On Thursday, August 28, 2014 1:03:04 PM UTC+1, Daniel Friedan wrote: > > (1) The file system is fine. Despite the name, '/NOT_ON_SSD/' is just a > subdirectory of '/' on the imac's internal disk, which is an SSD with a > Journaled HFS+ file system with 140GB free. > $ mount > /dev/disk0s2 on / (hfs, local, journaled, noatime) > > (2) '$ ./sage -f polybori' succeeded: > .... > Done installing PolyBoRi. > > real 2m12.677s > user 1m59.620s > sys 0m11.988s > Successfully installed polybori-0.8.3 > Deleting temporary build directory > > /NOT_ON_SSD/Software/sage_build/build2/sage/local/var/tmp/sage/build/polybori-0.8.3 > Finished installing polybori-0.8.3.spkg > > Warning: it might be needed to update the Sage library before > installed packages work: you should run 'make' from $SAGE_ROOT > before running Sage. > > Oh, sorry, I guess ./sage -f might still build it 32-bit. You need to do
$ SAGE64=yes ./sage -f polybori instead. > (3) I then tried (without running 'make distclean'): > $ export SAGE64=yes > $ export MAKE="make -j2" > $ $MAKE > > and got the same error: > Error building Sage. > The following package(s) may have failed to build: > package: polybori-0.8.3 > at that point I would have tried 'make' without 'j2', i.e. building using just one core. It could be a race condition in the parallel build that causes this. Next time, when your build ends up with this error, just do $ make > > (4) I'm now trying 'make' again from scratch. It will take 4-6 hours, > depending on whether the error recurs. > > Daniel > > On Thursday, August 28, 2014 8:42:44 AM UTC, Dima Pasechnik wrote: >> >> On 2014-08-27, Daniel Friedan <dfri...@gmail.com> wrote: >> > ------=_Part_4855_138320293.1409165879953 >> > Content-Type: multipart/alternative; >> > boundary="----=_Part_4856_1171858397.1409165879953" >> > >> > ------=_Part_4856_1171858397.1409165879953 >> > Content-Type: text/plain; charset=UTF-8 >> > >> > All the executables/libraries are 64bit. (per output from $ file >> > sage/local/lib/*; file sage/local/bin/*) >> > >> > Attached is the log for the package that failed: polybori-0.8.3.log >> >> weird, it ends up with an error in creating a manpage...: >> >> scons: *** >> [/NOT_ON_SSD/Software/sage_build/build2/sage/local/man/man1/PolyGUI2.1] >> Source >> `/NOT_ON_SSD/Software/sage_build/build2/sage/local/man/man1/PolyGUI.1' not >> found, needed by target >> `/NOT_ON_SSD/Software/sage_build/build2/sage/local/man/man1/PolyGUI2.1'. >> scons: building terminated because of errors. >> Error installing PolyBoRi. >> >> What kind of a filesystem do you have mounted on /NOT_ON_SSD/ ? >> >> Could you post the output of >> >> $ mount >> >> >> A wild guess might be that if it's some kind of vfat thing, then it might >> misbehave when >> you use a 64-bit tool, scons, there... >> (or you're just running out of space on the device...) >> >> If it's the case, you might try your build on a proper OSX filesystem >> instead. >> >> ------------------------------ >> >> PS. just to make sure you can reproduce it, can you run >> $ ./sage -f polybori >> >> in the directory you build Sage in, and see if you end up with the same >> error? >> >> >> -- You received this message because you are subscribed to the Google Groups "sage-support" group. To unsubscribe from this group and stop receiving emails from it, send an email to sage-support+unsubscr...@googlegroups.com. To post to this group, send email to sage-support@googlegroups.com. Visit this group at http://groups.google.com/group/sage-support. For more options, visit https://groups.google.com/d/optout.
