I don't understand your question - it's definitely a LAM error, but you mention open MPI.
I would highly recommend upgrading to open MPI and using it's native torque support.
Sent from my phone. No type good.
On Jun 30, 2011, at 10:11 PM, "zhuangchao" <freeoser_at_[hidden]> wrote:
> hello all ,
>
> I submited the following Torque/pbs script.
>
> #PBS -e /tmp/blast_19297.err
> #PBS -o /tmp/blast.output
> mpiexec -d -machinefile /tmp/nodes.19297.txt -np 3 /data1/bin/mpiblast -p tblastx -i /data1/cluster/sequences/seq_4.txt -d nt -o /data1/cluster/blast.out/blast.out.19297 -g T -m 0 -F F
>
>
>
> Then I got the following error from openmpi. But it is ok if I run the above mpiexec command
> in the bash shell. Can you help me ? Thank you very much !
>
> n-1<18242> ssi:boot:base: looking for boot schema in following directories:
> n-1<18242> ssi:boot:base: <current directory>
> n-1<18242> ssi:boot:base: $TROLLIUSHOME/etc
> n-1<18242> ssi:boot:base: $LAMHOME/etc
> n-1<18242> ssi:boot:base: /data1/lam-mpi/etc
> n-1<18242> ssi:boot:base: looking for boot schema file:
> n-1<18242> ssi:boot:base: /tmp/nodes.19297.txt
> n-1<18242> ssi:boot:base: found boot schema: Not found!
> -----------------------------------------------------------------------------
> The hostfile "/tmp/nodes.19297.txt" could not be found:
> Unix errno: 2
> No such file or directory
> Things to check:
> - ensure that the file exists
> try "ls -l /tmp/nodes.19297.txt"
> - ensure that you have read permissions on the file
> try "cat /tmp/nodes.19297.txt"
> -----------------------------------------------------------------------------
> -----------------------------------------------------------------------------
> It seems that there is no lamd running on the host node2.
> This indicates that the LAM/MPI runtime environment is not operating.
> The LAM/MPI runtime environment is necessary for the "lamhalt" 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.
> _______________________________________________
> This list is archived at http://www.lam-mpi.org/MailArchives/lam/
|
