Running CDFT Tutorial Calculation on Cluster

Nico Holmberg holmbe... at gmail.com
Sat Jul 21 08:39:50 UTC 2018


Hi,

My bad. I made some changes to the CDFT input structure between CP2K 
version 5.1 and 6.1 and I forgot to update the tutorial. You probably have 
different versions locally and on your cluster. All CDFT related stuff from 
the OUTER_SCF section was moved to a CDFT_OPT subsection 
<https://manual.cp2k.org/trunk/CP2K_INPUT/FORCE_EVAL/DFT/QS/CDFT/OUTER_SCF/CDFT_OPT.html>. 
If move the relevant keywords in the file becke_qs.inc, the calculation 
should run without issue.

I'll update the tutorial on Monday and upload new files for CP2K 6.1. 


BR,

Nico

perjantai 20. heinäkuuta 2018 18.24.22 UTC+3 Brian Day kirjoitti:
>
> Hi all,
>
> I am trying to run the following CDFT tutorial (
> https://www.cp2k.org/howto:cdft) for water dimers on my university's 
> cluster and am getting an error I do not understand. For clarity, the issue 
> is not with submitting the file, but with CP2K itself. I have also 
> successfully completed the tutorial locally, which is another reason I am 
> confused by the error.
>
> My process for running this on the cluster was as follows:
>      -Duplicate the energy.bash script and modify it to generate a series 
> of input files (one for each cp2k run)
>      -Modify the slurm file to submit each input file individually (i.e. 
> Rather than running them with a single bash script as when running locally, 
> I wanted to run each input with its own submission to the cluster)
>
> I have attached the relevant files.
>
> I am able to run the first calculation (energy_run_standard.inp) 
> successfully, but when I try to run the second one 
> (energy_run_cte_noadj.inp) I get the following error:
>
>  Possible matches for unknown keyword
>
>
>  MAX_LS
>
>
>    keyword MAX_LS in section 
> %__ROOT__%FORCE_EVAL%DFT%SCF%OUTER_SCF%CDFT_OPT score:  44
>
>    keyword MAX_LS in section 
> %__ROOT__%FORCE_EVAL%DFT%QS%CDFT%OUTER_SCF%CDFT_OPT score:  44
>
>    keyword MAX_LS in section 
> %__ROOT__%FORCE_EVAL%DFT%XAS%SCF%OUTER_SCF%CDFT_OPT score:  44
>
>    keyword MAX_SCF in section %__ROOT__%FORCE_EVAL%DFT%SCF%OUTER_SCF 
> score:  13
>
>    keyword MAX_SCF in section %__ROOT__%FORCE_EVAL%DFT%QS%CDFT%OUTER_SCF 
> score:  13
>
>
>  
> *******************************************************************************
>
>  *   ___                                                                 
>       *
>
>  *  /   \                                                                
>       *
>
>  * [ABORT]                                                               
>       *
>
>  *  \___/          found an unknown keyword MAX_LS in section OUTER_SCF  
>       *
>
>  *    |                                                                  
>       *
>
>  *  O/|                                                                  
>       *
>
>  * /| |                                                                  
>       *
>
>  * / \                                               input/input_parsing.F:246 
> *
>
>  
> *******************************************************************************
>
>
>
>  ===== Routine Calling Stack =====
>
>
>             7 section_vals_parse
>
>             6 section_vals_parse
>
>             5 section_vals_parse
>
>             4 section_vals_parse
>
>             3 section_vals_parse
>
>             2 section_vals_parse
>
>             1 read_input
>
> application called MPI_Abort(MPI_COMM_WORLD, 1) - process 0
>
> I checked my input files, and MAX_LS is never called in either of those 
> sections.
>
> Any help in understanding this would be greatly appreciated!
>
> Regards,
>      Brian
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20180721/28ce287f/attachment.htm>


More information about the CP2K-user mailing list