error: Incompatible library version for module. You need to rebuild GRASS or untangle multiple installations.

[Nikos Alexandris]

Hi dev's and advanced useRs,

I have seen this message for the X-th time now :-( while trying to load
(anything) data using readRAST6 or readVECT6.

I've rebuild grass-6.4.0svn 2-3 times now. I've re-installed R 2-3 times
as well. Not to mention how many attempts to get "spgrass6".

Which is the "incompatible library" and to which "module" does it refer
to? Is this an R error or a grass error? Any help is *higly*
appreciated.

Kind regards, Nikos

_______________________________________________
grass-stats mailing list
[hidden email]
http://lists.osgeo.org/mailman/listinfo/grass-stats

[Nikos Alexandris]

stephen sefick wrote:

> What computer architecture, operating system, what R version etc.?

Right, I assume that people know that I am using Ubuntu Jaunty 64-bit, R
2.9.1, latest rpy2, grass-6.4.0svn from source.

> What is the error message?

The message is the title of this thread.

> I am usingce 6.4RC5 GRASS and R 2.9.1 on mac
> OS X 10.5.7, and it is working fine. I remember there being a problem
> with something, but I don't remember exactly what.  If you were to
> provide the error message then maybe it would jog my memory.

Sorry, that's all the prog prints.

> Also, it may be a good idea to search the R-sig-geo list archive, as your
> problem may have already been solved (assuming it is the same one that
> I had).

Before doing countless compilations/installations I filtered the web for
an hour! I've found some similar messages in grass-dev's mailing-list
but nothing specific.

I thought that /etc/ld.so.conf is my problem... and I commented out all
grass-versions but grass-6.4.0svn. I even removed all other
grass-versions, (re-)installed X-times spgrass6 but no luck.

It's the first time I see this message. I also tried to do some
debugging but R requires, I guess, special treatment (meaning other than
gdb, etc.) !?

Nikos

_______________________________________________
grass-stats mailing list
[hidden email]
http://lists.osgeo.org/mailman/listinfo/grass-stats

[Nikos Alexandris]

On Tue, 2009-07-28 at 22:06 -0500, stephen sefick wrote:
> WOW!  I don't have a clue.  That is nowhere close to the problem that
> I had.  Sorry if I am asking obvious, or redundant questions.  Are you
> installing within R, or from the commandline (R CMD install)?

>From within R with root access (e.g. sudo R, then the rest).

>   I will
> help if I can, but I think this may be over my paygrade.
> good luck!

Thanks for your concern Stephen.
Kind regards, Nikos

_______________________________________________
grass-stats mailing list
[hidden email]
http://lists.osgeo.org/mailman/listinfo/grass-stats

[Roger Bivand]

On Wed, 29 Jul 2009, Nikos Alexandris wrote:

>
> stephen sefick wrote:
>
>> What computer architecture, operating system, what R version etc.?
>
> Right, I assume that people know that I am using Ubuntu Jaunty 64-bit, R
> 2.9.1, latest rpy2, grass-6.4.0svn from source.

Please always repeat everything every time; don't delete content from
threads.

I'm travelling and cannot handhold.

Build only from source, only one piece at a time, and check everything
every time. I'm sure that other people have seen dependency problems when
installing from binary on this platform. This is not seen on any other
platform (Linux, Windows, OSX). There are two underlying problems:
"dependency hell" and your inability to sort out your own system. I
suspect that the XML package in R does not have its dependencies
satisfied. It could also be GDAL/GRASS not having dependencies satisfied.
The reason to build from source is that then you control the dependencies,
rather than the reverse.

Please stop creating the impressionn that software that works for
effectively everyone else is broken only because you can't get your own
system to work, and expect others to do your work for you!

Roger

>
>> What is the error message?
>
> The message is the title of this thread.
>
>> I am usingce 6.4RC5 GRASS and R 2.9.1 on mac
>> OS X 10.5.7, and it is working fine. I remember there being a problem
>> with something, but I don't remember exactly what.  If you were to
>> provide the error message then maybe it would jog my memory.
>
> Sorry, that's all the prog prints.
>
>> Also, it may be a good idea to search the R-sig-geo list archive, as your
>> problem may have already been solved (assuming it is the same one that
>> I had).
>
> Before doing countless compilations/installations I filtered the web for
> an hour! I've found some similar messages in grass-dev's mailing-list
> but nothing specific.
>
> I thought that /etc/ld.so.conf is my problem... and I commented out all
> grass-versions but grass-6.4.0svn. I even removed all other
> grass-versions, (re-)installed X-times spgrass6 but no luck.
>
> It's the first time I see this message. I also tried to do some
> debugging but R requires, I guess, special treatment (meaning other than
> gdb, etc.) !?
>
> Nikos
>
> _______________________________________________
> grass-stats mailing list
> [hidden email]
> http://lists.osgeo.org/mailman/listinfo/grass-stats
>

--
Roger Bivand
Economic Geography Section, Department of Economics, Norwegian School of
Economics and Business Administration, Helleveien 30, N-5045 Bergen,
Norway. voice: [phone]; fax [phone]
e-mail: [hidden email]

_______________________________________________
grass-stats mailing list
[hidden email]
http://lists.osgeo.org/mailman/listinfo/grass-stats

[Glynn Clements]

Nikos Alexandris wrote:

> Hi dev's and advanced useRs,
>
> I have seen this message for the X-th time now :-( while trying to load
> (anything) data using readRAST6 or readVECT6.
>
> I've rebuild grass-6.4.0svn 2-3 times now. I've re-installed R 2-3 times
> as well. Not to mention how many attempts to get "spgrass6".
>
> Which is the "incompatible library"

libgrass_gis

> and to which "module" does it refer to?

Whatever is calling G_gisinit().

> Is this an R error or a grass error? Any help is *higly*
> appreciated.

It means that something was compiled using <grass/gis.h> from one
version of GRASS, but when run it uses libgrass_gis from a different
version.

It may be a problem with R. It can also be caused by a GDAL library
(or GDAL-GRASS plugin) which was built with GRASS support. You might
want to try using gdalinfo on a GRASS dataset to confirm or deny this
possibility.

--
Glynn Clements <[hidden email]>
_______________________________________________
grass-stats mailing list
[hidden email]
http://lists.osgeo.org/mailman/listinfo/grass-stats

[Nikos Alexandris]

On Wed, 2009-07-29 at 11:23 +0100, Glynn Clements wrote:

> Nikos Alexandris wrote:
>
> > Hi dev's and advanced useRs,
> >
> > I have seen this message for the X-th time now :-( while trying to load
> > (anything) data using readRAST6 or readVECT6.
> >
> > I've rebuild grass-6.4.0svn 2-3 times now. I've re-installed R 2-3 times
> > as well. Not to mention how many attempts to get "spgrass6".
> >
> > Which is the "incompatible library"
>
> libgrass_gis
>
> > and to which "module" does it refer to?
>
> Whatever is calling G_gisinit().
>
> > Is this an R error or a grass error? Any help is *higly*
> > appreciated.
>
> It means that something was compiled using <grass/gis.h> from one
> version of GRASS, but when run it uses libgrass_gis from a different
> version.
>
> It may be a problem with R. It can also be caused by a GDAL library
> (or GDAL-GRASS plugin) which was built with GRASS support. You might
> want to try using gdalinfo on a GRASS dataset to confirm or deny this
> possibility.

Thanks. It's, like most of the times, my ignorance on how to deal with
compiling/installing issues.

I guess you are Glynn. gdalinfo segfaults when trying to "query" a grass
raster (under cellhd).

Kind regards, Nikos

_______________________________________________
grass-stats mailing list
[hidden email]
http://lists.osgeo.org/mailman/listinfo/grass-stats

[Nikos Alexandris]

Nikos Alexandris:

> > Hi dev's and advanced useRs,
> >
> > I have seen this message for the X-th time now :-( while trying to load
> > (anything) data using readRAST6 or readVECT6.
> >
> > I've rebuild grass-6.4.0svn 2-3 times now. I've re-installed R 2-3 times
> > as well. Not to mention how many attempts to get "spgrass6".
> >
> > Which is the "incompatible library" and to which "module" does it refer
> > to? Is this an R error or a grass error? Any help is *higly*
> > appreciated.

Maris Nartiss wrote:

> In two words - You are running module compiled for different GRASS
> version than one You are running. Recompile affected module OR, if You
> have multiple GRASS installations, fix Your  LD_LIBRARY_PATH paths.

Thanks Maris, I am learning continuously ;-)

_______________________________________________
grass-stats mailing list
[hidden email]
http://lists.osgeo.org/mailman/listinfo/grass-stats

[Nikos Alexandris]

stephen sefick wrote:

> Nikos,
> I am sorry I couldn't be of more help.  When you resolve the issue
> would you mind telling me what you found.  My interest is peaked.

I am glad you took the time to read my post :-)
Nikos

_______________________________________________
grass-stats mailing list
[hidden email]
http://lists.osgeo.org/mailman/listinfo/grass-stats

[Nikos Alexandris]

stephen sefick wrote:

> >> What computer architecture, operating system, what R version etc.?

Nikos Alexandris wrote:

> > Right, I assume that people know that I am using Ubuntu Jaunty 64-bit, R
> > 2.9.1, latest rpy2, grass-6.4.0svn from source.

> Please always repeat everything every time; don't delete content from
> threads. I'm travelling and cannot handhold.

Just try to do them more easy to read :-). Maybe I overdid it :-O

> Build only from source, only one piece at a time, and check everything
> every time. I'm sure that other people have seen dependency problems when
> installing from binary on this platform. This is not seen on any other
> platform (Linux, Windows, OSX). There are two underlying problems:
> "dependency hell" and your inability to sort out your own system. I
> suspect that the XML package in R does not have its dependencies
> satisfied. It could also be GDAL/GRASS not having dependencies satisfied.
> The reason to build from source is that then you control the dependencies,
> rather than the reverse.
>
> Please stop creating the impressionn that software that works for
> effectively everyone else is broken only because you can't get your own
> system to work, and expect others to do your work for you!

True, it's my ignorance on how to deal with compilation/installation
issues.

Nevertheless, I strongly believe that there is always some gain by
asking questions, even if they are not "smart". No offense, but
unfortunately of fortunately I can only promise that, for as long as I
am interested to understand/learn, I'll keep asking :-p

Kindest regards, Nikos

> >
> >> What is the error message?
> >
> > The message is the title of this thread.
> >
> >> I am usingce 6.4RC5 GRASS and R 2.9.1 on mac
> >> OS X 10.5.7, and it is working fine. I remember there being a problem
> >> with something, but I don't remember exactly what.  If you were to
> >> provide the error message then maybe it would jog my memory.
> >
> > Sorry, that's all the prog prints.
> >
> >> Also, it may be a good idea to search the R-sig-geo list archive, as your
> >> problem may have already been solved (assuming it is the same one that
> >> I had).
> >
> > Before doing countless compilations/installations I filtered the web for
> > an hour! I've found some similar messages in grass-dev's mailing-list
> > but nothing specific.
> >
> > I thought that /etc/ld.so.conf is my problem... and I commented out all
> > grass-versions but grass-6.4.0svn. I even removed all other
> > grass-versions, (re-)installed X-times spgrass6 but no luck.
> >
> > It's the first time I see this message. I also tried to do some
> > debugging but R requires, I guess, special treatment (meaning other than
> > gdb, etc.) !?
> >
> > Nikos

_______________________________________________
grass-stats mailing list
[hidden email]
http://lists.osgeo.org/mailman/listinfo/grass-stats

[Nikos Alexandris]

On Wed, 2009-07-29 at 03:39 +0200, Nikos Alexandris wrote:

> Hi dev's and advanced useRs,
>
> I have seen this message for the X-th time now :-( while trying to load
> (anything) data using readRAST6 or readVECT6.
>
> I've rebuild grass-6.4.0svn 2-3 times now. I've re-installed R 2-3 times
> as well. Not to mention how many attempts to get "spgrass6".
>
> Which is the "incompatible library" and to which "module" does it refer
> to? Is this an R error or a grass error? Any help is *higly*
> appreciated.

After so many posts of mine one more wont hurt :-p. Finally it was
gdal-grass (however I want to mention that I got rid of all other
grass-versions but 6.4.0sv).

Thanks, Nikos

_______________________________________________
grass-stats mailing list
[hidden email]
http://lists.osgeo.org/mailman/listinfo/grass-stats

[Jarosław Jasiewicz]

Nikos Alexandris pisze:

> On Wed, 2009-07-29 at 03:39 +0200, Nikos Alexandris wrote:
>  
>> Hi dev's and advanced useRs,
>>
>> I have seen this message for the X-th time now :-( while trying to load
>> (anything) data using readRAST6 or readVECT6.
>>
>> I've rebuild grass-6.4.0svn 2-3 times now. I've re-installed R 2-3 times
>> as well. Not to mention how many attempts to get "spgrass6".
>>
>> Which is the "incompatible library" and to which "module" does it refer
>> to? Is this an R error or a grass error? Any help is *higly*
>> appreciated.
>>    
>
> After so many posts of mine one more wont hurt :-p. Finally it was
> gdal-grass (however I want to mention that I got rid of all other
> grass-versions but 6.4.0sv).
>
> Thanks, Nikos
>  

it was mentioned and solved eriler
http://lists.osgeo.org/pipermail/grass-stats/2009-January/000897.html
Jarek
> _______________________________________________
> grass-stats mailing list
> [hidden email]
> http://lists.osgeo.org/mailman/listinfo/grass-stats
>  
_______________________________________________
grass-stats mailing list
[hidden email]
http://lists.osgeo.org/mailman/listinfo/grass-stats