GTH-POTENTIAL ERROR in cp2k 2026.1

[qier]

Hello all,

I am using cpk2k 2026.1 to test some old runs in cp2k 7.1 which worked before. However it keep showing error of:

The requested atomic potential <GTH-PBE-q6> for element <O> was not *

 *  \___/                       found in the potential file

I have checked the GTH-POTENTIAL file and O is there, and format is correct. I even removed O and it will show error for the next element not found. not sure why. I also copied the new GTH-POTENTAIL file from the data folder, but did not resolve this error. Any advise will be helpful! Here is more information on how that part is set up:

POTENTIAL_FILE_NAME /home/cp2k-2026.1/data/GTH_POTENTIALS

   &KIND O

      BASIS_SET DZVP-MOLOPT-SR-GTH

      POTENTIAL GTH-PBE-q6

    &END KIND

[Frederick Stein]

Dear qier,

Did you double-check the spelling of the files. I am asking because you used at least four different spellings of the same file. Can you also try what happens if you copy the file into your working directory and change the POTENTIAL_FILE_NAME ./GTH_POTENTIALS? You may also pass the data directory by calling `export CP2K_DATA_DIR=/path/to/data/dir` and just drop the path in your input file.

Best,

Frederick

[qier]

Hi Frederick,

Thanks for taking time to reply to my questions. Sorry about the typo. They are all GTH_POTENTIALS in the input file. I tried the two suggestions your have provided:

1.copy the file into working directory and change the POTENTIAL_FILE_NAME ./GTH_POTENTIALS

I got error like:

2.Pass the data directory by calling `export CP2K_DATA_DIR=/path/to/data/dir` and just drop the path in your input file

I still got error like:

I also tried to assign a different O potential like GTH-PB-q6 or GTH-BLYP-q6,  and the error is similar to the O potential not found.

Since this exact input file worked for cp2k 7.1 in a different HPC cluster, I am going to compile the 7.1 version and see what happens, unless you have other suggestions.

Best,

qier

--
You received this message because you are subscribed to the Google Groups "cp2k" group.
To unsubscribe from this group and stop receiving emails from it, send an email to cp2k+uns...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/cp2k/8250a8ce-ca8d-4ee2-90cd-ad25e3100932n%40googlegroups.com.

[Stein, Dr. Frederick]

Hi qier,
Thank you for the clarification. Just that we do not miss anything, could
you send the full input file and the corresponding output file with CP2K
2026.1 and version 7.1?
Best,
Frederick

>> <https://groups.google.com/d/msgid/cp2k/8250a8ce-ca8d-4ee2-90cd-ad25e3100932n%40googlegroups.com?utm_medium=email&utm_source=footer>
>> .

>>
>
> --
> You received this message because you are subscribed to the Google Groups
>"cp2k" group.
> To unsubscribe from this group and stop receiving emails from it, send an
>email to cp2k+uns...@googlegroups.com.
> To view this discussion visit

>https://groups.google.com/d/msgid/cp2k/CAL_y3k%3Di95rde6j6w2SxbE76wABY7kxCh1zE02%3DpXPqAQzNRow%40mail.gmail.com.

[qier]

Sure. Frederick. Please see files attached include the .in, .out, and the sbatch file used for directing the /data folder.

Thanks!

To view this discussion visit https://groups.google.com/d/msgid/cp2k/ximss-53294485%40cg5.fz-rossendorf.de.

[Frederick Stein]

Apparently, you compiled CP2K yourself. Have you run the regression test suite? If not, please do so. That is something which should be caught by the tests.

[qier]

yes, indeed I complied it myself. After running a regression test, seems there is no errors ？

To view this discussion visit https://groups.google.com/d/msgid/cp2k/072603f0-3f49-4be0-bb30-85bf7b09ecf4n%40googlegroups.com.

[Frederick Stein]

Still strange. What I would try is to take one of the regtests using GTH_POTENTIALS, copy it to your directory and modify it until you are at your requested INPUT. Then, we should be able to reconstruct what is going on.