Skip to content
This repository has been archived by the owner on Jun 3, 2024. It is now read-only.

RuntimeError: Timed out waiting for send operation to complete #4

Open
obaidullahzaland opened this issue Jul 28, 2023 · 0 comments
Open
Labels
bug Something isn't working

Comments

@obaidullahzaland
Copy link

Describe the bug
I am running the simple command python -m openfed.tools.simulator --nproc 6 examples/run.py as given in the repository just to check if the code was running and I encountered the following error.

(openfed) ozaland@prec3660c:~/OpenFed$ python -m openfed.tools.simulator --nproc 6 examples/run.py
  0%|                                                    | 0/10 [00:00<?, ?it/s]/home/ozaland/anaconda3/envs/openfed/lib/python3.7/site-packages/torch/distributed/distributed_c10d.py:430: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
  "torch.distributed.distributed_c10d._get_global_rank is deprecated "
/home/ozaland/anaconda3/envs/openfed/lib/python3.7/site-packages/torch/distributed/distributed_c10d.py:430: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
  "torch.distributed.distributed_c10d._get_global_rank is deprecated "
/home/ozaland/anaconda3/envs/openfed/lib/python3.7/site-packages/torch/distributed/distributed_c10d.py:430: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
  "torch.distributed.distributed_c10d._get_global_rank is deprecated "
/home/ozaland/anaconda3/envs/openfed/lib/python3.7/site-packages/torch/distributed/distributed_c10d.py:430: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
  "torch.distributed.distributed_c10d._get_global_rank is deprecated "
/home/ozaland/anaconda3/envs/openfed/lib/python3.7/site-packages/torch/distributed/distributed_c10d.py:430: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
  "torch.distributed.distributed_c10d._get_global_rank is deprecated "
/home/ozaland/anaconda3/envs/openfed/lib/python3.7/site-packages/torch/distributed/distributed_c10d.py:430: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
  "torch.distributed.distributed_c10d._get_global_rank is deprecated "
 10%|████                                    | 1/10 [30:02<4:30:18, 1802.07s/it]
Traceback (most recent call last):
  File "examples/run.py", line 99, in <module>
    simulate()
  File "examples/run.py", line 52, in simulate
    api.run()
  File "/home/ozaland/OpenFed/openfed/api.py", line 71, in run
    maintainer.step()
  File "/home/ozaland/OpenFed/openfed/core/maintainer.py", line 306, in step
    return self._aggregator_step(*args, **kwargs)
  File "/home/ozaland/OpenFed/openfed/core/maintainer.py", line 378, in _aggregator_step
    flag = self.upload()
  File "/home/ozaland/OpenFed/openfed/core/maintainer.py", line 298, in upload
    self.transfer(to=True)
  File "/home/ozaland/OpenFed/openfed/core/functional.py", line 33, in _fed_context
    return safe_call(self, *args, **kwargs)
  File "/home/ozaland/OpenFed/openfed/core/functional.py", line 24, in safe_call
    return func(*args, **kwargs)
  File "/home/ozaland/OpenFed/openfed/core/maintainer.py", line 253, in transfer
    self.pipe.upload(self.packaged_data)
  File "/home/ozaland/OpenFed/openfed/federated/pipe.py", line 164, in upload
    self.transfer(True, data)
  File "/home/ozaland/OpenFed/openfed/federated/pipe.py", line 233, in transfer
    self.push(data)
  File "/home/ozaland/OpenFed/openfed/federated/pipe.py", line 249, in push
    distributed_c10d.gather_object(data, None, dst=rank, group=self.pg)
  File "/home/ozaland/anaconda3/envs/openfed/lib/python3.7/site-packages/torch/distributed/distributed_c10d.py", line 1981, in gather_object
    all_gather(object_size_list, local_size, group=group)
  File "/home/ozaland/anaconda3/envs/openfed/lib/python3.7/site-packages/torch/distributed/distributed_c10d.py", line 2282, in all_gather
    work.wait()
RuntimeError: [../third_party/gloo/gloo/transport/tcp/unbound_buffer.cc:133] Timed out waiting 1800000ms for send operation to complete
/home/ozaland/OpenFed/openfed/core/functional.py:26: UserWarning: Failed to call <function Maintainer.transfer at 0x7fc6515eff80>
  warnings.warn(f'Failed to call {func}')
/home/ozaland/OpenFed/openfed/core/functional.py:26: UserWarning: Failed to call <function Maintainer.transfer at 0x7f2c126eff80>
  warnings.warn(f'Failed to call {func}')
/home/ozaland/OpenFed/openfed/core/functional.py:26: UserWarning: Failed to call <function Maintainer.transfer at 0x7f60a89eff80>
  warnings.warn(f'Failed to call {func}')
/home/ozaland/OpenFed/openfed/core/functional.py:26: UserWarning: Failed to call <function Maintainer.transfer at 0x7ff26edaff80>
  warnings.warn(f'Failed to call {func}')
/home/ozaland/OpenFed/openfed/core/functional.py:26: UserWarning: Failed to call <function Maintainer.transfer at 0x7f290cfaff80>
  warnings.warn(f'Failed to call {func}')
Killing subprocess 1740079
Killing subprocess 1740080
Killing subprocess 1740081
Killing subprocess 1740082
Killing subprocess 1740083
Killing subprocess 1740084
Traceback (most recent call last):
  File "/home/ozaland/anaconda3/envs/openfed/lib/python3.7/runpy.py", line 193, in _run_module_as_main
    "__main__", mod_spec)
  File "/home/ozaland/anaconda3/envs/openfed/lib/python3.7/runpy.py", line 85, in _run_code
    exec(code, run_globals)
  File "/home/ozaland/OpenFed/openfed/tools/simulator.py", line 221, in <module>
    main()
  File "/home/ozaland/OpenFed/openfed/tools/simulator.py", line 204, in main
    sigkill_handler(signal.SIGTERM, None)
  File "/home/ozaland/OpenFed/openfed/tools/simulator.py", line 138, in sigkill_handler
    returncode=last_return_code, cmd=cmd)
subprocess.CalledProcessError: Command '['/home/ozaland/anaconda3/envs/openfed/bin/python', '-u', 'examples/run.py', '--props=/tmp/collaborator-5.json']' returned non-zero exit status 1.

Environment (please complete the following information):

  • OS Platform and Distribution (e.g., Linux Ubuntu 22.04):
  • Python package versions: Pytorch v1.13.1, openfed v0.0.0, torchvision v0.14.1
  • Python version: 3.7
  • CUDA/cuDNN version: 11.7
@obaidullahzaland obaidullahzaland added the bug Something isn't working label Jul 28, 2023
Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Labels
bug Something isn't working
Projects
None yet
Development

No branches or pull requests

1 participant