[GRASS-dev] Impossible to kill r.in.gdal

[Markus Neteler]

On Mon, Apr 16, 2018 at 4:30 PM, Luca Delucchi <lucadeluge at gmail.com> wrote:
> Hi devs,
>
> I'm importing raster dataset with r.in.gdal but I have a problem
> because the processes seems to died but they didn't stop and I'm not
> neither able to kill them, any idea?
>
>  r.in.gdal input=NETCDF:/mnt/sas1/cosmorea6/converted/T_2M/T_2M_199705.nc
> output=T_2M_1997_05 memory=4048 -o
> No projection information available
> Over-riding projection check
> Importing 744 raster bands...
> Importing raster map <T_2M_1997_05.1>...
>  100%
> ^C^C^C^C^C^C^C^C^C^C^C^C^Z
> ^C
> ^C^C^C^C^Z^C
> ^Z^Z

It looks like that the sas1/ device is hanging.

> ps aux | grep r.in.gdal
> lucadelu  9486  0.0  0.1 415908 37880 ?        D    Apr12   0:00
> r.in.gdal input=NETCDF:/mnt/sas1/cosmorea6/converted/RELHUM_2M/RELHUM_2M_199901.nc
> memory=6072 title=COSMO REA6 offset=0 num_digits=0
> output=RELHUM_2M_1999_01 -o --q

--> D state
https://stackoverflow.com/questions/20423521/process-permanently-stuck-on-d-state


...
> killall -9 r.in.gdal

vs
...
> kill -9 9486 9493 16182 16185 43603 47014 47540

killall does not offer -9 but with D state the zombie process cannot
be killed anyway.

So: reboot is the only way (and fix the mounting of the device).

Markus