[Beowulf] PBS, LAM_MPI error

Reuti reuti at staff.uni-marburg.de
Mon Aug 22 17:22:49 EDT 2005


Hi Onur,

when using PBS you use the TM interface of it (and use daemons 
dedicated to each
job). Please have a look at the LAM/MPI installation guide on page 
24/36 and the
user guide guide on page 71 - you don't have to give a nodelist on your 
own. But
I think the executing user must exist on all the nodes anyway, often this is
done by using NIS.

Cheers - Reuti


Zitat von Onur Destano?lu <odestanoglu at gmail.com>:

> Hi all, i have finally find my correct eror report (this seems ironic :)) )
>
> eror report file===>
> [root at bee01 ~]# cat /var/spool/torque/undelivered/31.bee00.be.ER
> -----------------------------------------------------------------------------
> It seems that there is no lamd running on the host bee01.
>
> This indicates that the LAM/MPI runtime environment is not operating.
> The LAM/MPI runtime environment is necessary for the "mpirun" command.
>
> Please run the "lamboot" command the start the LAM/MPI runtime
> environment.  See the LAM/MPI documentation for how to invoke
> "lamboot" across multiple machines.
> -----------------------------------------------------------------------------
>
> lam mpi is running on master node (bee00) and it accept nodes files
> that includes cluster nodes.
> do i have to run LAM_MPI  on cluster nodes? this means that i have to
> create same accounts on cluster nodes and run "lamboot -v hosts" on
> each of them. This sounds ridiculous.
>
> _______________________________________________
> Beowulf mailing list, Beowulf at beowulf.org
> To change your subscription (digest mode or unsubscribe) visit 
> http://www.beowulf.org/mailman/listinfo/beowulf
>



_______________________________________________
Beowulf mailing list, Beowulf at beowulf.org
To change your subscription (digest mode or unsubscribe) visit http://www.beowulf.org/mailman/listinfo/beowulf



More information about the Beowulf mailing list