BASPRO error

Find answers or ask questions regarding Dalton calculations.
Please upload an output file showing the problem, if applicable.
(It is not necessary to upload input files, they can be found in the output file.)

Post Reply
Zeljka.San
Posts: 44
Joined: 13 Mar 2014, 16:24
First name(s): Zeljka
Last name(s): Sanader
Affiliation: University of Split
Country: Croatia

BASPRO error

Post by Zeljka.San » 29 Apr 2014, 16:00

Dear all,
I want to run two photon cross section calculation and I get error. I tested .dal file on smaller system and everything is fine. So I think I have to change this baspro number, but I cannot find how.
Thanks for any suggestions!
Zeljka

SYMGRP: Point group information
-------------------------------

Full point group is: C(1)
Represented as: C1

>>> BASPRO error, # of shells when adding Huckel orbitals 1044
current maximum number 1000

--- SEVERE ERROR, PROGRAM WILL BE ABORTED ---
Date and time (Linux) : Tue Apr 29 18:46:01 2014
Host name : backend1

@ MPI MASTER, node no.: 0
@ Reason: BASPRO: Too many shells

>>>> Total CPU time used in DALTON: 0.08 seconds
>>>> Total wall time used in DALTON: 0.08 seconds

bast
Posts: 1210
Joined: 26 Aug 2013, 13:22
First name(s): Radovan
Last name(s): Bast
Affiliation: none
Country: Germany

Re: BASPRO error

Post by bast » 29 Apr 2014, 18:38

dear Zeljka,

in DALTON/include/maxorb.h
increase MXSHEL
in line " PARAMETER (MXSHEL = 1000, MXPRIM = 8000 )"

in DALTON/include/infpar.h
increase MXSHEL
in line "#define MXSHEL 1000"

then compile the code:

Code: Select all

cd build
make
the error message is not so helpful - i will improve it.

good luck!
radovan

Zeljka.San
Posts: 44
Joined: 13 Mar 2014, 16:24
First name(s): Zeljka
Last name(s): Sanader
Affiliation: University of Split
Country: Croatia

Re: BASPRO error

Post by Zeljka.San » 02 May 2014, 12:49

Hi, Radovan!
Thanks for your help, but I still have the same problem.
I changed MXSHELL number to 2000 in both files you suggested and I ran "make" in build folder.
I got:

[ 1%] Built target gen1int
Scanning dependencies of target gen1int_interface
[ 1%] Building Fortran object gen1int/CMakeFiles/gen1int_interface.dir/gen1int_host.F90.o
Linking Fortran static library ../lib/libgen1int_interface.a
[ 1%] Built target gen1int_interface
[ 2%] Built target pelib
[ 2%] Built target generate_binary_info
Scanning dependencies of target dalton...............
.............................
[ 48%] Building Fortran object CMakeFiles/dalton.dir/binary_info.F90.o
Linking Fortran static library lib/libdalton.a
[100%] Built target dalton
Scanning dependencies of target dalton.x
[100%] Building Fortran object CMakeFiles/dalton.x.dir/DALTON/abacus/dalton.F.o
Linking Fortran executable dalton.x
[100%] Built target dalton.x



But I still get
>>> BASPRO error, # of shells when adding Huckel orbitals 1044
current maximum number 1000

--- SEVERE ERROR, PROGRAM WILL BE ABORTED ---

It looks like MXSHELL number is not changed.
I don't know what went wrong.I would appreciate any suggestion.

Greetings,
Zeljka

bast
Posts: 1210
Joined: 26 Aug 2013, 13:22
First name(s): Radovan
Last name(s): Bast
Affiliation: none
Country: Germany

Re: BASPRO error

Post by bast » 04 May 2014, 15:38

dear Zeljka,

difficult to say what went wrong. "normally" it should
work - i have tested this locally and it behaves as expected.

when you modify these parameters you should
see that many files get recompiled (all files which include
maxorb.h and infpar.h). i hope that this is what happened
during the "..." dots in your compilation output.

you can force the recompilation of all these files by updating the timestamps:

Code: Select all

touch DALTON/include/maxorb.h
touch DALTON/include/infpar.h
cd build
make
you can also force full recompilation with (normally never needed though):

Code: Select all

make clean
make
another thing to double check is whether the correct binary was used.
perhaps you recompiled the one in the build path but still use another one
in the install path?
happened to me several times that i was debugging the wrong binary
(having around 10 different dalton.x binaries on my harddisc).

good luck,
radovan

Zeljka.San
Posts: 44
Joined: 13 Mar 2014, 16:24
First name(s): Zeljka
Last name(s): Sanader
Affiliation: University of Split
Country: Croatia

Re: BASPRO error

Post by Zeljka.San » 06 May 2014, 11:14

Hi!
I used old binary. New binary solved everything. :)
After increasing MXSHEL I got new BASSPRO error, which I solved by increasing MXCORB number in maxorb.h
Well, these numbers are closely related so it was easy to solve.
Thanks a lot for your help!
Zeljka

bast
Posts: 1210
Joined: 26 Aug 2013, 13:22
First name(s): Radovan
Last name(s): Bast
Affiliation: none
Country: Germany

Re: BASPRO error

Post by bast » 06 May 2014, 11:56

great. glad it works. we have increased some of the parameters
for the next patch meaning that such edits won't be needed -
until they are needed again (today's limits are tomorrow's defaults).
radovan

ismailcakmak
Posts: 3
Joined: 04 Feb 2015, 19:28
First name(s): ISMAIL
Last name(s): CAKMAK
Affiliation: KAFKAS UNIVERSITY
Country: Turkey
Contact:

Re: BASPRO error

Post by ismailcakmak » 05 Mar 2015, 09:52

What is the maximum value of mxshell number? İs there any limitation?

I would be veyr appreciate if somebody would answer.

Joanna
Posts: 116
Joined: 27 Aug 2013, 16:38
First name(s): Joanna
Last name(s): Kauczor
Affiliation: Helsinki
Country: Finland

Re: BASPRO error

Post by Joanna » 05 Mar 2015, 10:00

ismailcakmak wrote:What is the maximum value of mxshell number? İs there any limitation?

I would be veyr appreciate if somebody would answer.
mxshell parameter is specified in
DALTON/include/maxorb.F
You can change it there and recompile the code.

ismailcakmak
Posts: 3
Joined: 04 Feb 2015, 19:28
First name(s): ISMAIL
Last name(s): CAKMAK
Affiliation: KAFKAS UNIVERSITY
Country: Turkey
Contact:

Re: BASPRO error

Post by ismailcakmak » 05 Mar 2015, 11:25

Thank you ery much for the message. But I have asked maximum value of the MXSHELL.

Joanna
Posts: 116
Joined: 27 Aug 2013, 16:38
First name(s): Joanna
Last name(s): Kauczor
Affiliation: Helsinki
Country: Finland

Re: BASPRO error

Post by Joanna » 05 Mar 2015, 11:32

ismailcakmak wrote:Thank you ery much for the message. But I have asked maximum value of the MXSHELL.
And you will find this value at the given location.
You can also increase it there, if necessary.
You must, however, recompile the code after changing it.

arnfinn
Posts: 231
Joined: 28 Aug 2013, 08:02
First name(s): Arnfinn
Middle name(s): Hykkerud
Last name(s): Steindal
Affiliation: UiT
Country: Norway
Location: UiT The Arctic University of Norway

Re: BASPRO error

Post by arnfinn » 05 Mar 2015, 11:40

ismailcakmak wrote:What is the maximum value of mxshell number? İs there any limitation?

I would be veyr appreciate if somebody would answer.
As far as I know there is no limitation per se, but the static memory allocated by the dalton executable might increase significantly. Normally it is around 160 MB. If I increased mxshell from 1500 to 5000 (both in maxorb.h and infpar.h) it became almost 700 MB. I also got an error when I tried to compile it at 15000, but have not tried to compile it with --int64.

taylor
Posts: 595
Joined: 15 Oct 2013, 05:37
First name(s): Peter
Middle name(s): Robert
Last name(s): Taylor
Affiliation: Tianjin University
Country: China

Re: BASPRO error

Post by taylor » 05 Mar 2015, 11:43

I think we are struggling to help you here. Are you asking "what is the largest I can set MXSHEL to"? That is, is there any fundamental limit on how large MXSHEL can be?

Joanna has told you (twice!) how to change it. If your question is actually what I state above, i.e., you're not asking how to change it, but what you can change it to, then the answer is completely dependent on your local situation and without (a lot) more information there is no unambiguous answer. How much memory do you have? What sort of basis (segmented vs contracted) are you using? What addressing (32-bit or 64-bit) are you using?

As with many forum postings, it would be much better if you told us what you actually are trying to do, as opposed to asking abstract questions like "how large can MXSHEL be"? This is pretty much akin to the old joke "how long is a piece of a string"?!!

Best regards
Pete

ismailcakmak
Posts: 3
Joined: 04 Feb 2015, 19:28
First name(s): ISMAIL
Last name(s): CAKMAK
Affiliation: KAFKAS UNIVERSITY
Country: Turkey
Contact:

Re: BASPRO error

Post by ismailcakmak » 05 Mar 2015, 13:44

Dear Pete
Thank you ver much for your answer, I apologize for incomplete questions.

I am trying to compute band gap value of a polymer. I had calculated band gap of monomer and dimer form. I received baspro error when I tried to compute trimer molecule unit of the polymer.
With best regards, I am.

Computer
UBUNTU 14.04
DALTON 2015
4 GB
64Bit
Attachments
p8c.out
(146.7 KiB) Downloaded 239 times
p8c.mol
(38.92 KiB) Downloaded 227 times
p8c.dal
(21.22 KiB) Downloaded 236 times

kennethruud
Posts: 269
Joined: 27 Aug 2013, 16:42
First name(s): Kenneth
Last name(s): Ruud
Affiliation: UiT The Arctic University of Norway
Country: Norway

Re: BASPRO error

Post by kennethruud » 06 Mar 2015, 10:53

You are trying to run a, by Dalton standards, very large molecule. I would suggest you consider trying LSDalton.

A way of reducing the size of the orbital space, is otherwise to turn off the use of Huckel starting guess, and instead use the one-electron core Hamiltonian

**WAVE FUNCTION
...
*ORBITAL INPUT
.MOSTART
H1DIAG

It is likely to give a poorer starting guess in your case, and it is not going to solve the problems occurring because of the size of the molecule you are trying to calculate, so the proper solution to the problem you want to solve remains, I believe, to switch to LSDalton.


Kenneth

Post Reply

Who is online

Users browsing this forum: No registered users and 5 guests