nvidia-nat 1.3.0a20250828__py3-none-any.whl → 1.3.0a20250830__py3-none-any.whl

nat/observability/mixin/file_mixin.py

@@ -103,7 +103,7 @@ class FileExportMixin(ResourceConflictMixin):
                     self._current_file_path.unlink()
                     logger.info("Cleaned up existing file: %s", self._current_file_path)
                 except OSError as e:
-                    logger.error("Error removing existing file %s: %s", self._current_file_path, e)
+                    logger.exception("Error removing existing file %s: %s", self._current_file_path, e)
 
     def _get_resource_identifiers(self) -> dict[str, Any]:
         """Return the file resources this instance will use.
@@ -154,10 +154,10 @@ class FileExportMixin(ResourceConflictMixin):
                     old_file.unlink()
                     logger.info("Cleaned up old log file during init: %s", old_file)
                 except OSError as e:
-                    logger.error("Error removing old file %s: %s", old_file, e)
+                    logger.exception("Error removing old file %s: %s", old_file, e)
 
         except Exception as e:
-            logger.error("Error during initialization cleanup: %s", e)
+            logger.exception("Error during initialization cleanup: %s", e)
 
     async def _should_roll_file(self) -> bool:
         """Check if the current file should be rolled based on size."""
@@ -191,7 +191,7 @@ class FileExportMixin(ResourceConflictMixin):
             await self._cleanup_old_files()
 
         except OSError as e:
-            logger.error("Error rolling file %s: %s", self._current_file_path, e)
+            logger.exception("Error rolling file %s: %s", self._current_file_path, e)
 
     async def _cleanup_old_files(self) -> None:
         """Remove old rolled files beyond the maximum count."""
@@ -209,10 +209,10 @@ class FileExportMixin(ResourceConflictMixin):
                     old_file.unlink()
                     logger.info("Cleaned up old log file: %s", old_file)
                 except OSError as e:
-                    logger.error("Error removing old file %s: %s", old_file, e)
+                    logger.exception("Error removing old file %s: %s", old_file, e)
 
         except Exception as e:
-            logger.error("Error during cleanup: %s", e)
+            logger.exception("Error during cleanup: %s", e)
 
     async def export_processed(self, item: str | list[str]) -> None:
         """Export a processed string or list of strings.
@@ -248,7 +248,7 @@ class FileExportMixin(ResourceConflictMixin):
                         await f.write("\n")
 
         except Exception as e:
-            logger.error("Error exporting event: %s", e, exc_info=True)
+            logger.exception("Error exporting event: %s", e)
 
     def get_current_file_path(self) -> Path:
         """Get the current file path being written to.

nat/observability/processor/batching_processor.py

@@ -193,14 +193,14 @@ class BatchingProcessor(CallbackProcessor[T, list[T]], Generic[T]):
                                 await self._done_callback(batch)
                                 logger.debug("Scheduled flush routed batch of %d items through pipeline", len(batch))
                             except Exception as e:
-                                logger.error("Error routing scheduled batch through pipeline: %s", e, exc_info=True)
+                                logger.exception("Error routing scheduled batch through pipeline: %s", e)
                         else:
                             logger.warning("Scheduled flush created batch of %d items but no pipeline callback set",
                                            len(batch))
         except asyncio.CancelledError:
             pass
         except Exception as e:
-            logger.error("Error in scheduled flush: %s", e, exc_info=True)
+            logger.exception("Error in scheduled flush: %s", e)
 
     async def _create_batch(self) -> list[T]:
         """Create a batch from the current queue."""
@@ -271,9 +271,7 @@ class BatchingProcessor(CallbackProcessor[T, list[T]], Generic[T]):
                                 "Successfully flushed final batch of %d items through pipeline during shutdown",
                                 len(final_batch))
                         except Exception as e:
-                            logger.error("Error routing final batch through pipeline during shutdown: %s",
-                                         e,
-                                         exc_info=True)
+                            logger.exception("Error routing final batch through pipeline during shutdown: %s", e)
                     else:
                         logger.warning("Final batch of %d items created during shutdown but no pipeline callback set",
                                        len(final_batch))
@@ -285,7 +283,7 @@ class BatchingProcessor(CallbackProcessor[T, list[T]], Generic[T]):
             logger.debug("BatchingProcessor shutdown completed successfully")
 
         except Exception as e:
-            logger.error("Error during BatchingProcessor shutdown: %s", e, exc_info=True)
+            logger.exception("Error during BatchingProcessor shutdown: %s", e)
             self._shutdown_complete = True
             self._shutdown_complete_event.set()
 

nat/observability/register.py

@@ -72,8 +72,10 @@ async def console_logging_method(config: ConsoleLoggingMethodConfig, builder: Bu
     """
     Build and return a StreamHandler for console-based logging.
     """
+    import sys
+
     level = getattr(logging, config.level.upper(), logging.INFO)
-    handler = logging.StreamHandler()
+    handler = logging.StreamHandler(stream=sys.stdout)
     handler.setLevel(level)
     yield handler
 

nat/profiler/calc/calc_runner.py

@@ -442,7 +442,7 @@ class CalcRunner:
                 runtime_fit=self.linear_analyzer.wf_runtime_fit  # May be None
             )
         except Exception as e:
-            logger.exception("Failed to plot concurrency vs. time metrics: %s", e, exc_info=True)
+            logger.exception("Failed to plot concurrency vs. time metrics: %s", e)
             logger.warning("Skipping plot of concurrency vs. time metrics")
 
     def write_output(self, output_dir: Path, calc_runner_output: CalcRunnerOutput):
@@ -506,11 +506,10 @@ class CalcRunner:
                 continue
             try:
                 calc_output = CalcRunnerOutput.model_validate_json(calc_runner_output_path.read_text())
-            except ValidationError as e:
+            except ValidationError:
                 logger.exception("Failed to validate calc runner output file %s. Skipping job %s.",
                                  calc_runner_output_path,
-                                 e,
-                                 exc_info=True)
+                                 job_dir.name)
                 continue
 
             # Extract sizing metrics from calc_data

nat/profiler/callbacks/agno_callback_handler.py

@@ -144,7 +144,7 @@ class AgnoProfilerHandler(BaseProfilerCallback):
                 return result
 
             except Exception as e:
-                logger.exception("Tool execution error: %s", e)
+                logger.error("Tool execution error: %s", e)
                 raise
 
         return wrapped_tool_execute

nat/profiler/callbacks/langchain_callback_handler.py

@@ -106,7 +106,7 @@ class LangchainProfilerHandler(AsyncCallbackHandler, BaseProfilerCallback):
         try:
             model_name = kwargs.get("metadata")["ls_model_name"]
         except Exception as e:
-            logger.exception("Error getting model name: %s", e, exc_info=True)
+            logger.exception("Error getting model name: %s", e)
 
         run_id = str(kwargs.get("run_id", str(uuid4())))
         self._run_id_to_model_name[run_id] = model_name
@@ -144,7 +144,7 @@ class LangchainProfilerHandler(AsyncCallbackHandler, BaseProfilerCallback):
         try:
             model_name = metadata["ls_model_name"] if metadata else kwargs.get("metadata")["ls_model_name"]
         except Exception as e:
-            logger.exception("Error getting model name: %s", e, exc_info=True)
+            logger.exception("Error getting model name: %s", e)
 
         run_id = str(run_id)
         self._run_id_to_model_name[run_id] = model_name
@@ -173,13 +173,13 @@ class LangchainProfilerHandler(AsyncCallbackHandler, BaseProfilerCallback):
         try:
             model_name = self._run_id_to_model_name.get(str(kwargs.get("run_id", "")), "")
         except Exception as e:
-            logger.exception("Error getting model name: %s", e, exc_info=True)
+            logger.exception("Error getting model name: %s", e)
 
         usage_metadata = {}
         try:
             usage_metadata = kwargs.get("chunk").message.usage_metadata if kwargs.get("chunk") else {}
         except Exception as e:
-            logger.exception("Error getting usage metadata: %s", e, exc_info=True)
+            logger.exception("Error getting usage metadata: %s", e)
 
         stats = IntermediateStepPayload(
             event_type=IntermediateStepType.LLM_NEW_TOKEN,
@@ -206,7 +206,7 @@ class LangchainProfilerHandler(AsyncCallbackHandler, BaseProfilerCallback):
             try:
                 model_name = self._run_id_to_model_name.get(str(kwargs.get("run_id", "")), "")
             except Exception as e_inner:
-                logger.exception("Error getting model name: %s from outer error %s", e_inner, e, exc_info=True)
+                logger.exception("Error getting model name: %s from outer error %s", e_inner, e)
 
         try:
             generation = response.generations[0][0]

nat/profiler/callbacks/llama_index_callback_handler.py

@@ -94,7 +94,7 @@ class LlamaIndexProfilerHandler(BaseCallbackHandler, BaseProfilerCallback):
             try:
                 model_name = payload.get(EventPayload.SERIALIZED)['model']
             except Exception as e:
-                logger.exception("Error getting model name: %s", e, exc_info=True)
+                logger.exception("Error getting model name: %s", e)
 
             llm_text_input = " ".join(prompts_or_messages) if prompts_or_messages else ""
 
@@ -159,13 +159,13 @@ class LlamaIndexProfilerHandler(BaseCallbackHandler, BaseProfilerCallback):
                     for block in response.message.blocks:
                         llm_text_output += block.text
                 except Exception as e:
-                    logger.exception("Error getting LLM text output: %s", e, exc_info=True)
+                    logger.exception("Error getting LLM text output: %s", e)
 
                 model_name = ""
                 try:
                     model_name = response.raw.model
                 except Exception as e:
-                    logger.exception("Error getting model name: %s", e, exc_info=True)
+                    logger.exception("Error getting model name: %s", e)
 
                 # Append usage data to NAT usage stats
                 with self._lock:

nat/profiler/callbacks/semantic_kernel_callback_handler.py

@@ -132,7 +132,7 @@ class SemanticKernelProfilerHandler(BaseProfilerCallback):
                         if "text" in item:
                             model_input += item["text"]
             except Exception as e:
-                logger.exception("Error in getting model input: %s", e, exc_info=True)
+                logger.exception("Error in getting model input: %s", e)
 
             input_stats = IntermediateStepPayload(event_type=IntermediateStepType.LLM_START,
                                                   framework=LLMFrameworkEnum.SEMANTIC_KERNEL,
@@ -232,7 +232,7 @@ class SemanticKernelProfilerHandler(BaseProfilerCallback):
                 return result
 
             except Exception as e:
-                logger.exception("ToolUsage._use error: %s", e)
+                logger.error("ToolUsage._use error: %s", e)
                 raise
 
         return patched_tool_call

nat/profiler/profile_runner.py

@@ -282,7 +282,7 @@ class ProfilerRunner:
                 fitted_model = model_trainer.train(all_steps)
                 logger.info("Fitted model for forecasting.")
             except Exception as e:
-                logger.exception("Fitting model failed. %s", e, exc_info=True)
+                logger.exception("Fitting model failed. %s", e)
                 return ProfilerResults()
 
             if self.write_output:

nat/profiler/utils.py

@@ -175,7 +175,7 @@ def create_standardized_dataframe(requests_data: list[list[IntermediateStep]]) -
                                  event_type=step.event_type).model_dump(), )
 
     except Exception as e:
-        logger.exception("Error creating standardized DataFrame: %s", e, exc_info=True)
+        logger.exception("Error creating standardized DataFrame: %s", e)
         return pd.DataFrame()
 
     if not all_rows:

nat/registry_handlers/local/local_handler.py

@@ -133,7 +133,7 @@ class LocalRegistryHandler(AbstractRegistryHandler):
                                                            "message": msg,
                                                            "action": ActionEnum.SEARCH
                                                        })
-            logger.exception(validated_search_response.status.message, exc_info=True)
+            logger.exception(validated_search_response.status.message)
 
             yield validated_search_response
 
@@ -168,7 +168,7 @@ class LocalRegistryHandler(AbstractRegistryHandler):
             validated_remove_response = RemoveResponse(status={
                 "status": StatusEnum.ERROR, "message": msg, "action": ActionEnum.REMOVE
             })  # type: ignore
-            logger.exception(validated_remove_response.status.message, exc_info=True)
+            logger.exception(validated_remove_response.status.message)
 
             yield validated_remove_response
 

nat/registry_handlers/package_utils.py

@@ -396,7 +396,7 @@ def get_transitive_dependencies(distribution_names: list[str]) -> dict[str, set[
                     except importlib.metadata.PackageNotFoundError:
                         pass
 
-                logger.error("Distribution %s not found (tried common variations)", dist_name)
+                logger.error("Distribution %s not found (tried common variations)", dist_name, exc_info=True)
                 result[dist_name] = set()
 
     return result

nat/registry_handlers/pypi/pypi_handler.py

@@ -85,7 +85,7 @@ class PypiRegistryHandler(AbstractRegistryHandler):
             validated_publish_response = PublishResponse(status={
                 "status": StatusEnum.ERROR, "message": msg, "action": ActionEnum.PUBLISH
             })
-            logger.exception(validated_publish_response.status.message, exc_info=True)
+            logger.exception(validated_publish_response.status.message)
 
             yield validated_publish_response
 
@@ -149,7 +149,7 @@ class PypiRegistryHandler(AbstractRegistryHandler):
             validated_pull_response = PullResponse(status={
                 "status": StatusEnum.ERROR, "message": msg, "action": ActionEnum.PULL
             })
-            logger.exception(validated_pull_response.status.message, exc_info=True)
+            logger.exception(validated_pull_response.status.message)
 
             yield validated_pull_response
 
@@ -212,7 +212,7 @@ class PypiRegistryHandler(AbstractRegistryHandler):
 
         except Exception as e:
             msg = f"Error searching for artifacts: {e}"
-            logger.exception(msg, exc_info=True)
+            logger.exception(msg)
             validated_search_response = SearchResponse(params=query,
                                                        status={
                                                            "status": StatusEnum.ERROR,

nat/registry_handlers/rest/rest_handler.py

@@ -88,7 +88,7 @@ class RestRegistryHandler(AbstractRegistryHandler):
             validated_publish_response = PublishResponse(status={
                 "status": StatusEnum.ERROR, "message": msg, "action": ActionEnum.PUBLISH
             })
-            logger.exception(validated_publish_response.status.message, exc_info=True)
+            logger.exception(validated_publish_response.status.message)
 
             yield validated_publish_response
 
@@ -155,7 +155,7 @@ class RestRegistryHandler(AbstractRegistryHandler):
             validated_pull_response = PullResponse(status={
                 "status": StatusEnum.ERROR, "message": msg, "action": ActionEnum.PULL
             })
-            logger.exception(validated_pull_response.status.message, exc_info=True)
+            logger.exception(validated_pull_response.status.message)
 
             yield validated_pull_response
 
@@ -193,7 +193,7 @@ class RestRegistryHandler(AbstractRegistryHandler):
                                                            "message": msg,
                                                            "action": ActionEnum.SEARCH
                                                        })
-            logger.exception(validated_search_response.status.message, exc_info=True)
+            logger.exception(validated_search_response.status.message)
 
             yield validated_search_response
 
@@ -228,7 +228,7 @@ class RestRegistryHandler(AbstractRegistryHandler):
             validated_remove_response = RemoveResponse(status={
                 "status": StatusEnum.ERROR, "message": msg, "action": ActionEnum.REMOVE
             })
-            logger.exception(validated_remove_response.status.message, exc_info=True)
+            logger.exception(validated_remove_response.status.message)
 
             yield validated_remove_response
 

nat/retriever/milvus/retriever.py

@@ -154,7 +154,7 @@ class MilvusRetriever(Retriever):
                 return _wrap_milvus_results(results, content_field=self.content_field)
 
         except Exception as e:
-            logger.exception("Exception when retrieving results from milvus for query %s: %s", query, e)
+            logger.error("Exception when retrieving results from milvus for query %s: %s", query, e)
             raise RetrieverError(f"Error when retrieving documents from {collection_name} for query '{query}'") from e
 
     async def _search(self,

nat/retriever/nemo_retriever/retriever.py

@@ -143,7 +143,7 @@ class NemoRetriever(Retriever):
                 return _wrap_nemo_results(output=output, content_field="content")
 
         except Exception as e:
-            logger.exception("Encountered an error when retrieving results from Nemo Retriever: %s", e)
+            logger.error("Encountered an error when retrieving results from Nemo Retriever: %s", e)
             raise CollectionUnavailableError(
                 f"Error when retrieving documents from {collection_name} for query '{query}'") from e
 

nat/runtime/loader.py

@@ -210,7 +210,7 @@ def discover_and_register_plugins(plugin_type: PluginTypes):
                 # Optionally, you can mark the plugin as unavailable or take other actions
 
             except Exception:
-                logger.exception("An error occurred while loading plugin '%s': {e}", entry_point.name, exc_info=True)
+                logger.exception("An error occurred while loading plugin '%s'", entry_point.name)
 
             finally:
                 count += 1

nat/runtime/runner.py

@@ -149,7 +149,7 @@ class Runner:
 
             return result
         except Exception as e:
-            logger.exception("Error running workflow: %s", e)
+            logger.error("Error running workflow: %s", e)
             event_stream = self._context_state.event_stream.get()
             if event_stream:
                 event_stream.on_complete()
@@ -181,7 +181,7 @@ class Runner:
                     event_stream.on_complete()
 
         except Exception as e:
-            logger.exception("Error running workflow: %s", e)
+            logger.error("Error running workflow: %s", e)
             event_stream = self._context_state.event_stream.get()
             if event_stream:
                 event_stream.on_complete()

nat/settings/global_settings.py

@@ -231,7 +231,7 @@ class Settings(HashableBaseModel):
             return True
 
         except Exception as e:
-            logger.exception("Unable to validate user settings configuration: %s", e, exc_info=True)
+            logger.exception("Unable to validate user settings configuration: %s", e)
             return False
 
     def print_channel_settings(self, channel_type: str | None = None) -> None:

nat/tool/code_execution/local_sandbox/local_sandbox_server.py

@@ -121,7 +121,7 @@ def execute_code_subprocess(generated_code: str, queue):
         resource.setrlimit(resource.RLIMIT_AS, (limit, limit))
         resource.setrlimit(resource.RLIMIT_DATA, (limit, limit))
     except Exception as e:
-        logger.error("Failed to set resource limits, PID: %s, error: %s", os.getpid(), e)
+        logger.exception("Failed to set resource limits, PID: %s, error: %s", os.getpid(), e)
 
     stdout_capture = StringIO()
     stderr_capture = StringIO()