Simple MPI Testing Fails

[WasimWani]

Hi 

So i am doing simple MPI testing on RHEL 8.3. The MPI works on first run then fails and then works if i reboot again only once. The MPI is as follows:

[root@localhost test.mpi]# cat mpi_hello_world.c
#include <mpi.h>
#include <stdio.h>

int main(int argc, char** argv) {
// Initialize the MPI environment
MPI_Init(NULL, NULL);

// Get the number of processes
int world_size;
MPI_Comm_size(MPI_COMM_WORLD, &world_size);

// Get the rank of the process
int world_rank;
MPI_Comm_rank(MPI_COMM_WORLD, &world_rank);

// Get the name of the processor
char processor_name[MPI_MAX_PROCESSOR_NAME];
int name_len;
MPI_Get_processor_name(processor_name, &name_len);

// Print off a hello world message
printf("Hello world from processor %s, rank %d out of %d processors\n",
processor_name, world_rank, world_size);

// Finalize the MPI environment.
MPI_Finalize();
}

[root@localhost test.mpi]#

I get the following error:

[root@localhost test.mpi]# cc mpi_hello_world.c
[root@localhost test.mpi]# ./a.out
CCE libu memcpy: WARNING: Intel cpuid detection failed
Using generic mem* routines instead of tuned routines
This message can be disabled by exporting RFE_811452_DISABLE to your environment
MPICH ERROR [Rank 0] [job id ] [Sat Apr 17 04:13:19 2021] [localhost.localdomain] - Abort(2161295) (rank 0 in comm 0): Fatal error in PMPI_Init: Other MPI error, error stack:
MPIR_Init_thread(147).......:
MPID_Init(495)..............:
MPIDI_OFI_mpi_init_hook(606):
open_fabric(1318)...........:
find_provider(1601).........: OFI fi_getinfo() failed (ofi_init.c:1601:find_provider:No data available)

aborting job:
Fatal error in PMPI_Init: Other MPI error, error stack:
MPIR_Init_thread(147).......:
MPID_Init(495)..............:
MPIDI_OFI_mpi_init_hook(606):
open_fabric(1318)...........:
find_provider(1601).........: OFI fi_getinfo() failed (ofi_init.c:1601:find_provider:No data available)
[root@localhost test.mpi]#

Your inputs are highly appreciated.

[Victor_G_Intel]

Hello WasimWani,

Thank you for posting on the Intel® communities.

1-Can you please let us know if what you are currently referring to as MPI testing is related to the information found in the link down below?

https://software.intel.com/content/www/us/en/develop/documentation/get-started-with-mpi-for-linux/top.html

2-Please provide the following report:

Intel® System Support Utility (Intel® SSU)

Download link: https://downloadcenter.intel.com/download/26735/Intel-System-Support-Utility-for-the-Linux-Operating-System

Instructions: https://downloadmirror.intel.com/26735/eng/ReleaseNotes.txt

I look forward to hearing from you.

Regards,

Victor G.

Intel Technical Support Technician  

[Victor_G_Intel]

Hello WasimWani,

Were you able to check the previous post?  

Please let me know if you need further assistance.  

Best regards,

Victor G. 

Intel Technical Support Technician

[WasimWani]

Hello Victor

I was able to resolve this. Thanks for sharing the post but i had some other problem in my case this is how it got solved.

export MPICH_OFI_USE_PROVIDER=sockets

Thanks for helping out Victor.

Regards
Wasim

[Victor_G_Intel]

Hello WasimWani,

Thank you for your response.

We are glad that you solve your issue and also appreciate that you took the time to share your fix in the forum. We will proceed to close this thread. If you need any additional information in the future, you are more than welcome to submit a new question, just remember that this thread will no longer be monitored.

Regards,

Victor G.

Intel Technical Support Technician