Skip to content

Abort logic for harbor trajectories #28

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.

Sign up for GitHub

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

Open
nightlessbaron wants to merge 2 commits into
base: prod
Choose a base branch
from
Open

Abort logic for harbor trajectories #28

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
cf8163b
add abort logic for harbor trajectories
nightlessbaron May 27, 2026
01debd2
Merge branch 'prod' into e2e/msa-bbq
nightlessbaron May 29, 2026
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: 8 additions & 0 deletions miles/rollout/session/sessions.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -265,6 +265,14 @@ async def chat_completions(request: Request, session_id: str):
session.append_record(record)
# --- lock released here ---

# Signal abort to the agent via a header so it can stop immediately,
# independent of how the caller's LLM client normalizes finish_reason.
# (litellm remaps "abort" → "stop" or "tool_calls" in some versions,
# masking the signal; reading raw response headers is the only
# reliable cross-version detection path.)
if choice.get("finish_reason") == "abort":
result["headers"]["x-sglang-aborted"] = "1"

return backend.build_proxy_response(result)
finally:
_inflight_chat["count"] -= 1
Expand Down
Loading