Slurm和Openmpi:ORTE守护程序在启动后和通信回mpirun之前意外失败

我在两个节点中安装了openmpi和slurm。我想使用slurm运行mpi作业。当我使用

srun

运行非mpi作业时，一切正常。但是，当我使用

salloc

运行mpi作业时，出现了一些错误。环境和规范如下

Env:

slurm 17.02.1-2

mpirun（开放MPI）2.1.0

test.sh

#!/bin/bash

MACHINEFILE="nodes.$SLURM_JOB_ID"

# Generate Machinefile for mpich such that hosts are in the same
#  order as if run via srun
#
srun -l /bin/hostname | sort -n | awk '{print $2}' > $MACHINEFILE

source /home/slurm/allreduce/tf/tf-allreduce/bin/activate

mpirun -np $SLURM_NTASKS -machinefile $MACHINEFILE test

rm $MACHINEFILE

命令

salloc -N2 -n2 bash test.sh

错误

salloc: Granted job allocation 97
--------------------------------------------------------------------------
An ORTE daemon has unexpectedly failed after launch and before
communicating back to mpirun. This could be caused by a number
of factors, including an inability to create a connection back
to mpirun due to a lack of common network interfaces and/or no
route found between them. Please check network connectivity
(including firewalls and network routing requirements).
--------------------------------------------------------------------------
salloc: Relinquishing job allocation 97

---

有人能帮忙吗？谢谢。

出于调试目的，能否在脚本中插入

srun-N$SLURM\u NNODES-N$SLURM\u NNODES$（这是orted）

作为旁注，您不需要

-np

或

机器文件

选项。您可以简单地

mpirun./test

（如果您简单地使用

mpirun test

，您可能会无意中使用

/usr/bin/test

）