[Frontend] Support custom request_id from request

[guoyuhong]

FILL IN THE PR DESCRIPTION HERE:

Currently, request_id is generated in function create_chat_completion, which cannot be controlled by high-level user.

In our scenario, we want to pass the custom request_id into create_chat_completion. Then, we can debug the end-to-end process using a unique request id.

This PR simply add a field request_id to ChatCompletionRequest with a default random_uuid() value. If the user passes a request_id to the request object, the backend will use it directly. Otherwise, the backend will generate one which is the same as current behavior.

Example w/o request_id:

Example w/ request_id:

BEFORE SUBMITTING, PLEASE READ THE CHECKLIST BELOW AND FILL IN THE DESCRIPTION ABOVE

---

PR Checklist (Click to Expand)

Thank you for your contribution to vLLM! Before submitting the pull request, please ensure the PR meets the following criteria. This helps vLLM maintain the code quality and improve the efficiency of the review process.

PR Title and Classification

Only specific types of PRs will be reviewed. The PR title is prefixed appropriately to indicate the type of change. Please use one of the following:

• [Bugfix] for bug fixes.
• [CI/Build] for build or continuous integration improvements.
• [Doc] for documentation fixes and improvements.
• [Model] for adding a new model or improving an existing model. Model name should appear in the title.
• [Frontend] For changes on the vLLM frontend (e.g., OpenAI API server, LLM class, etc.)
• [Kernel] for changes affecting CUDA kernels or other compute kernels.
• [Core] for changes in the core vLLM logic (e.g., LLMEngine, AsyncLLMEngine, Scheduler, etc.)
• [Hardware][Vendor] for hardware-specific changes. Vendor name should appear in the prefix (e.g., [Hardware][AMD]).
• [Misc] for PRs that do not fit the above categories. Please use this sparingly.

Note: If the PR spans more than one category, please include all relevant prefixes.

Code Quality

The PR need to meet the following code quality standards:

• We adhere to Google Python style guide and Google C++ style guide.
• Pass all linter checks. Please use format.sh to format your code.
• The code need to be well-documented to ensure future contributors can easily understand the code.
• Include sufficient tests to ensure the project to stay correct and robust. This includes both unit tests and integration tests.
• Please add documentation to docs/source/ if the PR modifies the user-facing behaviors of vLLM. It helps vLLM user understand and utilize the new features or changes.

Adding or changing kernels

Each custom kernel needs a schema and one or more implementations to be registered with PyTorch.

• Make sure custom ops are registered following PyTorch guidelines: Custom C++ and CUDA Operators and The Custom Operators Manual
• Custom operations that return Tensors require meta-functions. Meta-functions should be implemented and registered in python so that dynamic dims can be handled automatically. See above documents for a description of meta-functions.
• Use torch.libary.opcheck() to test the function registration and meta-function for any registered ops. See tests/kernels for examples.
• When changing the C++ signature of an existing op, the schema must be updated to reflect the changes.
• If a new custom type is needed, see the following document: Custom Class Support in PT2.

Notes for Large Changes

Please keep the changes as concise as possible. For major architectural changes (>500 LOC excluding kernel/data/config/test), we would expect a GitHub issue (RFC) discussing the technical design and justification. Otherwise, we will tag it with rfc-required and might not go through the PR.

What to Expect for the Reviews

The goal of the vLLM team is to be a transparent reviewing machine. We would like to make the review process transparent and efficient and make sure no contributor feel confused or frustrated. However, the vLLM team is small, so we need to prioritize some PRs over others. Here is what you can expect from the review process:

• After the PR is submitted, the PR will be assigned to a reviewer. Every reviewer will pick up the PRs based on their expertise and availability.
• After the PR is assigned, the reviewer will provide status update every 2-3 days. If the PR is not reviewed within 7 days, please feel free to ping the reviewer or the vLLM team.
• After the review, the reviewer will put an action-required label on the PR if there are changes required. The contributor should address the comments and ping the reviewer to re-review the PR.
• Please respond to all comments within a reasonable time frame. If a comment isn't clear or you disagree with a suggestion, feel free to ask for clarification or discuss the suggestion.

Thank You

Finally, thank you for taking the time to read these guidelines and for your interest in contributing to vLLM. Your contributions make vLLM a great tool for everyone!

[github-actions]

👋 Hi! Thank you for contributing to the vLLM project.
Just a reminder: PRs would not trigger full CI run by default. Instead, it would only run fastcheck CI which starts running only a small and essential subset of CI tests to quickly catch errors. You can run other CI tests on top of those by going to your fastcheck build on Buildkite UI (linked in the PR checks section) and unblock them. If you do not have permission to unblock, ping simon-mo or khluu to add you in our Buildkite org.

Once the PR is approved and ready to go, your PR reviewer(s) can run CI to test the changes comprehensively before merging.

To run CI, PR reviewers can do one of these:

• Add ready label to the PR
• Enable auto-merge.

🚀

[guoyuhong]

The PR is ready for review.

[cjackal]

Request ID is usually passed to the header(X-Request-Id) for accessibility; otherwise model server admins should (parse and) log the response body, which is a huge burden. OpenAI also returns request id in response header, so how about supporting X-Request-Id header (on both request and response) instead?

[guoyuhong]

@cjackal I can get X-Request-Id from raw_request.headers["x-request-id"]. Does that means that if the X-Request-Id is set, we can set it to response.id?

[DarkLight1337]

Thanks for adding this!

[cjackal]

@cjackal I can get X-Request-Id from raw_request.headers["x-request-id"]. Does that means that if the X-Request-Id is set, we can set it to response.id?

I'd suggest to implement it. Like:

AS-IS(v0.6.3.post1): X-Request-Id in request header is not passed back to response header

$ curl -v -X POST http://localhost:8000/v1/chat/completions \
> -d '{"model":"meta-llama/Llama-3.2-1B-Instruct","messages":[{"role":"user","content":"Hi"}]}' \
> -H 'X-Request-Id: aaaa' \
> -H 'Content-Type: Application/json'
*   Trying 127.0.0.1:8000...
* TCP_NODELAY set
* Connected to localhost (127.0.0.1) port 8000 (#0)
> POST /v1/chat/completions HTTP/1.1
> Host: localhost:8000
> User-Agent: curl/7.68.0
> Accept: */*
> X-Request-Id: aaaa
> Content-Type: application/json
> Content-Length: 88
> 
* upload completely sent off: 88 out of 88 bytes
* Mark bundle as not supporting multiuse
< HTTP/1.1 200 OK
< date: Tue, 22 Oct 2024 15:58:16 GMT
< server: uvicorn
< content-length: 230
< content-type: application/json
...

TO-BE:

$ curl -v -X POST http://localhost:8000/v1/chat/completions \
> -d '{"model":"meta-llama/Llama-3.2-1B-Instruct","messages":[{"role":"user","content":"Hi"}]}' \
> -H 'X-Request-Id: aaaa' \
> -H 'Content-Type: Application/json'
*   Trying 127.0.0.1:8000...
* TCP_NODELAY set
* Connected to localhost (127.0.0.1) port 8000 (#0)
> POST /v1/chat/completions HTTP/1.1
> Host: localhost:8000
> User-Agent: curl/7.68.0
> Accept: */*
> X-Request-Id: aaaa
> Content-Type: application/json
> Content-Length: 88
> 
* upload completely sent off: 88 out of 88 bytes
* Mark bundle as not supporting multiuse
< HTTP/1.1 200 OK
< date: Tue, 22 Oct 2024 15:58:16 GMT
< server: uvicorn
< content-length: 230
< content-type: application/json
< x-request-id: aaa
...

This is kind of standard approach for server log correlation (production level server usually keep logging incoming/outgoing headers, so no additional cost to pay) and easy to implement (simple middleware on fastapi side).

Of course my feature request is completely orthogonal to this PR; request id in chat completion stream body is independent of the request id in the header.

[DarkLight1337]

@cjackal I can get X-Request-Id from raw_request.headers["x-request-id"]. Does that means that if the X-Request-Id is set, we can set it to response.id?

I'd suggest to implement it. Like:

AS-IS(v0.6.3.post1): X-Request-Id in request header is not passed back to response header

$ curl -v -X POST http://localhost:8000/v1/chat/completions \
> -d '{"model":"meta-llama/Llama-3.2-1B-Instruct","messages":[{"role":"user","content":"Hi"}]}' \
> -H 'X-Request-Id: aaaa' \
> -H 'Content-Type: Application/json'
*   Trying 127.0.0.1:8000...
* TCP_NODELAY set
* Connected to localhost (127.0.0.1) port 8000 (#0)
> POST /v1/chat/completions HTTP/1.1
> Host: localhost:8000
> User-Agent: curl/7.68.0
> Accept: */*
> X-Request-Id: aaaa
> Content-Type: application/json
> Content-Length: 88
> 
* upload completely sent off: 88 out of 88 bytes
* Mark bundle as not supporting multiuse
< HTTP/1.1 200 OK
< date: Tue, 22 Oct 2024 15:58:16 GMT
< server: uvicorn
< content-length: 230
< content-type: application/json
...

TO-BE:

$ curl -v -X POST http://localhost:8000/v1/chat/completions \
> -d '{"model":"meta-llama/Llama-3.2-1B-Instruct","messages":[{"role":"user","content":"Hi"}]}' \
> -H 'X-Request-Id: aaaa' \
> -H 'Content-Type: Application/json'
*   Trying 127.0.0.1:8000...
* TCP_NODELAY set
* Connected to localhost (127.0.0.1) port 8000 (#0)
> POST /v1/chat/completions HTTP/1.1
> Host: localhost:8000
> User-Agent: curl/7.68.0
> Accept: */*
> X-Request-Id: aaaa
> Content-Type: application/json
> Content-Length: 88
> 
* upload completely sent off: 88 out of 88 bytes
* Mark bundle as not supporting multiuse
< HTTP/1.1 200 OK
< date: Tue, 22 Oct 2024 15:58:16 GMT
< server: uvicorn
< content-length: 230
< content-type: application/json
< x-request-id: aaa
...

This is kind of standard approach for server log correlation (production level server usually keep logging incoming/outgoing headers, so no additional cost to pay) and easy to implement (simple middleware on fastapi side).

Of course my feature request is completely orthogonal to this PR; request id in chat completion stream body is independent of the request id in the header.

Sorry I missed this earlier. Thanks for bringing this up, can you open a new issue for this specifically so we can have a more comprehensive discussion about this?

[cjackal]

Sorry I missed this earlier. Thanks for bringing this up, can you open a new issue for this specifically so we can have a more comprehensive discussion about this?

I should say sorry for getting off the PR and spamming notifications; issue and PR for request id header are on the way.