Hi, I upgraded my open-mpi version to 4.00 on a Cray Aries cluster (GNI/uGNI). Every time I run mpi, I get the following warning. Is there any way to suppress this message? I am not seeing this in 3.1.3 version of open-mpi. Any idea why openfabrics device is getting initialized when cray GNI transport is actively used? (i.e. looks like it is related to UCX, but I assume this could potentially be a configuration issue even though GNI transport/btl seems to get detected correctly and works by default without any explicit --mca parameters)
By default, for Open MPI 4.0 and later, infiniband ports on a device are not used by default. The intent is to use UCX for these devices. You can override this policy by setting the btl_openib_allow_ib MCA parameter to true. Local host: nid00301 Local adapter: ibgni Local port: 1 -------------------------------------------------------------------------- -------------------------------------------------------------------------- WARNING: There was an error initializing an OpenFabrics device. Local host: nid00300 Local device: ibgni ... [program output] ... [program output] ... [login1:02032] 3 more processes have sent help message help-mpi-btl-openib.txt / ib port not selected [login1:02032] Set MCA parameter "orte_base_help_aggregate" to 0 to see all help / error messages [login1:02032] 3 more processes have sent help message help-mpi-btl-openib.txt / error in device init Thanks, Udayanga
_______________________________________________ users mailing list users@lists.open-mpi.org https://lists.open-mpi.org/mailman/listinfo/users