On Thu, Jan 21, 2010 at 1:15 PM, Erik Lane <erikl...@gmail.com> wrote: > Issues with the install: > > README.txt needs updating. At least for Ubuntu ranlib doesn't show up as an > option for apt-get, but it is part of the binutils package. Also needs to be > updated re: the new requirements for gfortran. It still lists the old info. > GCC needs to be all lowercase for apt-get to find it on Ubuntu.
Thanks. Can you post a new README.txt? > > The sagenb.org still lists itself as being 4.3, but at least some of the > options listed as new for 4.3.1 work on it, so I assume that was just > overlooked? It's fully 4.3.1. However, I'm not going to restart it so that it shows "4.3.1" until sometime late tonight, to minimize inconvenience for users, since restarting causes downtime. > I tried to build sage-4.3.1 from source with a clean install of Ubuntu > 32-bit on my faster computer. It's a quad core AMD Phenom 9650 with 8 gig of > memory. It made it quite a ways, but then failed at building r-2.9.2 > > I used the options from the README.txt file: > > export SAGE_CHECK="yes" > export MAKE="make -j4" > > This seems to be the relevant part of the install log: > > Successfully installed r-2.9.2 > Running the test suite. > make[2]: Entering directory > `/home/erik/Downloads/sage-4.3.1/spkg/build/r-2.9.2/src' > make[3]: Entering directory > `/home/erik/Downloads/sage-4.3.1/spkg/build/r-2.9.2/src/tests' > make[4]: Entering directory > `/home/erik/Downloads/sage-4.3.1/spkg/build/r-2.9.2/src/tests' > make[4]: warning: -jN forced in submake: disabling jobserver mode. > make[5]: Entering directory > `/home/erik/Downloads/sage-4.3.1/spkg/build/r-2.9.2/src/tests/Examples' > make[5]: warning: -jN forced in submake: disabling jobserver mode. > > Collecting examples for package ‘base’ > Running examples in package ‘base’ > > Collecting examples for package ‘tools’ > Running examples in package ‘tools’ > > Collecting examples for package ‘utils’ > Running examples in package ‘utils’ > Error: testing 'utils' failed > Execution halted > make[5]: *** [test-Examples-Base] Error 1 > make[5]: Leaving directory > `/home/erik/Downloads/sage-4.3.1/spkg/build/r-2.9.2/src/tests/Examples' > make[4]: *** [test-Examples] Error 2 > make[4]: Leaving directory > `/home/erik/Downloads/sage-4.3.1/spkg/build/r-2.9.2/src/tests' > make[3]: *** [test-all-basics] Error 1 > make[3]: Leaving directory > `/home/erik/Downloads/sage-4.3.1/spkg/build/r-2.9.2/src/tests' > make[2]: *** [check] Error 2 > make[2]: Leaving directory > `/home/erik/Downloads/sage-4.3.1/spkg/build/r-2.9.2/src' > Error while running the R testsuite ... exiting > ************************************* > Error testing package ** r-2.9.2 ** > ************************************* > sage: An error occurred while testing r-2.9.2 > Please email sage-devel http://groups.google.com/group/sage-devel > explaining the problem and send the relevant part of > of /home/erik/Downloads/sage-4.3.1/install.log. Describe your computer, > operating system, etc. > > So it looks like it compiled r, but then found an error in the test suite > and quit the make process because of that? Yes. You should try again with unset SAGE_CHECK Sage has never *ever* built with SAGE_CHECK set, as far as I know. See http://trac.sagemath.org/sage_trac/ticket/7090 http://trac.sagemath.org/sage_trac/ticket/7091 etc. We've hand-inspected the R failures and they are all because of missing optional R packages that we don't include with Sage. -- William > > The entire log is 13.6MB, but if someone wants I can email the entire file > offlist. > > Thanks, > Erik > > -- > To post to this group, send an email to sage-devel@googlegroups.com > To unsubscribe from this group, send an email to > sage-devel+unsubscr...@googlegroups.com > For more options, visit this group at > http://groups.google.com/group/sage-devel > URL: http://www.sagemath.org > > -- William Stein Associate Professor of Mathematics University of Washington http://wstein.org
-- To post to this group, send an email to sage-devel@googlegroups.com To unsubscribe from this group, send an email to sage-devel+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/sage-devel URL: http://www.sagemath.org
