EDWAV internal error bug

[andrew_rosen]

In VASP 6.3.0, I ran a meta-GGA (M06-L) calculation and received the following output. Unfortunately, I can't upload the structure in a public forum since it is an ongoing research project. Would I be able to send it to a VASP developer via private message? The INCAR is below for reference.

-----------------------------------------------------------------------------
| _ ____ _ _ _____ _ |
| | | | _ \ | | | | / ____| | | |
| | | | |_) | | | | | | | __ | | |
| |_| | _ < | | | | | | |_ | |_| |
| _ | |_) | | |__| | | |__| | _ |
| (_) |____/ \____/ \_____| (_) |
| |
| internal error in: rot.F at line: 801 |
| |
| EDWAV: internal error, the gradient is not orthogonal 1 1 1.27e-4 |
| |
| If you are not a developer, you should not encounter this problem. |
| Please submit a bug report. |
| |
-----------------------------------------------------------------------------

NUPDOWN = 24
LASPH = .TRUE.
SIGMA = 0.010000
EDIFFG = -3.00e-02
SYMPREC = 1.00e-08
ALGO = All
METAGGA = M06L
IBRION = 2
ISIF = 2
ISMEAR = 0
ISYM = 0
LORBIT = 11
NSW = 2000
NCORE = 16
ISPIN = 2
LMAXMIX = 4
LWAVE = .TRUE.
LCHARG = .FALSE.
LAECHG = .FALSE.
MAGMOM = 80*0.0 8*3.0 48*0.0 36*0.0 12*0.0 2*0.0

[andreas.singraber]

Hello!

Thank you for posting this bug report! Do you see any chance of reproducing the error with a different structure which is not problematic to share? It is much easier for us to handle support if all files can be found directly in the forum. Also, it helps other users who may encounter the same problem in the future. Alternatively, if sharing is not at all possible please send the files to support@vasp.at. Thank you!

Best,
Andreas Singraber

[andrew_rosen]

Thanks for the reply.

Unfortunately, I am not able to reproduce the bug on simpler systems. I will contact support via email.

Strangely, it seems the error goes away when I exclude the geometry optimization flags (e.g. ISIF, EDIFFG) even though the error occurs during the first SCF cycle. Very bizarre.