aioredlock very fast error. Is this a bug?

[centosredhat]

import  asyncio
import  aioredis
from  aioredlock import Aioredlock, LockError, Sentinel

redis_instances = [('127.0.0.1', 6379)]

async def test():
    await asyncio.sleep(0.1)
    return 'test'

async def get_lock():
    try:
        lock_manager = Aioredlock(redis_instances)
        async with await lock_manager.lock("resource_name") as lock:
            print(lock.valid)
            result = await test()
            print(result)
        assert lock.valid is False

    except Exception as e:
        print('Lock not acquired')


async def all_tasks():
    tasks = [asyncio.create_task(get_lock()) for i in range(10000)]
    await asyncio.wait(tasks)
asyncio.run(all_tasks())

error log →
task: <Task pending name='Task-84846' coro=<Aioredlock._set_lock.<locals>.cleanup() running at /usr/local/lib/python3.8/site-packages/aioredlock/algorithm.py:99> wait_for=<_GatheringFuturepending cb=[<TaskWakeupMethWrapper object at 0x7f8c476b9c70>()]>>
Task was destroyed but it is pending!
task: <Task pending name='Task-84847' coro=<Aioredlock._set_lock.<locals>.cleanup() running at /usr/local/lib/python3.8/site-packages/aioredlock/algorithm.py:99> wait_for=<_GatheringFuturepending cb=[<TaskWakeupMethWrapper object at 0x7f8c487d0dc0>()]>>
Connection <RedisConnection [db:0]> has pending commands, closing it.
Connection <RedisConnection [db:0]> has pending commands, closing it.
Future exception was never retrieved
future: <Future finished exception=ConnectionForcedCloseError()>
aioredis.errors.ConnectionForcedCloseError
Future exception was never retrieved
future: <Future finished exception=ConnectionForcedCloseError()>
aioredis.errors.ConnectionForcedCloseError
Future exception was never retrieved
future: <Future finished exception=ConnectionForcedCloseError()>
aioredis.errors.ConnectionForcedCloseError
Future exception was never retrieved
[get_lock.log](https://github.com/joanvila/aioredlock/files/6438662/get_lock.log)

[gtmanfred]

This is not a bug with aioredlock. Looks like for redis is closing the connection, and aioredis is raising an exception. Possibly because you are trying to open 10k connections at once.

[centosredhat]

2K connections at once will have the same problem. You can try.

This is not a bug with aioredlock. Looks like for redis is closing the connection, and aioredis is raising an exception. Possibly because you are trying to open 10k connections at once.

[gtmanfred]

Can you do 2000 commands with aioredis like this? The error is not coming from aioredlock. It is coming from your redis instance closing the connections.

[centosredhat]

aioredis with 2000 set command not error .(set 'resource_name' uuid ex 10 nx)

[centosredhat]

aioredlock with 2000 set command.
error →
task: <Task pending name='Task-29924' coro=<Aioredlock._set_lock..cleanup() running at /usr/local/lib/python3.8/site-packages/aioredlock/algorithm.py:99> wait_for=<_GatheringFuture pending cb=[<TaskWakeupMethWrapper object at 0x7faba88ed910>()]>>
Task was destroyed but it is pending!
task: <Task pending name='Task-29925' coro=<Aioredlock._set_lock..cleanup() running at /usr/local/lib/python3.8/site-packages/aioredlock/algorithm.py:99> wait_for=<_GatheringFuture pending cb=[<TaskWakeupMethWrapper object at 0x7fabcd3f4f10>()]>>
Task was destroyed but it is pending!
task: <Task pending name='Task-29926' coro=<Aioredlock._set_lock..cleanup() running at /usr/local/lib/python3.8/site-packages/aioredlock/algorithm.py:99> wait_for=<_GatheringFuture pending cb=[<TaskWakeupMethWrapper object at 0x7faba88e98e0>()]>>
Task was destroyed but it is pending!
task: <Task pending name='Task-29927' coro=<Aioredlock._set_lock..cleanup() running at /usr/local/lib/python3.8/site-packages/aioredlock/algorithm.py:99> wait_for=<_GatheringFuture pending cb=[<TaskWakeupMethWrapper object at 0x7faba86f7fd0>()]>>
Task was destroyed but it is pending!
task: <Task pending name='Task-29928' coro=<Aioredlock._set_lock..cleanup() running at /usr/local/lib/python3.8/site-packages/aioredlock/algorithm.py:99> wait_for=<_GatheringFuture pending cb=[<TaskWakeupMethWrapper object at 0x7fabcd101700>()]>>
Task was destroyed but it is pending!
task: <Task pending name='Task-29929' coro=<Aioredlock._set_lock..cleanup() running at /usr/local/lib/python3.8/site-packages/aioredlock/algorithm.py:99> wait_for=<_GatheringFuture pending cb=[<TaskWakeupMethWrapper object at 0x7faba8c00e50>()]>>
Task was destroyed but it is pending!
task: <Task pending name='Task-29930' coro=<Aioredlock._set_lock..cleanup() running at /usr/local/lib/python3.8/site-packages/aioredlock/algorithm.py:99> wait_for=<_GatheringFuture pending cb=[<TaskWakeupMethWrapper object at 0x7faba967ac70>()]>>
Task was destroyed but it is pending!
task: <Task pending name='Task-29931' coro=<Aioredlock._set_lock..cleanup() running at /usr/local/lib/python3.8/site-packages/aioredlock/algorithm.py:99> wait_for=<_GatheringFuture pending cb=[<TaskWakeupMethWrapper object at 0x7faba90278e0>()]>>
Task was destroyed but it is pending!
task: <Task pending name='Task-29932' coro=<Aioredlock._set_lock..cleanup() running at /usr/local/lib/python3.8/site-packages/aioredlock/algorithm.py:99> wait_for=<_GatheringFuture pending cb=[<TaskWakeupMethWrapper object at 0x7fabcd11a070>()]>>

[centosredhat]

/usr/local/lib/python3.8/site-packages/aioredlock/algorithm.py
.....
async def destroy(self):
self.log.debug('Destroying %s', repr(self))
for resource, lock in self._locks.items():
if lock.valid:
try:
await self.unlock(lock)
except Exception:
self.log.exception('Can not unlock "%s"', resource)

    self._locks.clear()
    self._watchdogs.clear()

    await self.redis.clear_connections()

self._locks.clear() self._watchdogs.clear()
Could this be the cause.

[gtmanfred]

those just clear the dictionaries of all keys https://www.geeksforgeeks.org/python-dictionary-clear/