nicolargo/glances · error · RuntimeError

AsyncIO event loop creation failed: {self._loop_exception}

Error message

AsyncIO event loop creation failed: {self._loop_exception}

What it means

When the background thread that creates the asyncio event loop fails (e.g. asyncio.new_event_loop() raises), the exception is stored in _loop_exception and re-raised as RuntimeError from __init__. The message embeds the original exception text, so it's a wrapper around the real cause.

Source

Thrown at glances/exports/export_asyncio.py:54

        """Init the AsyncIO export interface."""
        super().__init__(config=config, args=args)

        # AsyncIO event loop management
        self.loop = None
        self._loop_ready = threading.Event()
        self._loop_exception = None
        self._shutdown = False

        # Start the background event loop thread
        self._loop_thread = threading.Thread(target=self._run_event_loop, daemon=True)
        self._loop_thread.start()

        # Wait for the loop to be ready
        if not self._loop_ready.wait(timeout=10):
            raise RuntimeError("AsyncIO event loop failed to start within timeout")

        if self._loop_exception:
            raise RuntimeError(f"AsyncIO event loop creation failed: {self._loop_exception}")

        if self.loop is None:
            raise RuntimeError("AsyncIO event loop is None after initialization")

        # Call child class AsyncIO initialization
        future = asyncio.run_coroutine_threadsafe(self._async_init(), self.loop)
        try:
            future.result(timeout=10)
            logger.debug(f"{self.export_name} AsyncIO export initialized successfully")
        except Exception as e:
            logger.warning(f"{self.export_name} AsyncIO initialization failed: {e}. Will retry in background.")

    def _run_event_loop(self):
        """Run event loop in background thread."""
        try:
            self.loop = asyncio.new_event_loop()
            asyncio.set_event_loop(self.loop)
            self._loop_ready.set()

View on GitHub (pinned to a240d8dfb3)

Solutions

1. Read the embedded exception text — it names the actual failure (e.g. 'too many file descriptors').
2. Fix the root cause: raise fd limits (ulimit -n), remove conflicting monkeypatches, or upgrade Python.
3. Disable the failing async export plugin in glances.conf if it's not needed.

try:
    export = MyAsyncExport(...)
except RuntimeError as e:
    if 'event loop creation failed' in str(e):
        root_cause = str(e).split(':', 1)[-1]  # embedded original exception
        logger.error('async export loop failed: %s', root_cause)

When it happens

Trigger: Anything making event loop creation fail inside _run_event_loop: ResourceWarning/ResourceError from too many open FDs, a broken selector implementation, or a monkeypatched/broken asyncio module. The stored exception is raised from the constructor of any async export right after _loop_ready.wait() succeeds (thread sets the event, then records the exception) or after timeout.

Common situations: Environments with fd exhaustion (ulimit -n), CI runners with restricted asyncio selectors, or conflicting asyncio polyfills (gevent/eventlet patches).

Related errors

• AsyncIO event loop failed to start within timeout
• AsyncIO event loop is None after initialization

AI-assisted analysis of nicolargo/glances@a240d8dfb3 (2026-08-27). Data as JSON: /api/errors/f9ddb6bf2930f490. Report an issue: GitHub.