[molpro-user] Problem with CI calculation

Andy May MayAJ1 at cardiff.ac.uk
Wed Apr 29 09:09:54 BST 2015


Ulli,

As far as I can tell that error comes when there is not enough remaining 
memory to do the next stage of the calculation. In the input file there is:

MEMORY,2400,M;

which restricts how much each Molpro process can use (note this is mega 
words, see:

http://www.molpro.net/pipermail/molpro-user/2010-April/003723.html

for details). Perhaps try increasing this value, you might need to run 
with a few less processes to ensure that each has enough memory.

Best wishes,

Andy

On 23/04/15 09:38, Ulrike Nitzsche wrote:
> Dear molpro users,
>
> I'm the system administrator of a 2012.1.18 Installation. One of my
> users has a problem with a CI calculation, the output ended with the
> following lines:
>
> **
>     Spin-orbit matrix elements
>     ==========================
>
>   Wavefunction restored from record  5126.2  Symmetry=1  S= 2.0  NSTATE=5
>   Wavefunction restored from record  5127.2  Symmetry=1  S= 1.0  NSTATE=6
>   ? Error
>   ? CILSDPS 1 PAGE ERROR
>   ? The problem occurs in CILSDPS
> **
>
> The job was running on one host using 12 cores. RAM is not the issue, the
> job took about 220GB and 256GB are available.
>
> Please can you help us to fix the problem? First question is to localize the
> problem: is it the installation, the input file, a chemical problem...
>
> I attach the input file to this mail, if you need the full output or any
> further information please ask me by email.
>
> Thank you very much in advance and best regards,
> Ulli.
>
>
>
> _______________________________________________
> Molpro-user mailing list
> Molpro-user at molpro.net
> http://www.molpro.net/mailman/listinfo/molpro-user
>



More information about the Molpro-user mailing list