udev-config md5 sum

2010-01-18 Thread Alexander Kozlov
development udev-config changes all the time: current udev-config-20090925 has changed md5 sum and it doesn't match the one from the svn book now.. Would it be more logical to update the timestamp in the package name each time it is rebuilt? or add a revision tag -2, -3, etc. to avoid confusion

bash 4.1 needs readline 6.1

2010-01-02 Thread Alexander Kozlov
On my machine, bash 4.1 builds happily with new readline and fails with the 'old' 6.0. If it's not me only, this should be mentioned in the "Host System Requirements", I think. Alex. -- http://linuxfromscratch.org/mailman/listinfo/lfs-support FAQ: http://www.linuxfromscratch.org/lfs/faq.html Un

coreutils 8.2 tests in Chap.6

2009-12-16 Thread Alexander Kozlov
To run tests as user 'nobody' in Sec. 6.21.1, I have to change permissions of the entire directories ./tests, ./gnulib-tests and ./src chown -Rv nobody ./{tests,gnulib-tests,src} and not of the {gnulib-tests,lib,src}/.deps as in the book now. This is an overkill but I did not find bett

LC_ALL omitted in chroot?

2009-12-13 Thread Alexander Kozlov
As far as I can see, LFS omits setting LC_ALL in the chroot environment in Chap.6 but does it in Chap.4 for the toolchain. Is this correct? Alex. -- http://linuxfromscratch.org/mailman/listinfo/lfs-support FAQ: http://www.linuxfromscratch.org/lfs/faq.html Unsubscribe: See the above information

Re: plsce for killall

2009-12-12 Thread Alexander Kozlov
>> I do. And I do install psmisc. I'm just thinking how important >> it is? Debian considers it optional (it is misc after all) and >> installs to /usr/bin and Slackware builds both killall and fuser >> in procps. > > Oh, so both of those are available there? Are they more or less > compatible w

Re: plsce for killall

2009-12-11 Thread Alexander Kozlov
> Slackware builds both killall and fuser in procps. hmm.. apparently from the same psmisc-*.tar.gz source as LFS does. Sorry for the noise, I seem to stay with psmisc. Question about `dirname $(which killall)` is still valid though. Alex. -- http://linuxfromscratch.org/mailman/listinfo/lfs-sup

Re: plsce for killall

2009-12-11 Thread Alexander Kozlov
>> Another question, who needs psmisc? Slackware does not use them, >> neither lfs-bootscripts. > > You don't use killall or fuser? > > Simon. I do. And I do install psmisc. I'm just thinking how important it is? Debian considers it optional (it is misc after all) and installs to /usr/bin and Sl

plsce for killall

2009-12-10 Thread Alexander Kozlov
>> psmisc 22.8 configured with --prefix=/usr installs killall to >> /usr/bin. >> >> No problem to leave it there, as debian does, but I move it to >> /bin to keep my old init scripts unchanged. Anyway, the command >> "ln -sv killall /bin/pidof" in Sec. 6.52 should be corrected. > > if this is wro

place for killall

2009-12-10 Thread Alexander Kozlov
psmisc 22.8 configured with --prefix=/usr installs killall to /usr/bin. No problem to leave it there, as debian does, but I move it to /bin to keep my old init scripts unchanged. Anyway, the command "ln -sv killall /bin/pidof" in Sec. 6.52 should be corrected. Alexander Kozlov LF

Re: glibc overwrites scsi.h

2009-11-19 Thread Alexander Kozlov
This is due to the "[SCSI] FC Pass Thru support" patch in v2.6.31 which "also exports several headers from include/scsi such that they can be available to user-space applications: include/scsi/scsi.h include/scsi/scsi_netlink.h include/scsi/scsi_netlink_fc.h include/sc

Re: glibc overwrites scsi.h

2009-11-19 Thread Alexander Kozlov
> Does it matter that glibc-2.11 overwrites > /usr/include/scsi/scsi.h > that has previously been installed by Linux-2.6.31.5 API Headers? > > (Linux From Scratch - Version SVN-20091109) > > Brett I'm asking the same question. As far as I can see in my Debian sid environment, starting with v2