apps/FastDeploy
1
0
Fork 0
mirror of https://github.com/PaddlePaddle/FastDeploy.git synced 2025-10-05 16:48:03 +08:00
Code Issues Actions 2 Packages Projects Releases Wiki Activity

[BugFix] fix control signal release failed (#3374)

Browse Source 
* [BugFix]

* [BugFix]

* [BugFix]

* [BugFix]

* fix

* fix

---------

Co-authored-by: YuBaoku <49938469+EmmonsCurse@users.noreply.github.com>
Co-authored-by: Jiang-Jia-Jun <163579578+Jiang-Jia-Jun@users.noreply.github.com>
This commit is contained in:
ltd0924
2025-08-14 17:01:25 +08:00
committed by GitHub
parent b2df0311b8
commit 03347626a6
4 changed files with 51 additions and 44 deletions
Download Patch File Download Diff File Expand all files Collapse all files

58
fastdeploy/entrypoints/openai/serving_chat.py
View File

@@ -78,45 +78,45 @@ class OpenAIServingChat:
api_server_logger.error(err_msg)
return ErrorResponse(message=err_msg, code=400)
if request.user is not None:
request_id = f"chatcmpl-{request.user}-{uuid.uuid4()}"
else:
request_id = f"chatcmpl-{uuid.uuid4()}"
api_server_logger.info(f"create chat completion request: {request_id}")
text_after_process = None
try:
current_req_dict = request.to_dict_for_infer(request_id)
current_req_dict["arrival_time"] = time.time()
prompt_token_ids = self.engine_client.format_and_add_data(current_req_dict)
text_after_process = current_req_dict.get("text_after_process")
if isinstance(prompt_token_ids, np.ndarray):
prompt_token_ids = prompt_token_ids.tolist()
except Exception as e:
return ErrorResponse(code=400, message=str(e))
del current_req_dict
try:
api_server_logger.debug(f"{self.engine_client.semaphore.status()}")
if self.max_waiting_time < 0:
await self.engine_client.semaphore.acquire()
else:
await asyncio.wait_for(self.engine_client.semaphore.acquire(), timeout=self.max_waiting_time)
except Exception:
return ErrorResponse(code=408, message=f"Request queued time exceed {self.max_waiting_time}")
api_server_logger.debug(f"current waiting request {self.engine_client.semaphore.status()}")
if request.stream:
return self.chat_completion_stream_generator(
request, request_id, request.model, prompt_token_ids, text_after_process
)
else:
if request.user is not None:
request_id = f"chatcmpl-{request.user}-{uuid.uuid4()}"
else:
request_id = f"chatcmpl-{uuid.uuid4()}"
api_server_logger.info(f"create chat completion request: {request_id}")
text_after_process = None
try:
return await self.chat_completion_full_generator(
request, request_id, request.model, prompt_token_ids, text_after_process
)
current_req_dict = request.to_dict_for_infer(request_id)
current_req_dict["arrival_time"] = time.time()
prompt_token_ids = self.engine_client.format_and_add_data(current_req_dict)
text_after_process = current_req_dict.get("text_after_process")
if isinstance(prompt_token_ids, np.ndarray):
prompt_token_ids = prompt_token_ids.tolist()
except Exception as e:
return ErrorResponse(code=400, message=str(e))
del current_req_dict
if request.stream:
return self.chat_completion_stream_generator(
request, request_id, request.model, prompt_token_ids, text_after_process
)
else:
try:
return await self.chat_completion_full_generator(
request, request_id, request.model, prompt_token_ids, text_after_process
)
except Exception as e:
return ErrorResponse(code=400, message=str(e))
except Exception:
return ErrorResponse(code=408, message=f"Request queued time exceed {self.max_waiting_time}")
def _create_streaming_error_response(self, message: str) -> str:
error_response = ErrorResponse(
code=400,