*** error in GROUP_CONSTRUCTOR_REL*** impossible group size <= 0.

16 views
Skip to first unread message

Giorgio Visentin

unread,
May 11, 2023, 8:15:35 AM5/11/23
to dirac...@googlegroups.com
Dear SIrs,
I am trying to run a KRCI calculation on NH using the LUCIAREL module. However, the calculation crashes upon reporting the following error:

*** error in GROUP_CONSTRUCTOR_REL*** impossible group size <= 0.

The input and xyz files read as follows:

*DIRAC
.WAVE FUNCTION
.ANALYZE
**GENERAL
*PARALLEL
.NTASK
2
**INTEGRALS
*READIN
.UNCONTRACT
**HAMILTONIAN
.X2C
**MOLECULE
*BASIS
.DEFAULT
dyall.cv3z
#*SYMMETRY
#.NOSYM
**WAVE FUNCTIONS
.SCF
.KRCI
*KRCI
.CI PROGRAM
LUCIAREL
.GAS SHELLS
5
2 2 / 1
4 4 / 1
4 6 / 1
4 8 / 2
8 8 / 40
.CIROOTS
 2 4
.CIROOTS
 4 3
.MAX CI
 600
.MXCIVE
 30
.NOOCCN
.ANALYZ
.TRDM
.DIPMOM
**ANALYZE
.MULPOP
*MULPOP
.VECPOP
1..oo
*END OF INPUT

2
xyz
N   0.0000   0.0000   0.0000
H   0.0000   0.0000   0.8

This is instead the outcome I get from the last part of the output file:

 (RSETCI) Number of determinants:       294982 (dimension       294982)

 CPU (Wall) time for generation of determinants:  1.86163100s( 1.99824405s)

 (RSETWOP) Orbital classes:
     Inactive orbitals   :
     Active orbitals     :    1..  45
     Secondary orbitals  :   46..  80

 (ROPTST)  Start orbitals read from CHECKPOINT file

 -> Gram-Schmidt orthonormalizing MO coefficients.

  Vector containing mj-values saved as MJVEC on KRMCSCF

 (RTRACTL1) Starting 4-index integral transformation.
 - Integral class 1 : (LL|??)
   - Beginning task   1 of   7 after      0. seconds and      0. CPU-seconds
   - Beginning task   2 of   7 after      0. seconds and      0. CPU-seconds
   - Beginning task   3 of   7 after      0. seconds and      0. CPU-seconds
   - Beginning task   4 of   7 after      0. seconds and      0. CPU-seconds
   - Beginning task   5 of   7 after      1. seconds and      0. CPU-seconds
   - Beginning task   6 of   7 after      1. seconds and      1. CPU-seconds
   - Beginning task   7 of   7 after      1. seconds and      1. CPU-seconds

 - Starting symmetrization after           0.99 seconds
 - Finished symmetrization after           0.99 seconds


------ Timing report (in CPU seconds) of module  integral transformation

 Time in  Initializing MS4IND file            0.028 seconds
 Time in  Computing+transform. integral       0.699 seconds
 Time in  Symmetrizing MO integrals           0.000 seconds

------ End of timing report ------


 (RTRACTL1) Total CPU (wall) time used :   00:00:01 (   00:00:01)
            Transformation ended Thu May 11 13:57:15 2023
  //  LUCIAREL called
  //  CIRUN = KR-CI
  //  Running large-scale CI calculation

* REAFCK: Fock matrix read from file /home/xi58teb/DIRAC_scratch_directory/xi58teb/DIRAC_NHTtest_
* Heading :DIRAC: No title specified !!!                     Thu May 11 13:57:11 2023
 *** error in GROUP_CONSTRUCTOR_REL*** impossible group size:
           -4294967295

Notice that I can run this same input file using GASCIP. However I would prefer to use LUCIAREL.
How may I solve this issue?
I thank you for your time and attention.
Kind regards,
Giorgio

P.S. I managed somehow to avoid, for now, the previous error related to pam after exporting its parent directory to $PATH.

Reply all
Reply to author
Forward
0 new messages