On Fri, 27 Dec 2002, Johnathan Bailes wrote: > On Fri, 2002-12-27 at 12:29, Giulio Orsero wrote: > > On Fri, 27 Dec 2002 11:53:26 -0500 (EST), "Robert P. J. Day" > > <[EMAIL PROTECTED]> wrote: > > > > > > >> Did you by any chance added a disk or partition, so that now there's more > > >> than one partition with label = /? > > >> Does "e2label /dev/hda1" show the expected label? > > > > >i see no reason for that to make a difference. both kernel entries > > >in grub.conf use the "root=LABEL=/" syntax. the earlier kernel boots > > >fine, the newer kernel fails and requires me to edit the kernel > > >line and substitute "root=/dev/hda1". > > > > == begin pure speculation > > If you use initrd the root fs is mounted this way > > > > echo Mounting root filesystem > > mount --ro -t ext3 /dev/root /sysroot > > umount /proc > > pivot_root /sysroot /sysroot/initrd > > > > so there's a mount binary (actually the static shell run as "mount") that > > knows about LABEL. > > > > If you boot without initrd maybe the mount is done by kernel which does not > > know about LABEL. > > > > So this would be another reason for the need of initrd. > > > > == end pure speculation > > > > -- > [EMAIL PROTECTED] > > I had the same problem on my laptop. I updated the kernel and had to > edit the grub.conf under the /boot dir to get the thing to boot up > correctly. root=/dev/hda1 Have no idea why.
i tried a couple other things -- no difference, so it's in bugzilla, bug 80540, if you care. regarding initrd, since when this worked fine, my initrd.img file contained only a single module for "ext3.o", and now that's built into the kernel, i'm not sure how initrd could affect this. i'll just leave this in bugzilla's lap and move on. rday -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED]?subject=unsubscribe https://listman.redhat.com/mailman/listinfo/redhat-list
