[R] mclust, f2c, vanilla. Was: Check error

Christian Hennig fm3a004 at math.uni-hamburg.de
Mon Jul 24 16:06:41 CEST 2000


Dear list,

On Mon, 24 Jul 2000, Prof Brian D Ripley wrote:

> On 24 Jul 2000, Peter Dalgaard BSA wrote:
> 
> > Christian Hennig <fm3a004 at math.uni-hamburg.de> writes:
> > 
> > > > # save all data
> > > > save(list = ls(), file= "all.Rdata")
> > > > 
> > > > # restore the saved values to the current environment
> > > > load("all.Rdata")
> > > Error in load("all.Rdata") : restore file corrupted -- no data loaded
> > > Execution halted
> > 
> > This would be the culprit. Does this also happen if you repeat those
> > lines inside R?

Yes, but it does not happen if I run 
.../bin/R --vanilla < .../R-1.1.0/tests/Examples/base-Ex.R
from a directory of my home area in the network.
Now it seems to me that I did not get all rights properly from our system 
administrator to perform "save" and "load". So it seems to be a problem of
our network, and not of R. 

Is there a way to perform the whole check from another directory?

make /usr/applic/packages/R/R-1.1.0/bin/check
yields
make: Nothing to be done for `/usr/applic/packages/R/R-1.1.0/bin/check'.

Or have I already performed everything by .../bin/R --vanilla on base-Ex.R?

> > > * In my work with this R-1.1.0 up to now everything worked fine except
> > > 
> > > > library(mclust)
> > > /usr/applic/packages/R/lib/R/bin/R.bin: can't resolve symbol '__moddi3'
> > > /usr/applic/packages/R/lib/R/bin/R.bin: can't resolve symbol '__divdi3'
> > > Error in dyn.load(x, as.logical(local), as.logical(now)) : 
> > >         unable to load shared library "/usr/applic/packages/R/lib/R/library/mclust/libs/mclust.so":
> > >   Unable to resolve symbol
> > > Error in library(mclust) : .First.lib failed
> > > 
> > > (This happened with R-0.64.0 on the same system as well, and also under 
> > > Solaris. I assume that f2c has to be installed to prevent this, because
> > > mclust works on the only machine in our institute where f2c is installed.)
> > 
> > Hmm. What about the built-in packages (e.g. ctest)? Did you use R
> > INSTALL correctly to install mclust (or install.packages('mclust'))?
> 
> I suspect this just means that you do not have the current mclust.
> It happened often prior to mclust_1.1-1.tar.gz (around 0.64 time).

install.packages("mclust") on July 15th. I should have the current mclust,
shouldn`t I?
Perhaps it is because SuSe 5.2 is too old for that?

> 
> > In general, you need to link the shared library with the fortran libs,
> > either  -lg2c or -lf2c. Older compilers may not be getting this right.
> 
> If that is the problem then make check on modreg should not work.
> What is FLIBS in .../etc/Makeconf? Does it contain -lg2c?

FLIBS =  -lf2c -lm -L /usr/X11R6/lib -L/usr/lib/gcc-lib/i486-linux/2.7.2.1 -L/usr/i486-linux/lib -lm

Thank you for help,
Christian

***

Christian Hennig
Fachbereich Mathematik - SPST
 (Schwerpunkt Mathematische Statistik und Stochastische Prozesse,
 Zentrum fuer Modellierung und Simulation)
Bundesstrasse 55
D-20146 Hamburg
Tel: x40/42838 4907, privat x40/631 62 79
hennig at math.uni-hamburg.de
http://www.math.uni-hamburg.de/home/hennig/



-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-help-request at stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._



More information about the R-help mailing list