Ilias,
at first glance, you are using the PGI preprocessor (!)
can you re-run configure with CPP=cpp,
or after removing all PGI related environment variables,
and see it it helps ?
Cheers,
Gilles
On Wednesday, May 11, 2016, Ilias Miroslav wrote:
>
On May 10, 2016, at 5:05 PM, Gilles Gouaillardet
wrote:
>
> I was basically suggesting you open a few ports to anyone (e.g. any IP
> address), and Jeff suggests you open all ports to a few trusted IP addresses.
+1
--
Jeff Squyres
jsquy...@cisco.com
For
I was basically suggesting you open a few ports to anyone (e.g. any IP
address), and Jeff suggests you open all ports to a few trusted IP
addresses.
btw, how many network ports do you have ?
if you have two ports (e.g. eth0 for external access and eth1 for private
network) and MPI should only use
Hello Jeff,
I think what you suggest is likely exactly what we want to see happen. We
run the interop tests with at least two servers, sometimes more. We also
have other devices (InfiniBand or RoCE switches) between the servers.
I will have to ask a stupid question here but when you suggest that
Open MPI generally needs to be able to communicate on random TCP ports between
machines in the MPI job (and the machine where mpirun is invoked, if that is a
different machine).
You could also open your firewall to trust random TCP connections just between
the servers in your cluster.
> On
Hello Orion,
I actually rather like the new CentOS 7.2 system better and would like to
not remove firewalld. We will try Gilles' suggestion and see what happens.
I thank you.
--
Llolsten
-Original Message-
From: users [mailto:users-boun...@open-mpi.org] On Behalf Of Orion Poplawski
On 05/10/2016 09:24 AM, Llolsten Kaonga wrote:
> Hello Durga,
>
> As I mentioned earlier, up to version 1.8.2, we would just disable SELinux and
> the IPv4 firewall and things run smoothly. It was only when we installed
> version 1.10.2 (CentOS 7.2) that we run into these troubles. CentOS 7.2 no
This usually indicates that the remote process is using a different OMPI
version. You might check to ensure that the paths on the remote nodes are
correct.
On Tue, May 10, 2016 at 8:46 AM, lzfneu wrote:
> Hi everyone,
>
> I have a problem to consult you, when I cd to the
On 05/09/2016 04:59 PM, dpchoudh . wrote:
Hi Gus
Thanks for your suggestion. But I am not using any resource manager
(i.e. I am launching mpirun from the bash shell.). In fact, both of the
two clusters I talked about run CentOS 7 and I launch the job the same
way on both of these, yet one of
https://www.open-mpi.org/community/lists/users/2016/05/29148.php
Dear Gill,
Ooops, I don't have SIZEOF_PTRDIFF_T in config.status/config.log.
I am attaching config.log, config.status files.
This is virtual machine :
Linux grid.ui.savba.sk 2.6.32-573.26.1.el6.x86_64 #1 SMP Tue May 3 14:22:07
Hi everyone,
I have a problem to consult you, when I cd to the /examples folder contained in
the openmpi-1.8.4 package, and test the hello_c example program with mpirun
command errors occured:
Here are the command and the error messages in details:
[user@localhost examples]$ mpirun -np 2
Hello Durga,
These are dedicated servers we use to run Interoperability testing and it is
just easier for us to use a privileged user account.
As I mentioned earlier, up to version 1.8.2, we would just disable SELinux and
the IPv4 firewall and things run smoothly. It was only when we
btr files are indeed created by open mpi's backtrace mechanism. I think we
should revisit it at some point but for now the only effective way i have found
to prevent it is to restore the default signal handlers after MPI_Init.
Excuse the quoting style. Good sucks.
except if you #include the libc header in your app, *and* your send
function has a different prototype, I do not see how clang can issue a
warning
(except of course if clang "knows" all the libc subroutines ...)
Cheers,
Gilles
On Tuesday, May 10, 2016, Devon Hollowood
Siegmar,
this issue was previously reported at
http://www.open-mpi.org/community/lists/devel/2016/05/18923.php
i just pushed the patch
Cheers,
Gilles
On 5/10/2016 2:27 PM, Siegmar Gross wrote:
Hi,
I tried to install openmpi-dev-4010-g6c9d65c on my "SUSE Linux
Enterprise Server 12
you can direct OpenMPI to only use a specific range of ports (that
should be open in your firewall configuration)
mpirun --mca oob_tcp_static_ipv4_ports - ...
if you use the tcp btl, you can (also) use
mpirun --mca btl_tcp_port_min_v4 --mca btl_tcp_port_range_v4
...
Cheers,
Gilles
That worked perfectly. Thank you. I'm surprised that clang didn't emit a
warning about this!
-Devon
On Mon, May 9, 2016 at 3:42 PM, Gilles Gouaillardet <
gilles.gouaillar...@gmail.com> wrote:
> Devon,
>
> send() is a libc function that is used internally by Open MPI, and it uses
> your user
17 matches
Mail list logo