feat: Unify logging and add request middleware

src/agent.py

@@ -258,10 +258,7 @@ async def async_response_stream(
         logger.info("Response generated", log_prefix=log_prefix, response=full_response)
 
     except Exception as e:
-        logger.error("Exception in streaming", error=str(e))
-        import traceback
-
-        traceback.print_exc()
+        logger.exception("[AGENT] Streaming failed")
         raise
 
 
@@ -316,10 +313,7 @@ async def async_response(
                     msg = error_msg
             raise ValueError(msg)
     except Exception as e:
-        logger.error("Exception in non-streaming response", error=str(e))
-        import traceback
-
-        traceback.print_exc()
+        logger.exception("[AGENT] Non-streaming response failed")
         raise
 
 
@@ -387,10 +381,7 @@ async def async_langflow_stream(
             yield chunk
         logger.debug("Langflow stream completed")
     except Exception as e:
-        logger.error("Exception in langflow stream", error=str(e))
-        import traceback
-
-        traceback.print_exc()
+        logger.exception("[AGENT] Langflow stream failed")
         raise
 
 

src/api/auth.py

@@ -4,6 +4,9 @@
 from fastapi.responses import JSONResponse
 from utils.telemetry import TelemetryClient, Category, MessageId
 from utils.version_utils import OPENRAG_VERSION
+from utils.logging_config import get_logger
+
+logger = get_logger(__name__)
 
 from dependencies import (
     get_auth_service,
@@ -50,9 +53,7 @@ async def auth_init(
         return JSONResponse(result)
 
     except Exception as e:
-        import traceback
-
-        traceback.print_exc()
+        logger.exception("[AUTH] OAuth init failed")
         return JSONResponse(
             {"error": f"Failed to initialize OAuth: {str(e)}"}, status_code=500
         )
@@ -90,9 +91,7 @@ async def auth_callback(
             return JSONResponse(result)
 
     except Exception as e:
-        import traceback
-
-        traceback.print_exc()
+        logger.exception("[AUTH] OAuth callback failed")
         await TelemetryClient.send_event(Category.AUTHENTICATION, MessageId.ORB_AUTH_OAUTH_FAILED)
         return JSONResponse({"error": f"Callback failed: {str(e)}"}, status_code=500)
 

src/api/chat.py

@@ -122,9 +122,7 @@ async def langflow_endpoint(
             return JSONResponse(result)
 
     except Exception as e:
-        import traceback
-        traceback.print_exc()
-        logger.error("Langflow request failed", error=str(e))
+        logger.exception("[CHAT] Langflow request failed")
         return JSONResponse(
             {"error": f"Langflow request failed: {str(e)}"}, status_code=500
         )
@@ -139,6 +137,7 @@ async def chat_history_endpoint(
         history = await chat_service.get_chat_history(user.user_id)
         return JSONResponse(history)
     except Exception as e:
+        logger.exception("[CHAT] Failed to get chat history")
         return JSONResponse(
             {"error": f"Failed to get chat history: {str(e)}"}, status_code=500
         )
@@ -153,6 +152,7 @@ async def langflow_history_endpoint(
         history = await chat_service.get_langflow_history(user.user_id)
         return JSONResponse(history)
     except Exception as e:
+        logger.exception("[CHAT] Failed to get langflow history")
         return JSONResponse(
             {"error": f"Failed to get langflow history: {str(e)}"}, status_code=500
         )

src/api/connectors.py

@@ -106,7 +106,7 @@ async def list_connectors(
         )
         return JSONResponse({"connectors": connector_types})
     except Exception as e:
-        logger.info("Error listing connectors", error=str(e))
+        logger.error("[CONNECTOR] Error listing connectors", error=str(e))
         return JSONResponse({"connectors": []})
 
 
@@ -549,15 +549,10 @@ async def connector_webhook(
             )
 
         except Exception as e:
-            logger.error(
-                "Failed to process webhook for connection",
+            logger.exception(
+                "[CONNECTOR] Failed to process webhook",
                 connection_id=connection.connection_id,
-                error=str(e),
             )
-            import traceback
-
-            traceback.print_exc()
-
             return JSONResponse(
                 {
                     "status": "error",

src/api/knowledge_filter.py

@@ -418,12 +418,9 @@ async def knowledge_filter_webhook(
         )
 
     except Exception as e:
-        logger.error(
-            "Failed to process knowledge filter webhook",
+        logger.exception(
+            "[API] Knowledge filter webhook failed",
             filter_id=filter_id,
             subscription_id=subscription_id,
-            error=str(e),
         )
-        import traceback
-        traceback.print_exc()
         return JSONResponse({"error": f"Webhook processing failed: {str(e)}"}, status_code=500)

src/api/provider_validation.py

@@ -316,7 +316,7 @@ async def _test_openai_completion_with_tools(api_key: str, llm_model: str) -> No
 
             # If max_tokens doesn't work, try with max_completion_tokens
             if response.status_code != 200:
-                logger.info("max_tokens parameter failed, trying max_completion_tokens instead")
+                logger.warning("[API] max_tokens parameter failed, trying max_completion_tokens instead")
                 payload = {**base_payload, "max_completion_tokens": 50}
                 response = await client.post(
                     "https://api.openai.com/v1/chat/completions",

src/api/settings.py

@@ -1609,7 +1609,7 @@ async def update_onboarding_state(
         if not success:
             raise HTTPException(status_code=500, detail="Failed to update onboarding state")
 
-        logger.info(f"Onboarding state updated: {body}")
+        logger.info("[CONFIG] Onboarding state updated", fields=list(body.model_fields_set))
 
         return OnboardingStateResponse(
             message="Onboarding state updated successfully",
@@ -1705,7 +1705,7 @@ async def rollback_onboarding(
                 {"error": "No onboarding configuration to rollback"}, status_code=400
             )
 
-        logger.info("Rolling back onboarding configuration due to file failures")
+        logger.warning("[CONFIG] Rolling back onboarding configuration due to file failures")
 
         # Get all tasks for the user
         all_tasks = task_service.get_all_tasks(user.user_id)

src/api/upload.py

@@ -16,6 +16,9 @@
     get_current_user,
 )
 from session_manager import User
+from utils.logging_config import get_logger
+
+logger = get_logger(__name__)
 
 
 class UploadPathBody(BaseModel):
@@ -55,8 +58,10 @@ async def upload(
             "AuthenticationException" in error_msg
             or "access denied" in error_msg.lower()
         ):
+            logger.warning("[INGEST] Upload rejected — access denied", error=error_msg)
             return JSONResponse({"error": error_msg}, status_code=403)
         else:
+            logger.exception("[INGEST] Upload failed")
             return JSONResponse({"error": error_msg}, status_code=500)
 
 

src/bootstrap.py

@@ -1,17 +1,23 @@
 from pathlib import Path
 from dotenv import load_dotenv
-from utils.logging_config import get_logger
-logger = get_logger(__name__)
 
 # This module is intended to be imported as the first line in entry points
-# to ensure environment variables are available to all subsequent imports.
+# to ensure environment variables and logging are available to all subsequent imports.
 
 def load_env():
-    """Load environment variables from the project root .env file."""
-    # .env is located in the project root (one level up from src/)
-    logger.info("Application startup span: loading environment variables")
+    """Load .env then immediately configure structured logging so all subsequent
+    module-level log calls use the correct formatter and processors."""
     env_path = Path(__file__).parent.parent / ".env"
     load_dotenv(dotenv_path=env_path, override=False)
 
+    # Configure logging immediately after env vars are available so every
+    # subsequent import gets a properly configured logger.
+    from utils.logging_config import configure_from_env
+    configure_from_env()
+
 # Execute immediately on import
 load_env()
+
+from utils.logging_config import get_logger  # noqa: E402 — after configure_from_env
+logger = get_logger(__name__)
+logger.info("Application startup span: environment loaded")

src/config/config_manager.py

@@ -243,7 +243,7 @@ def load_config(self) -> OpenRAGConfig:
             logger.info("Upgrading unencrypted secrets in config.yaml to AES-256-GCM")
             self.save_config_file(self._config, preserve_edited=True)
 
-        logger.debug("Configuration loaded", config=self._config.to_dict())
+        logger.debug("[CONFIG] Configuration loaded successfully")
         return self._config
 
     def _load_env_overrides(

src/config/settings.py

@@ -213,7 +213,7 @@ async def get_langflow_api_key(force_regenerate: bool = False):
 
     # If forcing regeneration, clear the cached key
     if force_regenerate and LANGFLOW_KEY:
-        logger.info("Forcing Langflow API key regeneration due to auth failure")
+        logger.warning("[LF] Forcing Langflow API key regeneration due to auth failure")
         LANGFLOW_KEY = None
 
     # Use default langflow/langflow credentials if auto-login is enabled and credentials not set

src/connectors/ibm_cos/connector.py

@@ -368,8 +368,3 @@ async def cleanup_subscription(self, subscription_id: str) -> bool:
 
 
 
-if __name__ == "__main__":
-    connector = IBMCOSConnector({})
-    print(connector.authenticate())
-    print(connector.list_files())
-    # print(connector.get_file_content("test_cos.py"))

src/connectors/onedrive/connector.py

@@ -168,9 +168,7 @@ async def authenticate(self) -> bool:
             self._authenticated = authenticated
             return authenticated
         except Exception as e:
-            logger.error(f"OneDrive authentication failed: {e}")
-            import traceback
-            traceback.print_exc()
+            logger.exception("[CONNECTOR] OneDrive authentication failed")
             self._authenticated = False
             return False
 
@@ -238,13 +236,11 @@ async def _detect_onedrive_url(self) -> Optional[str]:
                     else:
                         logger.warning("_detect_onedrive_url: webUrl is empty in response")
                 else:
-                    logger.warning(f"_detect_onedrive_url: Failed to get drive info: {response.status_code}, response: {response.text[:500]}")
+                    logger.warning("[CONNECTOR] OneDrive detect URL failed", status_code=response.status_code)
 
         except Exception as e:
-            logger.error(f"_detect_onedrive_url: Exception during detection: {e}")
-            import traceback
-            traceback.print_exc()
-
+            logger.exception("[CONNECTOR] OneDrive URL detection failed")
+
         return None
 
     def sync_once(self) -> None:

src/connectors/onedrive/oauth.py

@@ -135,9 +135,7 @@ async def load_credentials(self) -> bool:
             return False
 
         except Exception as e:
-            logger.error(f"Failed to load OneDrive credentials: {e}")
-            import traceback
-            traceback.print_exc()
+            logger.exception("[CONNECTOR] OneDrive credential load failed")
             return False
 
     async def _refresh_from_json_token(self, token_data: dict) -> bool:
@@ -185,9 +183,7 @@ async def _refresh_from_json_token(self, token_data: dict) -> bool:
             return False
 
         except Exception as e:
-            logger.error(f"Exception during refresh from JSON token: {e}")
-            import traceback
-            traceback.print_exc()
+            logger.exception("[CONNECTOR] OneDrive JSON token refresh failed")
             return False
 
     async def save_cache(self):
@@ -299,9 +295,7 @@ async def is_authenticated(self) -> bool:
             return False
 
         except Exception as e:
-            logger.error(f"OneDrive is_authenticated: Exception: {e}")
-            import traceback
-            traceback.print_exc()
+            logger.exception("[CONNECTOR] OneDrive is_authenticated failed")
             return False
 
     def get_access_token(self) -> str:
@@ -331,9 +325,7 @@ def get_access_token(self) -> str:
             raise ValueError(f"Failed to acquire access token: {error_msg}")
 
         except Exception as e:
-            logger.error(f"OneDrive get_access_token: Exception: {e}")
-            import traceback
-            traceback.print_exc()
+            logger.exception("[CONNECTOR] OneDrive get_access_token failed")
             raise
 
     async def revoke_credentials(self):

src/connectors/service.py

@@ -61,7 +61,7 @@ async def process_connector_document(
             with open(tmp_path, "wb") as f:
                 f.write(document.content)
 
-            logger.debug("Processing connector document", document_id=document.id)
+            logger.info("[CONNECTOR] Processing document", document_id=document.id, connector_type=connector_type, filename=document.filename)
 
             # Process using consolidated processing pipeline
             from models.processors import TaskProcessor
@@ -80,7 +80,7 @@ async def process_connector_document(
                 acl=document.acl,
             )
 
-            logger.debug("Document processing result", result=result)
+            logger.info("[CONNECTOR] Document processed", document_id=document.id, status=result.get("status"))
 
             # If successfully indexed or already exists, update the indexed documents with connector metadata
             if result["status"] in ["indexed", "unchanged"]:

src/connectors/sharepoint/connector.py

@@ -175,9 +175,7 @@ async def authenticate(self) -> bool:
             self._authenticated = authenticated
             return authenticated
         except Exception as e:
-            logger.error(f"SharePoint authentication failed: {e}")
-            import traceback
-            traceback.print_exc()
+            logger.exception("[CONNECTOR] SharePoint authentication failed")
             self._authenticated = False
             return False
 
@@ -252,13 +250,11 @@ async def _detect_sharepoint_url(self) -> Optional[str]:
                     else:
                         logger.warning("_detect_sharepoint_url: webUrl is empty in response")
                 else:
-                    logger.warning(f"_detect_sharepoint_url: Failed to get drive info: {response.status_code}, response: {response.text[:500]}")
+                    logger.warning("[CONNECTOR] SharePoint detect URL failed", status_code=response.status_code)
 
         except Exception as e:
-            logger.error(f"_detect_sharepoint_url: Exception during detection: {e}")
-            import traceback
-            traceback.print_exc()
-
+            logger.exception("[CONNECTOR] SharePoint URL detection failed")
+
         return None
 
     def sync_once(self) -> None: