error when graph is executing.

[zhufeijuanjuan]

This is an occasional situation where an error is reported when calling the large model to generate the final answer

Exception in callback FuturesDict.on_done(PregelExecuta... subgraphs=[]))(<Task finishe...的用户。'}], ...}>)
handle: <Handle FuturesDict.on_done(PregelExecuta... subgraphs=[]))(<Task finishe...的用户。'}], ...}>)>
Traceback (most recent call last):
File "/usr/local/python3/lib/python3.12/asyncio/events.py", line 88, in _run
self._context.run(self._callback, *self._args)
File "/apps/.local/lib/python3.12/site-packages/langgraph/pregel/runner.py", line 103, in on_done
self.callback()(task, _exception(fut)) # type: ignore[misc]
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
TypeError: 'NoneType' object is not callable

[BastinFlorian]

I got the same issue trying load testing my langgraph api

[nathan-brewiq]

Hi everyone,

I’m running into exactly the same TypeError: 'NoneType' object is not callable in FuturesDict.on_done when using Pregel to drive my graph execution. Here are my environment details:

• Python: 3.12.x
• langgraph: 0.3.22
• langchain: 0.3.14
• OS: Windows (as shown by the traceback path)

[kongjiellx]

+1

[zbn1120]

hey,guys. Did you resolve this problem? I also got the error.

[mmatvienko]

what context are you running this in? i am running behind FastAPI.
I think, if you can, just upgrade you langgraph to the latest version. #4996

[zbn1120]

Yes! I upgrade langgraph to v0.5.2, and the problem is resolved!

[OnlyBrinTy]

+1