Out of curiosity, why do they install two instances of R in a Docker
container?  On a traditional physical tin server, sure, you can't avoid
that sort of thing, but why in Docker?  I didn't see a rationale for it on
the Rocker wiki or website.

Greg

On Mon, 11 Aug 2025 at 16:05, Ott Toomet <otoo...@gmail.com> wrote:

> Thanks.
>
> I thought this was the issue, and even manually deleted /usr/bin/R and
> /usr/bin/Rscript, but the problem persisted.  But now I am thinking that
> /usr/bin/R may not be the relevant executables, the development version may
> pull up something directly from /usr/lib/R/bin instead (the devel is
> installed in /usr/local/lib/R ).
>
> Will try!
>
> Ott
>
> On Sun, Aug 10, 2025 at 9:57 PM Tim Taylor <
> tim.tay...@hiddenelephants.co.uk>
> wrote:
>
> > You may be hitting the following issue in tinytest:
> > https://github.com/markvanderloo/tinytest/issues/135
> >
> > There is a linked PR you could try that hopefully fixes the problem:
> > https://github.com/markvanderloo/tinytest/pull/134
> >
> > Tim
> >
> > On 11 Aug 2025, at 05:39, Ott Toomet <otoo...@gmail.com> wrote:
> >
> > Hey everyone who is more familiar with R-devel and docker.
> >
> > I am trying to test the new version of my package (maxLik) on R-devel,
> > using the docker container.  I am using rocker/r-devel as the base.  Note
> > that R-devel is called "RD" in the image.  I am copying in my package
> > sources and a few scripts, and run "RD CMD build" and "RD CMD check".
> > However, the latter fails with complaints about missing packages that are
> > installed and accessible when I run RD from command line.
> >
> > Here are a few results:
> >
> > * After a successful build inside the container, I run check:
> > root@f4024e015396:/# RD CMD check --as-cran maxLik_1.6-3.tar.gz
> > * using log directory ‘//maxLik.Rcheck’
> > * using R Under development (unstable) (2025-08-11 r88561)
> > * using platform: x86_64-pc-linux-gnu
> > * R was compiled by
> >    gcc (Debian 14.2.0-19) 14.2.0
> >    GNU Fortran (Debian 14.2.0-19) 14.2.0
> > * running under: Debian GNU/Linux trixie/sid
> > * using session charset: UTF-8
> > * using option ‘--as-cran’
> > * checking for file ‘maxLik/DESCRIPTION’ ... OK
> > * this is package ‘maxLik’ version ‘1.6-3’
> > * checking CRAN incoming feasibility ... [6s/35s] OK
> > * checking package namespace information ... OK
> > * checking package dependencies ... OK
> > * checking if this is a source package ... OK
> > * checking if there is a namespace ... OK
> > * checking for executable files ... OK
> > * checking for hidden files and directories ... OK
> > * checking for portable file names ... OK
> > * checking for sufficient/correct file permissions ... OK
> > * checking serialization versions ... OK
> > * checking whether package ‘maxLik’ can be installed ... ERROR
> > Installation failed.
> > See ‘//maxLik.Rcheck/00install.out’ for details.
> > * DONE
> >
> > Status: 1 ERROR
> > See
> >  ‘//maxLik.Rcheck/00check.log’
> > for details.
> >
> > * Take a look at the log:
> >
> > root@f4024e015396:/# cat maxLik.Rcheck/00install.out
> > * installing *source* package ‘maxLik’ ...
> > ** this is package ‘maxLik’ version ‘1.6-3’
> > ** using staged installation
> > ** R
> > ** inst
> > ** byte-compile and prepare package for lazy loading
> > Error in loadNamespace(i, c(lib.loc, .libPaths()), versionCheck =
> vI[[i]])
> > :
> >  there is no package called ‘lattice’
> > Calls: <Anonymous> ... loadNamespace -> withRestarts -> withOneRestart ->
> > doWithOneRestart
> > Execution halted
> > ERROR: lazy loading failed for package ‘maxLik’
> > * removing ‘/maxLik.Rcheck/maxLik’
> >
> > Apparently 'lattice' is missing???
> >
> > * But it is there and I can load it with no errors:
> >
> > root@f4024e015396:/# RDscript -e "packageVersion('lattice')"
> > [1] ‘0.22.7’
> > root@f4024e015396:/# RDscript -e "library(lattice)"
> >
> > Formerly I got similar missing package messages for other packages, I
> have
> > also tried to manually overwrote the R_LIB_USR environment variable (that
> > resulted in another error a bit further down in the line), and to install
> > my own package into different locations.  So far everything seems to lead
> > to the same problem.
> >
> > Note that it works if I run it inside the container using the stable R
> ("R"
> > instead of "RD").  Currently it is based on r88561 (2025-08-11) version
> of
> > R-devel as you can see above.
> >
> > For reference, here is the Dockerfile:
> >
> > # -*- dockerfile-mode -*-
> > FROM rocker/r-devel
> > RUN apt-get install -y texlive-science
> > ENV INSIDE_DOCKER=true
> > COPY packages.R .  # install script for necessary packages
> > RUN RDscript packages.R  # install dependencies/suggestions etc
> > COPY pkg ./pkg  # copy the maxlik package source
> > COPY build-check .  # building/checking script
> > CMD ["bash"]
> >
> >
> > Any ideas?
> >
> > Ott
> >
> >    [[alternative HTML version deleted]]
> >
> > ______________________________________________
> > R-package-devel@r-project.org mailing list
> > https://stat.ethz.ch/mailman/listinfo/r-package-devel
> >
> >
>
>         [[alternative HTML version deleted]]
>
> ______________________________________________
> R-package-devel@r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-package-devel
>

        [[alternative HTML version deleted]]

______________________________________________
R-package-devel@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-package-devel

Reply via email to