Event loop stalls even with an idle bot

[nicolaxe]

Summary

The event loop stalls and results in "heartbeat blocked" warnings even with an entirely idle bot

Reproduction Steps

1. Make a bot that does literally anything (including nothing at all)
2. Run it
3. See warnings like:

Shard ID None heartbeat blocked for more than 10 seconds.
Loop thread traceback (most recent call last):
  File "/tmp/test.py", line 13, in <module>
    asyncio.run(main())
  File "/usr/lib/python3.13/asyncio/runners.py", line 195, in run
    return runner.run(main)
  File "/usr/lib/python3.13/asyncio/runners.py", line 118, in run
    return self._loop.run_until_complete(task)
  File "/usr/lib/python3.13/asyncio/base_events.py", line 712, in run_until_complete
    self.run_forever()
  File "/usr/lib/python3.13/asyncio/base_events.py", line 683, in run_forever
    self._run_once()
  File "/usr/lib/python3.13/asyncio/base_events.py", line 2012, in _run_once
    event_list = self._selector.select(timeout)
  File "/usr/lib/python3.13/selectors.py", line 452, in select
    fd_event_list = self._selector.poll(timeout, max_ev)

Minimal Reproducible Code

import discord, asyncio, logging

logging.basicConfig(
        format="%(asctime)s | %(levelname)s | %(message)s",
        datefmt="%Y-%m-%d %H:%M:%S",
        level=logging.INFO)

bot = discord.Client()

async def main():
    await bot.start("TOKEN")

asyncio.run(main())

Expected Results

The bot runs correctly and without stall warnings

Actual Results

After about ~30 seconds the bot starts logging stall warnings every ~10 seconds.

Intents

None (it doesn't matter)

System Information

• Python v3.13.7-final
• py-cord v2.6.1-final
• aiohttp v3.13.0
• system info: Linux 6.17.2-arch1-1 #\1 SMP PREEMPT_DYNAMIC Sun, 12 Oct 2025 12:45:18 +0000

Checklist

• I have searched the open issues for duplicates.
• I have shown the entire traceback, if possible.
• I have removed my token from display, if visible.

Additional Context

The issue seems to be specific to running the bot using discord.Client.start() in a preexisting event loop. When running using discord.Client.run() it works as expected.

[Paillat-dev]

Can repro

Test Code

import discord, asyncio, logging
from os import getenv
from dotenv import load_dotenv

load_dotenv()

logging.basicConfig(
        format="%(asctime)s | %(levelname)s | %(message)s",
        datefmt="%Y-%m-%d %H:%M:%S",
        level=logging.INFO)

bot = discord.Client()

async def main():
    await bot.start(getenv("TOKEN_2"))

asyncio.run(main())

2025-10-18 14:20:19 | WARNING | PyNaCl is not installed, voice will NOT be supported
2025-10-18 14:20:19 | INFO | logging in using static token
2025-10-18 14:20:20 | INFO | Shard ID None has sent the IDENTIFY payload.
2025-10-18 14:20:21 | INFO | Shard ID None has connected to Gateway: ["gateway-prd-arm-us-east1-b-rgpb",{"micros":112440,"calls":["id_created",{"micros":652,"calls":[]},"session_lookup_time",{"micros":4112,"calls":[]},"session_lookup_finished",{"micros":8,"calls":[]},"discord-sessions-prd-2-37",{"micros":107246,"calls":["start_session",{"micros":96760,"calls":["discord-api-rpc-7447f9b866-prncs",{"micros":50727,"calls":["get_user",{"micros":5730},"get_guilds",{"micros":3739},"send_scheduled_deletion_message",{"micros":13},"guild_join_requests",{"micros":2150},"authorized_ip_coro",{"micros":11},"pending_payments",{"micros":37167},"apex_experiments",{"micros":303},"user_activities",{"micros":5},"played_application_ids",{"micros":3},"linked_users",{"micros":10}]}]},"starting_guild_connect",{"micros":39,"calls":[]},"presence_started",{"micros":8994,"calls":[]},"guilds_started",{"micros":84,"calls":[]},"lobbies_started",{"micros":1,"calls":[]},"guilds_connect",{"micros":2,"calls":[]},"presence_connect",{"micros":1338,"calls":[]},"connect_finished",{"micros":1344,"calls":[]},"build_ready",{"micros":11,"calls":[]},"clean_ready",{"micros":1,"calls":[]},"optimize_ready",{"micros":0,"calls":[]},"split_ready",{"micros":0,"calls":[]}]}]}] (Session ID: c75c1aed94aebfcd0e431a57d6fac079).
2025-10-18 14:21:10 | WARNING | Can't keep up, shard ID None websocket is 49.8s behind.
2025-10-18 14:21:12 | WARNING | Shard ID None heartbeat blocked for more than 10 seconds.
Loop thread traceback (most recent call last):
  File "C:\Users\jerem\Documents\pycord\pycord\test\test.py", line 17, in <module>
    asyncio.run(main())
  File "C:\Users\jerem\AppData\Roaming\uv\python\cpython-3.13.7-windows-x86_64-none\Lib\asyncio\runners.py", line 195, in run
    return runner.run(main)
  File "C:\Users\jerem\AppData\Roaming\uv\python\cpython-3.13.7-windows-x86_64-none\Lib\asyncio\runners.py", line 118, in run
    return self._loop.run_until_complete(task)
  File "C:\Users\jerem\AppData\Roaming\uv\python\cpython-3.13.7-windows-x86_64-none\Lib\asyncio\base_events.py", line 712, in run_until_complete
    self.run_forever()
  File "C:\Users\jerem\AppData\Roaming\uv\python\cpython-3.13.7-windows-x86_64-none\Lib\asyncio\base_events.py", line 683, in run_forever
    self._run_once()
  File "C:\Users\jerem\AppData\Roaming\uv\python\cpython-3.13.7-windows-x86_64-none\Lib\asyncio\base_events.py", line 2012, in _run_once
    event_list = self._selector.select(timeout)
  File "C:\Users\jerem\AppData\Roaming\uv\python\cpython-3.13.7-windows-x86_64-none\Lib\asyncio\windows_events.py", line 446, in select
    self._poll(timeout)
  File "C:\Users\jerem\AppData\Roaming\uv\python\cpython-3.13.7-windows-x86_64-none\Lib\asyncio\windows_events.py", line 775, in _poll
    status = _overlapped.GetQueuedCompletionStatus(self._iocp, ms)

2025-10-18 14:21:22 | WARNING | Shard ID None heartbeat blocked for more than 20 seconds.

System:

• Python v3.13.7-final
• py-cord v2.6.1-final
  OR
• py-cord v2.7.None-candidate
  • py-cord importlib.metadata: v2.7.0rc2.dev49+gacbe07bdb
• aiohttp v3.13.1
• system info: Windows 11 10.0.26200

Note: this doesn't repro in the python debugger for some reason

[Paillat-dev]

Hello ! Thank you for your issue. This is currently fixed by #2645.
In the meantime I found that a workaround to your issue is adding bot.loop = get_running_loop() like follows to make sure that the bot uses the correct loop:

from asyncio import get_running_loop

# ...
async def main():
    bot.loop = get_running_loop()
    await bot.start(getenv("TOKEN_2"))

asyncio.run(main())

Please let me know if this workaround works for you

[nicolaxe]

Please let me know if this workaround works for you

Seems to work fine with this. Thanks

[nicolaxe]

After some more testing I have ran into an issue with this workaround. The typing context manager can break when processing takes a while and the bot stops typing. After terminating the program I sometimes get:

task: <Task cancelling name='Task-8' coro=<Typing.do_typing() running at /usr/lib/python3.13/site-packages/discord/context_managers.py:54> cb=[_typing_done_callback() at /usr/lib/python3.13/site-packages/discord/context_managers.py:41]>
<sys>:0: RuntimeWarning: coroutine 'Typing.do_typing' was never awaited

(this can happen even long after the context should have been closed)

This issue also appears to only happen when running Client.start()

[Paillat-dev]

I am going to take a look.

[Paillat-dev]

@mik111111 Sorry for the late reply. Is this issue still happening with the workaround ? If yes, could you share the minimal code snippet to reproduce it ? Thanks !

[nicolaxe]

Yes it still happens with the workaround. Some minimal code that causes it:

import discord, asyncio, logging

logging.basicConfig(
        format="%(asctime)s | %(levelname)s | %(message)s",
        datefmt="%Y-%m-%d %H:%M:%S",
        level=logging.INFO)

bot = discord.Client()

@bot.event
async def on_message(message):
    async with message.channel.typing():
        for i in range(1, 30):
            print(i)
            await asyncio.sleep(1)

async def main():
    bot.loop = asyncio.get_running_loop()
    await bot.start('TOKEN')


asyncio.run(main())

The bot stops typing after around 10 seconds. I can't reproduce the specific error message I shared previously anymore, but the issue still happens. Once again this seems to only happen when using bot.start() instead of bot.run()