[GRASS-dev] DBMI troubles on Win32

Hi all,

I am currently doing a lot of testing for GRASS 6.3. on the mingw/msys
platform (the same that comes with QGIS).

There seems to be a problem related to the DBMI protocol. Modules that
access attribute data via the GRASS DBMI lib have a habit of hanging
once in a while. I have seen this happen on half a dozen different
machines with Win2000 and XP.

An example is v.info -c. The module simply hangs and has to be killed
and restarted until it does its work.
The strange thing is that I cannot find any pattern here: sometimes
dozens of runs show no problems at all and sometimes I have to restart a
module several times until it starts running.
Also, this seems to concern only the first DB access. I have a module
here that does hundreds of DB accesses during each run. If the first one
goes OK, all the others will be fine too ...

I am using the latest 6.3 CVS sources for my tests.

I am at a loss here. Can anyone confirm this problem?

Thanks,

Benjamin

--
Benjamin Ducke, M.A.
Archäoinformatik
(Archaeoinformation Science)
Institut für Ur- und Frühgeschichte
(Inst. of Prehistoric and Historic Archaeology)
Christian-Albrechts-Universität zu Kiel
Johanna-Mestorf-Straße 2-6
D 24098 Kiel
Germany

Tel.: ++49 (0)431 880-3378 / -3379
Fax : ++49 (0)431 880-7300
www.uni-kiel.de/ufg

On Fri, March 9, 2007 09:18, Benjamin Ducke wrote:

Hi all,

I am currently doing a lot of testing for GRASS 6.3. on the mingw/msys
platform (the same that comes with QGIS).

There seems to be a problem related to the DBMI protocol. Modules that
access attribute data via the GRASS DBMI lib have a habit of hanging
once in a while. I have seen this happen on half a dozen different
machines with Win2000 and XP.

An example is v.info -c. The module simply hangs and has to be killed
and restarted until it does its work.
The strange thing is that I cannot find any pattern here: sometimes
dozens of runs show no problems at all and sometimes I have to restart a
module several times until it starts running.
Also, this seems to concern only the first DB access. I have a module
here that does hundreds of DB accesses during each run. If the first one
goes OK, all the others will be fine too ...

I am using the latest 6.3 CVS sources for my tests.

I am at a loss here. Can anyone confirm this problem?

I can't with the native winGRASS (not QGIS), at least not using v.info -c
as a test case within spearfish. But there tables are very small. Which
type of files did you try this on ?

Moritz