topic Re: mpirun error in Intel® MPI Library

mpirun error

GoodLuck — Tue, 01 Nov 2022 17:11:33 GMT

Code:

import mpi4py import time import numpy as np from mpi4py import MPI comm = MPI.COMM_WORLD rank = comm.Get_rank() print("rank",rank) if __name__ == '__main__': if rank == 0: mem = np.array([0], dtype='i') win = MPI.Win.Create(mem, comm=comm) else: win = MPI.Win.Create(None, comm=comm) print(rank, "end")

(py3.6.8) ➜ ~ mpirun -n 2 python -u test.py
rank 0
rank 1
0 end
1 end
Abort(806449679): Fatal error in internal_Finalize: Other MPI error, error stack:
internal_Finalize(50)...........: MPI_Finalize failed
MPII_Finalize(345)..............:
MPID_Finalize(511)..............:
MPIDI_OFI_mpi_finalize_hook(895):
destroy_vni_context(1137).......: OFI domain close failed (ofi_init.c:1137:destroy_vni_context:Device or resource busy)

Why is this happening? How to debug? This error is not reported on the other machine.

Re: mpirun error

SantoshY_Intel — Fri, 04 Nov 2022 04:08:21 GMT

Hi,

Thanks for posting in the Intel forums.

New window objects can be created by calling the "MPI.Win.Create()" method within a communicator and by specifying a memory buffer.

When a window instance is no longer needed, the "MPI.Win.Free()" method should be called.

So we should use MPI.Win.Free() method to free the new window objects that have been created. Please refer to the below code:

import mpi4py import time import numpy as np from mpi4py import MPI #comm = MPI.COMM_WORLD #rank = comm.Get_rank() #print("rank",rank) if __name__ == '__main__': comm = MPI.COMM_WORLD rank = comm.Get_rank() print("rank",rank) if rank == 0: mem = np.array([0], dtype='i') win = MPI.Win.Create(mem, comm=comm) MPI.Win.Free(win) else: win = MPI.Win.Create(None, comm=comm) MPI.Win.Free(win) print(rank, "end")

Command to run the code:

mpirun -n 2 python -u test.py

Observed output:

Thanks & Regards,

Santosh

Re: mpirun error

GoodLuck — Fri, 04 Nov 2022 04:14:41 GMT

It works. thank you very much.

Re: mpirun error

SantoshY_Intel — Fri, 04 Nov 2022 04:32:10 GMT

Hi,

Glad to know that your issue is resolved. If you need any additional information, please post a new question as this thread will no longer be monitored by Intel.

Since your issue is resolved, make sure to accept my previous post as a solution. This would help others with similar issues. Thank you!

Best Regards,

Santosh