Molcas Forum

kaoru

Member

Registered: 2022-05-19

Posts: 1

I/O error in MS-RASPT2 calculation

Dear OpenMolcas developers,

I am trying to run an core-hole state calculation on an aromatic molecule (C7H6O) with MS-RASPT2/SAPPORO-2012-TZP using OpenMolcas version 21.10.
I want to calculate ~ 60 states for each irrep and SA-RASSCF runs well. However, the MS-RASPT2 did not start accordingly and end up with an I/O error as follows:

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

                                              &CASPT2

                       only a single process is used, running in SERIAL mode
                       available to each process: 20 GB of memory, 9 threads
                                            pid: 214351
()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()
 
 ###############################################################################
 ###############################################################################
 ###                                                                         ###
 ###                                                                         ###
 ###    ERROR: I/O error when reading line.                                  ###
 ###                                                                         ###
 ###                                                                         ###
 ###############################################################################
 ###############################################################################
 Last line read from input: 
 End of input                                                                   
                                                  
--- Stop Module: caspt2 at Thu May 19 13:29:51 2022 /rc=_RC_INPUT_ERROR_ ---
*** files: xmldump
    saved to directory /lustre/save/users/uv5/tropone/xps/220518/10e9o/c1s/q_0/core-hole-states/a1

.####################.
.# User input error #.
.####################.

Increasing the size of memory and disk won't help at all. However, this error disappears when I reduce the number of target states (i.e. 60 to 20)
with keeping the other part of the input file unchanged, and MS-RASPT2 runs well. I suspect it would be due to some bug. The input file is as follows:

&GATEWAY
 Title= sapporo basis test
 coord = /lustre/save/users/uv5/tropone/xps/220518/10e9o/tropone_neutral_dflmp2_def2tzvp.xyz
 BASLIB = /lustre/save/users/uv5/tropone/xps/220426/q_0/basisset/
 basis =TEST
&SEWARD
CHOLesky

&SCF

* rasscf reference
&rasscf
spin = 1
charge = 0
ras2 = 0  5 3 1 
nactel = 10 0 0 
inactive = 14 0 0 9
rasscf = 0 0 

* rasscf with orbital lotation: Include C(1s) to ras1 
&rasscf
ras1 = 4 0 0 3
ras2 = 0 5 3 1
inactive = 10 0 0 6
alter
7
1 2 11
1 3 12
1 4 13
1 5 14
4 1 7
4 2 8
4 3 9
supsym
*a1
1
4 2 3 4 5
*b1
0
*a2
0
*b2
1
3 1 2 3 
hexs
1
1
spin =2
symmetry = 1
ciroot = 60 60 1
nactel = 23 1 0

&caspt2
frozen = 1 0 0 0
maxiter = 2000
imag=0.20
multistate 
60 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Could you let me know possible solutions ?

Best regards,
Kaoru Yamazaki, Ph. D
Research Scientist
RAP, RIKEN, Japan

Ignacio

Administrator

From: Uppsala

Registered: 2015-11-03

Posts: 1,085

Re: I/O error in MS-RASPT2 calculation

Most likely you've hit a maximum line length. You can split the states in several lines, or just write "multistate = all".