path.

Written by the indexing model from the issue text.

Assessment

Tech stack
go
Domain
api, backend
Issue type
Bug
Difficulty
2/5
Estimated time
1-3 hours
Activity status
Active
Clarity
Mostly clear
Newbie friendliness
68/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.