Fwd: [its-cluster-admin] Fwd: [gmx-users] gmx_blast error when attempting to run in parallel

Erik Marklund erikm at xray.bmc.uu.se
Thu Jun 11 21:20:08 CEST 2009


Justin A. Lemkul skrev:
>
>
> Erik Marklund wrote:
>> Your admin compiled another version thah the one you're using. Look 
>> closely at tle path.
>>
>> /Erik
>
> <snip>
>
>>> export CPPFLAGS="/usr/local/fftw-3.1.2/include"
>
> In addition, this is not correct.  It should read:
>
> export CPPFLAGS="-I/usr/local/fftw-3.1.2/include"
Indeed.
>
>>> ./configure --prefix=/usr/local/gromacs/gromacs-${version} 
>>> --with-fft=fftw3  --enable-mpi
>>> make > make.out 2>&1
>>> make install
>
> I don't know if this makes any difference or not (probably not), but 
> if you just want the mdrun executable, the commands should be "make 
> mdrun" and "make install-mdrun"
>
> For diagnostic purposes, it may be more useful to run the installation 
> manually, instead of with a script.  That way, if there are warnings 
> along the way (configure should have complained about the 
> CPPFLAGS...), then you will see them instead of going on auto-pilot.
>
I disagree. One should always sift through the configure-/make-output, 
obviously (and therefore a tee would be better than the redirection 
above), but I find scripting is superior to manual compilation. It makes 
it easy to see exactly what options and stuff you used and to alter when 
needed. Plus, I can just copy the script when compiling with a slightly 
different setup or with another version of gromacs.

/Erik
> -Justin
>
>>>
>>> The name of executable should not really matter. Any way I will try 
>>> to build new version later today or tomorrow. You can give a try 
>>> yourself. If you are succeded or I make progress let me know. Also 
>>> could you please sene me some of your simple test files.
>>>
>>> Thanks.
>>> Tula
>>>
>>>  
>>>
>>>
>>>
>>> On Thu, Jun 11, 2009 at 10:29 AM, Sashank Karri <srk18 at case.edu 
>>> <mailto:srk18 at case.edu>> wrote:
>>>
>>>     Hi Tula,
>>>
>>>     I received two e-mails informing me that gromacs was not compiled
>>>     with MPI.  Here is the second, more informative e-mail.  Could you
>>>     possibly try compiling gromacs-4.0.5 this time?  I'm around
>>>     Fridays with nothing to do now, if you want someone with a little
>>>     bit of gromacs experience around while you're working on it.
>>>
>>>     Thanks,
>>>
>>>     Sashank
>>>
>>>     ---------- Forwarded message ----------
>>>     From: *Justin A. Lemkul* <jalemkul at vt.edu <mailto:jalemkul at vt.edu>>
>>>     Date: Thu, Jun 11, 2009 at 8:24 AM
>>>     Subject: Re: [gmx-users] gmx_blast error when attempting to run in
>>>     parallel
>>>     To: Discussion list for GROMACS users <gmx-users at gromacs.org
>>>     <mailto:gmx-users at gromacs.org>>
>>>
>>>
>>>
>>>
>>>
>>>     Sashank Karri wrote:
>>>
>>>         Hi, I'm running gromacs-4.0.3 on a cluster.  I am testing
>>>         gromacs on it.  We are currently getting this error when I run
>>>         with 4 nodes with one dedicated to PME calculations.
>>>         Back Off! I just backed up md.log to ./#md.log.5#
>>>         Reading file ionsol_minim96-1.tpr, VERSION 4.0.3 (single
>>>         precision)
>>>
>>>         ------------------------------
>>>         -------------------------
>>>         Program mdrun, VERSION 4.0.3
>>>         Source code file: network.c, line: 357
>>>
>>>         Routine should not have been called:
>>>         gmx_bast
>>>         -------------------------------------------------------
>>>
>>>         "I'll Match Your DNA" (Red Hot Chili Peppers).
>>>
>>>         Here is my job that I'm submitting to the cluster.
>>>
>>>         #PBS -N ionsol_karri
>>>         #PBS -l walltime=24:00:00
>>>         #PBS -l nodes=4:ppn=4:quad
>>>         #PBS -j oe
>>>
>>>         module load mpich
>>>         module load gromacs-4.0.3
>>>         cd /home/srk18/newplcre/
>>>         mpirun -nodes 4 /usr/local/gromacs/gromacs-4.0.3/bin/mdrun
>>>         -npme 1 -s ionsol_minim96-1.tpr -o finminim96_traj.trr -x
>>>         finminim96_traj.xtc -c final_minim.g96 -e enermin_fin.edr -cpo
>>>         state96.cpt
>>>
>>>
>>>     Is your mdrun truly MPI-enabled?  Typically the configure script
>>>     advises you to append _mpi to the parallel version.
>>>
>>>     Also, upgrade to the newest version of Gromacs, 4.0.5, and try
>>>     again.  That way, if it is a bug, then changes can be made to the
>>>     newest version, and not one that is five months old.
>>>
>>>     -Justin
>>>
>>>         When discussing the installation of gromacs earlier with the
>>>         cluster admin, he sent the following message:
>>>
>>>          >I remember I seeing your error (not the present error) while
>>>         I tried to make the dynamic version of the GROMACS. So I used
>>>         the --enable-static while building parallel version of the
>>>         GROMACS. That build the executable and most of the inbuilt
>>>         test were successful. If you think we need dynamic build let
>>>         me know, I might have to struggle a bit. Also It would be good
>>>          if you share information you have regarding building.
>>>
>>>         Do you folks have any idea as to how to fix the above error?
>>>
>>>         Thanks,
>>>         Sashank Karri
>>>
>>>
>>>         
>>> ------------------------------------------------------------------------ 
>>>
>>>
>>>         _______________________________________________
>>>         gmx-users mailing list    gmx-users at gromacs.org
>>>         <mailto:gmx-users at gromacs.org>
>>>         http://lists.gromacs.org/mailman/listinfo/gmx-users
>>>         Please search the archive at http://www.gromacs.org/search
>>>         before posting!
>>>         Please don't post (un)subscribe requests to the list. Use the
>>>         www interface or send it to gmx-users-request at gromacs.org
>>>         <mailto:gmx-users-request at gromacs.org>.
>>>         Can't post? Read http://www.gromacs.org/mailing_lists/users.php
>>>
>>>
>>>     --     ========================================
>>>
>>>     Justin A. Lemkul
>>>     Ph.D. Candidate
>>>     ICTAS Doctoral Scholar
>>>     Department of Biochemistry
>>>     Virginia Tech
>>>     Blacksburg, VA
>>>     jalemkul[at]vt.edu <http://vt.edu> | (540) 231-9080
>>>     http://www.bevanlab.biochem.vt.edu/Pages/Personal/justin
>>>
>>>     ========================================
>>>
>>>     _______________________________________________
>>>     gmx-users mailing list    gmx-users at gromacs.org
>>>     <mailto:gmx-users at gromacs.org>
>>>     http://lists.gromacs.org/mailman/listinfo/gmx-users
>>>     Please search the archive at http://www.gromacs.org/search before
>>>     posting!
>>>     Please don't post (un)subscribe requests to the list. Use the www
>>>     interface or send it to gmx-users-request at gromacs.org
>>>     <mailto:gmx-users-request at gromacs.org>.
>>>     Can't post? Read http://www.gromacs.org/mailing_lists/users.php
>>>
>>>
>>>
>>>
>>> -- 
>>> Tula R Paudel Ph.D
>>> 10900 Euclid Ave
>>> Rockfeller 104A / Crawford 403
>>> Cleveland OH-44106
>>> @216 368 4035, 216 368 0395
>>>
>>> ------------------------------------------------------------------------ 
>>>
>>>
>>> _______________________________________________
>>> gmx-users mailing list    gmx-users at gromacs.org
>>> http://lists.gromacs.org/mailman/listinfo/gmx-users
>>> Please search the archive at http://www.gromacs.org/search before 
>>> posting!
>>> Please don't post (un)subscribe requests to the list. Use the www 
>>> interface or send it to gmx-users-request at gromacs.org.
>>> Can't post? Read http://www.gromacs.org/mailing_lists/users.php
>>
>>
>


-- 
-----------------------------------------------
Erik Marklund, PhD student
Laboratory of Molecular Biophysics,
Dept. of Cell and Molecular Biology, Uppsala University.
Husargatan 3, Box 596,    75124 Uppsala, Sweden
phone:    +46 18 471 4537        fax: +46 18 511 755
erikm at xray.bmc.uu.se    http://xray.bmc.uu.se/molbiophys




More information about the gromacs.org_gmx-users mailing list