Missing hostname or invalid host/port description in business card

[StreamAzure]

Hi,

I'm using Intel 2022.1.2 on CentOS 7 and trying to run the MPI tests in the path intel/oneapi/mpi/latest/test on two machines but having the problem as follows:

$ export I_MPI_CC=icx
$ mpiicc test.c
$ mpirun -np 2 -ppn 1 -hosts host01,host02 ./a.out 
Abort(1614991) on node 0 (rank 0 in comm 0): Fatal error in PMPI_Init: Other MPI error, error stack:
MPIR_Init_thread(143)........: 
MPID_Init(1310)..............: 
MPIDI_OFI_mpi_init_hook(1953): 
MPIDU_bc_table_create(320)...: Missing hostname or invalid host/port description in business card

And it can run without `-ppn`, but all processes will be executed on the first mechine that `-hosts` identifies. For example, if I run `mpirun -np 2 -hosts host01,host02 ./a.out ` on host02, both 2 processes will be executed on host01.

I also tried `mpirun -np 2 -ppn 1 -hosts host01,host02 hostname`. It worked successfully and the two hostnames are printed.

Any help will be appreciated.

Thanks,
Mok

[StreamAzure · Accepted solution]

We made adjustment on our cluster and it seems that something has been fixed. Now it works successfully as expected.

I guess the problem is caused by some bad wires in the machine.

[SantoshY_Intel]

Hi,

Thanks for reaching out to us.

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

Thanks & Regards,

Santosh