Skip to content

[BugFix] reset exist tasks signal in clear_data #7111

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

Merged
Jiang-Jia-Jun merged 4 commits into from
Mar 31, 2026
+6 −2
Merged

[BugFix] reset exist tasks signal in clear_data #7111

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
aa80f75
[BugFix] reset exist tasks signal in clear_data
liyonghua0910 Mar 31, 2026
37a3ed9
[Fix] fix stale exist tasks signal after weight update
liyonghua0910 Mar 31, 2026
325ed2e
[Chore] downgrade detected new requests log to DEBUG level
liyonghua0910 Mar 31, 2026
c2c57a4
[fix] adjust continue place
liyonghua0910 Mar 31, 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
4 changes: 4 additions & 0 deletions fastdeploy/inter_communicator/engine_worker_queue.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -837,6 +837,10 @@ def clear_data(self):
self.lock.acquire()
self.tasks[:] = list()
self.client_read_flag[:] = [1] * self.num_client
if self.is_single_node:
self.exist_tasks_intra_signal.value[0] = 0
else:
self.exist_tasks_inter_signal.set(0)
self.lock.release()
llm_logger.info("clear data for engine worker queue")

Expand Down
4 changes: 2 additions & 2 deletions fastdeploy/worker/worker_process.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -567,10 +567,10 @@ def event_loop_normal(self) -> None:
self.model_weights_status.value[0] = (
ModelWeightsStatus.UPDATING
) # 所有 Rank 已同步唤醒，启动权重更新流程
continue
continue

if self.exist_task_signal.value[0] == ExistTaskStatus.EXIST or self.task_queue.read_finish_flag.get() == 1:
logger.info(f"Rank: {self.local_rank} Detected new requests.")
logger.debug(f"Rank: {self.local_rank} Detected new requests.")
self.engine_forward_signal.value[0] = 1
tasks, read_finish = self.task_queue.get_tasks()
# Only one of all tp_size client will get read_finish == True.
Expand Down
Loading