[diffusion]: fix scheduler crash on ZMQ messages with unexpected frame counts

[Ratish1]

Motivation

This PR improves the the multimodal generation scheduler's request receiving loop. Previously, the scheduler would crash with a ValueError if it received ZMQ messages with an unexpected number of parts. Fixes #17874

Modifications

• Replaced ZMQ unpacking with flexible frame extraction to support multiple socket types.
• Added exception guards for malformed or non-pickle payloads to prevent the event loop from terminating on invalid traffic.
• Ensured the scheduler remains alive and responsive in environments with active network scanners or health probes.

Accuracy Tests

Benchmarking and Profiling

Checklist

• Format your code according to the Format code with pre-commit.
• Add unit tests according to the Run and add unit tests.
• Update documentation according to Write documentations.
• Provide accuracy and speed benchmark results according to Test the accuracy and Benchmark the speed.
• Follow the SGLang code style guidance.

Review Process

1. Ping Merge Oncalls to start the PR flow. See the PR Merge Process.
2. Get approvals from CODEOWNERS and other reviewers.
3. Trigger CI tests with comments or contact authorized users to do so.
   • /tag-run-ci-label, /rerun-failed-ci, /tag-and-rerun-ci
4. After green CI and required approvals, ask Merge Oncalls to merge.

[gemini-code-assist]

Summary of Changes

Hello @Ratish1, I'm Gemini Code Assist¹! I'm currently reviewing this pull request and will post my feedback shortly. In the meantime, here's a summary to help you and other reviewers quickly get up to speed!

This pull request significantly enhances the stability and resilience of the multimodal generation scheduler. By implementing more flexible ZMQ message parsing and robust error handling for unexpected or malformed payloads, it prevents critical crashes and ensures continuous operation, even in the presence of invalid network traffic.

Highlights

• Scheduler Stability Fix: Fixed a critical crash in the multimodal generation scheduler that occurred when processing ZMQ messages with an unexpected number of frames.
• Flexible ZMQ Message Handling: Replaced rigid ZMQ unpacking with a more flexible approach to support both REQ (3 parts) and DEALER (2 parts) socket types, improving compatibility.
• Robust Payload Processing: Added comprehensive exception guards for pickle.UnpicklingError, IndexError, and EOFError to prevent the event loop from terminating due to malformed or non-pickle payloads.
• Enhanced Responsiveness: Ensures the scheduler remains alive and responsive even when encountering invalid traffic from network scanners or health probes.

🧠 New Feature in Public Preview: You can now enable Memory to help Gemini Code Assist learn from your team's feedback. This makes future code reviews more consistent and personalized to your project's style. Click here to enable Memory in your admin console.

Using Gemini Code Assist

The full guide for Gemini Code Assist can be found on our documentation page, here are some quick tips.

Invoking Gemini

You can request assistance from Gemini at any point by creating a comment using either /gemini <command> or @gemini-code-assist <command>. Below is a summary of the supported commands on the current page.

Feature	Command	Description
Code Review	/gemini review	Performs a code review for the current pull request in its current state.
Pull Request Summary	/gemini summary	Provides a summary of the current pull request in its current state.
Comment	@gemini-code-assist	Responds in comments when explicitly tagged, both in pull request comments and review comments.
Help	/gemini help	Displays a list of available commands.

Customization

To customize Gemini Code Assist for GitHub experience, repository maintainers can create a configuration file and/or provide a custom code review style guide (such as PEP-8 for Python) by creating and adding files to a .gemini/ folder in the base of the repository. Detailed instructions can be found here.

Limitations & Feedback

Gemini Code Assist may make mistakes. Please leave feedback on any instances where its feedback is incorrect or counter productive. You can react with 👍 and 👎 on @gemini-code-assist comments. If you're interested in giving your feedback about your experience with Gemini Code Assist for Github and other Google products, sign up here.

You can also get AI-powered code generation, chat, as well as code reviews directly in the IDE at no cost with the Gemini Code Assist IDE Extension.

Footnotes

1. Review the Privacy Notices, Generative AI Prohibited Use Policy, Terms of Service, and learn how to configure Gemini Code Assist in GitHub here. Gemini can make mistakes, so double check it and use code with caution. ↩

[gemini-code-assist]

Code Review

This pull request enhances the robustness of the ZMQ message receiving loop in the scheduler. The changes correctly handle messages with unexpected frame counts by using flexible frame extraction and expanding the exception handling to cover malformed payloads. This prevents the scheduler from crashing on invalid network traffic, such as from network scanners. The logic is sound and effectively addresses the issue. I have one suggestion to refactor the code for improved clarity.

[adarshxs]

/tag-and-rerun-ci

[adarshxs]

cc @mickqian this LGTM

[adarshxs]

/rerun-failed-ci

[Ratish1]

Hey @mickqian , I wanted to follow up on this PR. thanks

[BBuf]

/tag-and-rerun-ci