[multi-lora] 1/n - 7, test: pin update_weight_version abort_all_requests=False regression - #2713
Closed
yushengsu-thu wants to merge 1 commit into
Closed
[multi-lora] 1/n - 7, test: pin update_weight_version abort_all_requests=False regression#2713yushengsu-thu wants to merge 1 commit into
yushengsu-thu wants to merge 1 commit into
Conversation
PR #2589 switched the engine weight-version bump to abort_all_requests=False so in-flight requests survive a weight update, but shipped no test; this pins the exact wire payload so a revert or refactor cannot silently bring the aborts back.
yushengsu-thu
changed the title
yushengsu-thu
marked this pull request as ready for review
August 24, 2026 00:21
![claude[bot]](https://avatars.githubusercontent.com/in/1236702?s=60&v=4){kind=small}
There was a problem hiding this comment.
Claude Code Review
This repository is configured for manual code reviews. Comment @claude review for a one-time review, or @claude review always to subscribe this PR to a review on every future push.
Tip: disable this comment in your organization's Code Review settings.
Collaborator
Author
|
Closing: not an extraction from #2273 (the stack carries no sglang_utils production diff), so it does not belong in the 1/n merge train. |
yushengsu-thu
added a commit
that referenced
this pull request
Aug 24, 2026
…uted sync LoRA sync sends only adapter tensors and never refills base weights; opening the session anyway makes begin/end_weight_update restore and re-pack the quantized base buffers with nothing loaded in between, corrupting the frozen base (reproduced on Kimi-K2.5 W4A16, TP8). Also re-adds the update_weight_version abort_all_requests=False wire pin so main #2589's no-abort behavior cannot silently regress. Absorbed from closed PRs #2715 and #2713.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
PR #2589 changed the engine's weight-version bump to send
abort_all_requests: Falseso in-flight requests survive a weight update, but the behavior landed without a test. This adds a fast CPU test that pins the exact/update_weight_versionwire payload, so a revert or refactor cannot silently reintroduce the aborts. Test-only; no production changes.Part of the [multi-lora] 1/n merge train: #2273 carries the same (or dependent) content and rebases after this merges.