PollerCompletionQueue._handle_events BlockingIOError spam in LangGraph Cloud logs

We’re seeing hundreds of these errors per day in LangGraph Cloud, firing on graph invocations -

[ERROR] Exception in callback functools.partial(<bound method PollerCompletionQueue._handle_events of <grpc._cython.cygrpc.PollerCompletionQueue object at 0x7ff3947ac660>>, <uvloop.Loop
running=True closed=False debug=False>)
Traceback (most recent call last):
File “uvloop/cbhandles.pyx”, line 61, in uvloop.loop.Handle._run
File “src/python/grpcio/grpc/_cython/_cygrpc/aio/completion_queue.pyx.pxi”, line 154, in grpc._cython.cygrpc.PollerCompletionQueue._handle_events
BlockingIOError: [Errno 11] Resource temporarily unavailable

The gRPC dependency comes from langgraph-api itself (grpcio >= 1.75.0, grpcio-health-checking, grpcio-tools), not our application code. This appears to be a grpcio + uvloop incompatibility where
the PollerCompletionQueue pipe fd triggers EAGAIN under uvloop.

Graphs execute correctly — the errors seem cosmetic — but they pollute logs at ERROR level and trigger alert fatigue in monitoring tools.

Environment: langgraph 1.0.7, langgraph-sdk 0.3.3, grpcio 1.76.0, Python 3.12, deployed to LangGraph Cloud.

Is this a known platform issue? Any planned fix or recommended workaround?

Any guidance appreciated — happy to provide more logs or details.

hello @sjay,
Will be happy to help you out but can you please provide more logs and also tell me the configuration you are using for Langgraph Cloud?

The logs were removed in 0.7.85

Thanks both!

Just redeployed and it picked up 0.7.86 – thanks Will!

-salman

1 Like

Do let us know if you see any other issues!