Molcas Forum

Support and discussions for Molcas and OpenMolcas users and developers

You are not logged in.

Announcement

Welcome to the Molcas forum. You can choose an avatar and change the default style by going to "Profile" → "Personality" or "Display".

Pages: 1

#1 2021-12-21 11:34:01

Freeman
Member
Registered: 2021-12-15
Posts: 3

Unknown error in CASPT2

Hi
I am new to molcas and am testing molcas's CASPT2. But an error appeared.
My input is:

&gateway
  title
    123

  coord
    molecule.xyz

  basis
    cc-pvdz
    ricd

&seward

&scf
  GSSRunfile
  prorbitals
    0
&rasscf
  inactive
    43
  ras2
    6

&caspt2
  shift
    0.2

HF and CASSCF tasks completed normally but CASPT2 crashed. The last few lines of the output is:

--------------------------------------------------------------------------------
 Estimated memory requirements:
  POLY3 :                   262414
  RHS:                     2174404
  SIGMA :                  6345656
  PRPCTL:                        0
 Available workspace:    230312539
 
++ CASPT2 computation for group   1
 
********************************************************************************
 Multi-State initialization phase begins for group   1
--------------------------------------------------------------------------------
********************************************************************************
 Compute H0 matrices for state    1
--------------------------------------------------------------------------------
--------------------------------------------------------------------------------
 H0 matrices have been computed.
 
********************************************************************************
  CASPT2 EQUATION SOLUTION
--------------------------------------------------------------------------------
--- Stop Module: caspt2 at Tue Dec 21 18:17:32 2021 /rc=2 ---
*** files: xmldump
    saved to directory /home/yzhangnn/g16job/test/cas
--- Module caspt2 spent 6 seconds ---

.########################.
.# Non-zero return code #.
.########################.

    Timing: Wall=158.76 User=0.32 System=0.60

Because the server is using SLURM, there is also a slurm output file:

Abort(805902850) on node 2 (rank 2 in comm 0): Fatal error in PMPI_Alltoallv: Invalid count, error stack:
PMPI_Alltoallv(368): MPI_Alltoallv(sbuf=0x3d512c0, scnts=0x7ffe5ca1b840, sdispls=0x7ffe5ca1b800, dtype=0x4c000829, rbuf=0x3d69ce0, rcnts=0x7ffe5ca1b7c0, rdispls=0x7ffe5ca1b780, datatype=dtype=0x4c000829, comm=MPI_COMM_WORLD) failed
PMPI_Alltoallv(306): Negative count, value is -1135233048
Abort(269031938) on node 4 (rank 4 in comm 0): Fatal error in PMPI_Alltoallv: Invalid count, error stack:
PMPI_Alltoallv(368): MPI_Alltoallv(sbuf=0x34aa8a0, scnts=0x7ffc2d9ba940, sdispls=0x7ffc2d9ba900, dtype=0x4c000829, rbuf=0x34c32c0, rcnts=0x7ffc2d9ba8c0, rdispls=0x7ffc2d9ba880, datatype=dtype=0x4c000829, comm=MPI_COMM_WORLD) failed
PMPI_Alltoallv(306): Negative count, value is -1907353624
Abort(940120578) on node 6 (rank 6 in comm 0): Fatal error in PMPI_Alltoallv: Invalid count, error stack:
PMPI_Alltoallv(368): MPI_Alltoallv(sbuf=0x2f7e340, scnts=0x7ffc37b159c0, sdispls=0x7ffc37b15980, dtype=0x4c000829, rbuf=0x2f96d60, rcnts=0x7ffc37b15940, rdispls=0x7ffc37b15900, datatype=dtype=0x4c000829, comm=MPI_COMM_WORLD) failed
PMPI_Alltoallv(309): Negative count, value is -769271880
Abort(805902850) on node 9 (rank 9 in comm 0): Fatal error in PMPI_Alltoallv: Invalid count, error stack:
PMPI_Alltoallv(368): MPI_Alltoallv(sbuf=0x37df9c0, scnts=0x7fff87008d40, sdispls=0x7fff87008d00, dtype=0x4c000829, rbuf=0x37f83e0, rcnts=0x7fff87008cc0, rdispls=0x7fff87008c80, datatype=dtype=0x4c000829, comm=MPI_COMM_WORLD) failed
PMPI_Alltoallv(309): Negative count, value is -167553096
Abort(940120578) on node 13 (rank 13 in comm 0): Fatal error in PMPI_Alltoallv: Invalid count, error stack:
PMPI_Alltoallv(368): MPI_Alltoallv(sbuf=0x2ea3980, scnts=0x7ffdc2eb91c0, sdispls=0x7ffdc2eb9180, dtype=0x4c000829, rbuf=0x2ebc3a0, rcnts=0x7ffdc2eb9140, rdispls=0x7ffdc2eb9100, datatype=dtype=0x4c000829, comm=MPI_COMM_WORLD) failed
PMPI_Alltoallv(306): Negative count, value is -1184393240
Abort(940120578) on node 14 (rank 14 in comm 0): Fatal error in PMPI_Alltoallv: Invalid count, error stack:
PMPI_Alltoallv(368): MPI_Alltoallv(sbuf=0x413b720, scnts=0x7fff5e1f8540, sdispls=0x7fff5e1f8500, dtype=0x4c000829, rbuf=0x4154140, rcnts=0x7fff5e1f84c0, rdispls=0x7fff5e1f8480, datatype=dtype=0x4c000829, comm=MPI_COMM_WORLD) failed
PMPI_Alltoallv(306): Negative count, value is -131336216

I compiled molcas with GlobalArray_5.7.2, hdf5_1.10.5, Intel_2021.4, and the configuration is:

CC=mpiicc FC=mpiifort CXX=mpiicpc cmake -DMPI_Fortran_COMPILER=mpiifort -DMPI_C_COMPILER=mpiicc -DMPI_CXX_COMPILER=mpiicpc -DMPI=ON -DGA=ON -DLINALG=MKL -DCMAKE_INSTALL_PREFIX=[my path to molcas] ..

Any help will be appreciated.

Offline

#2 2021-12-22 13:44:49

Ignacio
Administrator
From: Uppsala
Registered: 2015-11-03
Posts: 1,202

Re: Unknown error in CASPT2

That looks like an integer size problem. Make sure you compile GlobalArrays with 64-bit integers and with a 64-bit integer interface to blas/lapack.

Offline

#3 2021-12-22 18:17:22

Freeman
Member
Registered: 2021-12-15
Posts: 3

Re: Unknown error in CASPT2

Ignacio wrote:

That looks like an integer size problem. Make sure you compile GlobalArrays with 64-bit integers and with a 64-bit integer interface to blas/lapack.

Thank you for your attention. My configuration of GlobalArray is:

./configure MPICXX=mpiicpc MPIF77=mpiifort MPICC=mpiicc --prefix=[dir to ga-build] --enable-i8 --with-openib \
--with-blas8="-L$MKLROOT/lib/intel64 -lmkl_intel_ilp64 -lmkl_intel_thread -lmkl_core -liomp5 -lpthread -lm" \
--with-scalapack8="-L$MKLROOT/lib/intel64 -lmkl_scalapack_ilp64 -lmkl_intel_ilp64 -lmkl_intel_thread -lmkl_core -lmkl_blacs_intelmpi_ilp64 -liomp5 -lpthread -lm"

It seems that I have already compiled ga with 64-bit. Would you please show me how to fix it?

Last edited by Freeman (2021-12-22 18:21:24)

Offline

#4 2021-12-24 12:58:35

cnu-hjp
Member
Registered: 2021-11-29
Posts: 1

Re: Unknown error in CASPT2

Hello, I also encountered the same error when calculating CASPT2 - "non-zero return code". Both SCF and CASSCF also end normally, and the last few lines of CASPT2 output are similar to your output file. I would like to ask you, have you solved this problem? Any help would be greatly appreciated. The output file is shown below.

Constructing G3/F3
   memory avail:     18.387536416 GB
   memory used:       0.000810432 GB
  Sym:   1, #Tasks:     64
--------------------------------------------------------------------------------
 H0 matrices have been computed.
 
********************************************************************************
  CASPT2 EQUATION SOLUTION
--------------------------------------------------------------------------------

 Computing the S/B matrices
 --------------------------
 
  Construct S matrices
 
  Construct B matrices
 
  Find transformation matrices to eigenbasis of block-diagonal part of H0.
  Eliminate linear dependency. Thresholds for:
   Initial squared norm  :  0.1000E-09
   Eigenvalue of scaled S:  0.1000E-07
 
  Condition numbers are computed after diagonal scaling and after removal of
  linear dependency. Resulting sizes, condition numbers, and times:
    CASE(SYM)       NASUP       NISUP      NINDEP     COND NR  CPU (s)
   VJTU  (1)          512          46         512   0.38E+08        0
   VJTIP (1)           36        1081          36   0.31E+06        0
   VJTIM (1)           28        1035          28   0.82E+03        0
   ATVX  (1)          512         668         512   0.29E+06        0
   AIVX  (1)          128       30728         128   0.13E+05        0
   VJAIP (1)            8      722108           8   0.19E+03        0
   VJAIM (1)            8      691380           8   0.19E+03        0
   BVATP (1)           36      223446          36   0.10E+04        0
   BVATM (1)           28      222778          28    65.            0
   BJATP (1)            8    10278516           8    11.            0
   BJATM (1)            8    10247788           8    11.            0
 
  Total nr of CASPT2 parameters:
   Before reduction:   666287180
   After  reduction:   666287180

 Computing the right-hand side (RHS) elements
 --------------------------------------------
  Using RHSALL2+ADDRHS algorithm
 
 The contributions to the second order correlation energy in atomic units.
-----------------------------------------------------------------------------------------------------------------------------
  IT.      VJTU        VJTI        ATVX        AIVX        VJAI        BVAT        BJAT        BJAI        TOTAL       RNORM  
-----------------------------------------------------------------------------------------------------------------------------
   1    -0.001972   -0.003980   -0.035346   -0.133896   -0.109366   -0.064981   -0.536719   -2.770862   -3.657123    0.006039
   2    -0.002012   -0.004036   -0.035651   -0.134418   -0.109684   -0.065121   -0.537092   -2.771326   -3.659340    0.000751
   3    -0.002011   -0.004037   -0.035649   -0.134433   -0.109680   -0.065123   -0.537095   -2.771307   -3.659335    0.000108
   4    -0.002012   -0.004037   -0.035647   -0.134425   -0.109676   -0.065121   -0.537088   -2.771298   -3.659303    0.000014
   5    -0.002012   -0.004037   -0.035647   -0.134425   -0.109676   -0.065121   -0.537088   -2.771298   -3.659304    0.000002
-----------------------------------------------------------------------------------------------------------------------------
 
  FINAL CASPT2 RESULT:
--- Stop Module: caspt2 at Fri Dec 24 17:58:41 2021 /rc=-1 ---
*** files: xmldump
    saved to directory /media/HD2T/jingping/dpndcn/excited-sp/t1
--- Module caspt2 spent 5 hours 0 minutes 44 seconds ---

.########################.
.# Non-zero return code #.
.########################.

Offline

#5 2021-12-26 17:09:12

Freeman
Member
Registered: 2021-12-15
Posts: 3

Re: Unknown error in CASPT2

cnu-hjp wrote:

Hello, I also encountered the same error when calculating CASPT2 - "non-zero return code". Both SCF and CASSCF also end normally, and the last few lines of CASPT2 output are similar to your output file. I would like to ask you, have you solved this problem? Any help would be greatly appreciated. The output file is shown below.

Constructing G3/F3
   memory avail:     18.387536416 GB
   memory used:       0.000810432 GB
  Sym:   1, #Tasks:     64
--------------------------------------------------------------------------------
 H0 matrices have been computed.
 
********************************************************************************
  CASPT2 EQUATION SOLUTION
--------------------------------------------------------------------------------

 Computing the S/B matrices
 --------------------------
 
  Construct S matrices
 
  Construct B matrices
 
  Find transformation matrices to eigenbasis of block-diagonal part of H0.
  Eliminate linear dependency. Thresholds for:
   Initial squared norm  :  0.1000E-09
   Eigenvalue of scaled S:  0.1000E-07
 
  Condition numbers are computed after diagonal scaling and after removal of
  linear dependency. Resulting sizes, condition numbers, and times:
    CASE(SYM)       NASUP       NISUP      NINDEP     COND NR  CPU (s)
   VJTU  (1)          512          46         512   0.38E+08        0
   VJTIP (1)           36        1081          36   0.31E+06        0
   VJTIM (1)           28        1035          28   0.82E+03        0
   ATVX  (1)          512         668         512   0.29E+06        0
   AIVX  (1)          128       30728         128   0.13E+05        0
   VJAIP (1)            8      722108           8   0.19E+03        0
   VJAIM (1)            8      691380           8   0.19E+03        0
   BVATP (1)           36      223446          36   0.10E+04        0
   BVATM (1)           28      222778          28    65.            0
   BJATP (1)            8    10278516           8    11.            0
   BJATM (1)            8    10247788           8    11.            0
 
  Total nr of CASPT2 parameters:
   Before reduction:   666287180
   After  reduction:   666287180

 Computing the right-hand side (RHS) elements
 --------------------------------------------
  Using RHSALL2+ADDRHS algorithm
 
 The contributions to the second order correlation energy in atomic units.
-----------------------------------------------------------------------------------------------------------------------------
  IT.      VJTU        VJTI        ATVX        AIVX        VJAI        BVAT        BJAT        BJAI        TOTAL       RNORM  
-----------------------------------------------------------------------------------------------------------------------------
   1    -0.001972   -0.003980   -0.035346   -0.133896   -0.109366   -0.064981   -0.536719   -2.770862   -3.657123    0.006039
   2    -0.002012   -0.004036   -0.035651   -0.134418   -0.109684   -0.065121   -0.537092   -2.771326   -3.659340    0.000751
   3    -0.002011   -0.004037   -0.035649   -0.134433   -0.109680   -0.065123   -0.537095   -2.771307   -3.659335    0.000108
   4    -0.002012   -0.004037   -0.035647   -0.134425   -0.109676   -0.065121   -0.537088   -2.771298   -3.659303    0.000014
   5    -0.002012   -0.004037   -0.035647   -0.134425   -0.109676   -0.065121   -0.537088   -2.771298   -3.659304    0.000002
-----------------------------------------------------------------------------------------------------------------------------
 
  FINAL CASPT2 RESULT:
--- Stop Module: caspt2 at Fri Dec 24 17:58:41 2021 /rc=-1 ---
*** files: xmldump
    saved to directory /media/HD2T/jingping/dpndcn/excited-sp/t1
--- Module caspt2 spent 5 hours 0 minutes 44 seconds ---

.########################.
.# Non-zero return code #.
.########################.

I have not solved the problem yet, and I do not think the problem with your caspt2 task is the same as that with mine. There will always be a "non-zero return code" regardless of where the problem comes from. I suggest you open a new issue for your question.

Offline

Pages: 1

Links: OpenMolcas | Molcas | Manual | Facebook

Board footer

Powered by FluxBB 1.5.11

Last refresh: Today 08:37:10