"LAPACK: Routine ZPOTRF failed" on Cray XT3

[borchert]

Hello. I install VASP for our users. On our Cray XT3, we have a user with an issue running VASP 4.6.28 on a very small input. Output is below, but the error is “LAPACK: Routine ZPOTRF failed�. The same input on our SGI Origin stops in the same place but gives this message instead “ERROR FEXCP: supplied Exchange-correletion table is too small, maximal index : 2147483647�. The user tells me that the code crashes when IWAVPR = 2 is used. I can provide his input files if needed. I hope someone on the forum has knowledge of this issue.

Thanks,
Chris

Code: Select all

> yod -np 8 $WORKDIR/vasp.4.6-mpi/vasp
 running on    8 nodes
 distr:  one band on    1 nodes,    8 groups
 vasp.4.6.28 25Jul05 complex
 POSCAR found :  2 types and   14 ions
 LDA part: xc-table for Ceperly-Alder, standard interpolation
 POSCAR, INCAR and KPOINTS ok, starting setup
 FFT: planning ...            1
 reading WAVECAR
 prediction of wavefunctions initialized - no I/O
 entering main loop
       N       E                     dE             d eps       ncg     rms          rms(c)
RMM:   1     0.257110684240E+03    0.25711E+03   -0.66168E+03   512   0.416E+02
RMM:   2     0.542260330471E+02   -0.20288E+03   -0.22272E+03   512   0.117E+02
RMM:   3    -0.209237073829E+02   -0.75150E+02   -0.78091E+02   512   0.802E+01
RMM:   4    -0.512862710105E+02   -0.30363E+02   -0.24864E+02   512   0.534E+01
RMM:   5    -0.604342554867E+02   -0.91480E+01   -0.81863E+01   512   0.305E+01
RMM:   6    -0.639606371480E+02   -0.35264E+01   -0.30751E+01   512   0.191E+01
RMM:   7    -0.653312726206E+02   -0.13706E+01   -0.12794E+01   512   0.118E+01
RMM:   8    -0.659387291604E+02   -0.60746E+00   -0.56097E+00   512   0.784E+00
RMM:   9    -0.664261562861E+02   -0.48743E+00   -0.47733E+00  1194   0.503E+00
RMM:  10    -0.664457885097E+02   -0.19632E-01   -0.21726E-01  1302   0.127E+00
RMM:  11    -0.664470230231E+02   -0.12345E-02   -0.12460E-02  1195   0.231E-01
RMM:  12    -0.664470249490E+02   -0.19260E-05   -0.12383E-04  1208   0.553E-02    0.102E+01
RMM:  13    -0.635837645668E+02    0.28633E+01   -0.70831E+00  1026   0.764E+00    0.170E+00
RMM:  14    -0.634626945503E+02    0.12107E+00   -0.53221E-01  1053   0.227E+00    0.470E-01
RMM:  15    -0.634338682456E+02    0.28826E-01   -0.83545E-02  1067   0.817E-01    0.242E-01
RMM:  16    -0.634205023438E+02    0.13366E-01   -0.25201E-02  1058   0.422E-01    0.200E-01
RMM:  17    -0.634083709340E+02    0.12131E-01   -0.14890E-02  1033   0.311E-01    0.125E-01
RMM:  18    -0.634005700848E+02    0.78008E-02   -0.17008E-02  1035   0.352E-01    0.613E-02
RMM:  19    -0.634008881412E+02   -0.31806E-03   -0.49093E-03  1027   0.202E-01    0.313E-02
RMM:  20    -0.634032494445E+02   -0.23613E-02   -0.20939E-03  1024   0.187E-01
   1 F= -.63403249E+02 E0= -.63432026E+02  d E =-.634032E+02
 curvature:   0.00 expect dE= 0.000E+00 dE for cont linesearch  0.000E+00
 trial: gam= 0.00000 g(F)=  0.271E+01 g(S)=  0.000E+00 ort = 0.000E+00 (trialstep = 0.100E+01)
 search vector abs. value=  0.271E+01
 bond charge predicted
       N       E                     dE             d eps       ncg     rms          rms(c)
RMM:   1    -0.622380488977E+02   -0.62238E+02   -0.80856E+01  1034   0.250E+01    0.257E+00
RMM:   2    -0.623627374812E+02   -0.12469E+00   -0.37779E+00  1092   0.575E+00    0.182E+00
RMM:   3    -0.620302994366E+02    0.33244E+00   -0.62853E-01  1090   0.197E+00    0.544E-01
RMM:   4    -0.620175914927E+02    0.12708E-01   -0.91897E-02  1109   0.776E-01    0.183E-01
RMM:   5    -0.620171460096E+02    0.44548E-03   -0.11059E-02  1156   0.288E-01    0.566E-02
RMM:   6    -0.620170080846E+02    0.13792E-03   -0.10960E-03  1121   0.987E-02    0.194E-02
RMM:   7    -0.620170386643E+02   -0.30580E-04   -0.20009E-04  1090   0.475E-02    0.106E-02
RMM:   8    -0.620171699583E+02   -0.13129E-03   -0.11087E-04  1043   0.360E-02    0.547E-03
RMM:   9    -0.620173525229E+02   -0.18256E-03   -0.10615E-04  1017   0.333E-02    0.302E-03
RMM:  10    -0.620175143218E+02   -0.16180E-03   -0.13935E-04   997   0.327E-02    0.126E-03
RMM:  11    -0.620175602818E+02   -0.45960E-04   -0.61231E-05   962   0.176E-02    0.775E-04
RMM:  12    -0.620175725729E+02   -0.12291E-04   -0.29721E-05   871   0.100E-02    0.408E-04
RMM:  13    -0.620175758387E+02   -0.32659E-05   -0.69854E-06   773   0.448E-03    0.284E-04
RMM:  14    -0.620175801452E+02   -0.43064E-05   -0.57090E-07   748   0.415E-03    0.136E-04
RMM:  15    -0.620175808738E+02   -0.72864E-06    0.12260E-06   654   0.218E-03
   2 F= -.62017581E+02 E0= -.62066802E+02  d E =0.138567E+01
 trial-energy change:    1.385669  1 .order    0.301372   -2.714160    3.316904
 step:   0.2486(harm=  0.4500)  dis= 0.08163  next Energy=   -63.723907 (dE=-0.321E+00)
Information: wavefunction orthogonal band   28  0.8265
Information: wavefunction orthogonal band   31  0.8884
Information: wavefunction orthogonal band   32  0.7410
Information: wavefunction orthogonal band   32  0.8199
Information: wavefunction orthogonal band   32  0.8947
Information: wavefunction orthogonal band   32  0.8917
Information: wavefunction orthogonal band   31  0.8413
Information: wavefunction orthogonal band   32  0.8751
Information: wavefunction orthogonal band   32  0.7849
Information: wavefunction orthogonal band   28  0.8995
Information: wavefunction orthogonal band   31  0.8844
Information: wavefunction orthogonal band   32  0.8960
Information: wavefunction orthogonal band   29  0.7967
Information: wavefunction orthogonal band   32  0.8830
Information: wavefunction orthogonal band   32  0.7903
Information: wavefunction orthogonal band   28  0.8881
Information: wavefunction orthogonal band   32  0.8822
Information: wavefunction orthogonal band   31  0.8943
Information: wavefunction orthogonal band   27  0.8857
Information: wavefunction orthogonal band   30  0.8459
Information: wavefunction orthogonal band   31  0.8960
Information: wavefunction orthogonal band   32  0.8219
Information: wavefunction orthogonal band   30  0.8979
Information: wavefunction orthogonal band   31  0.8921
Information: wavefunction orthogonal band   32  0.8013
 bond charge predicted
 prediction of wavefunctions
       N       E                     dE             d eps       ncg     rms          rms(c)
 LAPACK: Routine ZPOTRF failed!            1
 LAPACK: Routine ZPOTRF failed!            1
 LAPACK: Routine ZPOTRF failed!            1
 LAPACK: Routine ZPOTRF failed!            1
 LAPACK: Routine ZPOTRF failed!            1
 LAPACK: Routine ZPOTRF failed!            1
 LAPACK: Routine ZPOTRF failed!            1
 LAPACK: Routine ZPOTRF failed!            1
FORTRAN STOP
FORTRAN STOP
FORTRAN STOP
FORTRAN STOP
FORTRAN STOP
FORTRAN STOP
FORTRAN STOP
FORTRAN STOP

[admin]

1) most probably, the forces of the first+second ionic steps lead to an unreasonable geometry for the third step (huge differences in the total energies!), which could not be converged electronically. (ie this is no LAPACK error, but just an unreasonable setup for that ionic step)
to check:
--) please have a look at the FORCES on the atoms at the end of ionic step #1 and #2, and the corresponding geometry changes which gave the input geometry for step #3 (the input geometries of each step are listed in XDATCAR). If the ions move too much, the wave function prediction may fail)
--> please reduce POTIM and/or use a different ionic relaxation algorithm
(IBRION)
[ Edited Wed Nov 08 2006, 12:28PM ]