Em 13-03-2012 13:56, Bruce Dubbs escreveu: > Fernando de Oliveira wrote: >> Since yesterday, I am trying to build LFS7.1-svn (today: Version >> SVN-20120311) using jhalfs. >> >> Build stops in Chapter 5 (tools) at expect >> > Just some suggestions.
Thank you very much for the suggestions, Bruce. > Did you run the command from the sources directory: md5sum -c md5sums ? No. So, two failures: udev-config-20100128.tar.bz2 and lfs-bootscripts. They differ in size but have the same name (should not they have different names?): $ find -name "udev*bz2" | xargs ls -lrt | cut -d" " -f5- 6295 2012-03-13 04:01 ./old-2012.03.14/udev-config-20100128.tar.bz2 6286 2012-03-14 04:01 ./old-2012.03.15/udev-config-20100128.tar.bz2 6279 2012-03-15 04:01 ./udev-config-20100128.tar.bz2 $ find -name "lfs*bz2" | xargs ls -lrt | cut -d" " -f5- 32734 2012-03-13 04:01 ./old-2012.03.14/lfs-bootscripts-20120229.tar.bz2 32723 2012-03-15 04:01 ./lfs-bootscripts-20120229.tar.bz2 After your post, I reloaded the filed files, same problm again. Next day, zeroed book, md5sums, those two files, started again, still problem, now with gzip. > Do you have plenty of free disk space? Yes. $ df -h /media/LFS71b Sist. Arq. Tam. Usado Disp. Uso% Montado em /dev/sdc1 19G 1,8G 16G 10% /media/LFS71b > Are you using -j1? Yes. > The scripts are doing: > !/bin/bash > set +h > set -e > > cd $PKGDIR > ... > > So $PKGDIR seems to getting confused. > > The Makefile is doing: > > @$(call remove_existing_dirs,dejagnu-1.5.tar.gz) > @$(call unpack,dejagnu-1.5.tar.gz) > @$(call get_pkg_root_LUSER) > @source ~/.bashrc && \ > $(CMDSDIR)/chapter05/$@ >> logs/$@ 2>&1 && \ > $(PRT_DU) >>logs/$@ > > where $@ should be 041-dejagnu > > -- Bruce Previous directory had been removed. Eager to start again, I deleted jhalf directory, stupid me. I tried again jhalfs from scratch, but, as I said above, the failure occurred in gzip-1.4. By hand, it compiled. One thing I noticed: it missed installing 039-tcl 040-expect 041-dejagnu, jumping from 038-gcc-pass2 to 042-check. In previous book, the problem was in 040-expect. I am assuming a mistake mine or in my system. However, it is intriguing that problems appear at about the same place. Bruce, I gave up at the moment. I thank you again and will come back to this again and ask for help, later in a couple of months, maybe before. The reason is that I built LFS 7.1 svn, executing each jhalfs generated script by hand. The system did not boot successfully for two days, when I remembered kmod, and modified kernel's configuration for support, as is very well remarked in LFS ch08. I will make a full working system (to my needs) and then come back to learn jhalfs (so many things to study, including security, forking bombs...), this time using lfs to build lfs, so that I can be sure that all requirements are installed correctly. As a very pleased user, I must say that both LFS and BLFS have been always great, but recently, from about a year, they have become much, much, increasingly better!!! Congratulations to you all for the very impressive effort, and for the new editors! -- []s, Fernando -- http://linuxfromscratch.org/mailman/listinfo/lfs-support FAQ: http://www.linuxfromscratch.org/lfs/faq.html Unsubscribe: See the above information page
