RAGFlow is an agent-native database: a Retrieval-Augmented Generation (RAG) engine that grew into context infrastructure.
Go agent-harnessagent-native-databaseagentic-aiagentic-retrievalagentic-searchaiai-agentscontext-enginecontext-engineeringcontext-managementharness-engineeringknowledge-compilationragretrieval-augmented-generationretrieval-harnesssearch-harness
Fix: set chunk_num before set_progress, avoid race condition (#17724)
Lynn51af9e2eae89
diff --git a/rag/svr/task_executor.py b/rag/svr/task_executor.pyindex 4f3ac2361c..b4d53808fe 100644--- a/rag/svr/task_executor.py+++ b/rag/svr/task_executor.py@@ -934,8 +934,12 @@ async def run_dataflow(task: dict):time_cost = timer() - start_tstask_time_cost = timer() - task_start_ts+ try:+ ret = DocumentService.increment_chunk_num(doc_id, task_dataset_id, embedding_token_consumption, len(chunks), task_time_cost)+ except Exception:+ logging.exception("increment_chunk_num failed for doc %s", doc_id)+ ret = Noneset_progress(task_id, prog=1.0, msg="Indexing done ({:.2f}s). Task done ({:.2f}s)".format(time_cost, task_time_cost))- ret = DocumentService.increment_chunk_num(doc_id, task_dataset_id, embedding_token_consumption, len(chunks), task_time_cost)get_recording_context().save_func_return_value("DocumentService.increment_chunk_num", ret)logging.info("[Done], chunks({}), token({}), elapsed:{:.2f}".format(len(chunks), embedding_token_consumption, task_time_cost))get_recording_context().record("dataflow_chunks", chunks)diff --git a/rag/svr/task_executor_refactor/dataflow_service.py b/rag/svr/task_executor_refactor/dataflow_service.pyindex 447a8f5590..ef4839019b 100644--- a/rag/svr/task_executor_refactor/dataflow_service.py+++ b/rag/svr/task_executor_refactor/dataflow_service.py@@ -166,13 +166,21 @@ class DataflowService:time_cost = timer() - start_tstask_time_cost = timer() - task_start_ts- self._progress(prog=1.0, msg="Indexing done ({:.2f}s). Task done ({:.2f}s)".format(time_cost, task_time_cost))- # Update document stats+ # Update document stats (chunk counters) BEFORE marking the task+ # as done, so the async _sync_progress loop never observes DONE+ # with stale chunk_num=0. If the stats update fails, the task+ # is still marked DONE — chunk counters are not critical to the+ # parse result.if ctx.write_interceptor:ctx.write_interceptor.intercept("DocumentService.increment_chunk_num")else:- DocumentService.increment_chunk_num(doc_id, task_dataset_id, embedding_token_consumption, len(chunks), task_time_cost)+ try:+ DocumentService.increment_chunk_num(doc_id, task_dataset_id, embedding_token_consumption, len(chunks), task_time_cost)+ except Exception:+ logging.exception("increment_chunk_num failed for doc %s", doc_id)++ self._progress(prog=1.0, msg="Indexing done ({:.2f}s). Task done ({:.2f}s)".format(time_cost, task_time_cost))logging.info("[Done], chunks({}), token({}), elapsed:{:.2f}".format(len(chunks), embedding_token_consumption, task_time_cost))ctx.recording_context.record("dataflow_chunks", chunks)