[Wien] WIEN2k LAPW1 crashes sometimes during eigenvector calculation for NMR Chemical Shift
Michael Fechtelkord
Michael.Fechtelkord at ruhr-uni-bochum.de
Thu Jul 31 10:40:48 CEST 2025
Hello everybody,
I am currently calculating 27Al and 1H NMR chemical shifts for
Diaspore AlO(OH)
Spacegroup 62 P b n m, a0 = 4.4007 A, b0 = 9.4253 A, c0 = 2.8452 A
Al 0.04472 -0.14450 -0.25000 RMT = 1.6
O1 0.71234 0.19892 -0.25000 RMT = 1.1
O2 0.19714 0.05338 -0.25000 RMT = 1.1
H 0.40950 0.08760 -0.25000 RMT = 0.6
WIEN2k version 24.1, ifx version 2025.2, gcc version 15.1.1, linux
kernel 6.15-8.1
SCF cycles worked fine with init_lapw -b -prec 1 -ecut 0.999 and a 8%
RMT reduction (see RMTs in atom table) and RKMAX= 3.0 to use -min in scf
with -ec = 0.0001 -cc = 0.0001 and -fc = 0.5
I used a reciprocal k-mesh with a delta-k of 0.025 A-1 (3072 k-points
generated, ndiv= 31 15 47)
NMR Initialiazation for all atoms (no reduced Ls)
During calculation of the eigenvectors in the k-mesh lapw1 partially
crashes for some shifted meshes (sometimes one time, sometimes in
different meshes, crashes do not always appear in same mesh) e.g.,
cd ./nmr_mqy ... x lapw1 -nmr -p -scratch /scratch/WIEN2k/
starting parallel lapw1 at Di 29. Jul 15:49:39 CEST 2025
-> starting parallel LAPW1 jobs at Di 29. Jul 15:49:39 CEST 2025
running LAPW1 in parallel mode (using .machines)
granularity set to 1 because of nonlocal SCRATCH variable
4 number_of_parallel_jobs
[1] 199672
[2] 199690
[3] 199710
[4] 199728
forrtl: severe (168): Program Exception - illegal instruction
Image PC Routine Line Source
libc.so.6 000014C20C042260 Unknown Unknown Unknown
lapw1 00000000004209A0 coors 120 coors.f
lapw1 0000000000418FB7 calkpt 140 calkpt_tmp_.F
lapw1 0000000000455764 lapw1 65 lapw1_tmp_.F
lapw1 000000000040796D Unknown Unknown Unknown
libc.so.6 000014C20C02B37B Unknown Unknown Unknown
libc.so.6 000014C20C02B44B __libc_start_main Unknown Unknown
lapw1 0000000000407875 Unknown Unknown Unknown
[4] Fertig ( cd $PWD; $t $exe
${def}_$loop.def; rm -f .lock_$lockfile[$p] ) >> .time1_$loop
LAPW1 END
LAPW1 END
LAPW1 END
[3] Fertig ( cd $PWD; $t $exe
${def}_$loop.def; rm -f .lock_$lockfile[$p] ) >> .time1_$loop
[2] + Fertig ( cd $PWD; $t $exe
${def}_$loop.def; rm -f .lock_$lockfile[$p] ) >> .time1_$loop
[1] + Fertig ( cd $PWD; $t $exe
${def}_$loop.def; rm -f .lock_$lockfile[$p] ) >> .time1_$loop
localhost(768) 8973.515u 80.966s 1:19:04.33 190.8% 0+0k
65592+96185144io 12pf+0w
localhost(768) 8970.035u 78.348s 1:19:01.57 190.8% 0+0k
448+96014968io 0pf+0w
localhost(768) 8975.787u 80.321s 1:19:03.86 190.9% 0+0k
344+95760256io 0pf+0w
localhost(768) 4349.782u 36.343s 38:03.20 192.1% 0+0k
344+40091368io 1pf+0w
Summary of lapw1para:
localhost k=3072 user=31269.1 wallclock=2520.2
** LAPW1 crashed!
31279.657u 289.667s 1:19:06.22 665.1% 0+0k 67904+328053032io 30pf+0w
error: command /usr/local/WIEN2k/lapw1para -scratch /scratch/WIEN2k/
lapw1.def failed
Has anyone and idea what possibly went wrong?
Best regards,
Michael
--
Dr. Michael Fechtelkord
Institut für Geowissenschaften
Ruhr-Universität Bochum
Universitätsstr. 150
D-44780 Bochum
Phone: +49 (234) 32-24380
Fax: +49 (234) 32-04380
Email: Michael.Fechtelkord at ruhr-uni-bochum.de
Web Page: https://www.ruhr-uni-bochum.de/kristallographie/kc/mitarbeiter/fechtelkord/
More information about the Wien
mailing list