[GRASS-user] Installation problem
gotnidea! Business Development Centre
gotnidea at gmail.com
Wed Oct 25 11:35:20 EDT 2006
I have installed postgresql on my D: ( NTFS one).
Now I have one more problem. I searched it over mailing list archives. It is
quite similar to the one at:
The only problem is that I couldn't get any workable solution. I am working
on spearfish data. I would love if any one could guide me to any tutorial on
On 10/25/06, Moritz Lennert <mlennert at club.worldonline.be> wrote:
> gotnidea! Business Development Centre wrote:
> > Thanks Moritz,
> > I could partially solve my problem with your help. I changed the c:
> > option to d:.
> > But now, as you say, I can install GRASS on my C: with FAT32 structure,
> > I am more interested in doing so. Earlier I had read at many places that
> > GRASS could only work with NTFS.
> > Now to be precise, my problem with installation on C: (FAT32) starts
> > while I install postgresql-8.1.4-1 and try to initialise database
> > structure. I get data directory error telling me that data directory
> > should be on NTFS formatted volume. It clearly mentions:
> > 'If you wish to install the data directory on another type of partition
> > (which is not recommended, and is unsupported) you must initialise the
> > database cluster manually by running initdb.exe.'
> > How can I do so?
> Can't you tell postgres where you want your data directory during the
> installation ? You could tell it to use your NTFS drive as data directory.
> If this is not possible, then just follow their advice and use
> initdb.exe. See the postgresql docs for more info. But see below.
> > And how it is done. Would it affect the functioning of
> > GRASS on FAT32?
> You actually don't need postgresql to use GRASS, but the current windows
> binary was compiled against grass, so unfortunately you need the pg.dll
> to run any module. Even if you were not able to create a postgresql data
> directory during installation, you should still be able to run GRASS
> without any problems. Have you tried ?
> The issue of the PostgreSQL dependency in wingrass was discussed here:
> It seems clear that GRASS should not depend on PostgreSQL...
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the grass-user