LSDALTON- geometry optimisation - segmentation fault and corrupted double-linked list

Find answers or ask questions regarding LSDalton calculations
Post Reply
zozonmr
Posts: 10
Joined: 25 Jul 2017, 13:54
First name(s): Zoltan
Last name(s): Takacs
Affiliation: C4X Discovery
Country: United Kingdom

LSDALTON- geometry optimisation - segmentation fault and corrupted double-linked list

Post by zozonmr » 31 Jul 2017, 15:58

Hi,

I am attempting to run lsdalton geometry optimisation (first without PCM solvent model ) with the following set up:
**GENERAL
.NOGCBASIS
**OPTIMIZE
.BFGS
.MAX IT
1000
.MAX RE
100
**WAVE FUNCTIONS
.DFT
B3LYP
*END OF INPUT

with:
BASISSET: 6-311+G*

The first run suggested in the out pout to include this:
**INTEGRALS
.THRESH
1.0d-13

which I did. since then I get 2 types of errors:
Error 1.
Program received signal SIGSEGV: Segmentation fault - invalid memory reference.

Backtrace for this error:

Error 2.
lsdalton.x': corrupted double-linked list: 0x0000000008eb5790 ***

I complied lsdalton with omp only and running it on 12 to 48 cores.

I am not sure what file to include since the errors do not appear in the lSDALTON.OUT file, Could you please advise me which file to send to make these errors easier to track?

zozonmr
Posts: 10
Joined: 25 Jul 2017, 13:54
First name(s): Zoltan
Last name(s): Takacs
Affiliation: C4X Discovery
Country: United Kingdom

Re: LSDALTON- geometry optimisation - segmentation fault and corrupted double-linked list

Post by zozonmr » 02 Aug 2017, 09:48

well in the meantime I got a third error:

*** Error in lsdalton.x': malloc(): memory corruption (fast): 0x0000000007d44790 ***

In the scratch folder there is an out file and also there has been some iteration done since there are 4 molecule.xxx files. Is it possible to automatically restart the calc? I just copied the last molecule.xxx and used that as an input with the same *.dal file.

Post Reply

Who is online

Users browsing this forum: No registered users and 2 guests