[molpro-user] CIPT2 segmentation violation

[Hans-Joachim Werner]

Dear Jacky,
I forgot to write that the second calculation (which crashed) equals an MRCI (if there are no correlated closed-shell orbitals, i.e. close=core, then cipt2 is the same as mrci). For mrci the crash should not occur.
Best wishes
Joachim
---
Prof. Dr. Hans-Joachim Werner
Institut für Theoretische Chemie
Universität Stuttgart
Pfaffenwaldring 55
70569 Stuttgart, Germany
e-mail: werner at theochem.uni-stuttgart.de




> Am 08.03.2018 um 15:33 schrieb Jacky LIEVIN <jlievin at ulb.ac.be>:
> 
> Dear all,
> 
> I get an unexpected segmentation violation error in the first CIPT2 calculations of the input below.
> The calculation converges properly and the ci vectors are printed, but the job crashes before printing the energy results.
> Surprisingly, the second CIPT2 calculation terminates normally. The only difference is the number of core orbitals  (10 and 12 in the first and second calculation, respectively).
> 
> Please note that the error message is different when running a sequential or mpi calculation:
> sequential:
> Received signal 11 Segmentation violation
> 0:fehler:Received an Error in Communication
> application called MPI_Abort(comm=0x84000000, 1) - process 0
> 
> mpi:
> forrtl: severe (174): SIGSEGV, segmentation fault occurred
> Image              PC                Routine            Line        Source
> molpro_initial.ex  0000000004A1C5C1  Unknown               Unknown  Unknown
> molpro_initial.ex  0000000004A1AD17  Unknown               Unknown  Unknown
> molpro_initial.ex  00000000049A7E94  Unknown               Unknown  Unknown
> 
> Many thanks for any advice or comments
> 
> Jacky
> 
> ***,cipt2                                         
> memory,500,M                                      
> basis=VDZ                                        
>         geometry= {7
>   1  1  0.0053
>  C          0.0000000000        0.0000000000       -3.2562129132
>  C          0.0000000000        0.0000000000       -2.0296033386
>  H          0.0000000000        0.0000000000       -4.3284842729
>  C          0.0000000000        0.0000000000       -0.6586947557
>  C          0.0000000000        0.0000000000        0.6014520701
>  C          0.0000000000        0.0000000000        1.9862803722
>  N          0.0000000000        0.0000000000        3.1899819218
>                   }
>  hf
> {casscf                                           
> closed,12,0;occ,13,4,4                              
> wf,37,2,1;state,1                                 
> wf,37,3,1;state,1                                 
> canonical                                       
> expec2,lzz}                                       
> {cipt2;                
> core,10;closed,12;occ,13,4,4                       
> wf,37,2,1;state,1}
> {cipt2;                
> core,12;closed,12;occ,13,4,4                       
> wf,37,2,1;state,1}.
> 
> _____________________________________ 
> Prof. Jacky Liévin
> Service de Chimie Quantique et Photophysique
> Université Libre de Bruxelles, CPi 160/09
> 50 av F.D. Roosevelt
> B-1050 Bruxelles
> Belgium
> _____________________________________
> 
> _______________________________________________
> Molpro-user mailing list
> Molpro-user at molpro.net
> http://www.molpro.net/mailman/listinfo/molpro-user