error when do a groud state calculation [SOLVED]

[jianzhifu]

Dear All, when I do a groud state calculation, there is a warning, and the calculation stop. Could you give some advices? Thanks!
#######################################################################
getcut: wavevector= 0.0000 0.0000 0.0000 ngfft= 128 128 128
ecut(hartree)= 30.000 => boxcut(ratio)= 2.05280

--- !WARNING
message: |
Density went too small (lower than xc_denpos) at112806 points
and was set to xc_denpos= 1.00E-14. Lowest was -0.34E-12.
Likely due to too low boxcut or too low ecut for pseudopotential core charge.
src_file: mkdenpos.F90
src_line: 176
...


ITER STEP NUMBER 1
vtorho : nnsclo_now= 2, note that nnsclo,dbl_nnsclo,istep= 0 0 1
--------------------------------------------------------------------------
mpirun noticed that process rank 3 with PID 19531 on node myu1404-Lenovo exited on signal 9 (Killed).
--------------------------------------------------------------------------
#######################################################################

The input and output files are in the attachments,Thanks!

[sheng]

Not sure if this is your case, but I got that error when my memory is insufficient. Try not cramp a lot of process inside one node to gain more memory usage for one process.

[jianzhifu]

Thank Sheng! I will try it.

[sheng]

Any reason to split ddk response calculation into 3 directions for dataset 3,4,5 respectively instead of using rf 1 1 1?

[jianzhifu]

Thanks Sheng, when decrease the cpu numbers and increase the ecut, the warning disappear.
The calculation is about the transport properties of liquid neon,
like the example inabinit-7.10.4\tests\v3\Input t78.in t79.in.