[Pw_forum] Want test crash

Andrea Ferretti ferretti.andrea at unimore.it
Mon Jul 2 11:57:28 CEST 2007


Dear Mashao,

it seems the espresso datafile cannot be read by WanT, some reasons might 
be:
-  the file ~test01/SCRATCH/silicon.save/datafile.xml is 
   corrupted or empty
-  some portability issue due to the compiler/architecture combination 
   occurred (even if I think this is somehow unlike since espresso is 
   working)

could you please post the make.sys file (from the WanT home directory) as 
well as a description of your machine ??

by the way:

1) did you run 
  ./run.sh bands
  ?
  this would not run the whole test but only a specific 
  calculation (bands interpolation using wannier functions);
  instead, use 
  ./run.sh dft       (to run espresso calculations)
  ./run.sh want      (to compute the remaining part of the test)

2) WanT-2.1-RC1 is a "release compliant" version of the code (i.e. a 
   pre-release): consider to update it to the stable version WanT-2.1
   (to be released in few days) 


hope it helps

andrea


> Dear Sir:
>           I am a newbie for pwscf and want. I have compiled the parallel  version 
>           of PWSCF(espresso-3.2) and WANT(want-2.1-RC1). I excute the 
>           tests of PWSCF and everything is OK! However, when I excute 
>           the tests of WANT,  the error appeared .
>           when I excute test01 for run.sh bands, the bands.x crashed and 
>           the error in bands.out as following. 
>   
> DFT-data fmt automaticaly detected: qexml
> 
>  *** from PE    :     0
>  *** in routine : lattice_read_ext
>  *** error msg. : reading lattice
>  *** error code :     1
>  *** aborting ***
> 


--
Andrea Ferretti
National Research Center S3, CNR-INFM  ( http://s3.infm.it )
Dip. Fisica, Univ. di Modena e Reggio E. (Italy)
Tel: +39 059 2055301;  Fax: +39 059 374794;  Skype: andrea_ferretti
URL: http://www.nanoscience.unimo.it

Please, if possible, don't send me MS Word or PowerPoint attachments
Why? See:  http://www.gnu.org/philosophy/no-word-attachments.html



More information about the Pw_forum mailing list