CFD Online Logo CFD Online URL
www.cfd-online.com
[Sponsors]
Home > Forums > Software User Forums > OpenFOAM > OpenFOAM Installation

[OpenFOAM.org] supercomputer parallel problem (I can not subjob!)

Register Blogs Community New Posts Updated Threads Search

Reply
 
LinkBack Thread Tools Search this Thread Display Modes
Old   October 12, 2015, 09:45
Default supercomputer parallel problem (I can not subjob!)
  #1
Senior Member
 
Freedom
Join Date: May 2014
Posts: 209
Rep Power: 12
wenxu is on a distinguished road
Dear foamers,

I am installing OF2.3.0 on supercomputer. But I suffer the same problem as others who had posted like: http://www.cfd-online.com/Forums/ope...entos-5-a.html. But the thing of my case is different , so I decide to post a new thread and wish someone reply me accordingly.

OF can run successfully with mpirun -np 4 interFoam -parallel on main(manage) node, but when I submit job to computing node the following error emerging.
Quote:
It looks like opal_init failed for some reason; your parallel process is
likely to abort. There are many reasons that a parallel process can
fail during opal_init; some of which are due to configuration or
environment problems. This failure appears to be an internal failure;
here's some additional information (which may only be relevant to an
Open MPI developer):

opal_shmem_base_select failed
--> Returned value -1 instead of OPAL_SUCCESS
--------------------------------------------------------------------------
[c12n5-ib:27601] [[INVALID],INVALID] ORTE_ERROR_LOG: Error in file runtime/orte_init.c at line 79
[c12n5-ib:27601] [[INVALID],INVALID] ORTE_ERROR_LOG: Error in file orterun.c at line 694
The compiler is : gcc 4.7.0
The openmpi is : openmpi 1.6.5
The supercomputer has been installed icc 14.0.0 itself. The bash file in my home is like this:
Code:
export LANG=C
export LD_LIBRARY_PATH=$HOME/software/gmp/lib:$HOME/software/mpfr/lib:$HOME/software/mpc/lib:$LD_LIBRARY_PATH
export LD_LIBRARY_PATH=$HOME/software/gcc/lib64:$LD_LIBRARY_PATH
export PATH=$HOME/software/gcc/bin:$PATH
export PATH=$HOME/software/cmake-2.8.12.2/bin:$PATH

#######openmpi#####
export PATH=$HOME/OpenFOAM/ThirdParty-2.3.0/openmpi-1.6.5/bin:$PATH
export MPI_PATH=$HOME/software/openmpi-1.6.5/lib
export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:$HOME/software/openmpi-1.6.5/lib
source $HOME/OpenFOAM/OpenFOAM-2.3.0/etc/bashrc
Any reply is highly appreciated!

Best regards,
Wen
wenxu is offline   Reply With Quote

Old   October 12, 2015, 17:37
Default
  #2
Retired Super Moderator
 
Bruno Santos
Join Date: Mar 2009
Location: Lisbon, Portugal
Posts: 10,975
Blog Entries: 45
Rep Power: 128
wyldckat is a name known to allwyldckat is a name known to allwyldckat is a name known to allwyldckat is a name known to allwyldckat is a name known to allwyldckat is a name known to all
Quick answer: Without and example of the job schedule script it's a bit hard to diagnose the problem, but my guess is that you're missing the source command to OpenFOAM's "bashrc" file for activating OpenFOAM's shell environment.
In essence:
Code:
source $HOME/OpenFOAM/OpenFOAM-2.3.0/etc/bashrc
mpirun -np 4 interFoam -parallel
For more details: http://openfoamwiki.net/index.php/In...with_the_Shell
wyldckat is offline   Reply With Quote

Old   October 12, 2015, 21:25
Default
  #3
Senior Member
 
Freedom
Join Date: May 2014
Posts: 209
Rep Power: 12
wenxu is on a distinguished road
Thank you for your quick answer, Bruno. Yes, I have sourced the bash of OF and the mpirun works fine on the main node. But when I submit the job to the computing node, the error in my last thread appears (The supercomputer is a distributed memory system.). Furthermore, I run foamInstallation command, the checking is OK.

Best regards,
Wen
wenxu is offline   Reply With Quote

Old   October 17, 2015, 10:36
Default
  #4
Retired Super Moderator
 
Bruno Santos
Join Date: Mar 2009
Location: Lisbon, Portugal
Posts: 10,975
Blog Entries: 45
Rep Power: 128
wyldckat is a name known to allwyldckat is a name known to allwyldckat is a name known to allwyldckat is a name known to allwyldckat is a name known to allwyldckat is a name known to all
Quick answer: OK, then what are the exact detailed steps you use for running the job?
One of the points of failure is somewhere in how you launch the job, but if you don't describe all of the steps you take, it's really hard to diagnose.
The other possibility is that the installation is not available in the supercomputer, because the installation folder might not be visible for all other nodes.
wyldckat is offline   Reply With Quote

Reply


Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are On
Refbacks are On


Similar Threads
Thread Thread Starter Forum Replies Last Post
problem during mpi in server: expected Scalar, found on line 0 the word 'nan' muth OpenFOAM Running, Solving & CFD 3 August 27, 2018 04:18
damBreak case parallel run problem behzad-cfd OpenFOAM Running, Solving & CFD 5 August 2, 2015 17:18
problem for parallel processing minaret OpenFOAM Running, Solving & CFD 14 January 19, 2015 23:41
Problem on Parallel Run Setup Hamidur Rahman CFX 0 September 23, 2007 17:11
cfx4 parallel sover Problem!! HELP Richard.A CFX 2 August 3, 2002 03:54


All times are GMT -4. The time now is 12:19.