Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[BugFix][Core] Multistep Fix Crash on Request Cancellation #8059

Merged
merged 6 commits into from
Aug 31, 2024
Merged
Changes from 5 commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
8 changes: 7 additions & 1 deletion vllm/engine/output_processor/multi_step.py
Original file line number Diff line number Diff line change
Expand Up @@ -88,9 +88,15 @@ def process_outputs(self,
# TODO: Add support for async if necessary
assert not is_async

# Sequences can only be in the RUNNING and FINISHED_ABORTED state
# once they are scheduled. A sequence is moved to FINSIHED_ABORTED
# client disconnects from the server, which can occur while
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Seems incomplete?

Copy link
Collaborator Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

my b

seqs = sequence_group.get_seqs(status=SequenceStatus.RUNNING)
if seqs is None:
seqs = sequence_group.get_seqs(
status=SequenceStatus.FINISHED_ABORTED)

assert seqs, "expected running sequences"
assert seqs, "Expected RUNNING or FINISHED_ABORTED sequences"
assert len(seqs) == 1, (
"Beam search not supported in multi-step decoding.")
seq = seqs[0]
Expand Down
Loading