diff --git a/fastdeploy/cache_manager/cache_messager.py b/fastdeploy/cache_manager/cache_messager.py
index 456ba1c342..409941f7d8 100644
--- a/fastdeploy/cache_manager/cache_messager.py
+++ b/fastdeploy/cache_manager/cache_messager.py
@@ -17,6 +17,7 @@
 import math
 import threading
 import time
+import traceback
 
 import numpy as np
 import paddle
@@ -309,4 +310,4 @@ def _prefill_layerwise_send_cache_thread(self):
                     self.last_layer_idx = prefilled_layer_idx
 
         except Exception as e:
-            logger.error(f"prefill layerwise send cache thread has exception: {e}")
+            logger.error(f"prefill layerwise send cache thread has exception: {e}, {str(traceback.format_exc())}")
diff --git a/fastdeploy/cache_manager/cache_transfer_manager.py b/fastdeploy/cache_manager/cache_transfer_manager.py
index 34ccf144ca..5078a513dd 100644
--- a/fastdeploy/cache_manager/cache_transfer_manager.py
+++ b/fastdeploy/cache_manager/cache_transfer_manager.py
@@ -19,6 +19,7 @@
 import json
 import queue
 import time
+import traceback
 
 import numpy as np
 import paddle
@@ -342,7 +343,7 @@ def do_data_transfer(self):
                     if self.rank == 0:
                         self.cache_task_queue.barrier3.reset()
             except Exception as e:
-                logger.info(f"do_data_transfer: error: {e}")
+                logger.info(f"do_data_transfer: error: {e}, {str(traceback.format_exc())}")
 
     def _transfer_data(
         self,
diff --git a/fastdeploy/cache_manager/prefix_cache_manager.py b/fastdeploy/cache_manager/prefix_cache_manager.py
index f033a565c9..e57f0f43b8 100644
--- a/fastdeploy/cache_manager/prefix_cache_manager.py
+++ b/fastdeploy/cache_manager/prefix_cache_manager.py
@@ -20,6 +20,7 @@
 import sys
 import threading
 import time
+import traceback
 import uuid
 from collections import defaultdict
 from concurrent.futures import ThreadPoolExecutor
@@ -469,7 +470,7 @@ def update_cache_blocks(self, task, block_size):
                 self.leaf_req_map[leaf_node].add(req_id)
                 self.cache_info[req_id] = (leaf_node, input_ids)
         except Exception as e:
-            logger.error(f"update_cache_blocks, error: {type(e)} {e}")
+            logger.error(f"update_cache_blocks, error: {type(e)} {e}, {str(traceback.format_exc())}")
             raise e
 
     def request_match_blocks(self, task, block_size, *args):
@@ -555,7 +556,7 @@ def request_match_blocks(self, task, block_size, *args):
                 )
                 return common_block_ids, matched_token_num, hit_info
             except Exception as e:
-                logger.error(f"request_block_ids: error: {type(e)} {e}")
+                logger.error(f"request_block_ids: error: {type(e)} {e}, {str(traceback.format_exc())}")
                 raise e
 
     def request_block_ids(self, task, block_size, dec_token_num, *args):
@@ -660,7 +661,7 @@ def request_block_ids(self, task, block_size, dec_token_num, *args):
                 )
                 return common_block_ids, unique_block_ids, hit_info
             except Exception as e:
-                logger.error(f"request_block_ids: error: {type(e)} {e}")
+                logger.error(f"request_block_ids: error: {type(e)} {e}, {str(traceback.format_exc())}")
                 raise e
 
     def release_block_ids_async(self, task):
@@ -709,7 +710,7 @@ def release_block_ids(self, task):
                 )
                 return
             except Exception as e:
-                logger.error(f"release_block_ids: error: {type(e)} {e}")
+                logger.error(f"release_block_ids: error: {type(e)} {e}, {str(traceback.format_exc())}")
                 raise e
 
     def _handle_free_gpu_node_without_cpu(self, node):
@@ -899,7 +900,7 @@ def free_block_ids_async(self, need_block_num):
                 else:
                     self.gpu_free_task_future = None
             except Exception as e:
-                logger.error(f"free_block_ids_async: error: {type(e)} {e}")
+                logger.error(f"free_block_ids_async: error: {type(e)} {e}, {str(traceback.format_exc())}")
                 raise e
 
     def free_cpu_block_ids(self, need_block_num):
@@ -1218,5 +1219,5 @@ def recv_data_transfer_result(self):
                     + f"task_cpu_block_id {task_cpu_block_id} event_type {event_type} done"
                 )
             except Exception as e:
-                logger.warning(f"recv_data_transfer_result: error: {e}")
+                logger.warning(f"recv_data_transfer_result: error: {e}, {str(traceback.format_exc())}")
                 raise e
diff --git a/fastdeploy/engine/engine.py b/fastdeploy/engine/engine.py
index cb9eef2b8a..3494186fa4 100644
--- a/fastdeploy/engine/engine.py
+++ b/fastdeploy/engine/engine.py
@@ -597,7 +597,7 @@ def receiver_loop():
                         time.sleep(0.001)
 
                 except Exception as e:
-                    llm_logger.error(f"Error in main loop: {e}")
+                    llm_logger.error(f"Error in main loop: {e}, {str(traceback.format_exc())}")
                     time.sleep(0.1)
 
         threading.Thread(target=receiver_loop, daemon=True).start()
@@ -985,7 +985,9 @@ def _exit_sub_services(self):
                 try:
                     os.killpg(p.pid, signal.SIGTERM)
                 except Exception as e:
-                    print(f"Error extracting file: {e}")
+                    console_logger.error(
+                        f"Error killing cache manager process {p.pid}: {e}, {str(traceback.format_exc())}"
+                    )
         self.worker_ready_signal.clear()
         self.exist_task_signal.clear()
         self.exist_swapped_task_signal.clear()
@@ -998,7 +1000,7 @@ def _exit_sub_services(self):
             try:
                 os.killpg(self.worker_proc.pid, signal.SIGTERM)
             except Exception as e:
-                print(f"Error extracting sub services: {e}")
+                console_logger.error(f"Error extracting sub services: {e}, {str(traceback.format_exc())}")
 
         self.engine_worker_queue.cleanup()
         if hasattr(self, "zmq_server") and self.zmq_server is not None:
@@ -1173,7 +1175,7 @@ def generate(self, prompts, stream):
         try:
             req_id = self._format_and_add_data(prompts)
         except Exception as e:
-            llm_logger.error(f"Error happend while adding request, details={e}")
+            llm_logger.error(f"Error happend while adding request, details={e}, {str(traceback.format_exc())}")
             raise EngineError(str(e), error_code=400)
 
         # Get the result of the current request
diff --git a/fastdeploy/engine/expert_service.py b/fastdeploy/engine/expert_service.py
index 9cf5f97f7f..2ed5f8924a 100644
--- a/fastdeploy/engine/expert_service.py
+++ b/fastdeploy/engine/expert_service.py
@@ -269,7 +269,7 @@ def receiver_loop():
                         time.sleep(0.001)
                         continue
                 except Exception as e:
-                    llm_logger.error(f"get decode tasks error: {e}")
+                    llm_logger.error(f"get decode tasks error: {e}, {str(traceback.format_exc())}")
 
         threading.Thread(target=receiver_loop, daemon=True).start()
 
@@ -378,4 +378,4 @@ def start_expert_service(cfg, local_data_parallel_id, ipc_signal_suffix):
         expert_service.start(ipc_signal_suffix, local_data_parallel_id)
         expert_service.split_connector.start_receiver()
     except Exception as e:
-        llm_logger.exception(f"Expert service failed to start: {e}")
+        llm_logger.exception(f"Expert service failed to start: {e}, {str(traceback.format_exc())}")
diff --git a/fastdeploy/engine/sched/resource_manager_v1.py b/fastdeploy/engine/sched/resource_manager_v1.py
index 26eba4ae09..ec8703ee0d 100644
--- a/fastdeploy/engine/sched/resource_manager_v1.py
+++ b/fastdeploy/engine/sched/resource_manager_v1.py
@@ -16,6 +16,7 @@
 
 import threading
 import time
+import traceback
 from collections import deque
 from collections.abc import Iterable
 from concurrent.futures import ThreadPoolExecutor
@@ -389,7 +390,7 @@ def get_prefix_cached_blocks(self, request: Request):
             request.cache_prepare_time = time.time() - cache_prepare_time
             return True
         except Exception as e:
-            llm_logger.error(f"prefix match blocks error: {e}, waiting reschedule...")
+            llm_logger.error(f"prefix match blocks error: {e}, {str(traceback.format_exc())} waiting reschedule...")
             return False
 
     def add_request(self, request: Request) -> None:
@@ -441,4 +442,4 @@ def finish_requests(self, request_ids: Union[str, Iterable[str]]):
                     self.stop_flags[request.idx] = True
                     del self.requests[req_id]
         except Exception as e:
-            llm_logger.error(e)
+            llm_logger.error(f"finish_request err: {e}, {str(traceback.format_exc())}")
diff --git a/fastdeploy/entrypoints/api_server.py b/fastdeploy/entrypoints/api_server.py
index f27c008314..4f4d7f2250 100644
--- a/fastdeploy/entrypoints/api_server.py
+++ b/fastdeploy/entrypoints/api_server.py
@@ -15,6 +15,7 @@
 """
 
 import json
+import traceback
 
 import uvicorn
 from fastapi import FastAPI
@@ -114,7 +115,7 @@ def launch_api_server(args) -> None:
             log_level="info",
         )  # set log level to error to avoid log
     except Exception as e:
-        api_server_logger.error(f"launch sync http server error, {e}")
+        api_server_logger.error(f"launch sync http server error, {e}, {str(traceback.format_exc())}")
 
 
 def main():
diff --git a/fastdeploy/entrypoints/engine_client.py b/fastdeploy/entrypoints/engine_client.py
index daed93b8f9..cf1ebdd297 100644
--- a/fastdeploy/entrypoints/engine_client.py
+++ b/fastdeploy/entrypoints/engine_client.py
@@ -15,6 +15,7 @@
 """
 
 import time
+import traceback
 import uuid
 
 import numpy as np
@@ -141,7 +142,7 @@ def add_requests(self, task):
             work_process_metrics.prompt_tokens_total.inc(input_ids_len)
             work_process_metrics.request_prompt_tokens.observe(input_ids_len)
         except Exception as e:
-            api_server_logger.error(e)
+            api_server_logger.error(f"add_requests error: {e}, {str(traceback.format_exc())}")
             raise EngineError(str(e), error_code=400)
 
         if input_ids_len + min_tokens >= self.max_model_len:
@@ -194,7 +195,7 @@ def add_requests(self, task):
             else:
                 self.zmq_client.send_pyobj(task)
         except Exception as e:
-            api_server_logger.error(e)
+            api_server_logger.error(f"zmq_client send task error: {e}, {str(traceback.format_exc())}")
             raise EngineError(str(e), error_code=400)
 
     def vaild_parameters(self, data):
diff --git a/fastdeploy/entrypoints/llm.py b/fastdeploy/entrypoints/llm.py
index 850c97333e..0dc8e2949b 100644
--- a/fastdeploy/entrypoints/llm.py
+++ b/fastdeploy/entrypoints/llm.py
@@ -346,7 +346,7 @@ def _build_sample_logprobs(self, logprobs_lists: LogprobsLists, topk_logprobs: i
             return result
 
         except Exception as e:
-            llm_logger.error(f"Error building sample logprobs from LogprobsLists: {e}")
+            llm_logger.error(f"Error building sample logprobs from LogprobsLists: {e}, {str(traceback.format_exc())}")
 
     def _run_engine(self, req_ids: list[str], use_tqdm: bool, topk_logprobs: Optional[int] = None):
         """
diff --git a/fastdeploy/entrypoints/openai/api_server.py b/fastdeploy/entrypoints/openai/api_server.py
index 98bb071d31..ca0b45e7f9 100644
--- a/fastdeploy/entrypoints/openai/api_server.py
+++ b/fastdeploy/entrypoints/openai/api_server.py
@@ -18,6 +18,7 @@
 import os
 import threading
 import time
+import traceback
 from collections.abc import AsyncGenerator
 from contextlib import asynccontextmanager
 from multiprocessing import current_process
@@ -159,7 +160,7 @@ async def lifespan(app: FastAPI):
         multiprocess.mark_process_dead(os.getpid())
         api_server_logger.info(f"Closing metrics client pid: {pid}")
     except Exception as e:
-        api_server_logger.warning(e)
+        api_server_logger.warning(f"exit error: {e}, {str(traceback.format_exc())}")
 
 
 app = FastAPI(lifespan=lifespan)
@@ -355,7 +356,7 @@ def launch_api_server() -> None:
             log_level="info",
         )  # set log level to error to avoid log
     except Exception as e:
-        api_server_logger.error(f"launch sync http server error, {e}")
+        api_server_logger.error(f"launch sync http server error, {e}, {str(traceback.format_exc())}")
 
 
 metrics_app = FastAPI()
diff --git a/fastdeploy/entrypoints/openai/serving_chat.py b/fastdeploy/entrypoints/openai/serving_chat.py
index e941719703..ba277a3874 100644
--- a/fastdeploy/entrypoints/openai/serving_chat.py
+++ b/fastdeploy/entrypoints/openai/serving_chat.py
@@ -101,7 +101,9 @@ async def create_chat_completion(self, request: ChatCompletionRequest):
                 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))
+                error_msg = f"request[{request_id}] generator error: {str(e)}, {str(traceback.format_exc())}"
+                api_server_logger.error(error_msg)
+                return ErrorResponse(code=400, message=error_msg)
 
             del current_req_dict
 
@@ -115,11 +117,19 @@ async def create_chat_completion(self, request: ChatCompletionRequest):
                         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}")
+                    error_msg = f"request[{request_id}]full generator error: {str(e)}, {str(traceback.format_exc())}"
+                    api_server_logger.error(error_msg)
+                    return ErrorResponse(code=408, message=error_msg)
+        except Exception as e:
+            error_msg = (
+                f"request[{request_id}] waiting error: {str(e)}, {str(traceback.format_exc())}, "
+                f"max waiting time: {self.max_waiting_time}"
+            )
+            api_server_logger.error(error_msg)
+            return ErrorResponse(code=408, message=error_msg)
 
     def _create_streaming_error_response(self, message: str) -> str:
+        api_server_logger.error(message)
         error_response = ErrorResponse(
             code=400,
             message=message,
@@ -336,7 +346,9 @@ async def chat_completion_stream_generator(
                 yield f"data: {chunk.model_dump_json(exclude_unset=True)}\n\n"
 
         except Exception as e:
-            error_data = self._create_streaming_error_response(str(e))
+            error_data = self._create_streaming_error_response(
+                f"request[{request_id}] generate stream error: {str(e)}, {str(traceback.format_exc())}"
+            )
             yield f"data: {error_data}\n\n"
         finally:
             dealer.close()
@@ -556,6 +568,6 @@ def _build_logprobs_response(
             return LogProbs(content=[sampled_entry])
 
         except Exception as e:
-            api_server_logger.error("Error in _build_logprobs_response: %s", e)
-            api_server_logger.error(traceback.format_exc())
+            error_msg = f"Error in _build_logprobs_response: {e}, {str(traceback.format_exc())}"
+            api_server_logger.error(error_msg)
             return None
diff --git a/fastdeploy/entrypoints/openai/serving_completion.py b/fastdeploy/entrypoints/openai/serving_completion.py
index 43336dac69..fdcf106ba6 100644
--- a/fastdeploy/entrypoints/openai/serving_completion.py
+++ b/fastdeploy/entrypoints/openai/serving_completion.py
@@ -16,6 +16,7 @@
 
 import asyncio
 import time
+import traceback
 import uuid
 from typing import List, Optional
 
@@ -92,7 +93,9 @@ async def create_completion(self, request: CompletionRequest):
             else:
                 raise ValueError("Prompt must be a string, a list of strings or a list of integers.")
         except Exception as e:
-            return ErrorResponse(message=str(e), code=400)
+            error_msg = f"OpenAIServingCompletion create_completion: {e}, {str(traceback.format_exc())}"
+            api_server_logger.error(error_msg)
+            return ErrorResponse(message=error_msg, code=400)
 
         if request_prompt_ids is not None:
             request_prompts = request_prompt_ids
@@ -106,8 +109,13 @@ async def create_completion(self, request: CompletionRequest):
                 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}")
+        except Exception as e:
+            error_msg = (
+                f"OpenAIServingCompletion waiting error: {e}, {str(traceback.format_exc())}, "
+                f"max waiting time: {self.max_waiting_time}"
+            )
+            api_server_logger.error(error_msg)
+            return ErrorResponse(code=408, message=error_msg)
 
         try:
             for idx, prompt in enumerate(request_prompts):
@@ -121,6 +129,8 @@ async def create_completion(self, request: CompletionRequest):
                     text_after_process_list.append(current_req_dict.get("text_after_process"))
                     prompt_batched_token_ids.append(prompt_token_ids)
                 except Exception as e:
+                    error_msg = f"OpenAIServingCompletion format error: {e}, {str(traceback.format_exc())}"
+                    api_server_logger.error(error_msg)
                     return ErrorResponse(message=str(e), code=400)
 
                 del current_req_dict
@@ -147,10 +157,16 @@ async def create_completion(self, request: CompletionRequest):
                         text_after_process_list=text_after_process_list,
                     )
                 except Exception as e:
-                    return ErrorResponse(code=400, message=str(e))
+                    error_msg = (
+                        f"OpenAIServingCompletion completion_full_generator error: {e}, {str(traceback.format_exc())}"
+                    )
+                    api_server_logger.error(error_msg)
+                    return ErrorResponse(code=400, message=error_msg)
 
         except Exception as e:
-            return ErrorResponse(message=str(e), code=400)
+            error_msg = f"OpenAIServingCompletion create_completion error: {e}, {str(traceback.format_exc())}"
+            api_server_logger.error(error_msg)
+            return ErrorResponse(message=error_msg, code=400)
 
     async def completion_full_generator(
         self,
@@ -431,6 +447,7 @@ async def completion_stream_generator(
                     choices = []
 
         except Exception as e:
+            api_server_logger.error(f"Error in completion_stream_generator: {e}, {str(traceback.format_exc())}")
             yield f"data: {ErrorResponse(message=str(e), code=400).model_dump_json(exclude_unset=True)}\n\n"
         finally:
             del request
@@ -614,5 +631,5 @@ def _build_logprobs_response(
             )
 
         except Exception as e:
-            api_server_logger.error("Error in _build_logprobs_response: %s", e)
+            api_server_logger.error(f"Error in _build_logprobs_response: {str(e)}, {str(traceback.format_exc())}")
             return None
diff --git a/fastdeploy/entrypoints/openai/tool_parsers/ernie_x1_tool_parser.py b/fastdeploy/entrypoints/openai/tool_parsers/ernie_x1_tool_parser.py
index cec1f68401..6f0534cf1e 100644
--- a/fastdeploy/entrypoints/openai/tool_parsers/ernie_x1_tool_parser.py
+++ b/fastdeploy/entrypoints/openai/tool_parsers/ernie_x1_tool_parser.py
@@ -14,6 +14,7 @@
 
 import json
 import re
+import traceback
 import uuid
 from collections.abc import Sequence
 from typing import Union
@@ -162,10 +163,12 @@ def extract_tool_calls(self, model_output: str, request: ChatCompletionRequest)
                                 }
                             )
                     except Exception as e:
-                        data_processor_logger.debug(f"Failed to parse tool call: {str(e)}")
+                        data_processor_logger.error(
+                            f"Failed to parse tool call: {str(e)}, {str(traceback.format_exc())}"
+                        )
                         continue
                 except Exception as e:
-                    data_processor_logger.debug(f"Failed to parse tool call: {str(e)}")
+                    data_processor_logger.error(f"Failed to parse tool call: {str(e)}, {str(traceback.format_exc())}")
                     continue
 
             if not function_call_arr:
@@ -211,7 +214,9 @@ def extract_tool_calls(self, model_output: str, request: ChatCompletionRequest)
             )
 
         except Exception as e:
-            data_processor_logger.error(f"Error in extracting tool call from response: {str(e)}")
+            data_processor_logger.error(
+                f"Error in extracting tool call from response: {str(e)}, {str(traceback.format_exc())}"
+            )
             return ExtractedToolCallInformation(tools_called=False, tool_calls=None, content=model_output)
 
     def extract_tool_calls_streaming(
@@ -302,7 +307,9 @@ def extract_tool_calls_streaming(
                                 self.streamed_args_for_tool[self.current_tool_id] = args_json
                                 return delta
                     except Exception as e:
-                        data_processor_logger.debug(f"Partial arguments parsing: {str(e)}")
+                        data_processor_logger.error(
+                            f"Partial arguments parsing: {str(e)}, {str(traceback.format_exc())}"
+                        )
 
             if "</tool_call>" in self.buffer:
                 end_pos = self.buffer.find("</tool_call>")
@@ -316,5 +323,7 @@ def extract_tool_calls_streaming(
             return delta
 
         except Exception as e:
-            data_processor_logger.error(f"Error in streaming tool call extraction: {str(e)}")
+            data_processor_logger.error(
+                f"Error in streaming tool call extraction: {str(e)}, {str(traceback.format_exc())}"
+            )
             return None
diff --git a/fastdeploy/input/ernie_vl_processor.py b/fastdeploy/input/ernie_vl_processor.py
index 5c64952a14..e7aad6aa64 100644
--- a/fastdeploy/input/ernie_vl_processor.py
+++ b/fastdeploy/input/ernie_vl_processor.py
@@ -14,6 +14,8 @@
 # limitations under the License.
 """
 
+import traceback
+
 import numpy as np
 from paddleformers.generation import GenerationConfig
 
@@ -152,7 +154,7 @@ def _parse_processor_kwargs(self, kwargs):
             return kwargs
 
         except Exception as e:
-            data_processor_logger.warning(f"Invalid mm-processor-kwargs format: {e}")
+            data_processor_logger.warning(f"Invalid mm-processor-kwargs format: {e}, {str(traceback.format_exc())}")
             return {}
 
     def _parse_limits(self, limits):
diff --git a/fastdeploy/inter_communicator/engine_cache_queue.py b/fastdeploy/inter_communicator/engine_cache_queue.py
index 03fae97d7d..6f56550386 100644
--- a/fastdeploy/inter_communicator/engine_cache_queue.py
+++ b/fastdeploy/inter_communicator/engine_cache_queue.py
@@ -16,6 +16,7 @@
 
 import threading
 import time
+import traceback
 from multiprocessing.managers import (
     AcquirerProxy,
     BaseManager,
@@ -275,5 +276,5 @@ def empty(self):
         try:
             return len(self.transfer_task_queue) == 0
         except Exception as e:
-            logger.error(f"empty function meets error: {e}")
+            logger.error(f"empty function meets error: {e}, {str(traceback.format_exc())}")
             raise e
diff --git a/fastdeploy/inter_communicator/zmq_client.py b/fastdeploy/inter_communicator/zmq_client.py
index 5143d9d47a..6affcd8e7a 100644
--- a/fastdeploy/inter_communicator/zmq_client.py
+++ b/fastdeploy/inter_communicator/zmq_client.py
@@ -17,6 +17,7 @@
 import os
 import threading
 import time
+import traceback
 
 import msgpack
 import zmq
@@ -143,7 +144,7 @@ def send_multipart(self, req_id, data):
             llm_logger.error(f"[{req_id}] zmq error: {e}")
             self.req_dict[req_id] = -1
         except Exception as e:
-            llm_logger.error(f"Send result to zmq client failed: {e}")
+            llm_logger.error(f"Send result to zmq client failed: {e}, {str(traceback.format_exc())}")
 
         if data[-1].finished:
             with self.mutex:
@@ -163,7 +164,7 @@ def receive_json_once(self, block=False):
             return None, None
         except Exception as e:
             self.close()
-            llm_logger.warning(f"{e}")
+            llm_logger.warning(f"{e}, {str(traceback.format_exc())}")
             return str(e), None
 
     def receive_pyobj_once(self, block=False):
@@ -179,7 +180,7 @@ def receive_pyobj_once(self, block=False):
             return None, None
         except Exception as e:
             self.close()
-            llm_logger.warning(f"{e}")
+            llm_logger.warning(f"{e}, {str(traceback.format_exc())}")
             return str(e), None
 
     def _clear_ipc(self, name):
@@ -214,7 +215,7 @@ def close(self):
             self._clear_ipc(self.file_name)
             self._clear_ipc(self.router_path)
         except Exception as e:
-            llm_logger.warning(f"Failed to close ZMQ connection - {e}")
+            llm_logger.warning(f"Failed to close ZMQ connection - {e}, {str(traceback.format_exc())}")
             return
 
     def __exit__(self, exc_type, exc_val, exc_tb):
diff --git a/fastdeploy/model_executor/guided_decoding/base_guided_decoding.py b/fastdeploy/model_executor/guided_decoding/base_guided_decoding.py
index 7baf2fe971..b23d0c85d8 100644
--- a/fastdeploy/model_executor/guided_decoding/base_guided_decoding.py
+++ b/fastdeploy/model_executor/guided_decoding/base_guided_decoding.py
@@ -15,6 +15,7 @@
 """
 
 import os
+import traceback
 from concurrent.futures import ThreadPoolExecutor
 
 from fastdeploy.config import ErnieArchitectures, FDConfig
@@ -300,7 +301,7 @@ def _get_tokenizer_hf(self):
 
             return tokenizer
         except Exception as e:
-            raise Exception(f"Fail to initialize hf tokenizer: {e}")
+            raise Exception(f"Fail to initialize hf tokenizer: {e}, {str(traceback.format_exc())}")
 
     def add_cache(self, schemata_key: tuple[str, str], processor: LogitsProcessorBase) -> None:
         """
diff --git a/fastdeploy/model_executor/guided_decoding/xgrammar_backend.py b/fastdeploy/model_executor/guided_decoding/xgrammar_backend.py
index f702a1085e..0d448d4293 100644
--- a/fastdeploy/model_executor/guided_decoding/xgrammar_backend.py
+++ b/fastdeploy/model_executor/guided_decoding/xgrammar_backend.py
@@ -16,6 +16,7 @@
 
 import json
 import re
+import traceback
 from typing import Any, List, Optional
 
 import paddle
@@ -263,7 +264,7 @@ def _json_processor(self, schemata: str) -> Optional[XGrammarProcessor]:
         try:
             compiled_grammar = self.grammar_compiler.compile_json_schema(schemata, any_whitespace=self.any_whitespace)
         except Exception as e:
-            llm_logger.error(f"Failed to compile json schema: {e}")
+            llm_logger.error(f"Failed to compile json schema: {e}, {str(traceback.format_exc())}")
             return None
         return self._create_processor(compiled_grammar)
 
@@ -280,7 +281,7 @@ def _regex_processor(self, schemata: str) -> Optional[XGrammarProcessor]:
         try:
             compiled_grammar = self.grammar_compiler.compile_regex(schemata)
         except Exception as e:
-            llm_logger.error(f"Failed to compile regex schema: {e}")
+            llm_logger.error(f"Failed to compile regex schema: {e}, {str(traceback.format_exc())}")
             return None
         return self._create_processor(compiled_grammar)
 
@@ -297,7 +298,7 @@ def _grammar_processor(self, schemata: str) -> Optional[XGrammarProcessor]:
         try:
             compiled_grammar = self.grammar_compiler.compile_grammar(schemata)
         except Exception as e:
-            llm_logger.error(f"Failed to compile ebnf schema: {e}")
+            llm_logger.error(f"Failed to compile ebnf schema: {e}, {str(traceback.format_exc())}")
             return None
         return self._create_processor(compiled_grammar)
 
@@ -324,7 +325,7 @@ def _structural_tag_processor(self, schemata: str) -> Optional[XGrammarProcessor
 
             compiled_grammar = self.grammar_compiler.compile_structural_tag(tags, structural_tag["triggers"])
         except Exception as e:
-            llm_logger.error(f"Failed to compile structural tags schema: {e}")
+            llm_logger.error(f"Failed to compile structural tags schema: {e}, {str(traceback.format_exc())}")
             return None
         return self._create_processor(compiled_grammar)
 
diff --git a/fastdeploy/output/token_processor.py b/fastdeploy/output/token_processor.py
index ebb64cebc7..36ab0c362b 100644
--- a/fastdeploy/output/token_processor.py
+++ b/fastdeploy/output/token_processor.py
@@ -201,7 +201,7 @@ def process_metrics():
                         self.prefill_time_signal.value[current_index] = 0
                     current_index += 1
             except Exception as e:
-                llm_logger.error(f"Error processing prefill metrics: {e}")
+                llm_logger.error(f"Error processing prefill metrics: {e}, {str(traceback.format_exc())}")
 
         self.executor.submit(process_metrics)
 
@@ -215,7 +215,7 @@ def postprocess(self, batch_result):
         try:
             self.cached_generated_tokens.put_results(batch_result)
         except Exception as e:
-            llm_logger.error(f"Error in TokenProcessor's postprocess: {e}")
+            llm_logger.error(f"Error in TokenProcessor's postprocess: {e}, {str(traceback.format_exc())}")
 
     def _recycle_resources(self, task_id, index, task, result=None, is_prefill=False):
         """
diff --git a/fastdeploy/platforms/cuda.py b/fastdeploy/platforms/cuda.py
index 6676d3c0f5..38504134a1 100644
--- a/fastdeploy/platforms/cuda.py
+++ b/fastdeploy/platforms/cuda.py
@@ -14,6 +14,8 @@
 # limitations under the License.
 """
 
+import traceback
+
 import paddle
 
 from fastdeploy.utils import console_logger as logger
@@ -40,7 +42,8 @@ def available(self):
             logger.warning(
                 "You are using GPU version PaddlePaddle, but there is no GPU "
                 "detected on your machine. Maybe CUDA devices is not set properly."
-                f"\n Original Error is {e}"
+                f"\n Original Error is {e}, "
+                f"{str(traceback.format_exc())}"
             )
             return False
 
diff --git a/fastdeploy/platforms/dcu.py b/fastdeploy/platforms/dcu.py
index bfd848335c..c18c45aca4 100644
--- a/fastdeploy/platforms/dcu.py
+++ b/fastdeploy/platforms/dcu.py
@@ -14,6 +14,8 @@
 """
 dcu platform file
 """
+import traceback
+
 import paddle
 from paddleformers.utils.log import logger
 
@@ -39,7 +41,8 @@ def available(self):
             logger.warning(
                 "You are using GPU version PaddlePaddle, but there is no GPU "
                 "detected on your machine. Maybe CUDA devices is not set properly."
-                f"\n Original Error is {e}"
+                f"\n Original Error is {e}, "
+                f"{str(traceback.format_exc())}"
             )
             return False
 
diff --git a/fastdeploy/platforms/gcu.py b/fastdeploy/platforms/gcu.py
index e812113e1e..76bb170b54 100644
--- a/fastdeploy/platforms/gcu.py
+++ b/fastdeploy/platforms/gcu.py
@@ -14,6 +14,8 @@
 # limitations under the License.
 """
 
+import traceback
+
 import paddle
 
 from fastdeploy.utils import console_logger as logger
@@ -40,7 +42,8 @@ def available(self):
             logger.warning(
                 "You are using GCUPlatform, but there is no GCU "
                 "detected on your machine. Maybe GCU devices is not set properly."
-                f"\n Original Error is {e}"
+                f"\n Original Error is {e}, "
+                f"{str(traceback.format_exc())}"
             )
             return False
 
diff --git a/fastdeploy/platforms/maca.py b/fastdeploy/platforms/maca.py
index f695a3d01a..250cebf6e1 100644
--- a/fastdeploy/platforms/maca.py
+++ b/fastdeploy/platforms/maca.py
@@ -17,6 +17,7 @@
 """
 maca platform file
 """
+import traceback
 
 import paddle
 from paddleformers.utils.log import logger
@@ -43,7 +44,8 @@ def available(self):
             logger.warning(
                 "You are using GPU version PaddlePaddle, but there is no GPU "
                 "detected on your machine. Maybe CUDA devices is not set properly."
-                f"\n Original Error is {e}"
+                f"\n Original Error is {e}, "
+                f"{str(traceback.format_exc())}"
             )
             return False
 
diff --git a/fastdeploy/platforms/xpu.py b/fastdeploy/platforms/xpu.py
index 2f31107423..8bc8236359 100644
--- a/fastdeploy/platforms/xpu.py
+++ b/fastdeploy/platforms/xpu.py
@@ -11,6 +11,8 @@
 # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 # See the License for the specific language governing permissions and
 # limitations under the License.
+import traceback
+
 import paddle
 
 from fastdeploy.utils import console_logger as logger
@@ -38,7 +40,8 @@ def available(self):
             logger.warning(
                 "You are using XPU version PaddlePaddle, but there is no XPU "
                 "detected on your machine. Maybe CUDA devices is not set properly."
-                f"\n Original Error is {e}"
+                f"\n Original Error is {e}, "
+                f"{str(traceback.format_exc())}"
             )
             return False
 
diff --git a/fastdeploy/scheduler/global_scheduler.py b/fastdeploy/scheduler/global_scheduler.py
index 8d9b67a6a8..f3962992cc 100644
--- a/fastdeploy/scheduler/global_scheduler.py
+++ b/fastdeploy/scheduler/global_scheduler.py
@@ -237,7 +237,7 @@ def _keep_alive(self):
                 )
                 time.sleep(self.keep_alive_duration / 2)
             except Exception as e:
-                scheduler_logger.error(f"Scheduler keep alive failed: {e}")
+                scheduler_logger.error(f"Scheduler keep alive failed: {e}, {str(traceback.format_exc())}")
                 time.sleep(min(3, self.keep_alive_duration / 4))
 
     def _scheduler_name_from_request_queue(self, request_queue: str) -> str:
diff --git a/fastdeploy/scheduler/splitwise_scheduler.py b/fastdeploy/scheduler/splitwise_scheduler.py
index 61dbd22309..ab1799f440 100644
--- a/fastdeploy/scheduler/splitwise_scheduler.py
+++ b/fastdeploy/scheduler/splitwise_scheduler.py
@@ -20,6 +20,7 @@
 import random
 import threading
 import time
+import traceback
 from collections import deque
 from typing import List
 
@@ -379,7 +380,7 @@ def run(self):
                 if total == 0:
                     time.sleep(0.01)
             except Exception as e:
-                logger.error(f"ResultsReader{self.idx} sync results error: {e!s}")
+                logger.error(f"ResultsReader{self.idx} sync results error: {e!s}, {str(traceback.format_exc())}")
 
     def sync_results(self, keys):
         """
@@ -402,7 +403,7 @@ def sync_results(self, keys):
                     result = RequestOutput.from_dict(data)
                     self.data.appendleft(result)
                 except Exception as e:
-                    logger.error(f"Parse Result Error:{e}, {result}")
+                    logger.error(f"Parse Result Error:{e}, {str(traceback.format_exc())}, {result}")
         return total
 
 
@@ -498,7 +499,7 @@ def loop_schedule(self):
             except IndexError:
                 continue
             except Exception as e:
-                logger.error(f"APIScheduler Schedule req error: {e!s}")
+                logger.error(f"APIScheduler Schedule req error: {e!s}, {str(traceback.format_exc())}")
 
     def schedule(self, req, pnodes, dnodes, mnodes, group=""):
         """
@@ -573,8 +574,8 @@ def loop_clear_expired_nodes(self):
                     # logger.info(f"clear expired nodes: {nodeid}")
                     self.client.hdel(self.cluster_key, nodeid)
                 time.sleep(self.clear_expired_nodes_period)
-            except Exception:
-                logger.error("APIScheduler clear expired nodes error: {str(e)}")
+            except Exception as e:
+                logger.error(f"APIScheduler clear expired nodes error: {str(e)}, {str(traceback.format_exc())}")
 
     def select_pd(self, req, nodes, role):
         """
@@ -664,7 +665,7 @@ def run(self):
                     # e = time.time()
                     # logger.info(f"Lpush {self.idx}: {key} used {e-s} {len(items)} items")
             except Exception as e:
-                logger.error(f"ResultWriter write error: {e!s}")
+                logger.error(f"ResultWriter write error: {e!s}, {str(traceback.format_exc())}")
 
 
 class InferScheduler:
@@ -723,7 +724,7 @@ def routine_report(self):
                 self.client.hset(self.cluster_key, self.nodeid, info)
                 time.sleep(self.sync_period / 1000.0)
             except Exception as e:
-                logger.error(f"InferScheduler routine report error: {e!s}")
+                logger.error(f"InferScheduler routine report error: {e!s}, {str(traceback.format_exc())}")
 
     def loop_expire_reqs(self):
         """
@@ -733,8 +734,8 @@ def loop_expire_reqs(self):
             try:
                 self.node.expire_reqs(self.release_load_expire_period)
                 time.sleep(60)
-            except Exception:
-                logger.error("InferScheduler expire reqs error: {e}")
+            except Exception as e:
+                logger.error(f"InferScheduler expire reqs error: {e}, {str(traceback.format_exc())}")
 
     def loop_get_reqs(self):
         """
@@ -772,7 +773,7 @@ def select_writer(req):
                     else:
                         self.node.add_req(req.request_id, 1)
             except Exception as e:
-                logger.error(f"InferScheduler loop get reqs error: {e!s}")
+                logger.error(f"InferScheduler loop get reqs error: {e!s}, {str(traceback.format_exc())}")
 
     def get_requests(
         self,
@@ -807,7 +808,8 @@ def get_requests(
                     return reqs
                 # logger.info(f"Get Requests from Scheduler: {req.request_id}")
                 reqs.append(req)
-            except Exception:
+            except Exception as e:
+                logger.error(f"InferScheduler get requests error: {e}, {str(traceback.format_exc())}")
                 return reqs
         return reqs
 
diff --git a/fastdeploy/splitwise/splitwise_connector.py b/fastdeploy/splitwise/splitwise_connector.py
index 6b4c8ce04d..8924c00f56 100644
--- a/fastdeploy/splitwise/splitwise_connector.py
+++ b/fastdeploy/splitwise/splitwise_connector.py
@@ -16,6 +16,7 @@
 
 import json
 import time
+import traceback
 from concurrent.futures import ThreadPoolExecutor
 from typing import Dict
 
@@ -97,7 +98,7 @@ def start_receiver(self):
                 time.sleep(0.001)
 
             except Exception as e:
-                logger.error(f"Receiver error: {e}")
+                logger.error(f"Receiver error: {e}, {str(traceback.format_exc())}")
                 time.sleep(1)
 
     def _get_push_socket(self, addr):
@@ -152,7 +153,7 @@ def _send_message(self, addr, msg_type: str, payload):
             except zmq.Again:
                 logger.warning(f"Send queue full for {addr}")
             except Exception as e:
-                logger.error(f"Send to {addr} failed: {e}")
+                logger.error(f"Send to {addr} failed: {e}, {str(traceback.format_exc())}")
                 self._close_connection(addr)
 
         except Exception as e:
@@ -433,7 +434,7 @@ def _process_message(self, message: bytes):
                 self.engine_worker_queue.put_cache_info(payload)
 
         except Exception as e:
-            logger.error(f"Message processing failed: {e}")
+            logger.error(f"Message processing failed: {e}, {str(traceback.format_exc())}")
 
     def _handle_prefill(self, tasks):
         """
diff --git a/fastdeploy/worker/utils.py b/fastdeploy/worker/utils.py
index bf727c3bbf..7554c7c08a 100644
--- a/fastdeploy/worker/utils.py
+++ b/fastdeploy/worker/utils.py
@@ -15,6 +15,7 @@
 """
 
 import os
+import traceback
 
 
 def check_safetensors_model(model_dir: str):
@@ -45,5 +46,5 @@ def check_safetensors_model(model_dir: str):
             sum(flags) == safetensors_num
         ), f"Number of safetensor files should be {len(model_files)}, but now it's {sum(flags)}"
     except Exception as e:
-        raise Exception(f"Failed to check unified checkpoint, details: {e}.")
+        raise Exception(f"Failed to check unified checkpoint, details: {e}, {str(traceback.format_exc())}.")
     return is_safetensors
diff --git a/test/ce/accuracy_cases/gsm8k.py b/test/ce/accuracy_cases/gsm8k.py
index 4ccfd24821..75356ca7b9 100644
--- a/test/ce/accuracy_cases/gsm8k.py
+++ b/test/ce/accuracy_cases/gsm8k.py
@@ -6,6 +6,7 @@
 
 import os
 import re
+import traceback
 from concurrent.futures import ThreadPoolExecutor, as_completed
 from urllib.parse import urlparse, urlunparse
 
@@ -122,7 +123,7 @@ def query_model(prompt):
         )
         return response.choices[0].message.content.strip()
     except Exception as e:
-        return f"[Error] {e}"
+        return f"[Error] {e}, {str(traceback.format_exc())}"
 
 
 # ========== 评估函数 ==========
diff --git a/test/ce/deploy/deploy.py b/test/ce/deploy/deploy.py
index aa305360b8..3947d22288 100644
--- a/test/ce/deploy/deploy.py
+++ b/test/ce/deploy/deploy.py
@@ -7,6 +7,7 @@
 import subprocess
 import sys
 import time
+import traceback
 
 import requests
 import yaml
@@ -175,7 +176,7 @@ def stop_server(signum=None, frame=None):
         # 终止进程组（包括所有子进程）
         os.killpg(os.getpgid(pid_port["PID"]), signal.SIGTERM)
     except Exception as e:
-        print(f"Failed to stop server: {e}")
+        print(f"Failed to stop server: {e}, {str(traceback.format_exc())}")
 
     for port in [FD_API_PORT, FD_ENGINE_QUEUE_PORT, FD_METRICS_PORT]:
         try:
@@ -184,7 +185,7 @@ def stop_server(signum=None, frame=None):
                 os.kill(int(pid), signal.SIGKILL)
                 print(f"Killed process on port {port}, pid={pid}")
         except Exception as e:
-            print(f"Failed to killed process on port: {e}")
+            print(f"Failed to kill process on port: {e}, {str(traceback.format_exc())}")
     # 若log目录存在，则重命名为log_timestamp
     if os.path.isdir("./log"):
         os.rename("./log", "./log_{}".format(time.strftime("%Y%m%d%H%M%S")))
@@ -229,8 +230,10 @@ def start_service():
         # 构建命令
         cmd = build_command(final_config)
     except Exception as e:
+        error_msg = f"Failed to start service: {e}, {str(traceback.format_exc())}"
+        print(error_msg)
         return Response(
-            json.dumps({"status": "error", "message": str(e)}, ensure_ascii=False),
+            json.dumps({"status": "error", "message": error_msg}, ensure_ascii=False),
             status=500,
             content_type="application/json",
         )
@@ -264,8 +267,10 @@ def start_service():
 
         return Response(json.dumps(json_data, ensure_ascii=False), status=200, content_type="application/json")
     except Exception as e:
+        error_msg = f"Failed to start service: {e}, {str(traceback.format_exc())}"
+        print(error_msg)
         return Response(
-            json.dumps({"status": "error", "message": str(e)}, ensure_ascii=False),
+            json.dumps({"status": "error", "message": error_msg}, ensure_ascii=False),
             status=500,
             content_type="application/json",
         )
@@ -295,8 +300,10 @@ def switch_service():
         # 构建命令
         cmd = build_command(final_config)
     except Exception as e:
+        error_msg = f"Failed to switch service: {e}, {str(traceback.format_exc())}"
+        print(error_msg)
         return Response(
-            json.dumps({"status": "error", "message": str(e)}, ensure_ascii=False),
+            json.dumps({"status": "error", "message": error_msg}, ensure_ascii=False),
             status=500,
             content_type="application/json",
         )
@@ -330,8 +337,10 @@ def switch_service():
 
         return Response(json.dumps(json_data, ensure_ascii=False), status=200, content_type="application/json")
     except Exception as e:
+        error_msg = f"Failed to switch service: {e}, {str(traceback.format_exc())}"
+        print(error_msg)
         return Response(
-            json.dumps({"status": "error", "message": str(e)}, ensure_ascii=False),
+            json.dumps({"status": "error", "message": error_msg}, ensure_ascii=False),
             status=500,
             content_type="application/json",
         )
@@ -406,8 +415,10 @@ def get_config():
         )
 
     except Exception as e:
+        error_msg = f"{e}, {str(traceback.format_exc())}"
+        print(error_msg)
         return Response(
-            json.dumps({"message": "api_server.log解析失败，请检查log", "error": str(e)}, ensure_ascii=False),
+            json.dumps({"message": "api_server.log解析失败，请检查log", "error": error_msg}, ensure_ascii=False),
             status=500,
             content_type="application/json",
         )
@@ -447,7 +458,7 @@ def tail_file(path, lines=50):
                         with open(path, "r", encoding="utf-8", errors="ignore") as f:
                             return "".join(f.readlines()[-lines:])
                     except Exception as e:
-                        return f"[无法读取 {path}]: {e}\n"
+                        return f"[无法读取 {path}]: {e}, {str(traceback.format_exc())}\n"
 
                 result = f"服务启动超时，耗时：[{timeout}s]\n\n"
                 result += "==== server.log tail 50 ====\n"