[Wien] Problems when running in a cluster

[Martin Gelfand]

Dear Daniel,
  Have you been able to run any other Wien2K calculations on
the linux cluster in question?  (Or, even if you have in the past,
can you try a test case, like the TiC example, now?)
I.e., are you sure the problem is specific to this calculation,
or could it be a configuration problem that affects all calculations?
  Something else that can sometimes help is to run the specific
calculation in the foreground---just 'x lapw0' in an appropriately
initialized shell.  Since the calculation will (presumably) die
quickly you won't have to wait around long to see a hint of what's wrong.

Martin Gelfand
Dept of Physics
Colorado State University

On Mon, 5 Apr 2004, Daniel [iso-8859-1] Fernández Hevia wrote:

> Dear WIEN users,
>
> I am trying to run a calculation with a 35 atoms AlN supercell (one Al atom
> is missing). When I run the calculation in a laptop, everything goes well.
> The initialization runs fine, no warnings, no problems. I used a small
> value for Rmt*Kmax in order to speed up the calculation, as my intention
> running in the laptop was just to check that everything is ok. Then, I try
> to run the same calculation (exactly the same: I just copy the input file)
> in a remote LINUX cluster (not a parallel calculation). All the
> initialization runs perfectly: all the files are created without any
> problem and everything goes well. But then, at the beginning of the scf
> cycle, the calculation crashes. No error messages, no indication
> whatsoever. I only find a lapw0.error file with a brief message “Error in
> lapw0”. That’s all. No dayfile, no scf file, nothing. The command I send to
> the remote machine in order to run the scf cycle is "run_lapw -cc 0.0001",
> exactly the same as in the laptop where everything goes well. Do you have
> any idea about the problem? I do not even know which files can I send: I’m
> sending the structure file, as well as an abridged version of the output
> dstart. I do not send the output of lstart because is quite a big file,
> but, of course, If you need to look into any other file, just tell me.
>
> Thanks in advance for your always fruitful help.
>
> Daniel
>
> ----------------------------------------------------------
> Daniel Fernández Hevia
> School of Physics
> The University of Sydney
> Sydney 2006, Australia
> Phone: +61 2 9036 5301
> email:
> <https://correo.etsit.upm.es/twig//twig/index.php3?&s[mailbox]=mail%2FEnviados1&s[mainGroup]=%2A&s[mailtree]=0%7C&s[mailGroup]=%2A&s[sortby]=date&s[sortbyway]=1&s[delete-return]=msgview&c[f]=mail&c[a]=compose&form[to]=dhevia@physics.usyd.edu.au>dhevia at physics.usyd.edu.au
> ----------------------------------------------------------
>
>
>
>
>