TURBOMOLE Users Forum

TURBOMOLE Modules => Mpshift => Topic started by: dzd on December 02, 2008, 01:31:34 PM

Title: Strange behaviour of MPSHIFT
Post by: dzd on December 02, 2008, 01:31:34 PM

Hello,

I've experienced some problem with mpshift calculations. After a succesfull dscf run, mpshift terminates abnormally. However magnetic shieldings values are written in log file. control file is updated too. Last few lines of log file consist of:

Code: [Select]

***  nmr shielding constants written onto general input/output file!  ***


 Deallocation failure for xterm, xaij, points, dcart2, fb in <mpshift>

 
 MODTRACE: no modules on stack

  abnormal termination
 mpshift ended abnormally
What is going on? Are those shieldings values reliable?

Reagards,
dzd

Title: Re: Strange behaviour of MPSHIFT
Post by: uwe on December 03, 2008, 10:59:05 AM

Hi,

never seen this before. Which Turbomole version are you using ? The version number is given in the header of the output.

Regards,

Uwe

Title: Re: Strange behaviour of MPSHIFT
Post by: dzd on December 03, 2008, 02:22:42 PM

I'm using TURBOMOLE V5-10.

dzd

Title: Re: Strange behaviour of MPSHIFT
Post by: uwe on December 03, 2008, 03:48:27 PM

Hi,

now, this is more than strange. Seems that you have a DFT input, right? My guess would be that you try to run an MP2 chemical shift calculation on top of DFT. This would explain the error output, but mpshift does recognize that case and stops right at the start, telling you that you cannot mix DFT and MP2...

Please send your input to the Turbomole support (turbomole at cosmologic de)!

Regards,

Uwe

Title: Re: Strange behaviour of MPSHIFT
Post by: klickklack on March 01, 2009, 02:26:44 PM

Hello!

Was this problem solved?
I also encountered exactly this strange error message and would thus be interested in the solution.

Thanks for sharing,

klickklack

Title: Re: Strange behaviour of MPSHIFT
Post by: uwe on March 02, 2009, 11:44:34 AM

Hi,

no, unfortunately dzd did not send anything to the support. So without an input that causes problems, we can not debug and fix it. We do not even know if it is a bug or a problem with input files or was caused by system/OS settings.

Dear Turbomole users (at least all who read this):

If you encounter a problem and a program really crashes, PLEASE send it to the Turbomole support team [ turbomole (at) cosmologic (dot) de ]. We try to deliver a program that does what it is meant to do. So even if you do not have support any more or if you think that you did anything wrong generating the input, we do not care! We just want to make Turbomole as bug free as possible (do not expect miracles, however)...

Please note also that this here is a user forum, this is NOT the support. If you have technical problems, the support is the right place to ask questions to!

Regards,

Uwe