[OMPI users] Problem in static linking and usage of openmpi with icc (and gcc)

2011-06-29 Thread rohan nigam
Hello Everyone,
I am trying to statically build open mpi libraries from a server which has 
intel and pgi compilers installed and copy the resulting binaries to be used on 
another test server since I do not have intel and pgi compilers on these test 
servers (only gcc). First of all, please tell me if this is possible.
So far these are the three different options I have tried and ended up getting 
the respective errors while running the mpicc or mpirun on the test server:
Option 1:  Using LDFLAGS=-Bstatic
./configure CC=icc CXX=icpc F77=ifort FC=ifort LDFLAGS=-Bstatic 
--without-memory-manager --without-libnuma --enable-static --disable-shared  
--prefix=/opt/openmpi/openmpi-intel --enable-mpi-f77 --enable-mpi-f90 2>&1 | 
tee configure-intel.log
Error on the test server:         ld: attempted static link of dynamic object 
`/cm/shared/apps/torque/2.4.11/lib/libtorque.so'

make[2]: *** [orte-clean]
Error 1 

make[2]: Leaving directory
`/opt/openmpi/openmpi-1.4.3-intel/openmpi-1.4.3/orte/tools/orte-clean' 

make[1]: *** [all-recursive]
Error 1 

make[1]: Leaving directory
`/opt/openmpi/openmpi-1.4.3-intel/openmpi-1.4.3/orte' 

make: *** [all-recursive] Error 1
Option 2: Using LDFLAGS= -staticError on the test server:                       
                           root@ubuntu-server2:/opt/openmpi-intel/bin#
./mpicc

Cannot
open configuration file
/opt/openmpi/openmpi-intel/share/openmpi/mpicc-wrapper-data.txt 

Error parsing data file mpicc:
Not found 

root@ubuntu-server2:/opt/openmpi-intel/bin#
mpirun 

                                    mpirun:
error while loading shared libraries: libtorque.so.2: cannot open shared object
file: No such file or directory
Option 3: LDFLAGS=-static-intel
(same as deprecated i-static option)Error on the test server                    
         mpirun:
error while loading shared libraries: libtorque.so.2:   cannot open shared 
object
file: No such file or directory

                             
root@ubuntu-server2:/opt/openmpi-intel/bin#
./mpicc 

Cannot
open configuration file
/opt/openmpi/openmpi-intel/share/openmpi/mpicc-wrapper-data.txt 

Error parsing data file mpicc:
Not found
How do I exclude torque shared libraries. Will that solve the problem?
FYI, both servers (test and the original server) are 64-bit AMD opterons and 
the openmpi works perfectly fine on the original server on which they were 
built on.
Any comments or suggestions will help.
Thanks,Rohan

[OMPI users] Problem running mpirun with ssh on remote nodes - Daemon did not report back when launched problem

2010-04-02 Thread rohan nigam
Hello,

I am trying to run a simple hello world program before actually launching some 
very heavy load testing over the Xen SMP set up that I have. 

I am trying to run this command over four different hosts, Dom0  being the host 
where i am launching mpirun and rest three being xen guest domains.

I have password less login setup across all the hosts. These hosts are actually 
on AMD Opteron dual socket Quad core with 2 cores assigned to each host/domain.

mpirun --prefix /root/xentools/openmpi-gcc/ -mca
 plm_rsh_agent ssh --host localhost, xenguest1 -np 4 
/root/xentools/hello

I am able to run mpirun successfully when I launch this command from one of the 
guests and also when i lauch this command on dom0 (localhost)  alone. But when 
i launch mpirun from the Dom 0 on one or more guests there is no response from 
the guests and I am eventually having to kill the process which reports saying 
  xenguest1 - daemon did not report back when launched

Can someone point out where I am going wrong. I have seen people having similar 
problem in the list but no one posted how they got around this problem.

Note: I also tried setting the default agent launcher to ssh. Also, on every 
host the directory structure is exactly the same and also the Library paths and 
paths are also set up properly and the executable is also present at the exact 
same location..

Thanks,
Rohan Nigam
Research Asst, 
Univ. of Houston





  

Re: [OMPI users] Problem running mpirun with ssh on remote nodes -Daemon did not report back when launched problem

2010-04-08 Thread rohan nigam
Hi Jeff,

You were right.  One of the other admins of the server I am working on, had a 
script that runs the firewall every time I logged in. So even when I was 
turning it off manually, the firewall ran the next time i logged in and hence 
the error.

Thanks.

- Rohan


--- On Tue, 4/6/10, Jeff Squyres  wrote:

From: Jeff Squyres 
Subject: Re: [OMPI users] Problem running mpirun with ssh on remote nodes 
-Daemon did not report back when launched problem
To: "Open MPI Users" 
List-Post: users@lists.open-mpi.org
Date: Tuesday, April 6, 2010, 11:40 AM

Open MPI opens random TCP sockets during the startup phase of MPI processes -- 
mostly from the "orted" helper process that is started on each node (or VM) 
back to the initiating mpirun process.

Do you have firewalling or other TCP blocking software running?  Or are the 
appropriate TCP routes setup between all your VMs?


On Apr 2, 2010, at 5:00 PM, rohan nigam wrote:

> Hello,
> 
> I am trying to run a simple hello world program before actually launching 
> some very heavy load testing over the Xen SMP set up that I have. 
> 
> I am trying to run this command over four different hosts, Dom0  being the 
> host where i am launching mpirun and rest three being xen guest domains.
> 
> I have password less login setup across all the hosts. These hosts are 
> actually on AMD Opteron dual socket Quad core with 2 cores assigned to each 
> host/domain.
> 
> mpirun --prefix /root/xentools/openmpi-gcc/ -mca plm_rsh_agent ssh --host 
> localhost, xenguest1 -np 4 /root/xentools/hello
> 
> I am able to run mpirun successfully when I launch this command from one of 
> the guests and also when i lauch this command on dom0 (localhost)  alone. But 
> when i launch mpirun from the Dom 0 on one or more guests there is no 
> response from the guests and I am eventually having to kill the process which 
> reports saying 
>       xenguest1 - daemon did not report back when launched
> 
> Can someone point out where I am going wrong. I have seen people having 
> similar problem in the list but no one posted how they got around this 
> problem.
> 
> Note: I also tried setting the default agent launcher to ssh. Also, on every 
> host the directory structure is exactly the same and also the Library paths 
> and paths are also set up properly and the executable is also present at the 
> exact same location..
> 
> Thanks,
> Rohan Nigam
> Research Asst, 
> Univ. of Houston
> 
> 
> 
> ___
> users mailing list
> us...@open-mpi.org
> http://www.open-mpi.org/mailman/listinfo.cgi/users


-- 
Jeff Squyres
jsquy...@cisco.com
For corporate legal information go to:
http://www.cisco.com/web/about/doing_business/legal/cri/


___
users mailing list
us...@open-mpi.org
http://www.open-mpi.org/mailman/listinfo.cgi/users