I am experiencing this problem. If I run in GRASS 7.0.0 (2015):
r.mapcalc new_cumRAD = cumRAD + altavaltellina_rad10_250_05.00
I receive:
ERROR: Error reading raster data for row 210 of <cumRAD>
"cumRAD" is the result of previous calculation, where I summed all the
hourly global radiation maps (derived with "r.sun.hourly" function) one by
one with a Python script. Everything worked fine until I got stucked at
julianday 250 hour 04).
The problem seems to be in the "cumRAD" variable, but I cannot seem to
understand the reason of the error message.
In addition, this is my Python code:
# cumulative global radiation from julian day 152 to 259 (included)
import grass.script as grass
day_month=[range(152,182),range(182,213),range(213,244),range(244,260)] #
list julian days from 1 Jun to 16 Sept divided per month
hour = range(24) # list the hours of the day (0 to 23)
hour = ['{:02d}'.format(x) for x in hour] # change the length to 2
for m in xrange(len(day_month)):
for d in day_month[m]:
for h in hour:
grass.mapcalc('new_cumRAD = cumRAD + $globrad',
globrad = 'altavaltellina_rad'+ str(m+7)+ '_'+ str(d)+ '_'+ h+ '.00',
overwrite = True) # WARNING: the "+7" in "str(m+7)" should refer to
month 6 (June) but it's 7 here due to a previous mistake in the calculation
On Thu, Sep 3, 2015 at 4:43 PM, umberto.minora <umberto.minora@unimi.it> wrote:
I am experiencing this problem. If I run in GRASS 7.0.0 (2015):
r.mapcalc new_cumRAD = cumRAD + altavaltellina_rad10_250_05.00
I receive:
ERROR: Error reading raster data for row 210 of <cumRAD>
I used to get this type of error when running r.series on a large
number of maps, but seemingly at random. I would re-make the map that
"caused' the error, re-run r.series and it would usually work.
Time passed and I forgot about it. I assumed that updating to grass71
(trunk) was the solution.
This afternoon the error returned when using t.rast.series
Error reading raster data for row 292 of <gdd_139> !!!
Again, these errors seem to happen at random. In this case, I am
iterating over a sequence of years and computing growing degree days
via t.rast.series. Here is the output from several years:
1984
WARNING: Overwriting space time raster dataset <tmin_subset> and
unregistering all maps
WARNING: Overwriting space time raster dataset <tmax_subset> and
unregistering all maps
Starting temporal sampling...
Starting mapcalc computation...
Starting map registration in temporal database...
WARNING: Overwriting space time raster dataset <gdd> and unregistering all
maps
100%
1985
WARNING: Overwriting space time raster dataset <tmin_subset> and
unregistering all maps
WARNING: Overwriting space time raster dataset <tmax_subset> and
unregistering all maps
Starting temporal sampling...
Starting mapcalc computation...
Starting map registration in temporal database...
WARNING: Overwriting space time raster dataset <gdd> and unregistering all
maps
100%
ERROR: Error reading raster data for row 292 of <gdd_139>
ERROR: r.series failed. Check above error messages.
1986
WARNING: Overwriting space time raster dataset <tmin_subset> and
unregistering all maps
WARNING: Overwriting space time raster dataset <tmax_subset> and
unregistering all maps
Starting temporal sampling...
Starting mapcalc computation...
Starting map registration in temporal database...
WARNING: Overwriting space time raster dataset <gdd> and unregistering all
maps
... the stack of rasters called "gdd" is re-computed for each year.
Everything worked as expected in 1984, read error in 1985, fine again
in 1986. There is no MASK present.
I have only ever encountered these errors when using t.rast.series and
r.series. Does anyone have any ideas on how I can replicate or
otherwise debug these kind of errors? This is a serious problem for
long-running calculations.
Thanks!
Dylan
On Tue, Sep 8, 2015 at 3:49 AM, umberto.minora <umberto.minora@unimi.it> wrote:
Thank you, installing the newer version (7.0.1) solved the problem!