Dear OpenMPI users
Relative indexing error occurs in my CentOS small cluster.
What and where should I check ?
Environment:
- 4node GbE cluster (CentOS 6.7)
- OpenMPI 1.8.7 (builded usin system compiler, gcc version 4.4.7 20120313
and installed /usr/local/openmpi-1.8.7)
- use ssh without password
On 8 October 2015 at 14:54, simona bellavista wrote:
>
>>
>> I cannot figure out how spawn would work with a string-command. I tried
>> MPI.COMM_SELF.Spawn(cmd, args=None,maxproc=4) and it just hangs
>
MPI.COMM_SELF.Spawn("/bin/echo", args=["Hello",
"World!"],maxprocs=1).Disconnect()
Could you
2015-10-09 9:40 GMT+02:00 Lisandro Dalcin :
> On 8 October 2015 at 14:54, simona bellavista wrote:
> >
>
> >>
> >> I cannot figure out how spawn would work with a string-command. I tried
> >> MPI.COMM_SELF.Spawn(cmd, args=None,maxproc=4) and it just hangs
> >
>
> MPI.COMM_SELF.Spawn("/bin/echo",
On 9 October 2015 at 12:05, simona bellavista wrote:
>
>
> 2015-10-09 9:40 GMT+02:00 Lisandro Dalcin :
>>
>> On 8 October 2015 at 14:54, simona bellavista wrote:
>> >
>>
>> >>
>> >> I cannot figure out how spawn would work with a string-command. I tried
>> >> MPI.COMM_SELF.Spawn(cmd, args=None,ma
FWIW: OpenMPI does support spawning of both MPI and non-MPI jobs. If you are
spawning a non-MPI job, then you have to -tell- us that so we don’t hang trying
to connect the new procs to the spawning proc as per MPI requirements.
This is done by providing an info key to indicate that the child job
Ralph,
Here is the result running
mpirun --map-by slot:pe=4 -display-allocation ./affinity
== ALLOCATED NODES ==
c12-29: slots=4 max_slots=0 slots_inuse=0 state=UP
=
rank 0 @ compute-
Actually, you just confirmed the problem for me. You are correct in that it
says 4 slots. However, if you then tell us pe=4, we will consume all 4 of those
slots with the very first process.
What we need to see was that slurm was assigning us 16 slots to correspond to
16 cpus. Instead, it is tr
Thank you, Ralph. The world wan wait, no problem :)
Marcin
On 10/09/2015 03:27 PM, Ralph Castain wrote:
Actually, you just confirmed the problem for me. You are correct in that it
says 4 slots. However, if you then tell us pe=4, we will consume all 4 of those
slots with the very first proce
