Are you 100% sure that you are not accidentally mixing and matching multiple 
versions of Open MPI in the same job?  This type of error (PMIX bad param) is 
typical when you accidentally use Open MPI vXYZ on one node and Open MPI vABC 
on a different node.


> On Apr 4, 2018, at 12:20 AM, abhisek Mondal <abhisek.m...@gmail.com> wrote:
> 
> Hi,
> 
> I have recently upgraded my system to openmpi-3.0. But despite proper 
> installation and gpu integration I keep receiving this error, as I was also 
> receiving in openmpi-1.4:
> 
> $ mpirun -np 10 -bynode  `which relion_preprocess_mpi`  --i 
> input_micrographs.star --coord_dir "." --coord_suffix .coords.star 
> --part_star extra/output_particles.star --part_dir "." --extract  
> --extract_size 140 --bg_radius 52 --invert_contrast --norm
> 
> 00021:   
> --------------------------------------------------------------------------
> 00022:   The following command line options and corresponding MCA parameter 
> have
> 00023:   been deprecated and replaced as follows:
> 00024:   
> 00025:     Command line options:
> 00026:       Deprecated:  --bynode, -bynode
> 00027:       Replacement: --map-by node
> 00028:   
> 00029:     Equivalent MCA parameter:
> 00030:       Deprecated:  rmaps_base_bynode
> 00031:       Replacement: rmaps_base_mapping_policy=node
> 00032:   
> 00033:   The deprecated forms *will* disappear in a future version of Open 
> MPI.
> 00034:   Please update to the new syntax.
> 00035:   
> --------------------------------------------------------------------------
> 00036:   [localhost.localdomain:29946] PMIX ERROR: BAD-PARAM in file 
> ../../../../../../../opal/mca/pmix/pmix2x/pmix/src/dstore/pmix_esh.c at line 
> 1005
> 00037:   [localhost.localdomain:29951] PMIX ERROR: BAD-PARAM in file 
> ../../../../../../../opal/mca/pmix/pmix2x/pmix/src/dstore/pmix_esh.c at line 
> 1005
> 00038:   [localhost.localdomain:29948] PMIX ERROR: BAD-PARAM in file 
> ../../../../../../../opal/mca/pmix/pmix2x/pmix/src/dstore/pmix_esh.c at line 
> 1005
> 00039:   [localhost.localdomain:29952] PMIX ERROR: BAD-PARAM in file 
> ../../../../../../../opal/mca/pmix/pmix2x/pmix/src/dstore/pmix_esh.c at line 
> 1005
> 00040:   [localhost.localdomain:29944] PMIX ERROR: BAD-PARAM in file 
> ../../../../../../../opal/mca/pmix/pmix2x/pmix/src/dstore/pmix_esh.c at line 
> 1005
> 00041:   [localhost.localdomain:29950] PMIX ERROR: BAD-PARAM in file 
> ../../../../../../../opal/mca/pmix/pmix2x/pmix/src/dstore/pmix_esh.c at line 
> 1005
> 00042:   [localhost.localdomain:29949] PMIX ERROR: BAD-PARAM in file 
> ../../../../../../../opal/mca/pmix/pmix2x/pmix/src/dstore/pmix_esh.c at line 
> 1005
> 00043:   *** An error occurred in MPI_Init
> 00044:   *** on a NULL communicator
> 00045:   *** MPI_ERRORS_ARE_FATAL (processes in this communicator will now 
> abort,
> 00046:   ***    and potentially your MPI job)
> 00047:   *** An error occurred in MPI_Init
> 00048:   *** on a NULL communicator
> 00049:   *** MPI_ERRORS_ARE_FATAL (processes in this communicator will now 
> abort,
> 00050:   ***    and potentially your MPI job)
> 00051:   *** An error occurred in MPI_Init
> 00052:   *** on a NULL communicator
> 00053:   *** MPI_ERRORS_ARE_FATAL (processes in this communicator will now 
> abort,
> 00054:   ***    and potentially your MPI job)
> 00055:   [localhost.localdomain:29952] Local abort before MPI_INIT completed 
> completed successfully, but am not able to aggregate error messages, and not 
> able to guarantee that all other processes were killed!
> 
> I'm not sure what is causing this crash.
> 
> Please help me out.
> 
> Thank you
> 
> -- 
> Abhisek Mondal
> Senior Research Fellow
> Structural Biology and Bioinformatics Division
> CSIR-Indian Institute of Chemical Biology
> Kolkata 700032
> INDIA
> _______________________________________________
> users mailing list
> users@lists.open-mpi.org
> https://lists.open-mpi.org/mailman/listinfo/users


-- 
Jeff Squyres
jsquy...@cisco.com

_______________________________________________
users mailing list
users@lists.open-mpi.org
https://lists.open-mpi.org/mailman/listinfo/users

Reply via email to