[Users] Simfactory (Maxwell) issue on Lonestar

[Erik Schnetter]

P.

I am currently encountering the same error, and am engaged in a
discussing with their support to find the cause.

(The "module swap" should go into the *.run script instead since the
*.ini file should remain independent of the MPI stack used. However, I
encountered the same problem as you, and will commit by state once
things are working for me.)

-erik

On Fri, Apr 13, 2012 at 3:43 AM, P. Ajith <ajith_p at ligo.caltech.edu> wrote:
>
> Hello!
>
> I am trying to run the Tutorial for New Users
> (https://docs.einsteintoolkit.org/et-docs/Tutorial_for_New_Users) on
> Lonestar (this is the first time I am using the machine). My job fails with
> the following error (in the .out) file:
>
>   ....
>
> Simulation name: static_tov_8
> Running simulation static_tov_8
> Preparing:
> Checking:
> /scratch/01489/pajith/simulations/static_tov_8/output-0000-active
> c324-310.ls4.tacc.utexas.edu
> Fri Apr 13 01:46:47 CDT 2012
> Environment:
> Starting:
> TACC: Starting up job 541416
> TACC: Setting up parallel environment for OpenMPI mpirun.
> TACC: Setup complete. Running job script.
> TACC: starting parallel tasks...
> TACC: Could not determine MPI stack. Exiting!
> ...
>
>
> These are the modules that are loaded:
>
> login2$ module list
> Currently Loaded Modules:
>   1) TACC-paths  2) Linux  3) cluster-paths  4) intel/11.1  5) gzip/1.3.12
>  6) tar/1.22  7) cluster  8) TACC  9) openmpi/1.4.3
>
>
> And my defs.local.ini file looks like this:
>
> # Defaults for all machines
> [default]
> user            = pajith
> email           = ajith_p at ligo.caltech.edu
> allocation      = TG-MCA02N014
> thornlist       = manifest/einsteintoolkit.th
>
> [lonestar]
> user            = pajith
> email           = ajith_p at ligo.caltech.edu
> allocation      = TG-MCA02N014
> sourcebasedir   = /work/01489/@USER@
> basedir         = /scratch/01489/@USER@/simulations
> envsetup        = module purge && source /etc/profile.d/tacc_modules.sh &&
> module load TACC && module swap mvapich2 openmpi
>
>
> The extra bits were added to the envsetup (following the advice of Roland
> Haas) because by default the system was loading the mvapich2 module and
> *earlier* my jobs were failing with the following error:
>
> Error: You can only have one MPI module loaded at time.
> You already have mvapich2 loaded.
> To correct the situation, please enter the following command:
>
>       module swap mvapich2 openmpi/1.4.3
>
>
> Please let me know if you need any additional information.
>
> Thanks a lot!
> Ajith
>
> PS: Sorry for spamming you all. I am not sure if this mailing list is the
> appropriate place to bring this issue. I tried to submit a bug report
> at https://trac.einsteintoolkit.org/. But my ticket was rejected as Trac
> suspected it as spam!
>
>
>
>
>
> --
> P. Ajith, LIGO Laboratory and Theoretical Astrophysics
> California Institute of Technology, MS 100-36, Pasadena, CA 91125, USA.
> Tel. +1 626 395 6295 (LIGO), 6162 (TApIR), Fax. +1 626 584 5735
>
> What is your impact?
> http://www.nature.org/initiatives/climatechange/calculator
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
> _______________________________________________
> Users mailing list
> Users at einsteintoolkit.org
> http://lists.einsteintoolkit.org/mailman/listinfo/users
>



-- 
Erik Schnetter <schnetter at cct.lsu.edu>
http://www.perimeterinstitute.ca/personal/eschnetter/