# Cannot run MPI job across nodes using OpenMPI in F17

**URL:** <https://community.unix.com/t/cannot-run-mpi-job-across-nodes-using-openmpi-in-f17/321252>\
**Category:** Programming\
**Created:** [December 13, 2012, 1:37am UTC](https://community.unix.com/t/cannot-run-mpi-job-across-nodes-using-openmpi-in-f17/321252 "2012-12-13T01:37:57Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![nsw1216](https://community.unix.com/letter_avatar/nsw1216/32/5_5575768a8748004e209b776fc1b2916d.png) [@nsw1216](https://community.unix.com/u/nsw1216)\
**Post date:** [December 13, 2012, 1:37am UTC](https://community.unix.com/t/cannot-run-mpi-job-across-nodes-using-openmpi-in-f17/321252/1 "2012-12-13T01:37:57Z")

</div>

I am new in Linux and clustering. I am setting up a Beowulf Cluster using several PCs according to this guide "Beowulf How-To"

I have setup and configure accordingly except for NFS part. Because I am not requiring it for my application. I have set my ssh to login each other without password. I started with 2 nodes 1st. I can compile and run in my headnode using openmpi. But when I try to run my MPI application across nodes, there is nothing displaying. It just like hanging there.

Headnode: master  
client: slave4

The command I used to mpirun across nodes is as below:

```nohighlight
mpirun -np 4 --host slave4 output

```

Since I not using NFS, so I installed OpenMPI in every nodes with same locations.

I wondering I missed out any configurations or not.

Hope someone can help me out of this problem.

Thanks in advance.

---

<div class="post-metadata">

**Author:** ![DGPickett](https://community.unix.com/letter_avatar/dgpickett/32/5_5575768a8748004e209b776fc1b2916d.png) [@DGPickett](https://community.unix.com/u/DGPickett)\
**Post date:** [December 18, 2012, 2:51pm UTC](https://community.unix.com/t/cannot-run-mpi-job-across-nodes-using-openmpi-in-f17/321252/3 "2012-12-18T14:51:13Z")

</div>

Were there test or sample apps built when you built OpenMPI ? Do they run?