fix(api): attach exc_info=True to logger.warning sites that drop the traceback (#41068)

api/configs/remote_settings_sources/apollo/client.py

@@ -167,8 +167,8 @@ class ApolloClient:
                 old_value = old_kv.get(key)
                 if old_value is None:
                     self._change_listener("add", namespace, key, new_value)
-        except BaseException as e:
-            logger.warning(str(e))
+        except BaseException:
+            logger.warning("Apollo config change listener failed", exc_info=True)
 
     def _path_checker(self) -> None:
         if not os.path.isdir(self._cache_file_path):
@@ -235,8 +235,8 @@ class ApolloClient:
                     return
             else:
                 logger.warning("Sleep...")
-        except Exception as e:
-            logger.warning(str(e))
+        except Exception:
+            logger.warning("Apollo long polling failed", exc_info=True)
 
     def _get_net_and_set_local(self, namespace: str, n_id: int, call_change: bool = False) -> None:
         namespace_data = self.get_json_from_net(namespace)

api/core/app/apps/workflow/app_generator.py

@@ -686,8 +686,8 @@ class WorkflowAppGenerator(BaseAppGenerator):
             try:
                 with active_workflow_task(application_generate_entity.task_id):
                     runner.run()
-            except GenerateTaskStoppedError as e:
-                logger.warning("Task stopped: %s", str(e))
+            except GenerateTaskStoppedError:
+                logger.warning("Task stopped", exc_info=True)
                 pass
             except InvokeAuthorizationError:
                 queue_manager.publish_error(

api/core/app/features/annotation_reply/annotation_reply.py

@@ -90,8 +90,8 @@ class AnnotationReplyFeature:
                     )
 
                     return annotation
-        except Exception as e:
-            logger.warning("Query annotation failed, exception: %s.", str(e))
+        except Exception:
+            logger.warning("Query annotation failed.", exc_info=True)
             return None
 
         return None

api/core/entities/provider_configuration.py

@@ -486,8 +486,8 @@ class ProviderConfiguration(BaseModel):
             next_number = max(numbers, default=0) + 1
             return f"API KEY {next_number}"
 
-        except Exception as e:
-            logger.warning("Error generating next credential name: %s", str(e))
+        except Exception:
+            logger.warning("Error generating next credential name", exc_info=True)
             return "API KEY 1"
 
     def _get_provider_names(self):

api/core/model_manager.py

@@ -923,8 +923,8 @@ class LBModelManager:
                         provider=self._provider,
                         credential_type=PluginCredentialType.MODEL,
                     )
-            except Exception as e:
-                logger.warning("Load balancing config %s failed policy compliance check: %s", config.id, str(e))
+            except Exception:
+                logger.warning("Load balancing config %s failed policy compliance check", config.id, exc_info=True)
                 cooldown_load_balancing_configs.append(config)
                 if len(cooldown_load_balancing_configs) >= len(self._load_balancing_configs):
                     # all configs are in cooldown or failed policy compliance

api/core/rag/extractor/word_extractor.py

@@ -131,8 +131,8 @@ class WordExtractor(BaseExtractor):
                         continue
                     try:
                         response = remote_fetcher.make_request("GET", url)
-                    except Exception as e:
-                        logger.warning("Failed to download image from URL: %s: %s", url, str(e))
+                    except Exception:
+                        logger.warning("Failed to download image from URL: %s", url, exc_info=True)
                         continue
                     if response.status_code == 200:
                         image_ext = mimetypes.guess_extension(response.headers.get("Content-Type", ""))

api/core/rag/index_processor/processor/paragraph_index_processor.py

@@ -468,8 +468,8 @@ class ParagraphIndexProcessor(BaseIndexProcessor):
                         file, image_detail_config=ImagePromptMessageContent.DETAIL.LOW
                     )
                     prompt_message_contents.append(file_content)
-                except Exception as e:
-                    logger.warning("Failed to convert image file to prompt message content: %s", str(e))
+                except Exception:
+                    logger.warning("Failed to convert image file to prompt message content", exc_info=True)
                     continue
 
             # Add text content
@@ -572,8 +572,8 @@ class ParagraphIndexProcessor(BaseIndexProcessor):
                     access_controller=_file_access_controller,
                 )
                 file_objects.append(file_obj)
-            except Exception as e:
-                logger.warning("Failed to create File object from UploadFile %s: %s", upload_file.id, str(e))
+            except Exception:
+                logger.warning("Failed to create File object from UploadFile %s", upload_file.id, exc_info=True)
                 continue
 
         return file_objects
@@ -629,8 +629,8 @@ class ParagraphIndexProcessor(BaseIndexProcessor):
                     storage_key=upload_file.key,
                 )
                 file_objects.append(file_obj)
-            except Exception as e:
-                logger.warning("Failed to create File object from UploadFile %s: %s", upload_file.id, str(e))
+            except Exception:
+                logger.warning("Failed to create File object from UploadFile %s", upload_file.id, exc_info=True)
                 continue
 
         return file_objects

api/providers/trace/trace-aliyun/src/dify_trace_aliyun/data_exporter/traceclient.py

@@ -86,7 +86,7 @@ class TraceClient:
                 logger.warning("AliyunTrace API check failed: Unexpected status code: %s", response.status_code)
                 return False
         except httpx.RequestError as e:
-            logger.warning("AliyunTrace API check failed: %s", str(e))
+            logger.warning("AliyunTrace API check failed", exc_info=True)
             raise ValueError(f"AliyunTrace API check failed: {str(e)}")
 
     def get_project_url(self) -> str:

api/services/summary_index_service.py

@@ -243,8 +243,8 @@ class SummaryIndexService:
                 tokens_list = embedding_model.get_text_embedding_num_tokens([summary_content])
                 raw_embedding_tokens = tokens_list[0] if tokens_list else 0
                 embedding_tokens = raw_embedding_tokens if isinstance(raw_embedding_tokens, int) else 0
-        except Exception as e:
-            logger.warning("Failed to calculate embedding tokens for summary: %s", str(e))
+        except Exception:
+            logger.warning("Failed to calculate embedding tokens for summary", exc_info=True)
 
         # Create document with summary content and metadata
         summary_document = Document(
@@ -914,8 +914,8 @@ class SummaryIndexService:
                     try:
                         vector = Vector(dataset, session=session)
                         vector.delete_by_ids(summary_node_ids)
-                    except Exception as e:
-                        logger.warning("Failed to remove summary vectors: %s", str(e))
+                    except Exception:
+                        logger.warning("Failed to remove summary vectors", exc_info=True)
 
             # Disable summary records (don't delete)
             now = naive_utc_now()

api/tasks/add_document_to_index_task.py

@@ -128,8 +128,8 @@ def add_document_to_index_task(dataset_document_id: str):
                         dataset=dataset,
                         segment_ids=segment_ids_list,
                     )
-                except Exception as e:
-                    logger.warning("Failed to enable summaries for document %s: %s", dataset_document.id, str(e))
+                except Exception:
+                    logger.warning("Failed to enable summaries for document %s", dataset_document.id, exc_info=True)
 
             end_at = time.perf_counter()
             logger.info(

api/tasks/disable_segment_from_index_task.py

@@ -73,8 +73,8 @@ def disable_segment_from_index_task(segment_id: str):
                     segment_ids=[segment.id],
                     disabled_by=segment.disabled_by,
                 )
-            except Exception as e:
-                logger.warning("Failed to disable summary for segment %s: %s", segment.id, str(e))
+            except Exception:
+                logger.warning("Failed to disable summary for segment %s", segment.id, exc_info=True)
 
             end_at = time.perf_counter()
             logger.info(

api/tasks/disable_segments_from_index_task.py

@@ -81,8 +81,8 @@ def disable_segments_from_index_task(segment_ids: list, dataset_id: str, documen
                     segment_ids=segment_ids_list,
                     disabled_by=disabled_by,
                 )
-            except Exception as e:
-                logger.warning("Failed to disable summaries for segments: %s", str(e))
+            except Exception:
+                logger.warning("Failed to disable summaries for segments", exc_info=True)
 
             end_at = time.perf_counter()
             logger.info(click.style(f"Segments removed from index latency: {end_at - start_at}", fg="green"))

api/tasks/enable_segment_to_index_task.py

@@ -117,8 +117,8 @@ def enable_segment_to_index_task(segment_id: str):
                     dataset=dataset,
                     segment_ids=[segment.id],
                 )
-            except Exception as e:
-                logger.warning("Failed to enable summary for segment %s: %s", segment.id, str(e))
+            except Exception:
+                logger.warning("Failed to enable summary for segment %s", segment.id, exc_info=True)
 
             end_at = time.perf_counter()
             logger.info(click.style(f"Segment enabled to index: {segment.id} latency: {end_at - start_at}", fg="green"))

api/tasks/enable_segments_to_index_task.py

@@ -117,8 +117,8 @@ def enable_segments_to_index_task(segment_ids: list, dataset_id: str, document_i
                     dataset=dataset,
                     segment_ids=segment_ids_list,
                 )
-            except Exception as e:
-                logger.warning("Failed to enable summaries for segments: %s", str(e))
+            except Exception:
+                logger.warning("Failed to enable summaries for segments", exc_info=True)
 
             end_at = time.perf_counter()
             logger.info(click.style(f"Segments enabled to index latency: {end_at - start_at}", fg="green"))

api/tasks/remove_document_from_index_task.py

@@ -58,8 +58,8 @@ def remove_document_from_index_task(document_id: str):
                         segment_ids=segment_ids_list,
                         disabled_by=document.disabled_by,
                     )
-                except Exception as e:
-                    logger.warning("Failed to disable summaries for document %s: %s", document.id, str(e))
+                except Exception:
+                    logger.warning("Failed to disable summaries for document %s", document.id, exc_info=True)
 
             index_node_ids = [segment.index_node_id for segment in segments if segment.index_node_id]
             if index_node_ids:

api/tests/unit_tests/core/rag/indexing/processor/test_paragraph_index_processor.py

@@ -712,6 +712,32 @@ class TestParagraphIndexProcessor:
         assert files == []
         assert sum(1 for r in caplog.records if r.levelno == logging.WARNING) == 1
 
+    def test_extract_images_from_text_warning_carries_traceback(self, caplog: pytest.LogCaptureFixture) -> None:
+        """The build failure is logged at WARNING with the traceback attached, not just str(exception)."""
+        text = "![img](/files/11111111-1111-1111-1111-111111111111/image-preview)"
+        session = self.session
+        session.add(self._upload_file(file_id="11111111-1111-1111-1111-111111111111"))
+        session.flush()
+
+        with (
+            patch(
+                "core.rag.index_processor.processor.paragraph_index_processor.build_from_mapping",
+                side_effect=RuntimeError("build failed"),
+            ),
+            caplog.at_level(logging.WARNING, logger="core.rag.index_processor.processor.paragraph_index_processor"),
+        ):
+            ParagraphIndexProcessor._extract_images_from_text("tenant-1", text, session)
+
+        record = next(r for r in caplog.records if r.levelno == logging.WARNING)
+        assert record.exc_info is not None
+        _, exc_value, exc_traceback = record.exc_info
+        assert isinstance(exc_value, RuntimeError)
+        assert exc_traceback is not None
+        # The formatted output carries the call chain, and the message itself no longer inlines the exception.
+        assert "Traceback (most recent call last)" in caplog.text
+        assert "RuntimeError: build failed" in caplog.text
+        assert "build failed" not in record.getMessage()
+
     def test_extract_images_from_segment_attachments(self, caplog: pytest.LogCaptureFixture) -> None:
         session = self.session
         uploads = [