STAR-CCM+ MPI job failure

Category: 
Resolution: 
Resolved
Affected Software: 

STAR-CCM+ encounters errors when running MPI jobs with Intel MPI or OpenMPI, displaying the following message:

ib_iface.c:1139 UCX ERROR Invalid active_speed on mlx5_0:1: 128

This issue occurs because the UCX library (v1.8) bundled with STAR-CCM+ only supports Mellanox InfiniBand EDR, while Mellanox InfiniBand NDR (Next Data Rate) is used on Cardinal. As a result, STAR-CCM+ fails to communicate correctly over the newer fabric.

Affected versions

18.18.06.006, 19.04.009 and possibly later versions

Workaround

The solution for STAR-CCM+ is to bypass the UCX library for MPI communication by configuring the environment variables appropriately:


For Intel MPI:

export -n I_MPI_HYDRA_BOOTSTRAP_EXEC_EXTRA_ARGS
export I_MPI_HYDRA_BOOTSTRAP=ssh
export FI_PROVIDER=verbs

Set these variables before executing the starccm+ command.

For OpenMPI, append the following parameters to the starccm+ batch command:

-mpiflags "--mca pml ^ucx --mca --mca btl_openib_allow_ib 1"