Dear Ralph Castain,
Thank you for you reply!!!
Actually, I have adjusted my /etc/security/limits.conf file,
I modified the "soft nofile" and "hard nofile" values up to 65535, so these
days I tried another possible limits settings
another settings include "soft memlock" ,"hard memlock", and
Dear Sabuj Pattanayek,
After your reply, I try to disable my /etc/hosts.deny,
but unfortunately, It didn't work still
But I finally solve my problem,
The reason is my "soft nofile" and "hard nofile" values aren't set large
enough,
so I can't open too much file like that
Still thanks for your
On Jun 3, 2013, at 5:06 AM, Paul Kapinos wrote:
> It is more or less well-known that MPI_THREAD_MULTIPLE disable the OpenFabric
> / InfiniBand networking in Open MPI:
>
> http://www.open-mpi.org/faq/?category=supported-systems#thread-support
>
On 06/04/2013 03:23 AM, Jeff Squyres (jsquyres) wrote:
On Jun 3, 2013, at 5:06 AM, Paul Kapinos wrote:
It is more or less well-known that MPI_THREAD_MULTIPLE disable the OpenFabric /
InfiniBand networking in Open MPI:
Hi,
So, I was able to remove the "cannot open shared file or object" errors.
But I am not able to checkpoint yet. When I enter ompi-checkpoint PID of
mpirun, it does not return anything (not even a new prompt). In my
mca-params.conf file, I added
sstore=stage
Hello everyone,
After solving my first ssh_exchange_identification problem,
I feel embarrassed to ask my another problem... :'((
I got some "*ssh: connect to host XXX.XXX.XXX.XX port 22: connection timed
out*" errors
when I mpirun over 2000 times almost at the same time.
---
my bash shell script
I'd like to be able to force mpirun to require being run under a gridengine
environment. Any ideas on how to achieve this, if possible?
--
Orion Poplawski
Technical Manager 303-415-9701 x222
NWRA, Boulder/CoRA Office FAX: 303-415-9702
3380 Mitchell Lane
There is an Mca param to require an allocation
Sent from my iPhone
On Jun 4, 2013, at 11:18 AM, Orion Poplawski wrote:
> I'd like to be able to force mpirun to require being run under a gridengine
> environment. Any ideas on how to achieve this, if possible?
>
> --
>
Am 04.06.2013 um 20:38 schrieb Ralph Castain:
> There is an Mca param to require an allocation
But this can be requested (or not) at execution time?
Even a decicated compilation with a builtin test of an allocation won't give
the intended effect, as someone could use his own compilation of
On Jun 4, 2013, at 8:20 AM, W Spector wrote:
>> Yes, this is true -- MPI_THREAD_MULITPLE support is fairly incomplete in
>> Open MPI.
>
> One would hope a simple MPI_Barrier call would work though...
Underneath, I am pretty sure that barrier is doing an MPI_WAITALL.
Yes, current releases do not have a way of prohibiting user-override of MCA
params, so a user could indeed circumvent the directive to require an
allocation. The original intent of the parameter was to close a hole that
allowed users to mistakenly overload the head node of a cluster by
forgetting
I believe that HLRS had to run the mtt-relay when it was running MTT before.
See client/mtt-relay, and https://svn.open-mpi.org/trac/mtt/changeset/623.
On Jun 4, 2013, at 5:19 AM, sethi wrote:
>
> Hello!
> I am setting up mtt testing in my institute. Clusters in my
>
12 matches
Mail list logo