I would like to report a bug that appears when VASP is run with ISYM=3. With different ISYM, my system runs without any issues. With ISYM=3, when VASP enters the main loop (or sometimes after it does the few initial steps that don't print rms(c) it crashes and gives the following bug message:
-----------------------------------------------------------------------------
| _ ____ _ _ _____ _ |
| | | | _ \ | | | | / ____| | | |
| | | | |_) | | | | | | | __ | | |
| |_| | _ < | | | | | | |_ | |_| |
| _ | |_) | | |__| | | |__| | _ |
| (_) |____/ \____/ \_____| (_) |
| |
| internal error in: mkpoints_full.F at line: 137 |
| |
| internal error in CHECK_FULL_KPOINTS: KPOINTS_FULL not properly |
| initialised |
| |
| If you are not a developer, you should not encounter this problem. |
| Please submit a bug report. |
| |
-----------------------------------------------------------------------------
my INCAR file is
ISTART=1
ICHARG=0
LWAVE=.TRUE.
LCHARG=.TRUE.
NCORE=4
ENCUT=600
ISMEAR=2
SIGMA=0.2
EDIFF=1E-8
NELMIN=5
NELM=100
LREAL=Auto
PREC=Normal
LDIPOL=.TRUE.
ALGO=VeryFast
IDIPOL=3
EFIELD=0.00
IMIX=4
AMIN=.01
AMIX=.01
IBRION=-1
ISYM=3
My POSCAR is (although I don't think it matters, I have tried with several other different systems):
POSCAR file
1.0000000000000000
3.1572621919093828 0.0000000000000000 0.0000000000000000
0.0000000000000000 3.1572621919093824 0.0000000000000000
0.0000000000000000 -0.0000000000000000 35.2088082662093100
W
10
Direct
0.0000000000000000 -0.0000000000000000 0.0023041509514915
0.5000000000000000 0.5000000000000000 0.0372801702243210
-0.0000000000000000 -0.0000000000000000 0.0779305829997762
0.5000000000000000 0.5000000000000000 0.1169575304316279
0.0000000000000000 0.0000000000000000 0.1565838578111613
0.5000000000000000 0.5000000000000000 0.1961388777837641
-0.0000000000000000 0.0000000000000000 0.2356111222162369
0.5000000000000000 0.5000000000000000 0.2751661421888326
-0.0000000000000000 -0.0000000000000000 0.3147924695683731
0.5000000000000000 0.5000000000000000 0.3538194170002247
My KPOINTS:
KPT
0
Gamma
12 12 1
0.0 0.0 0.0
I want to use ISYM=3 for electric field slab calculations, since in a few cases when ISYM=1|2, VASP doesn't take into account he breaking of symmetry caused by the external electric field and forces a symmetrization of the density that is wrong.
Best regards,
Andreas Kyritsakis
Bug with ISYM=3
Moderators: Global Moderator, Moderator
-
- Global Moderator
- Posts: 126
- Joined: Mon May 08, 2023 4:08 pm
Re: Bug with ISYM=3
Hi Andreas,
Thank you for posting a bug report here on the forum. Before I start looking into it, could you please provide additional information regarding the VASP version used and the output? The easiest thing to do would be to provide a zip file according to the Forum Posting Guidelines that contains the relevant files.
Thank you for posting a bug report here on the forum. Before I start looking into it, could you please provide additional information regarding the VASP version used and the output? The easiest thing to do would be to provide a zip file according to the Forum Posting Guidelines that contains the relevant files.
Manuel
VASP developer
VASP developer
-
- Global Moderator
- Posts: 126
- Joined: Mon May 08, 2023 4:08 pm
Re: Bug with ISYM=3
I have found that this is a known issue that is not properly documented. ISYM=3 works only in certain contexts, in particular for Hartree-Fock calculations. To still run a regular DFT calculation with ISYM=3, you can try the following workaround: In the INCAR, set
which will activate a HF calculation but does not add any amount of exact exchange. For details, I refer you to the VASP wiki.
Thanks again for the report. We will update the online documentation on the VASP wiki to reflect the current situation. We will also improve the support for ISYM=3 in the future.
Code: Select all
ISYM=3
LHFCALC=true
AEXX=0
Thanks again for the report. We will update the online documentation on the VASP wiki to reflect the current situation. We will also improve the support for ISYM=3 in the future.
Manuel
VASP developer
VASP developer