vllm-project / vllm-project/production-stack

feature: Support CRD based configuration

Open
#204 0 comments 1 reaction 0 assignees View on GitHub

Nobody has claimed this yet.

feature request
Dominant language
Python
Stars
2.6k
Forks
503
Avg merge
4d 17h
Merged PRs (30d)
8

Description

Describe the feature

I understand that if i use the router in a kubernetes environment, the router uses service discovery to build a pool of backend vllm services. While this worry-free mechanism is good in many ways, i am wondering if we can add yet another way to explicitly declare a backend connectivity through a kubernetes crd. The idea is that, for certain use cases, the crd represents a persistent configuration that can be managed by controllers (while annotation and labels that are fed to service discovery are mostly ad hoc)

Why do you need this feature?

Support routing CRD so that a declarative management can be supported instead of using service discovery.

Additional context

(see the slack discussion https://vllm-dev.slack.com/archives/C089SMEAKRA/p1740756084674449)

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

The issue mentions the router, Kubernetes service discovery, annotations and labels, but names no files or tests. Start by locating the router's service-discovery integration and the Kubernetes configuration entry points, then determine the CRD shape and how it should represent backend connectivity. Done means the requested declarative CRD-based routing behavior is specified and implemented with coverage for the relevant integration.

Written by the indexing model from the issue text.

Assessment

Tech stack
kubernetes, python
Domain
backend-api-design, cloud, infrastructure
Issue type
Feature
Difficulty
5/5
Estimated time
Over a week
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
25/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.