Molcas Forum

tuancpacheco

Member

Registered: 2026-04-11

Posts: 6

UNEXPECTED INTERNAL ERROR IN THE MBPT2 MODULE

Hello there!

Before I describe the error I came through, I will give detailed information on the environment as well as the compilation settings of my OpenMolcas v. 26.02 installation. I am using Ubuntu 20.04 Focal Fossa. HDF5 normally installed, Global Arrays 5.9.2 was installed in /usr/local/ with no custom settings, OpenBLAS 0.3.8 installed in /opt with INTERFACE64=1 setting defined when I ran make. OpenMPI 4.0.3 version is installed. I have compiled OpenMolcas in the 'build' directory with the following settings:

export GAROOT=/usr/local/

cmake .. -DCMAKE_INSTALL_PREFIX=$path_to_installation -DLINALG=OpenBLAS -DOPENBLASROOT=/opt/OpenBLAS -DMPI=ON -DGA=ON -OPENMP=ON

Now, calculations of RASSCF, RASSI as well as SLAPAF and ALASKA modules all went well. However, the MBPT2 always falls into the same internal error message.
Here is the input for MP2 calculation

>> export MOLCAS_NPROCS=3
>> export MOLCAS_MEM=8000
>> export MOLCAS_WORKDIR=/scratch

&GATEWAY

        coords=v_et2.xyz
        basis=def2-tzvp
        group=nosym
        RICD

&SEWARD


&MBPT2

        PRPT

Seward module ends up with ALL_IS_WELL return code and then &MBPT2 fail with internal error messages both in the xmldump file

[ process      0]: xquit (rc =    128): _INTERNAL_ERROR_
[ process      2]: xquit (rc =    128): _INTERNAL_ERROR_
[ process      1]: xquit (rc =    128): _INTERNAL_ERROR_
#0  0x7fd28006ed11 in ???
#1  0x556ea37a1a5b in ???
#2  0x556ea36efc03 in ???
#3  0x556ea36e8d25 in ???
#4  0x556ea367af3d in ???
#5  0x556ea367ad78 in ???
#6  0x7fd27ec1d082 in __libc_start_main
	at ../csu/libc-start.c:308
#7  0x556ea367adbd in ???
#8  0xffffffffffffffff in ???
--------------------------------------------------------------------------
MPI_ABORT was invoked on rank 1 in communicator MPI_COMM_WORLD
with errorcode 128.

NOTE: invoking MPI_ABORT causes Open MPI to kill all MPI processes.
You may or may not see output from other processes, depending on
exactly when Open MPI kills them.
--------------------------------------------------------------------------

and output file

()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()

                                               &MBPT2

             launched 3 MPI processes, running in PARALLEL mode (work-sharing enabled)
                       available to each process: 4.5 GB of memory, 1 thread
                                         master pid: 17329
()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()

 ###############################################################################
 ###############################################################################
 ###                                                                         ###
 ###                                                                         ###
 ###    Location: get_cArray                                                 ###
 ###                                                                         ###
 ###                                                                         ###
 ###    Data not defined: Relax Method                                       ###
 ###                                                                         ###
 ###                                                                         ###
 ###############################################################################
 ###############################################################################
--- Stop Module: mbpt2 at Tue Jun  2 13:13:08 2026 /rc=_RC_INTERNAL_ERROR_ ---
*** files: xmldump
    saved to directory /home/tuan/COISAS_DO_TUAN/openmolcas/unbv/cas_et2/mp2

.########################################################.
.# Some internal inconsistency of the code was detected #.
.########################################################.

Any ideas on what is going on here?

Ignacio

Administrator

From: Uppsala

Registered: 2015-11-03

Posts: 1,204

Re: UNEXPECTED INTERNAL ERROR IN THE MBPT2 MODULE

The error message is not very helpful, but I think what you're missing is running SCF before MBPT2. Try with:

>> export MOLCAS_NPROCS=3
>> export MOLCAS_MEM=8000
>> export MOLCAS_WORKDIR=/scratch

&GATEWAY

        coords=v_et2.xyz
        basis=def2-tzvp
        group=nosym
        RICD

&SEWARD

&SCF

&MBPT2

        PRPT

(you may need to add keywords to SCF if your system is charged/radical)

tuancpacheco

Member

Registered: 2026-04-11

Posts: 6

Re: UNEXPECTED INTERNAL ERROR IN THE MBPT2 MODULE

It's a bit awkward, but yes... I forgot to run SCF optimization before running MP2 correction+diagonalization.

Thanks a lot!