seg fault during tutorial

[camattelaer]

Hi all,

I'm following the TMS part of the 2-methyl-1-pentene tutorial for calculation of an NMR spectrum (link). I've encountered a seg fault during the optimization with r2scan-3c:

forrtl: severe (174): SIGSEGV, segmentation fault occurred Image PC Routine Line Source
xtb 000000000046D41A Unknown Unknown Unknown xtb 00000000037F12A0 Unknown Unknown Unknown xtb 0000000000D05C22 xtb_extern_orca_m 198 orca.f90 xtb 00000000006B39A3 xtb_main_setup_mp 160 setup.f90 xtb 0000000000439607 xtb_prog_mainmp 526 main.F90 xtb 0000000000450642 MAIN__ 57 primary.f90 xtb 00000000004067F2 Unknown Unknown Unknown xtb 00000000037F2740 Unknown Unknown Unknown xtb 00000000004066D7 Unknown Unknown Unknown

i noticed in the opt-part2.out xtb is called via:

program call : /home/camattelaer/Programs/xtb-6.6.0/bin/xtb coord --opt lax --orca -I opt.inp

if i remove the --orca call and run manually (xtb coord --opt lax -I opt.inp) everything runs smoothly. Is there perhaps a bug with the --orca option? An external call to orca is already defined in the generated opt.inp so the --orca might be redundant?

Thanks for any advice on this issue and how to prevent or work around it!

Kind regards

Charles-Alexandre

[haneug]

What xtb version do you use? In xtb release 6.6.0 the orca interface has a bug. Either you use the current version or an older version.

[camattelaer]

aha! It's been a while since using xtb and upgraded to 6.6.0 since I got redirected from a 'latest stable release' link so I thought it would be a safe bet. Installed xtb-bleed and running as expected now.

Sorry for the inconvenience and thanks for the quick reply!

Charles-Alexandre