INCF / MUSIC

MUSIC, the MUltiSimulation Coordinator
GNU General Public License v3.0
37 stars 37 forks source link

Unable to determine process rank #64

Open cristianoalessandro opened 4 years ago

cristianoalessandro commented 4 years ago

I am trying to execute the simple MUSIC demo example from the nest tutorial, but I get the following error:

$ mpirun -np 4 music python.music
MUSIC: Unable to determine process rank.
MUSIC: Did you launch music using mpirun?
MUSIC: If so, check the comments about porting in README.

I get a little more information by trying to run the example without using the music binary:

$ mpirun -np 4 ./send.py --music-confing python.music --app-label from -np 4 ./receive.py --music-config python.music --app-label to
[...]
Fatal error in PMPI_Intercomm_create: Invalid rank, error stack:
PMPI_Intercomm_create(330): MPI_Intercomm_create(comm=0x84000004, local_leader=0, MPI_COMM_WORLD, remote_leader=-1, tag=0, newintercomm=0x7ffcb7760938) failed
PMPI_Intercomm_create(288): Error specifying remote_leader; rank given was -1 but must be in the range 0 to 3

The README says this is likely due to an unexpected version of MPI, but I am not sure how to solve it. I am running mpich 3.3~a2-4, and nest-2.18.0 (installed with music and mpi) on Ubuntu 16.04. Thanks in advance.

mdjurfeldt commented 4 years ago

I'll look at this tomorrow. Meanwhile, is it possible for you to use OpenMPI? In Debian-like distributions, you typically install the packages openmpi-bin and libopenmpi-dev.

mdjurfeldt commented 4 years ago

(Also, did you really write --music-confing on the command line, or did this just happen when writing the issue text?)

cristianoalessandro commented 4 years ago

Thanks! Openmpi used to give me issues with some other software I also need, but I can try again. As for your other question; there was indeed a typo in the command, but resolving the typo did not solve the problem (I get the very same error).

cristianoalessandro commented 4 years ago

The MUSIC example works with OpenMPI! No issue with both the following commands:

$ mpirun -np 4 music python.music
$ mpirun -np 2 ./send.py --music-config python.music --app-label from : -np 2 ./receive.py --music-config python.music --app-label to

OpenMPI is fine with my personal situation, but the issue seems to remain if one needs to use the mpich implementation of MPI.