[CP2K-user] [CP2K:18135] Re: Segmentation fault with Hirshfeld CDFT

[uca...@ucl.ac.uk]

Hi Lili,

I had a look at your input files and the issue is likely related to the use 
of 2 constraints (your unsuccessful job) instead of 1 constraint (your 
successful job). Optimising multiple constraints is very challenging, and 
therefore this has not been tested extensively.

If you provide your structure 'ex1-geo.xyz' then I can check your job and 
perhaps implement a bug fix if needed.

Regards,
Chris 

On Thursday, 1 December 2022 at 23:13:27 UTC rassoul... at gmail.com wrote:

> Dear all, 
> I am using 2022.1 version of cp2k. But I have the same problem as Chris. I 
> want to optimize the geometry of a periodic system using cDFT. I received 
> Segmentation fault error ("Program received signal SIGSEGV: Segmentation 
> fault - invalid memory reference.") with Hirshfeld constraints. Same as 
> Chris, it is sensitive to the combinations of atoms I use in &ATOM_GROUP. 
> In my case, it only works fine for one combination of atoms and not any 
> other combinations. The input and output files for the successful and 
> unsuccessful jobs are attached. Only the ATOM GROUP in the input files was 
> altered.
> I Changed the number of nodes, cores, and memory, but it didn't solve the 
> problem. 
> Any advice or tips on how to solve this issue would be greatly valued.
> Best regards,
> Lili
>
> On Tuesday, March 31, 2020 at 12:21:31 PM UTC-4 uca... at ucl.ac.uk wrote:
>
>> This bug has now been fixed in the latest development version of CP2K. 
>> See https://github.com/cp2k/cp2k/issues/847
>>
>>
>>
>> On Thursday, 19 March 2020 18:31:58 UTC, Chris Ahart wrote:
>>>
>>> Dear all,
>>>
>>> When performing CDFT with Hirshfeld I am getting the following error: 
>>> "Program received signal SIGSEGV: Segmentation fault - invalid memory 
>>> reference." This appears to be an issue reported previously (
>>> https://github.com/cp2k/cp2k/issues/560), however I am encountering it 
>>> again in CP2K 7.1. This error occurs on both my local machine and on a 
>>> cluster.
>>>
>>> I have found this error to be sensitive to the system and to the atoms 
>>> included in &ATOM_GROUP, as for certain combinations of atoms CP2K runs 
>>> while with other combinations it crashes. Becke constraint runs in all 
>>> cases, so this is isolated to Hirshfeld. I have attached two example input 
>>> and output files, one with an atom combination that runs while another 
>>> which fails. I have confirmed that Hirshfeld runs for the input files 
>>> included in the CP2K CDFT tutorial.
>>>
>>> Any guidance or insight to resolve this problem would be greatly 
>>> appreciated.
>>>
>>> Thank you for your help and time.
>>>
>>> Regards,
>>> Chris
>>>
>>

-- 
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+unsubscribe at googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/cp2k/f552ee16-28a1-4b74-b2f1-a75ccfd2e2d5n%40googlegroups.com.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20221202/e02bc9c5/attachment.htm>