fix: make end_all_traces be called at the correct moment (langflow-ai…

…#2516)

* fix(tracing/service.py): remove wait_for_all_tracers call

* refactor: set reasonable timeout for TelemetryService client

* fix: handle HTTP and request errors in TelemetryService

Handle HTTPStatusError and RequestError exceptions separately in the send_telemetry_data method of TelemetryService to provide more specific error messages. Also, catch any unexpected exceptions and log them with an appropriate error message.

* fix: cancel worker task and close client in TelemetryService stop method

Cancel the worker task and close the client in the stop method of TelemetryService to ensure proper cleanup and prevent potential resource leaks. Also handle any exceptions that may occur during the cleanup process and log appropriate error messages.

* style(telemetry/service.py): fix indentation issue in await statement to comply with PEP8 guidelines

* feat(graph): add method to remove vertex from runnables in Graph class

* fix(chat.py): fix issue where vertex was not being removed from runnables list to prevent duplication of results

* fix(chat.py): defines when the end_all_traces call should happen

(cherry picked from commit 15aa68a)

src/backend/base/langflow/api/v1/chat.py

@@ -111,6 +111,7 @@ async def retrieve_vertices_order(
         # which duplicates the results
         for vertex_id in first_layer:
             graph.remove_from_predecessors(vertex_id)
+            graph.remove_vertex_from_runnables(vertex_id)
 
         # Now vertices is a list of lists
         # We need to get the id of each vertex
@@ -160,7 +161,7 @@ async def build_vertex(
     Args:
         flow_id (str): The ID of the flow.
         vertex_id (str): The ID of the vertex to build.
-        background_tasks (BackgroundTasks): The background tasks object for logging.
+        background_tasks (BackgroundTasks): The background tasks dependency.
         inputs (Optional[InputValueRequest], optional): The input values for the vertex. Defaults to None.
         chat_service (ChatService, optional): The chat service dependency. Defaults to Depends(get_chat_service).
         current_user (Any, optional): The current user dependency. Defaults to Depends(get_current_active_user).
@@ -211,8 +212,6 @@ async def build_vertex(
             )
             top_level_vertices = graph.run_manager.get_top_level_vertices(graph, next_runnable_vertices)
 
-            result_data_response = ResultDataResponse(**result_dict.model_dump())
-
             result_data_response = ResultDataResponse.model_validate(result_dict, from_attributes=True)
         except Exception as exc:
             if isinstance(exc, ComponentBuildException):
@@ -265,7 +264,7 @@ async def build_vertex(
         if graph.stop_vertex and graph.stop_vertex in next_runnable_vertices:
             next_runnable_vertices = [graph.stop_vertex]
 
-        if not next_runnable_vertices:
+        if not graph.run_manager.vertices_to_run and not next_runnable_vertices:
             background_tasks.add_task(graph.end_all_traces)
 
         build_response = VertexBuildResponse(

src/backend/base/langflow/graph/graph/base.py

@@ -1469,6 +1469,9 @@ def find_runnable_predecessors_for_successors(self, vertex_id: str) -> List[str]
     def remove_from_predecessors(self, vertex_id: str):
         self.run_manager.remove_from_predecessors(vertex_id)
 
+    def remove_vertex_from_runnables(self, vertex_id: str):
+        self.run_manager.remove_vertex_from_runnables(vertex_id)
+
     def build_in_degree(self, edges: List[ContractEdge]) -> Dict[str, int]:
         in_degree: Dict[str, int] = defaultdict(int)
         for edge in edges:

src/backend/base/langflow/services/telemetry/service.py

@@ -1,4 +1,5 @@
 import asyncio
+import contextlib
 import os
 import platform
 from datetime import datetime, timezone
@@ -30,7 +31,7 @@ def __init__(self, settings_service: "SettingsService"):
         self.settings_service = settings_service
         self.base_url = settings_service.settings.telemetry_base_url
         self.telemetry_queue: asyncio.Queue = asyncio.Queue()
-        self.client = httpx.AsyncClient(timeout=None)
+        self.client = httpx.AsyncClient(timeout=10.0)  # Set a reasonable timeout
         self.running = False
         self.package = get_version_info()["package"]
 
@@ -63,8 +64,12 @@ async def send_telemetry_data(self, payload: BaseModel, path: str | None = None)
                 logger.error(f"Failed to send telemetry data: {response.status_code} {response.text}")
             else:
                 logger.debug("Telemetry data sent successfully.")
+        except httpx.HTTPStatusError as e:
+            logger.error(f"HTTP error occurred: {e}")
+        except httpx.RequestError as e:
+            logger.error(f"Request error occurred: {e}")
         except Exception as e:
-            logger.error(f"Failed to send telemetry data due to: {e}")
+            logger.error(f"Unexpected error occurred: {e}")
 
     async def log_package_run(self, payload: RunPayload):
         await self.telemetry_queue.put((self.send_telemetry_data, payload, "run"))
@@ -119,8 +124,10 @@ async def stop(self):
         try:
             self.running = False
             await self.flush()
-            self.worker_task.cancel()
             if self.worker_task:
-                await self.worker_task
+                self.worker_task.cancel()
+                with contextlib.suppress(asyncio.CancelledError):
+                    await self.worker_task
+            await self.client.aclose()
         except Exception as e:
             logger.error(f"Error stopping tracing service: {e}")

src/backend/base/langflow/services/tracing/service.py

@@ -7,7 +7,6 @@
 from typing import TYPE_CHECKING, Any, Dict, Optional
 from uuid import UUID
 
-from langchain.callbacks.tracers.langchain import wait_for_all_tracers
 from loguru import logger
 
 from langflow.schema.data import Data
@@ -292,5 +291,4 @@ def end(
             self._run_tree.add_metadata(metadata)
         self._run_tree.end(outputs=outputs, error=error)
         self._run_tree.post()
-        wait_for_all_tracers()
         self._run_link = self._run_tree.get_url()