Developing with asyncio

Asynchronous programming is different from classic “sequential” programming.

This page lists common mistakes and traps and explains how to avoid them.

Debug Mode

By default asyncio runs in production mode. In order to ease the development asyncio has a debug mode.

There are several ways to enable asyncio debug mode:

In addition to enabling the debug mode, consider also:

  • setting the log level of the asyncio logger to logging.DEBUG, for example the following snippet of code can be run at startup of the application:

    logging.basicConfig(level=logging.DEBUG)

  • configuring the warnings module to display ResourceWarning warnings. One way of doing that is by using the -W default command line option.

When the debug mode is enabled:

  • Many non-threadsafe asyncio APIs (such as loop.call_soon() and loop.call_at() methods) raise an exception if they are called from a wrong thread.

  • The execution time of the I/O selector is logged if it takes too long to perform an I/O operation.

  • Callbacks taking longer than 100 milliseconds are logged. The loop.slow_callback_duration attribute can be used to set the minimum execution duration in seconds that is considered “slow”.

Concurrency and Multithreading

An event loop runs in a thread (typically the main thread) and executes all callbacks and Tasks in its thread. While a Task is running in the event loop, no other Tasks can run in the same thread. When a Task executes an await expression, the running Task gets suspended, and the event loop executes the next Task.

To schedule a callback from another OS thread, the loop.call_soon_threadsafe() method should be used. Example:

loop.call_soon_threadsafe(callback, *args)

Almost all asyncio objects are not thread safe, which is typically not a problem unless there is code that works with them from outside of a Task or a callback. If there’s a need for such code to call a low-level asyncio API, the loop.call_soon_threadsafe() method should be used, e.g.:

loop.call_soon_threadsafe(fut.cancel)

To schedule a coroutine object from a different OS thread, the run_coroutine_threadsafe() function should be used. It returns a concurrent.futures.Future to access the result:

async def coro_func():
     return await asyncio.sleep(1, 42)

# Later in another OS thread:

future = asyncio.run_coroutine_threadsafe(coro_func(), loop)
# Wait for the result:
result = future.result()

To handle signals the event loop must be run in the main thread.

The loop.run_in_executor() method can be used with a concurrent.futures.ThreadPoolExecutor or InterpreterPoolExecutor to execute blocking code in a different OS thread without blocking the OS thread that the event loop runs in.

There is currently no way to schedule coroutines or callbacks directly from a different process (such as one started with multiprocessing). The Event Loop Methods section lists APIs that can read from pipes and watch file descriptors without blocking the event loop. In addition, asyncio’s Subprocess APIs provide a way to start a process and communicate with it from the event loop. Lastly, the aforementioned loop.run_in_executor() method can also be used with a concurrent.futures.ProcessPoolExecutor to execute code in a different process.

Running Blocking Code

Blocking (CPU-bound) code should not be called directly. For example, if a function performs a CPU-intensive calculation for 1 second, all concurrent asyncio Tasks and IO operations would be delayed by 1 second.

An executor can be used to run a task in a different thread, including in a different interpreter, or even in a different process to avoid blocking the OS thread with the event loop. See the loop.run_in_executor() method for more details.

Logging

asyncio uses the logging module and all logging is performed via the "asyncio" logger.

The default log level is logging.INFO, which can be easily adjusted:

logging.getLogger("asyncio").setLevel(logging.WARNING)

Network logging can block the event loop. It is recommended to use a separate thread for handling logs or use non-blocking IO. For example, see Dealing with handlers that block.

Detect never-awaited coroutines

When a coroutine function is called, but not awaited (e.g. coro() instead of await coro()) or the coroutine is not scheduled with asyncio.create_task(), asyncio will emit a RuntimeWarning:

import asyncio

async def test():
    print("never scheduled")

async def main():
    test()

asyncio.run(main())

Output:

test.py:7: RuntimeWarning: coroutine 'test' was never awaited
  test()

Output in debug mode:

test.py:7: RuntimeWarning: coroutine 'test' was never awaited
Coroutine created at (most recent call last)
  File "../t.py", line 9, in <module>
    asyncio.run(main(), debug=True)

  < .. >

  File "../t.py", line 7, in main
    test()
  test()

The usual fix is to either await the coroutine or call the asyncio.create_task() function:

async def main():
    await test()

Detect never-retrieved exceptions

If a Future.set_exception() is called but the Future object is never awaited on, the exception would never be propagated to the user code. In this case, asyncio would emit a log message when the Future object is garbage collected.

Example of an unhandled exception:

import asyncio

async def bug():
    raise Exception("not consumed")

async def main():
    asyncio.create_task(bug())

asyncio.run(main())

Output:

Task exception was never retrieved
future: <Task finished coro=<bug() done, defined at test.py:3>
  exception=Exception('not consumed')>

Traceback (most recent call last):
  File "test.py", line 4, in bug
    raise Exception("not consumed")
Exception: not consumed

Enable the debug mode to get the traceback where the task was created:

asyncio.run(main(), debug=True)

Output in debug mode:

Task exception was never retrieved
future: <Task finished coro=<bug() done, defined at test.py:3>
    exception=Exception('not consumed') created at asyncio/tasks.py:321>

source_traceback: Object created at (most recent call last):
  File "../t.py", line 9, in <module>
    asyncio.run(main(), debug=True)

< .. >

Traceback (most recent call last):
  File "../t.py", line 4, in bug
    raise Exception("not consumed")
Exception: not consumed

Asynchronous generators best practices

By asynchronous generator in this section we will mean an asynchronous generator iterator that returned by asynchronous generator function.

Manually close the generator

If an asynchronous generator happens to exit early by break, the caller task being cancelled, or other exceptions, the generator’s async cleanup code will run and possibly raise exceptions or access context variables in an unexpected context–perhaps after the lifetime of tasks it depends, or during the event loop shutdown when the async-generator garbage collection hook is called.

To prevent this, it is recommended to explicitly close the async generator by calling aclose() method, or using contextlib.aclosing() context manager:

import asyncio
import contextlib

async def gen():
  yield 1
  yield 2

async def func():
  async with contextlib.aclosing(gen()) as g:
    async for x in g:
      break

asyncio.run(func())

Only create a generator when a loop is already running

As said above, if an asynchronous generator is not resumed before it is finalized, then any finalization procedures will be delayed. The event loop handles this situation and doing it best to call async generator-iterator’s aclose() at the proper moment, thus allowing any pending finally clauses to execute.

Then it is recomended to create async generators only after the event loop has already been created.

Avoid iterating and closing the same generator concurrently

The async generators allow to be reentered while another aclose()/aclose()/aclose() call is in progress. This may lead to an inconsistent state of the async generator and cause errors.

Let’s consider following example:

import asyncio

async def consumer():
    for idx in range(100):
        await asyncio.sleep(0)
        message = yield idx
        print('received', message)

async def amain():
    agenerator = consumer()
    await agenerator.asend(None)

    fa = asyncio.create_task(agenerator.asend('A'))
    fb = asyncio.create_task(agenerator.asend('B'))
    await fa
    await fb

asyncio.run(amain())

Output:

received A
Traceback (most recent call last):
  File "test.py", line 38, in <module>
    asyncio.run(amain())
    ~~~~~~~~~~~^^^^^^^^^
  File "Lib\asyncio\runners.py", line 204, in run
    return runner.run(main)
           ~~~~~~~~~~^^^^^^
  File "Lib\asyncio\runners.py", line 127, in run
    return self._loop.run_until_complete(task)
           ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^^^^^^
  File "Lib\asyncio\base_events.py", line 719, in run_until_complete
    return future.result()
           ~~~~~~~~~~~~~^^
  File "test.py", line 36, in amain
    await fb
RuntimeError: anext(): asynchronous generator is already running

Therefore, it is recommended to avoid using the async generators in parallel tasks or in the multiple event loops.