[GRASS-dev] [GRASS GIS] #2764: corrupt data written to FCELL and DCELL rasters, hard to re-produce

[GRASS GIS]

#2764: corrupt data written to FCELL and DCELL rasters, hard to re-produce
---------------------+-------------------------
  Reporter:  dylan   |      Owner:  grass-dev@…
      Type:  defect  |     Status:  new
  Priority:  normal  |  Milestone:  7.2.3
 Component:  Raster  |    Version:  unspecified
Resolution:          |   Keywords:
       CPU:  x86-64  |   Platform:  Linux
---------------------+-------------------------

Comment (by mmetz):

 Replying to [comment:9 dylan]:
 > After applying the patch, I see this (more informative) error message:
 >
 > {{{
 > Decompression failed with error 0
 > }}}

 The actual error code is missing, the patch was wrong, now updated.

 Which compressor are you using?
 >
 > This was in the context of a corrupt (?) map generated by r.horizon,
 when run in parallel like this:
 >
 > {{{
 > seq 0 $step 330 | parallel --gnu --progress "bash make-hz-maps.sh {}
 $step $elev"
 > }}}
 >
 > make-hz-makes.sh is a simple wrapper around r.horizon:
 >
 > {{{
 > # start angle
 > start=$1
 > # angle step size
 > step=$2
 > # elevation map
 > map=$3
 >
 > # r.horizon requires start, step, and end angles
 > # compute end angle via bc
 > end=`echo "$start + $step" | bc`
 >
 > r.horizon --q --o elevation=$map step=$step start=$start end=$end
 output=hzangle
 > }}}
 >
 > Am I abusing GRASS raster libraries by invocation of multiple instances
 of the same module that all access the same elevation map?

 I don't think so, multiple instances of the same module should run
 independent of each other. The problem could be anywhere between the
 physical hard drive having problems with multiple simultaneous requests to
 read the same file, the OS handling such requests, and the GRASS file
 reading mechanism.

--
Ticket URL: <https://trac.osgeo.org/grass/ticket/2764#comment:10>
GRASS GIS <https://grass.osgeo.org>