[GRASS-user] trouble with r.terraflow

GRASS GRASS user
1

Dear folks,

I am using GRASS 6.4.1 (2011) on Mac (mac os x 10.7.3). Generally I am
running GRASS without any trouble but calling r.terraflow from the command
line generates the following error comments:

MM error: limit =0B. allocating 16B. limit exceeded by 24B.
Assertion failed: (0), function operator new, file mm.cc, line 326.
Abort trap: 6

using the gui generates preety much the same output:

tree = etree.fromstring(getInterfaceDescription(cmd))
  File "/Applications/GRASS-6.4.app/Contents/MacOS/etc/wxpyt
hon/gui_modules/menuform.py", line 1973, in
getInterfaceDescription

"Details: %s") % (cmd, cmderr)
gui_modules.gcmd
.
GException
:
Unable to fetch interface description for command
'r.terraflow'. Details: MM error: limit =0B. allocating 16B.
limit exceeded by 24B.
Assertion failed: (0), function operator new, file mm.cc,
line 326.

Does anybody knows whats going wrong here.
Thanks in advance

Thomas

--
View this message in context: http://osgeo-org.1560.n6.nabble.com/trouble-with-r-terraflow-tp4716732p4716732.html
Sent from the Grass - Users mailing list archive at Nabble.com.

2

tomte wrote:

Dear folks,

I am using GRASS 6.4.1 (2011) on Mac (mac os x 10.7.3). Generally I am
running GRASS without any trouble but calling r.terraflow from the command
line generates the following error comments:

MM error: limit =0B. allocating 16B. limit exceeded by 24B.
Assertion failed: (0), function operator new, file mm.cc, line 326.
Abort trap: 6

It would help a lot if you could provide exact the command line used
to invoke r.terraflow. A first guess is that the memory option was set
to something higher than 2048, because r.terraflow in 6.4.1 can only
use a maximum of 2GB of memory. Since r.terraflow is very efficient in
handling temporary data, you could just as well leave out the memory
option (default is 300MB). Alternatively, you can use r.watershed
which produces more realistic results.

Markus M

using the gui generates preety much the same output:

tree = etree.fromstring(getInterfaceDescription(cmd))
File "/Applications/GRASS-6.4.app/Contents/MacOS/etc/wxpyt
hon/gui_modules/menuform.py", line 1973, in
getInterfaceDescription

"Details: %s") % (cmd, cmderr)
gui_modules.gcmd
.
GException
:
Unable to fetch interface description for command
'r.terraflow'. Details: MM error: limit =0B. allocating 16B.
limit exceeded by 24B.
Assertion failed: (0), function operator new, file mm.cc,
line 326.

Does anybody knows whats going wrong here.
Thanks in advance

Thomas

--
View this message in context: http://osgeo-org.1560.n6.nabble.com/trouble-with-r-terraflow-tp4716732p4716732.html
Sent from the Grass - Users mailing list archive at Nabble.com.
_______________________________________________
grass-user mailing list
grass-user@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/grass-user

3

On Mon, Apr 9, 2012 at 8:05 PM, tomte <tomte.tumetot@gmx.ch> wrote:

Dear folks,

I am using GRASS 6.4.1 (2011) on Mac (mac os x 10.7.3). Generally I am
running GRASS without any trouble but calling r.terraflow from the command
line generates the following error comments:

MM error: limit =0B. allocating 16B. limit exceeded by 24B.
Assertion failed: (0), function operator new, file mm.cc, line 326.
Abort trap: 6

In GRASS 6.4.2 I see three memory issues fixed in r.terraflow:
http://trac.osgeo.org/grass/changeset/47647
http://trac.osgeo.org/grass/changeset/47666

I am not sure that they are related, though.

Please post the output of

g.region -p

Markus

4

below the output of g.region -p

GRASS 6.4.1 (kana_cat0206):~/GIS/kanada/modelling/nka_05m/catch_0206_SA >
g.region -p
projection: 1 (UTM)
zone: 11
datum: nad83
ellipsoid: grs80
north: 5652250
south: 5652130
west: 635585
east: 635705
nsres: 5
ewres: 5
rows: 24
cols: 24
cells: 576

Thomas

--
View this message in context: http://osgeo-org.1560.n6.nabble.com/trouble-with-r-terraflow-tp4716732p4716996.html
Sent from the Grass - Users mailing list archive at Nabble.com.

5

On Mon, Apr 9, 2012 at 9:09 PM, tomte <tomte.tumetot@gmx.ch> wrote:

below the output of g.region -p

GRASS 6.4.1 (kana_cat0206):~/GIS/kanada/modelling/nka_05m/catch_0206_SA >
g.region -p
projection: 1 (UTM)
zone: 11
datum: nad83
ellipsoid: grs80
north: 5652250
south: 5652130
west: 635585
east: 635705
nsres: 5
ewres: 5
rows: 24
cols: 24
cells: 576

Hm, 576 cells is really not a lot, r.terraflow should work. I am still
puzzled about the "MM error: limit =0B" error. Can you try to set the
memory option to 200 and report what happens? Alternatively, try
r.watershed.

Markus M

6

Markus Neteler wrote:

> I am using GRASS 6.4.1 (2011) on Mac (mac os x 10.7.3). Generally I am
> running GRASS without any trouble but calling r.terraflow from the command
> line generates the following error comments:
>
> MM error: limit =0B. allocating 16B. limit exceeded by 24B.

In GRASS 6.4.2 I see three memory issues fixed in r.terraflow:
http://trac.osgeo.org/grass/changeset/47647
http://trac.osgeo.org/grass/changeset/47666

I am not sure that they are related, though.

The relevant change is r38702, r48704, r48705, which relates to the
iostream library rather than r.terraflow.

--
Glynn Clements <glynn@gclements.plus.com>