SAPT0-D3MBJ (error writing to file)?

SAPT
1

Hello,

I am trying to do SAPT0-D3MBJ following this tutorial. It is a dimer with 80 total number of atoms. 120GB memory were allocated for this calculation. I am using Psi4/1.5.0 compiled by PACE on Phoenix. I believe that the dftd3 package was compiled altogether.

I would like to ask for some help to understand what the following error message is about:

  => Loading Basis Set <=

    Name: (JUN-CC-PVDZ AUX)
    Role: RIFIT
    Keyword: DF_BASIS_SAPT
    atoms 1-4, 41-44   entry SI         line   376 file /usr/local/pace-apps/manual/packages/psi4/1.5.0/Core/share/psi4/basis/jun-cc-pvdz-ri.gbs 
    atoms 5-8, 45-48   entry O          line   188 file /usr/local/pace-apps/manual/packages/psi4/1.5.0/Core/share/psi4/basis/jun-cc-pvdz-ri.gbs 
    atoms 9-16, 49-56  entry C          line   100 file /usr/local/pace-apps/manual/packages/psi4/1.5.0/Core/share/psi4/basis/jun-cc-pvdz-ri.gbs 
    atoms 17-40, 57-80 entry H          line    30 file /usr/local/pace-apps/manual/packages/psi4/1.5.0/Core/share/psi4/basis/jun-cc-pvdz-ri.gbs 


  //>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>//
  //            SAPT0-D3MBJ            //
  //<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<//


*** tstart() called on atl1-1-02-010-24-r.pace.gatech.edu
*** at Mon May  9 15:55:36 2022

        SAPT0  
    Ed Hohenstein
     6 June 2009

      Orbital Information
  --------------------------
    NSO        =       848
    NMO        =       848
    NRI        =      2912
    NOCC A     =        80
    NOCC B     =        80
    FOCC A     =         0
    FOCC B     =         0
    NVIR A     =       768
    NVIR B     =       768


Traceback (most recent call last):
  File "/usr/local/pace-apps/manual/packages/psi4/1.5.0/Core/bin/psi4", line 333, in <module>
    exec(content)
  File "<string>", line 111, in <module>
  File "/usr/local/pace-apps/manual/packages/psi4/1.5.0/Core/lib//python3.8/site-packages/psi4/driver/driver.py", line 599, in energy
    wfn = procedures['energy'][lowername](lowername, molecule=molecule, **kwargs)
  File "/usr/local/pace-apps/manual/packages/psi4/1.5.0/Core/lib//python3.8/site-packages/psi4/driver/procrouting/proc.py", line 4374, in run_sapt
    e_sapt = core.sapt(dimer_wfn, monomerA_wfn, monomerB_wfn)

RuntimeError: 
Fatal Error: PSIO_ERROR: 12 (error writing to file)

Error occurred in file: /scratch/psilocaluser/conda-builds/psi4-multiout_1638121835567/work/psi4/src/psi4/libpsio/error.cc on line: 133
The most recent 5 function calls were:

psi::PSIO::rw(unsigned long, char*, psi::psio_address, unsigned long, int)



Printing out the relevant lines from the Psithon --> Python processed input file:
    ""","D4_dimer")
    core.IO.set_default_namespace("D4_dimer")
    core.set_memory_bytes(120000000000)
    
    core.set_global_option("BASIS", "jun-cc-pvdz")
--> energy('sapt0-d3mbj')

!----------------------------------------------------------------------------------!
!                                                                                  !
! Fatal Error: PSIO_ERROR: 12 (error writing to file)                              !
! Error occurred in file: /scratch/psilocaluser/conda-                             !
!     builds/psi4-multiout_1638121835567/work/psi4/src/psi4/libpsio/error.cc on    !
!     line: 133                                                                    !
! The most recent 5 function calls were:                                           !
! psi::PSIO::rw(unsigned long, char*, psi::psio_address, unsigned long, int)       !
!                                                                                  !
!----------------------------------------------------------------------------------!

    Psi4 stopped on: Monday, 09 May 2022 03:58PM
    Psi4 wall time for execution: 0:06:31.68

*** Psi4 encountered an error. Buy a developer more coffee!
*** Resources and help at github.com/psi4/psi4.

Thank you,
JY

2

Thanks for the report. Would you be able to provide the full output file? That particular error usually means that your machine doesn’t have enough scratch space to perform the calculation. Did you configure the scratch directory as described here? Installation and Runtime Configuration

It sounds like you’re using Georgia Tech’s PACE HPC resources. In that case, this link might also be useful:
https://docs.pace.gatech.edu/storage/localTmp/

3

Hello,

I have the complete output file, and also the PBS job file (that previously worked for a DF-MP2/cc-PVTZ geometry optimization of a 60-atom molecule) uploaded to a folder on my Box drive. Please let me know if you are not able to access it.

4

Thanks, I can access your output file. It does appear that you haven’t set a reasonable scratch directory for psi4 to use. Line 143 of your output file says:

Scratch directory: /tmp/

The location /tmp/ is psi4’s default scratch location, and is too small for serious calculations. I’m surprised that your DF-MP2 calculation was able to run. You’ll want to change the scratch directory, which the two hyperlinks in my previous message should help you with. Probably, you’ll include something like:

 export PSI_SCRATCH=$TMPDIR

in your PBS script.

Best,
Zach

5

Hi Zach,

Thanks for this - it worked perfect.

Best wishes,
JY

6

This topic was automatically closed 60 days after the last reply. New replies are no longer allowed.