import logging import logging.handlers import os import queue import sys import threading import time import colorlog from ...utils import constants log_colors_config = { 'DEBUG': 'green', # cyan white 'INFO': 'white', 'WARNING': 'yellow', 'ERROR': 'red', 'CRITICAL': 'cyan', } # Log rotation configuration to prevent unbounded log file growth LOG_FILE_MAX_BYTES = 10 * 1024 * 1024 # 10MB per file LOG_FILE_BACKUP_COUNT = 5 # Keep 5 backup files (total ~50MB max) LOG_DIR = 'data/logs' CONSOLE_QUEUE_CAPACITY = 256 CONSOLE_RECORD_MAX_BYTES = 16 * 1024 CONSOLE_CLOSE_TIMEOUT = 0.2 class NonBlockingConsoleHandler(logging.Handler): """Isolate a stalled console from callers with at most 4 MiB of queued bytes. Only formatted UTF-8 bytes cross the thread boundary, so queued records do not retain arguments or exception frames. Overflow drops console copies; other handlers still receive the original records. """ def __init__(self, stream, *, capacity: int = CONSOLE_QUEUE_CAPACITY): if capacity <= 0: raise ValueError('Console queue capacity must be positive') # Drain startup prints before the worker can fill the pipe. Otherwise # interpreter finalization could block flushing the original stream. stream.flush() super().__init__() self._queue: queue.Queue[bytes | None] = queue.Queue(maxsize=capacity) # logging.shutdown holds Handler.lock while joining the worker. self._state_lock = threading.Lock() self._failed = threading.Event() self.dropped_records = 0 self.write_error: str | None = None # The worker owns this descriptor. Do not share a buffered Python stream # lock with the main thread or interpreter finalization. self._fd = os.dup(stream.fileno()) self._worker = threading.Thread(target=self._consume, name='langbot-console-log', daemon=True) try: self._worker.start() except Exception: os.close(self._fd) super().close() raise def emit(self, record: logging.LogRecord) -> None: with self._state_lock: if self._closed or self._failed.is_set() or self._queue.full(): self.dropped_records += 1 return try: text = self.format(record) payload = (text[:CONSOLE_RECORD_MAX_BYTES] + '\n').encode('utf-8', errors='backslashreplace') if len(text) > CONSOLE_RECORD_MAX_BYTES or len(payload) > CONSOLE_RECORD_MAX_BYTES: marker = b'\n[console log truncated]\n' payload = ( payload[: CONSOLE_RECORD_MAX_BYTES - len(marker)] .decode('utf-8', errors='ignore') .encode('utf-8') + marker ) self._queue.put_nowait(payload) except queue.Full: self.dropped_records += 1 except Exception: # Logging's default handleError writes to stderr, which may # share the same stalled container logging driver. self.dropped_records += 1 def _write(self, payload: bytes) -> None: remaining = memoryview(payload) while remaining: written = os.write(self._fd, remaining) if written == 0: raise OSError('Console write made no progress') remaining = remaining[written:] def _consume(self) -> None: reported_drops = 0 try: while (payload := self._queue.get()) is not None: self._write(payload) with self._state_lock: dropped = self.dropped_records if dropped > reported_drops: self._write( f'[WARNING] Console logging dropped {dropped - reported_drops} records.\n'.encode('utf-8') ) reported_drops = dropped except OSError as exc: with self._state_lock: self.write_error = str(exc) self._failed.set() self.dropped_records += 1 finally: os.close(self._fd) with self._state_lock: while True: try: if self._queue.get_nowait() is not None: self.dropped_records += 1 except queue.Empty: break def close(self) -> None: with self._state_lock: if self._closed: return super().close() # A full queue must not make shutdown wait for the writer. while True: try: self._queue.put_nowait(None) break except queue.Full: try: self._queue.get_nowait() self.dropped_records += 1 except queue.Empty: # The writer consumed the last record after Full; # retry inserting the shutdown sentinel. continue self._worker.join(timeout=CONSOLE_CLOSE_TIMEOUT) class DailyGroupedRotatingFileHandler(logging.handlers.RotatingFileHandler): """File handler that writes to ``data/logs/langbot-YYYY-MM-DD.log``. It combines two rotation triggers: * **Size** — within a single day the file is rotated once it exceeds ``maxBytes``, producing numbered backups (``langbot-DATE.log.1`` etc.), exactly like :class:`~logging.handlers.RotatingFileHandler`. * **Date** — when the local date changes, logging switches to a fresh ``langbot-.log`` file. This happens even within a single long-running process, so a bot started on day N keeps writing to that day's file and rolls over to day N+1's file at midnight, instead of appending every subsequent day's logs to the start-day file. The on-disk naming stays compatible with the log-retention cleanup in ``api/http/service/maintenance.py`` (``LOG_FILE_PATTERN``). """ def __init__(self, log_dir: str, max_bytes: int, backup_count: int, encoding: str = 'utf-8'): self.log_dir = log_dir self._current_date = self._today() super().__init__( self._build_path(self._current_date), maxBytes=max_bytes, backupCount=backup_count, encoding=encoding, ) @staticmethod def _today() -> str: return time.strftime('%Y-%m-%d', time.localtime()) def _build_path(self, date_str: str) -> str: return os.path.join(self.log_dir, 'langbot-%s.log' % date_str) def shouldRollover(self, record): # Roll over when the day changes, regardless of file size. if self._today() != self._current_date: return True return super().shouldRollover(record) def doRollover(self): today = self._today() if today != self._current_date: # Date changed: point the handler at the new day's file. # This is a date switch, not a size-based numbered rotation. if self.stream: self.stream.close() self.stream = None self._current_date = today self.baseFilename = os.path.abspath(self._build_path(today)) if not self.delay: self.stream = self._open() else: # Same day, file exceeded maxBytes: numbered rotation. super().doRollover() async def init_logging(extra_handlers: list[logging.Handler] = None) -> logging.Logger: # Remove all existing loggers for handler in logging.root.handlers[:]: logging.root.removeHandler(handler) level = logging.INFO if constants.debug_mode: level = logging.DEBUG qcg_logger = logging.getLogger('langbot') for handler in qcg_logger.handlers[:]: qcg_logger.removeHandler(handler) handler.close() qcg_logger.setLevel(level) color_formatter = colorlog.ColoredFormatter( fmt='%(log_color)s[%(asctime)s.%(msecs)03d] %(filename)s (%(lineno)d) - [%(levelname)s] : %(message)s', datefmt='%m-%d %H:%M:%S', log_colors=log_colors_config, ) stream_handler = NonBlockingConsoleHandler(sys.stdout) # Rotate by size within a day and switch files when the date changes, # so long-running processes still produce a log file for the current day. rotating_file_handler = DailyGroupedRotatingFileHandler( LOG_DIR, max_bytes=LOG_FILE_MAX_BYTES, backup_count=LOG_FILE_BACKUP_COUNT, encoding='utf-8', ) log_handlers: list[logging.Handler] = [ stream_handler, rotating_file_handler, ] log_handlers += extra_handlers if extra_handlers is not None else [] for handler in log_handlers: handler.setLevel(level) handler.setFormatter(color_formatter) qcg_logger.addHandler(handler) qcg_logger.debug('Logging initialized, log level: %s' % level) logging.basicConfig( level=logging.CRITICAL, # Set log output format format='[DEPR][%(asctime)s.%(msecs)03d] %(pathname)s (%(lineno)d) - [%(levelname)s] :\n%(message)s', # Log output format # -8 is a placeholder, left-align the output, and output length is 8 datefmt='%Y-%m-%d %H:%M:%S', # Time output format handlers=[logging.NullHandler()], ) return qcg_logger