On Wed, Dec 06, 2006 at 11:14:00PM +0100, m4jkel wrote: > On Tue, Dec 05, 2006 at 10:22:12PM +0000, Ken Moffat wrote: > > > Did you have the /dev/console and /dev/null devices before you > > tried to fix things ? It's hard to tell at a distance, but the > > symptoms sound a bit like a missing /dev/console. > > Yes, I had both /dev/console and /dev/null at the very begining as said > in 6.2.1. When there are only those two device nodes, booting stops with > error "/dev/hda11 not found". Because udev doesn't work! > Oh well, it was just a suggestion.
> I did some more research, by playing with init=/bin/bash. > What I found out is that udevtrigger does not do its job. I don't know > exactly how this version of udev should > work, but AFAIR after running udevtrigger every coldplugged devices (like > hda*, vcs) should appear in /dev. Well, udevtrigger produces no info to > stdout even with --verbose parameter. It isn't good, is it? I'm not commenting on the details of udev - I don't understand it sufficiently to fix things. What I will mention is that I've come badly unstuck before now when I tried to upgrade udev on an existing system. The version in the book works, at least with everything else in the book (and that includes the bootscripts). You said you are using your own intscripts - take a look at the book's udev script and compare what it does to what your script does. ĸen -- das eine Mal als Tragödie, das andere Mal als Farce -- http://linuxfromscratch.org/mailman/listinfo/lfs-support FAQ: http://www.linuxfromscratch.org/lfs/faq.html Unsubscribe: See the above information page
