Juan,

So you need to load the knem module
sudo modprobe knem
and then you can check it is correctly loaded with lsmod

Loading the module should automagically create /dev/knem, but maybe not
with the permissions you expect

Cheers,

Gilles

On Wednesday, January 18, 2017, Juan A. Cordero Varelaq <
bioinformatica-i...@us.es> wrote:

> Hi,
>
> knem and MOFED drivers are installed in /opt:
>
>    - /opt/knem-1.1.90mlnx2
>    - /opt/mellanox/fca
>    - /opt/mellanox/mxm
>    - /opt/mellanox/openshmem
>
> However /dev/knem does not exist.
>
> Cheers,
>
> Juan
> On 18/01/17 11:36, Gilles Gouaillardet wrote:
>
> Juan,
>
> You also need to make sure knem and MOFED drivers are loaded on all the
> compute nodes.
> You also need to double check the permissions of /dev/knem
>
> Cheers,
>
> Gilles
>
> On Wednesday, January 18, 2017, Juan A. Cordero Varelaq <
> bioinformatica-i...@us.es
> <javascript:_e(%7B%7D,'cvml','bioinformatica-i...@us.es');>> wrote:
>
>> Sure, I attach the config.log from the openmpi installation.
>>
>> On 17/01/17 17:55, Joshua Ladd wrote:
>>
>> Can you please attach your configure log. It looks like both MXM and the
>> Vader BTL (used for OSC) are complaining because they can't find your KNEM
>> installation.
>>
>> Josh
>>
>> On Tue, Jan 17, 2017 at 7:16 AM, Juan A. Cordero Varelaq <
>> bioinformatica-i...@us.es> wrote:
>>
>>> Hi, I am running on my SCG cluster the following script (using qsub):
>>>
>>> #!/bin/bash
>>> #$-cwd
>>> #$ -S /bin/bash
>>> #$ -V
>>> #$ -q normal
>>> #$ -pe mpi 40
>>> #$ -P Lab219
>>> #$ -o output
>>> #$ -e error
>>> module load PhyML/3.3
>>> mpirun --mca pml yalla -np 40 phyml-mpi -i proteic -b 10 -d aa
>>>
>>> where phyml-mpi is the parallel version for OMPI of the program PhyML.
>>> --mca pml yalla option is called to used MXM (I have mellanox OFED).
>>>
>>>
>>> It gives me lots of errors related to KNEM (see error and output files
>>> from qsub in the attachments). However, I specified the KNEM directory when
>>> installing OMPI. I can't really understand such errors, and would
>>> appreciate any hint on this issue. I have run open-mpi on an own script
>>> (just a loop running inside something as: command --help) and got no error.
>>>
>>>
>>> Thanks in advance
>>>
>>>
>>> _______________________________________________
>>> users mailing list
>>> users@lists.open-mpi.org
>>> https://rfd.newmexicoconsortium.org/mailman/listinfo/users
>>>
>>
>>
>>
>> _______________________________________________
>> users mailing 
>> listus...@lists.open-mpi.orghttps://rfd.newmexicoconsortium.org/mailman/listinfo/users
>>
>> _______________________________________________
> users mailing listus...@lists.open-mpi.org 
> <javascript:_e(%7B%7D,'cvml','users@lists.open-mpi.org');>https://rfd.newmexicoconsortium.org/mailman/listinfo/users
>
>
_______________________________________________
users mailing list
users@lists.open-mpi.org
https://rfd.newmexicoconsortium.org/mailman/listinfo/users

Reply via email to