[GRASS5] mysterious disk full error in r.resamp.rst

Markus Neteler neteler at itc.it
Mon Apr 11 09:05:43 EDT 2005


Hi,

we have an error when using r.resamp.rst to downsample
a raster map:

Input data:

r.info -s max11
nsres=10
ewres=10

g.region -p
projection: 99 (Transverse Mercator)
zone:       0
datum:      rome40
ellipsoid:  international
north:      5157110
south:      5059580
west:       1612510
east:       1728650
nsres:      10
ewres:      10
rows:       9753
cols:       11614


GRASS 6.0.cvs: > r.resamp.rst in=max11 ew_res=110 ns_res=110 elev=resmax11
...
mapset for input demetri

Processing all selected output files will require
3746688 bytes of disk space for temp files

zmin=56.751999,zmax=3760.000000
Temporarily changing the region to desired resolution...
Changing the region back to initial...
Percent complete: 40%
Not enough disk space--cannot write files
Not enough disk space--cannot write files
Not enough disk space--cannot write files
Not enough disk space--cannot write files
...

The lsof command below shows that (only) around 4MB are used as indicated
above.
The df -h command below shows that more than 3GB are available:

/usr/sbin/lsof -p 4729
...
r.resamp. 4729 demetri    3r   REG    8,3 187093184   49753 /hardmnt/vanga0/ssi/grassdata_demetri/pat/demetri/fcell/max11
r.resamp. 4729 demetri    4u   REG    8,3   3743744   49659 /hardmnt/vanga0/ssi/grassdata_demetri/pat/demetri/.tmp/vanga/4729.0

df -h /hardmnt/vanga0/ssi/grassdata_demetri/pat/demetri/.tmp/vanga/
Filesystem            Size  Used Avail Use% Mounted on
/dev/sda3              25G   21G  3.2G  87% /hardmnt/vanga0

Is it possible that the fwrite() in main.c of r.resamp.rst doesn't report
the correct error (since we have plenty of disk space)? Which error
could it be (in general the module seems to work)?

Thanks

 Markus




More information about the grass-dev mailing list