Fix the problem that async logging has hanging thread (#10374)

Signed-off-by: chenmoneygithub <[email protected]>

mlflow/utils/async_logging/async_logging_queue.py

@@ -47,9 +47,8 @@ def _at_exit_callback(self) -> None:
         try:
             # Stop the data processing thread
             self._stop_data_logging_thread_event.set()
-            # Waits till queue is drained.
-            self._run_data_logging_thread.result()
-            self._batch_logging_threadpool.shutdown(wait=False)
+            # Waits till logging queue is drained.
+            self._batch_logging_thread.join()
             self._batch_status_check_threadpool.shutdown(wait=False)
         except Exception as e:
             _logger.error(f"Encountered error while trying to finish logging: {e}")
@@ -132,14 +131,10 @@ def __getstate__(self):
             del state["_run_data_logging_thread"]
         if "_stop_data_logging_thread_event" in state:
             del state["_stop_data_logging_thread_event"]
-        if "_batch_logging_threadpool" in state:
-            del state["_batch_logging_threadpool"]
+        if "_batch_logging_thread" in state:
+            del state["_batch_logging_thread"]
         if "_batch_status_check_threadpool" in state:
             del state["_batch_status_check_threadpool"]
-        if "_run_data_logging_thread" in state:
-            del state["_run_data_logging_thread"]
-        if "_stop_data_logging_thread_event" in state:
-            del state["_stop_data_logging_thread_event"]
 
         return state
 
@@ -158,7 +153,7 @@ def __setstate__(self, state):
         self._queue = Queue()
         self._lock = threading.RLock()
         self._is_activated = False
-        self._batch_logging_threadpool = None
+        self._batch_logging_thread = None
         self._batch_status_check_threadpool = None
         self._stop_data_logging_thread_event = None
 
@@ -193,7 +188,6 @@ def log_batch_async(
         )
 
         self._queue.put(batch)
-
         operation_future = self._batch_status_check_threadpool.submit(self._wait_for_batch, batch)
         return RunOperations(operation_futures=[operation_future])
 
@@ -217,14 +211,17 @@ def activate(self) -> None:
             self._stop_data_logging_thread_event = threading.Event()
 
             # Keeping max_workers=1 so that there are no two threads
-            self._batch_logging_threadpool = ThreadPoolExecutor(max_workers=1)
-
-            self._batch_status_check_threadpool = ThreadPoolExecutor(max_workers=10)
-
-            self._run_data_logging_thread = self._batch_logging_threadpool.submit(
-                self._logging_loop
-            )  # concurrent.futures.Future[self._logging_loop]
+            self._batch_logging_thread = threading.Thread(
+                target=self._logging_loop,
+                name="MLflowAsyncLoggingLoop",
+                daemon=True,
+            )
 
+            self._batch_status_check_threadpool = ThreadPoolExecutor(
+                max_workers=10,
+                thread_name_prefix="MLflowAsyncLoggingStatusCheck",
+            )
+            self._batch_logging_thread.start()
             atexit.register(self._at_exit_callback)
 
             self._is_activated = True