apps/FastDeploy
1
0
Fork 0
mirror of https://github.com/PaddlePaddle/FastDeploy.git synced 2025-10-05 08:37:06 +08:00
Code Issues Actions 2 Packages Projects Releases Wiki Activity

feat(log):add_request_and_response_log (#3391)

Browse Source 
* feat(log):add_request_and_response_log

* [ci] Retrigger

* [ci] Retrigger
This commit is contained in:
xiaolei373
2025-08-14 19:12:42 +08:00
committed by GitHub
parent 1f5983290c
commit f0519aec67
3 changed files with 27 additions and 3 deletions
Download Patch File Download Diff File Expand all files Collapse all files

19
fastdeploy/entrypoints/openai/serving_completion.py
View File

@@ -190,8 +190,7 @@ class OpenAIServingCompletion:
valid_results[rid] = data
num_choices -= 1
break
return self.request_output_to_completion_response(
res = self.request_output_to_completion_response(
final_res_batch=valid_results,
request=request,
request_id=request_id,
@@ -200,6 +199,8 @@ class OpenAIServingCompletion:
prompt_batched_token_ids=prompt_batched_token_ids,
completion_batched_token_ids=completion_batched_token_ids,
)
api_server_logger.info(f"Completion response: {res.model_dump_json()}")
return res
except Exception as e:
api_server_logger.error(f"Error in completion_full_generator: {e}", exc_info=True)
raise
@@ -278,6 +279,9 @@ class OpenAIServingCompletion:
],
)
yield f"data: {chunk.model_dump_json(exclude_unset=True)}\n\n"
api_server_logger.info(
f"Completion Streaming response send_idx 0: {chunk.model_dump_json()}"
)
first_iteration[idx] = False
self.engine_client.data_processor.process_response_dict(res, stream=True)
@@ -313,6 +317,16 @@ class OpenAIServingCompletion:
output_tokens[idx] += 1
send_idx = output.get("send_idx")
# 只有当 send_idx 明确为 0 时才记录日志
if send_idx == 0 and not request.return_token_ids:
chunk_temp = chunk
chunk_temp.choices = choices
api_server_logger.info(
f"Completion Streaming response send_idx 0: {chunk_temp.model_dump_json()}"
)
del chunk_temp
if len(choices) == max_streaming_response_tokens or res["finished"]:
chunk = CompletionStreamResponse(
id=request_id,
@@ -337,6 +351,7 @@ class OpenAIServingCompletion:
),
)
yield f"data: {usage_chunk.model_dump_json(exclude_unset=True)}\n\n"
api_server_logger.info(f"Completion Streaming response last send: {chunk.model_dump_json()}")
if choices:
chunk.choices = choices
yield f"data: {chunk.model_dump_json(exclude_unset=True)}\n\n"