Program received signal SIGSEGV: Segmentation fault - invalid memory reference.

[Zac Smith]

Dear CP2K Community.
I'm running a job to optimize the calculation of excited states of a periodic system and I encountered this error Program received signal SIGSEGV: Segmentation fault - invalid memory reference. Do you have any suggestions for changes to avoid this error? I am attaching the submitted submit file and slurm.out file for your reference. Thanks.

[Frederick Stein]

Dear Zac,

without providing your CP2K input file "BN.inp" and the respective output file (not the Slurm input/output files), we can only guess. Depending on the size of your system, you might be running out of memory in which case you need a larger number of compute nodes or less MPI ranks per node while increasing the number of OpenMP threads appropriately. It could also be a bug which we can only identify if you provide the CP2K input and output files.

Best,

Frederick

[Zac Smith]

Dear Frederick,

I've attached the input, output files for your reference. BN__PBE.wfn is a first guess to facilitate accelerated PBE0 calculations. 

Have a good day

Zac

[Frederick Stein]

Dear Zac,

Thank you for the files. Unfortunately, I am not familiar with TDDFPT in CP2K and the execution stops right there. So, someone else will have to jump in.

Best,

Frederick

[Jürg Hutter]

Hi

your input runs on my system (48 cores, 256 GB memory).
The only change I did was to increase the truncation radius to 4 A.
I have some concerns on your system. It seems you have degenerate orbitals
and probably degenerate excited states. A geometry optimization might need
special care for these states.

regards
JH


________________________________________
From: cp...@googlegroups.com <cp...@googlegroups.com> on behalf of Zac Smith <qingxi...@gmail.com>
Sent: Monday, March 4, 2024 4:02 AM
To: cp2k
Subject: [CP2K:19986] Re: Program received signal SIGSEGV: Segmentation fault - invalid memory reference.

--
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<mailto:cp2k+uns...@googlegroups.com>.
To view this discussion on the web visit https://groups.google.com/d/msgid/cp2k/75fe5d2d-85f3-4218-8300-b1362718c950n%40googlegroups.com<https://groups.google.com/d/msgid/cp2k/75fe5d2d-85f3-4218-8300-b1362718c950n%40googlegroups.com?utm_medium=email&utm_source=footer>.

[Zac Smith]

No worries, I think that your suggestion are vey useful.

[Zac Smith]

Dear Jürg,

Thank you for the suggestion.The modification to increase the truncation radius to 4 A worked, and the calculations started running. Can you tell me how to determine the increase of the truncation radius to 4 A? 

Have a good day

Zac

[Jürg Hutter]

Hi

you can find information on the truncation radius in the original paper by
M. Guidon et al. You can also make some tests yourself.
I would say as a rule of thumb: 4.5 A is in the lower range of acceptable values,
6 A and above is usually save. Half the box length is another important value
(see discussion in the paper).

regards
JH

________________________________________
From: cp...@googlegroups.com <cp...@googlegroups.com> on behalf of Zac Smith <qingxi...@gmail.com>

Sent: Tuesday, March 5, 2024 4:26 AM
To: cp2k
Subject: Re: [CP2K:19996] Re: Program received signal SIGSEGV: Segmentation fault - invalid memory reference.

Dear Jürg,

To view this discussion on the web visit https://groups.google.com/d/msgid/cp2k/90b58263-ec08-4165-ba9c-9453aebcc5b2n%40googlegroups.com<https://groups.google.com/d/msgid/cp2k/90b58263-ec08-4165-ba9c-9453aebcc5b2n%40googlegroups.com?utm_medium=email&utm_source=footer>.

[Zac Smith]

Thanks for your help!