Am 13.10.2008 um 00:55 schrieb Sean Davis:
On Sat, Oct 11, 2008 at 6:48 PM, Reuti <re...@staff.uni-marburg.de>
wrote:
Am 12.10.2008 um 00:21 schrieb Sean Davis:
<snip>
Thanks, Pak. There is only one queue on the SGE system. Of course,
there are queue instances for each machine, which is the usual for
SGE.
I'll give the -masterq a look. And the messages files for the
involved machines are devoid of anything useful; in fact, there
is no
mention of these jobs, in general.
Hi,
to see more, you can set "loglevel log_info" in the scheduler
configuration.
Do you have more than one network card installed and gave them the
same
name?
Your defined "tmpdir" is local on each machine?
Do you redifine $TMPDIR in your .bashrc or anything else therein?
The tmpdir is the same on each machine and is local to the machine.
We do have two interfaces on each machine, one for a local subnet and
the other for an outside connection from each machine. The DNS is
resolved on the outside network. Why would behavior be different on
jobs that are run in a PE and only in $round_robin than for standard
serial jobs or jobs on only one node?
Hi Sean,
I found it in MPICH(1), and don't know whether it applies also to
Open MPI. The node where the master of the parallel job starts, i.e.
executes the mpirun command, will send in the startup message his own
name for the slave tasks. When this is the outside network, the nodes
will never find the master. Therefore a special variable must be set,
MPI_HOST, to the name of the internal interface before calling mpirun:
MPI_HOST=`grep $(hostname) $SGE_ROOT/default/common/host_aliases |
cut -f 1 -d " "`
The master on its own is aware of both names, but not the other nodes.
Did you setup host_aliases? Is the primary interface used for the
internal or the external connection?
--- Reuti
Sean
Date: Sat, 11 Oct 2008 07:56:02 -0400
From: Jeff Squyres <jsquy...@cisco.com>
Subject: Re: [OMPI users] SGE tight integration and ?tm?
protocol for
start
To: Open MPI Users <us...@open-mpi.org>
Message-ID: <3e62159b-14b9-4d44-96f6-0345079bc...@cisco.com>
Content-Type: text/plain; charset=US-ASCII; format=flowed;
delsp=yes
I don't know much/anything about SGE (I'll leave that to the
Sun folks
on
this list to reply), but I can tell you about the tm plugins:
tm is the
protocol used by the PBS/Torque family of launchers. It looks
like
your
Open MPI was built with TM support, but when you launch, it's
likely
unable
to find the support libraries that it needs to load those
plugins.
This is probably fine in your case, since you want to use SGE,
not TM.
On Oct 9, 2008, at 4:40 PM, Sean Davis wrote:
I am relatively new to OpenMPI and Sun Grid Engine parallel
integration. I have a small cluster that is running SGE6.2 on
linux
machines all using Intel Xeon processors. I have installed
OpenMPI
1.2.7 from source using the --with-sge switch. Now, I am
trying to
troubleshoot some problems I am having. I have created a
simple job
script:
The job script looks like:
#!/bin/bash
#$ -S /bin/bash
#$ -cwd
mpirun --mca pls_gridengine_verbose 1 -np $NSLOTS hostname
And the output on the error stream:
more junksub.sh.e3574
[shakespeare:05720] mca: base: component_find: unable to open
ras tm:
file not found (ignored)
[shakespeare:05720] mca: base: component_find: unable to open
pls tm:
file not found (ignored)
Starting server daemon at host "shakespeare.nci.nih.gov"
Starting server daemon at host "octopus.nci.nih.gov"
Server daemon successfully started with task id "1.shakespeare"
[shakespeare:05733] mca: base: component_find: unable to open
ras tm:
file not found (ignored)
[shakespeare:05733] mca: base: component_find: unable to open
pls tm:
file not found (ignored)
error: executing task of job 3576 failed: failed sending task to
ex...@octopus.nci.nih.gov: can't find connecti
on
[shakespeare:05720] ERROR: A daemon on node
octopus.nci.nih.gov failed
to start as expected.
[shakespeare:05720] ERROR: There may be more information
available
from
[shakespeare:05720] ERROR: the 'qstat -t' command on the Grid
Engine
tasks.
[shakespeare:05720] ERROR: If the problem persists, please
restart the
[shakespeare:05720] ERROR: Grid Engine PE job
[shakespeare:05720] ERROR: The daemon exited unexpectedly
with status
1.
However, there is no output in any output stream.
And if I log into shakespeare and qrsh -q all.q@octopus, I
immediately
get a slot, so there isn't a "direct" problem with connecting.
As I got a hint from folks on the SGE mailing list, it appears
that
qrsh is not being used for job submission. Any suggestions as
to why
this might be the case (or if it is the case)?
Thanks,
Sean
_______________________________________________
users mailing list
us...@open-mpi.org
http://www.open-mpi.org/mailman/listinfo.cgi/users
_______________________________________________
users mailing list
us...@open-mpi.org
http://www.open-mpi.org/mailman/listinfo.cgi/users
_______________________________________________
users mailing list
us...@open-mpi.org
http://www.open-mpi.org/mailman/listinfo.cgi/users
_______________________________________________
users mailing list
us...@open-mpi.org
http://www.open-mpi.org/mailman/listinfo.cgi/users
_______________________________________________
users mailing list
us...@open-mpi.org
http://www.open-mpi.org/mailman/listinfo.cgi/users
