Consultation on an error when calculating SHG: "Writing file ndb.Overlap; Variable NOT DEFINED; Permission denied"

[Scat]

Greeting!

This post is to ask help for an error when calculating SHG. Recently I am trying to use Quantum ESPRESSO + Yambo 5.1.2 to calculate SHG susceptibility for KH2PO4 (KDP) as a benchmark.

My procedure follows the tutorial of https://www.yambo-code.eu/wiki/index.ph ... with_Yambo and https://www.yambo-code.eu/wiki/index.ph ... r_response, and the code was run with the command

Code: Select all

mpirun -np 36 yambo_nl -F yambo.in -J SHG

.

The process succeeded in writing the "dipoles header" but it seems failed Overlap files.

Code: Select all

<02h-10m> [DIP] Writing dipoles header


 <02h-10m> [DIP] Writing overlaps header
[ERROR] STOP signal received while in[04] Dipoles
[ERROR] Writing File ./SHG//ndb.Overlap; Variable  NOT DEFINED; Permission denied

I have checked the file permissions, and the error file "ndb.Overlap" is "-rw-rw-r--", seemingly with no problems. I also found posts reporting similar errors in the forum, but they don't seem to help.

I would highly appreciate it if anyone could offer me any help!

Best regards,
Ming-Yu Guo

[Davide Sangalli]

Dear Ming-Yu Guo,
from the few info provided it is hard to say which is the issue.

This kind of error generally appears when the code tries to write a variable into a database (ndb.Overlap in this case) which was not correctly initialized.
This is done via specific calls to netcdf subroutines inside the source code.

Two suggestion
a) try running on fewer mpi tasks, to see if the issue remains there
b) try compiling the branch bug-fixes (https://github.com/yambo-code/yambo/tree/bug-fixes), to see if the issue was fixed in this more recent version of the code

Best,
D.

[Scat]

Dear Dr. Sangalli,

Thank you very much for your reply. I will try your suggestions.

Best regards,
M.Y.G.

[Scat]

Dear Ming-Yu Guo,
from the few info provided it is hard to say which is the issue.

This kind of error generally appears when the code tries to write a variable into a database (ndb.Overlap in this case) which was not correctly initialized.
This is done via specific calls to netcdf subroutines inside the source code.

Two suggestion
a) try running on fewer mpi tasks, to see if the issue remains there
b) try compiling the branch bug-fixes (https://github.com/yambo-code/yambo/tree/bug-fixes), to see if the issue was fixed in this more recent version of the code

Best,
D.

Dear Dr. Davide Sangalli

I succeeded in solve the problem by reduce the mpi tasks. Thank you again!

However, I am encountering aother error.

Code: Select all

[ERROR] STOP signal received while in[05] NL Initialization
[ERROR] Use ypp -y to repair the symmetry group and restart Yambo

I followed the tutorial and my file content of ypp.in is

Code: Select all

fixsyms                          # [R] Remove symmetries not consistent with an external perturbation
% Efield1
 1.000000 | 1.000000 | 0.000000 |        # First external Electric Field
%
% Efield2
 0.000000 | 0.000000 | 0.000000 |        # Additional external Electric Field
%
#BField= 0.000000           T     # [MAG] Magnetic field modulus
#Bpsi= 0.000000             deg   # [MAG] Magnetic field psi angle [degree]
#Btheta= 0.000000           deg   # [MAG] Magnetic field theta angle [degree]
#RmAllSymm                     # Remove all symmetries
RmTimeRev                     # Remove Time Reversal
#RmSpaceInv                    # Remove Spatial Inversion
#KeepKGrid                     # Do not expand the k-grid

The file was generated via "ypp -y" command. And "ypp" command and "yambo_nl" command in FixSymm folder were subsequently executed as instructions of the tutorial.

Would you please tell me how to solve this problem? Thank you!

Best regards,
M.Y.G.

[Davide Sangalli]

Please check that the direction of the electric field(s) selected in the input file for yambo_nl is consistent with the direction used in the input file for ypp.

This should be the reason for your warning message.

Best,
D.

[Scat]

Please check that the direction of the electric field(s) selected in the input file for yambo_nl is consistent with the direction used in the input file for ypp.

This should be the reason for your warning message.

Best,
D.

Dr. Davide Sangalli

Thank you for your reply. But I have checked my input file of yambo_nl, of which contents are

Code: Select all

nloptics
NLstep=0.0100 fs                         # [R] Non-linear spectroscopy
% NLBands
   150 | 250 |                         # [NL] Bands range
%
NLverbosity= "high"              # [NL] Verbosity level (low | high)
NLtime=-1.000000           fs    # [NL] Simulation Time
NLintegrator= "INVINT"           # [NL] Integrator ("EULEREXP/RK2/RK4/RK2EXP/HEUN/INVINT/CRANKNIC")
NLCorrelation= "IPA"             # [NL] Correlation ("IPA/HARTREE/TDDFT/LRC/LRW/JGM/SEX")
NLLrcAlpha= 0.000000             # [NL] Long Range Correction
% NLEnRange
 0.200000 | 8.000000 |         eV    # [NL] Energy range
%
NLEnSteps= 1                     # [NL] Energy steps
NLDamping= 0.200000        eV    # [NL] Damping (or dephasing)
Field1_kind= "SOFTSIN"           # [RT Field1] Kind(SIN|RES|ANTIRES|GAUSS|DELTA|QSSIN
% Field1_Dir
 1.000000 | 1.000000 | 0.000000 |        # [RT Field1] Versor
%

To my understanding, the "Field1_Dir" section is consistent with the part:

Code: Select all

% Efield1
 1.000000 | 1.000000 | 0.000000 |        # First external Electric Field
%

in file ypp.in

I have also tried to comment out the Efield2 part in ypp.in, but the same error was given.

So maybe electric field is not the problem. Is there any other possibilities? Thanks!

Best regards,
M.Y.G.

[claudio]

Deat Ming-Yu Guo

notice that when you run ypp to remove the symmetries,
the new wave-function without symmetries are in a new folder FixSymm.
You have to run calculations there and not in the main folder, that still contain symmetric wave-functions

best
Claudio

[Scat]

Deat Ming-Yu Guo

notice that when you run ypp to remove the symmetries,
the new wave-function without symmetries are in a new folder FixSymm.
You have to run calculations there and not in the main folder, that still contain symmetric wave-functions

best
Claudio

Dear Claudio,

Thank you for your kind information! This problem is now solved.

I also tried to re-compile the code with Davide's link because the error "Writing file ndb.Overlap; Variable NOT DEFINED; Permission denied" still occurs in my other systems. My re-compiling commands are

Code: Select all

make disclean
./configure CC=icc MPICC=mpiicc CXX=icpc F77=ifort FC=ifort --enable-memory-profile --disable-open-mp --enable-iotk --with-blas-libs="-lmkl_intel_lp64 -lmkl_sequential -lmkl_core"  --with-lapack-libs="-lmkl_intel_lp64 -lmkl_sequential -lmkl_core" --with-blacs-libs="-lmkl_blacs_intelmpi_lp64" --with-scalapack-libs="-lmkl_scalapack_lp64"
make all

However, now when I run "ypp" after "ypp -y" to create ypp.in, the error of the following occurred.

Code: Select all

[ERROR] STOP signal received while in[04] K-point grid
[ERROR] Error reading K-point grid, please run setup again!

The procedures were the same as before. I also tried to re-run QE codes, but the error still occurs.

Sorry for my so many questions. Thank you for your patient replies.

Best,
M.Y.G.

[Davide Sangalli]

It means you need to first run yambo without any input to perform the setup.
After that you should be able to run ypp.

Best,
D.

[Scat]

Dr. Davide Sangalli,

My problems are solved. I am enjoying studying Yambo code now.

Thank you very much!

Cheers,
M.Y.G.