Molcas Forum

Support and discussions for Molcas and OpenMolcas users and developers

You are not logged in.

Announcement

Welcome to the Molcas forum.

Please note: The forum's URL has changed. The new URL is: https://molcasforum.univie.ac.at. Please update your bookmarks!

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

Pages: 1

#1 2024-09-21 17:50:49

EMP
Member
Registered: 2017-09-27
Posts: 10

CASPT2 calculation for multiple states

Dear all,

I'm trying to run a CASPT2 calculation after running a RASSCF for an S = 6 multiplicity. The CASPT2 I want to run is for three groups of states however, the CASPT2 aborts. Can someone give me a hint about what the problem could be?

Thanks a lot for the help!


&RASSCF                
Spin=  6                
Nactel=  9 0 0          
Ras2=  7              
Inactive= 88          
CiRoot= 21 21 1
ORBA
FULL         

&CASPT2                
XMUL
3
11	1 2 3 4 5 6 7 8 9 10 11
7	12 13 14 15 16 17 18
3	19 20 21
IMAG
0.1
MAXI
100

Offline

#2 2024-09-23 08:04:10

Ignacio
Administrator
From: Uppsala
Registered: 2015-11-03
Posts: 1,079

Re: CASPT2 calculation for multiple states

It seems you're using an old syntax with "groups". This is no longer supported, because fundamentally it didn't work as intended. The XMUL syntax is now the same as for MULT, so you either do a single calculation with all 21 states or three separate calculations each with:

XMUL = 11 ; 1 2 3 4 5 6 7 8 9 10 11

XMUL = 7 ; 12 13 14 15 16 17 18

XMUL = 3 ; 19 20 21

Offline

#3 2024-09-23 11:34:06

EMP
Member
Registered: 2017-09-27
Posts: 10

Re: CASPT2 calculation for multiple states

Thanks a lot for your response!

Offline

#4 2024-09-23 20:23:50

EMP
Member
Registered: 2017-09-27
Posts: 10

Re: CASPT2 calculation for multiple states

Dear Ignacio, I tried running the CASPT2 calculation as you described, but I'm getting this error:

Multi-State initialization phase begins for group   1
--------------------------------------------------------------------------------
 ###############################################################################
 ###############################################################################
 ###                                                                         ###
 ###                                                                         ###
 ###    Location: AixRd                                                      ###
 ###    File: CHVEC1                                                         ###
 ###                                                                         ###
 ###                                                                         ###
 ###    Premature abort while reading buffer from disk                       ###
 ###                                                                         ###
 ###    End of file reached                                                  ###
 ###                                                                         ###
 ###                                                                         ###
 ###############################################################################
 ###############################################################################
--- Stop Module: caspt2 at Mon Sep 23 21:21:01 2024 /rc=-6 ---
--- Module caspt2 spent 35 seconds ---

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

Am I doing something wrong? This is my input:

&RASSCF                
Spin=  6                
Nactel=  9 0 0          
Ras2=  7              
Inactive= 143          
CiRoot= 21 21 1         
>> COPY $Project.JobIph  1_IPH       
>> COPY $Project.RasOrb  1.RasOrb

&CASPT2
XMUL = 11; 1 2 3 4 5 6 7 8 9 10 11 
IMAG
0.1
MAXI
100

Thanks again for your support.

Offline

#5 2024-09-24 08:58:46

Ignacio
Administrator
From: Uppsala
Registered: 2015-11-03
Posts: 1,079

Re: CASPT2 calculation for multiple states

That looks like a missing or inconsistent integral file. Did you run everything from scratch (GATEWAY + SEWARD + ...)?

Offline

#6 2024-09-24 09:18:55

EMP
Member
Registered: 2017-09-27
Posts: 10

Re: CASPT2 calculation for multiple states

Dear Ignacio, thanks for the response.

I ran the SEWARD module before. Is it necessary to run GATEWAY?

If so, then this might be the issue.

Offline

#7 2024-09-24 09:21:58

Ignacio
Administrator
From: Uppsala
Registered: 2015-11-03
Posts: 1,079

Re: CASPT2 calculation for multiple states

It depends on how you run it. Everything done by GATEWAY can be done by SEWARD, but things may be weird if you don't run with a clean scratch directory. If the problem persists, show a full reproducible input file.

Offline

#8 2024-09-24 09:46:40

EMP
Member
Registered: 2017-09-27
Posts: 10

Re: CASPT2 calculation for multiple states

I ran it with a clean scratch directory, but I'll double-check as soon as I can.

Again, thanks for your support!

Offline

#9 2024-09-24 17:22:17

EMP
Member
Registered: 2017-09-27
Posts: 10

Re: CASPT2 calculation for multiple states

Dear Ignacio,

I've ran the calculation once again, making sure the scratch directory was fine, but I still get the same error. This is the input file: 

&SEWARD
RICD
ANGMOM
0.34910000000	3.16080000000	12.74320000000
AMFI
DOUG

Basis Set
Dy.ANO-RCC-VTZP
Dy1	0.34910000000	3.16080000000	12.74320000000	Angstrom
End of Basis Set
 
Basis Set
C.ANO-RCC-VDZP
C1	1.92730000000	2.42640000000	10.87830000000	Angstrom
C2	1.47930000000	1.21050000000	11.43890000000	Angstrom
C3	0.11020000000	1.06290000000	11.14010000000	Angstrom
C4	-0.29940000000	2.21350000000	10.41370000000	Angstrom
C5	0.82840000000	3.04090000000	10.24140000000	Angstrom
C11	0.07310000000	4.81550000000	14.61500000000	Angstrom
C12	-1.25840000000	4.55340000000	14.21360000000	Angstrom
C13	-1.57290000000	3.23250000000	14.57130000000	Angstrom
C14	-0.44660000000	2.66850000000	15.18210000000	Angstrom
C15	0.56960000000	3.63020000000	15.21480000000	Angstrom
End of Basis Set

Basis Set
C.ANO-RCC-VDZ
C6	3.35330000000	2.94520000000	10.89140000000	Angstrom
C7	2.36790000000	0.15560000000	12.03660000000	Angstrom
C8	-0.69390000000	-0.20220000000	11.32330000000	Angstrom
C9	-1.65250000000	2.38650000000	9.74400000000	Angstrom
C10	0.85490000000	4.32860000000	9.45610000000	Angstrom
C16	1.31260000000	5.81850000000	14.52990000000	Angstrom
C17	1.11180000000	7.02770000000	15.32170000000	Angstrom
C18	1.66480000000	5.97410000000	13.05310000000	Angstrom
C19	-1.66910000000	5.94220000000	13.40210000000	Angstrom
C20	-2.83260000000	6.55680000000	14.20710000000	Angstrom
C21	-1.95530000000	5.60830000000	12.01920000000	Angstrom
C22	-3.06860000000	3.12740000000	14.01730000000	Angstrom
C23	-3.95430000000	3.06890000000	15.39810000000	Angstrom
C24	-3.28640000000	1.92750000000	13.22330000000	Angstrom
C25	-0.95950000000	1.28900000000	15.62270000000	Angstrom
C26	-0.89290000000	1.34490000000	17.25000000000	Angstrom
C27	-0.29420000000	0.11040000000	15.10140000000	Angstrom
C28	1.80490000000	2.90260000000	15.83870000000	Angstrom
C29	2.31420000000	3.59430000000	17.07110000000	Angstrom
C30	2.84870000000	2.67250000000	14.76770000000	Angstrom
End of Basis Set

Basis Set
H.ANO-RCC-VDZ
H1	3.36590000000	3.83510000000	11.24920000000	Angstrom
H2	3.69850000000	2.95580000000	9.99490000000	Angstrom
H3	3.89660000000	2.37180000000	11.43680000000	Angstrom
H4	3.13530000000	0.57200000000	12.43360000000	Angstrom
H5	2.65050000000	-0.45230000000	11.34950000000	Angstrom
H6	1.88230000000	-0.32720000000	12.71070000000	Angstrom
H7	-0.49110000000	-0.58930000000	12.17620000000	Angstrom
H8	-0.46940000000	-0.82610000000	10.62750000000	Angstrom
H9	-1.62940000000	0.00400000000	11.27750000000	Angstrom
H10	-2.34910000000	2.16700000000	10.37010000000	Angstrom
H11	-1.71250000000	1.80380000000	8.98490000000	Angstrom
H12	-1.75550000000	3.29630000000	9.45830000000	Angstrom
H13	0.05180000000	4.82610000000	9.63060000000	Angstrom
H14	0.90860000000	4.13040000000	8.51810000000	Angstrom
H15	1.61800000000	4.84870000000	9.71790000000	Angstrom
H16	2.06890000000	5.34890000000	14.94000000000	Angstrom
H17	0.33210000000	7.49060000000	15.00760000000	Angstrom
H18	1.87910000000	7.59700000000	15.23880000000	Angstrom
H19	0.98980000000	6.78820000000	16.24440000000	Angstrom
H20	2.06860000000	5.16400000000	12.73250000000	Angstrom
H21	2.28130000000	6.70310000000	12.94620000000	Angstrom
H22	0.86600000000	6.15370000000	12.55140000000	Angstrom
H23	-0.90760000000	6.55940000000	13.42610000000	Angstrom
H24	-3.62580000000	6.03400000000	14.07400000000	Angstrom
H25	-2.98900000000	7.45600000000	13.90820000000	Angstrom
H26	-2.60660000000	6.56470000000	15.14070000000	Angstrom
H27	-1.18120000000	5.20390000000	11.62220000000	Angstrom
H28	-2.18270000000	6.40780000000	11.53710000000	Angstrom
H29	-2.69240000000	4.99370000000	11.98430000000	Angstrom
H30	-3.29890000000	3.92690000000	13.50030000000	Angstrom
H31	-3.83750000000	2.21350000000	15.81690000000	Angstrom
H32	-4.88170000000	3.20060000000	15.18870000000	Angstrom
H33	-3.65900000000	3.76060000000	15.99570000000	Angstrom
H34	-2.87570000000	2.03530000000	12.36160000000	Angstrom
H35	-4.22910000000	1.78390000000	13.11200000000	Angstrom
H36	-2.89840000000	1.17190000000	13.67040000000	Angstrom
H37	-1.90450000000	1.23050000000	15.36970000000	Angstrom
H38	-0.11760000000	1.84240000000	17.52270000000	Angstrom
H39	-0.83760000000	0.45230000000	17.59900000000	Angstrom
H40	-1.68320000000	1.77190000000	17.58810000000	Angstrom
H41	-0.49060000000	0.02000000000	14.16780000000	Angstrom
H42	-0.60620000000	-0.66650000000	15.57260000000	Angstrom
H43	0.65380000000	0.19820000000	15.22360000000	Angstrom
H44	1.49280000000	2.01660000000	16.12230000000	Angstrom
H45	2.41170000000	4.53210000000	16.89450000000	Angstrom
H46	3.16560000000	3.22450000000	17.31760000000	Angstrom
H47	1.69070000000	3.46790000000	17.79100000000	Angstrom
H48	2.48030000000	2.11910000000	14.07620000000	Angstrom
H49	3.61260000000	2.23610000000	15.15380000000	Angstrom
H50	3.11760000000	3.51580000000	14.39460000000	Angstrom
End of Basis Set

&RASSCF                
Spin=  6                
Nactel=  9 0 0          
Ras2=  7              
Inactive= 143          
CiRoot= 21 21 1         
>> COPY $Project.JobIph  1_IPH       
>> COPY $Project.RasOrb  1.RasOrb

&CASPT2
XMUL = 11; 1 2 3 4 5 6 7 8 9 10 11 
IMAG
0.1
MAXI
100

>>  COPY  $Project.JobMix  $CurrDir/JobMix.1 
>>  COPY  $Project.JobMix  JobMix.1
 
&GRID_IT
NAME
Dy1

&RASSI
Nr of JobIph
1 21 128 130
1   2   3   4   5   6   7   8   9  10  11  12  13  14  15  16  17  18  19  20  21
IPHN                   
 1_IPH  
MEES                    
PROP                    
3
'ANGMOM' 1
'ANGMOM' 2
'ANGMOM' 3
EJOB
 JobMix.1
EPRG                    
7.0D-1

&SINGLE_ANISO          
CRYS
Dy
MLTP
8
2 2 2 2 2 2 2 2
UBAR
QUAX
1

End Of Input

Offline

#10 2024-09-25 11:45:16

Ignacio
Administrator
From: Uppsala
Registered: 2015-11-03
Posts: 1,079

Re: CASPT2 calculation for multiple states

It works for me so far:

********************************************************************************
 Multi-State initialization phase begins for group   1
--------------------------------------------------------------------------------

  H0 in the original model space basis:
                1               2               3               4               5
   1     -154.85578143      0.00000011      0.00006982     -0.00018113      0.00050824
   2        0.00000011   -154.85577584      0.00016698      0.00009148     -0.00034635
   3        0.00006982      0.00016698   -154.83644932     -0.00001417      0.00007422
   4       -0.00018113      0.00009148     -0.00001417   -154.83654213      0.00186590
   5        0.00050824     -0.00034635      0.00007422      0.00186590   -154.84278321
   6       -0.00035178     -0.00051167      0.00160008      0.00000120      0.00001493
   7       -0.00014650      0.00001902     -0.00125701      0.00001473      0.00000918
   8       -0.00001971     -0.00013846      0.00002641     -0.00118565     -0.00026688
   9       -0.00003316     -0.00009779     -0.00006758      0.00005808      0.00007614
  10        0.00010491     -0.00005721      0.00017071      0.00006282      0.00015153
  11        0.00015492      0.00000782     -0.00010696     -0.00000213      0.00004919

Maybe it's a problem with your compilation, or with the hardware or settings you're using. If you're running on a cluster, make sure the scratch directory (WorkDir) is in a local storage for the node running the calculation, and not in a network-shared storage.

Offline

#11 2024-09-25 15:25:08

EMP
Member
Registered: 2017-09-27
Posts: 10

Re: CASPT2 calculation for multiple states

Dear Ignacio, thanks for the support.

I'm running this code on my own computer, so the scratch directory is local. Could running the code using 16 processes cause any trouble?

Last edited by EMP (2024-09-25 15:34:08)

Offline

#12 2024-09-25 15:36:39

EMP
Member
Registered: 2017-09-27
Posts: 10

Re: CASPT2 calculation for multiple states

I tried on another computer and 16 processes and different compilation and still have the same issue. I also keep the scratch directory before and after calculations and still get the same error. I have no idea what is wrong...

Last edited by EMP (2024-09-25 19:17:41)

Offline

#13 2024-09-26 09:33:34

Ignacio
Administrator
From: Uppsala
Registered: 2015-11-03
Posts: 1,079

Re: CASPT2 calculation for multiple states

Well, I tried in serial, and with 2 processes, and it worked with both. As far as i know, the main use of running CASPT2 in parallel is using resources (memory, disk) from several nodes. If you are running in a single node, running with more that 1 process will probably not result in much benefit. Try with fewer processes (1, 2, 4), hopefully it will work and you'll see if it's worth at all.

Offline

#14 2024-09-29 19:01:14

EMP
Member
Registered: 2017-09-27
Posts: 10

Re: CASPT2 calculation for multiple states

Dear Ignacio, I have tried to run the calculation using a single core, and I still got the same error. To test whether the compilation and computer were the problems, what I did is to install Openmolcas on another computer and ran the verification suit (pymolcas verify). I got some errors for the CASPT2 module:

group grayzone from: /home/parallels/OpenMolcas/test/grayzone
3:grayzone:832 Failed! (scf)
3:grayzone:834 Skipped!
3:grayzone:898 Failed! (caspt2)
----------

Of course, any calculation fails.

I also ran the verification for antheolder version of Openmolcas (the first computer I tried to run the CASPT2 calculation) and I get the following:

additional:216 Failed! (caspt2)     
Runngrayzone:834 Skipped!               
grayzone:898 Failed! (caspt2)     
************************************************************************
A total of 2 test(s) failed, with 1 critical failure(s).
************************************************************************
Please check the directory:
  test/failed
for the .out/.err files of the failed tests,
and check the submit directory:
  test/tmp
for the working directories of the last run.

Are these errors related to my installation? weird enough, most of the modules pass fine the tests.

Thank you for the support!

Offline

Pages: 1

Links: OpenMolcas | Molcas | Manual | Facebook

Board footer

Powered by FluxBB 1.5.11

Last refresh: Today 13:26:22