Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

MPI initialization exits ungracefully #3123

Open
Helveg opened this issue Mar 1, 2024 · 1 comment
Open

MPI initialization exits ungracefully #3123

Helveg opened this issue Mar 1, 2024 · 1 comment
Labels
I: No breaking change Previously written code will work as before, no one should note anything changing (aside the fix) S: Normal Handle this with default priority stale Automatic marker for inactivity, please have another look here T: Discussion Still searching for the right way to proceed / suggestions welcome
Projects

Comments

@Helveg
Copy link
Contributor

Helveg commented Mar 1, 2024

Describe the bug
When you run mpirun python ... and import nest built without MPI support, the following message is displayed and the process exits abruptly:

[FATAL] [2024.3.1 12:15:12 /home/robin/git/nest-simulator/nestkernel/mpi_manager.cpp:203 @ MPIManager::initialize()] : You seem to be using NEST via an MPI launcher like mpirun, mpiexec or srun although NEST was not compiled with MPI support. Please see the NEST documentation about parallel and distributed computing. Exiting.

This causes many problems associated with abrupt exits (open file handles, killed threads, ...) and in my case led to a segmentation fault because of an open MPI.recv call on another thread, which obscured the original cause of the error.

Expected behavior
NEST should raise an exception instead, and this should be propagated to the Python layer, like other NEST exceptions.

Desktop/Environment (please complete the following information):

  • OS: [e.g. iOS, Ubuntu-18.04]: WSL2
  • Shell: bash
  • Python-Version: 3.11
  • NEST-Version: nest 3.6.0-post0.dev0
  • Installation: without MPI

(PS: A collaborator of mine had the same type of ungraceful exit when NEST is built with MUSIC support, but MPI has already been initialized by another software, such as mpi4py. Could anyone do a quick double check whether this could be solved the same way?)

@gtrensch gtrensch added T: Discussion Still searching for the right way to proceed / suggestions welcome S: Normal Handle this with default priority I: No breaking change Previously written code will work as before, no one should note anything changing (aside the fix) labels Mar 22, 2024
@gtrensch gtrensch added this to To do (open issues) in Kernel via automation Mar 22, 2024
Copy link

Issue automatically marked stale!

@github-actions github-actions bot added the stale Automatic marker for inactivity, please have another look here label May 22, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
I: No breaking change Previously written code will work as before, no one should note anything changing (aside the fix) S: Normal Handle this with default priority stale Automatic marker for inactivity, please have another look here T: Discussion Still searching for the right way to proceed / suggestions welcome
Projects
Kernel
  
To do (open issues)
Development

No branches or pull requests

2 participants