vasp5.3.3: Segmentation fault occured when writing wavefunction

Message

hatdau · #1 Post by **hatdau** » Tue Jun 11, 2013 3:18 pm

Dear all,
I'm using VASP 5.3.3 compiled with Intel 2013 and FFTW 3.3. Everything is ok except one case when I'm doing HSE06 calculation for Sb, hexagonal unitcell containing 6 atom. Everything is ok until VASP start writing wavefunction. Any idea would be greatly appreciated.

Thank you in advance,

ERROR:

Code: Select all

DMP:  19    -0.337785495550E+02   -0.32091E-03   -0.15400E-03   816   0.274E-03 0.290E-03
         N     E                     dE            rms(c)
   ROT:  1  -0.339225440658E+02   -0.86148E-04    0.151E-03
   ROT:  2  -0.339225442792E+02   -0.21345E-06    0.889E-04
 gam= 0.382 g(H,U,f)=  0.746E-04 0.505E-04 0.217E-08 ort(H,U,f) = 0.972E-04-0.280E-04 0.288E-08
DMP:  20    -0.337786419238E+02   -0.92369E-04   -0.60594E-04   816   0.125E-03 0.692E-04
   1 F= -.33778642E+02 E0= -.33778641E+02  d E =-.980525E-06
 writing wavefunctions
forrtl: severe (174): SIGSEGV, segmentation fault occurred
Image              PC                Routine            Line        Source             
libopen-pal.so.0   00002B8ED3B2A9FA  Unknown               Unknown  Unknown
libopen-pal.so.0   00002B8ED3B29945  Unknown               Unknown  Unknown
libmpi.so.0        00002B8ED342BF39  Unknown               Unknown  Unknown
libmpi.so.0        00002B8ED3427D5A  Unknown               Unknown  Unknown
libmpi.so.0        00002B8ED33C2E7E  Unknown               Unknown  Unknown
libmpi_f77.so.0    00002B8ED314D904  Unknown               Unknown  Unknown
vasp53.nzhalf      00000000004A44EA  m_sum_d_                 1031  mpi.F
vasp53.nzhalf      00000000005154BC  mrg_grid_rl_plane        1432  mgrid.F
vasp53.nzhalf      0000000000A6AE0D  fileio_mp_outchg_         764  fileio.F
vasp53.nzhalf      0000000000491A22  MAIN__                   4223  main.F
vasp53.nzhalf      000000000044ED6C  Unknown               Unknown  Unknown
libc.so.6          00002B8ED5392CDD  Unknown               Unknown  Unknown
vasp53.nzhalf      000000000044EC69  Unknown               Unknown  Unknown

KPOINTS: 6x6x3

INCAR:

Code: Select all

PREC = medium
ENCUT = 400
ISMEAR = 0
SIGMA = 0.05
NEDOS = 9999
NBANDS = 24
LREAL = .FALSE.
LMAXMIX = 4
LORBIT = 11

ISPIN = 1

ISTART = 0
ICHARG = 2

IBRION = -1
ISIF = 2

LELF = .TRUE.

NSIM = 1


LHFCALC = .TRUE.
ALGO = Damped
HFSCREEN = 0.2
TIME = 0.4
PRECFOCK = Normal

#2 Post by **admin** » Wed Jun 12, 2013 4:50 pm

this looks strange. just to exclude the following possible reason: have you made sure that you did not exceed any (filesize, disk-quota) limits when
WAVECAR was written?

hatdau · #3 Post by **hatdau** » Wed Jun 12, 2013 5:06 pm

It is unlikely the case because vasp has been successful with other jobs with even larger supercell. And I'm working on scratch space.
Best,