Re: boot-floppies 2.2.9 and Debian CD's...
On Fri Mar 24, 2000 at 09:28:54AM -0500, Ben Collins wrote:
> On Fri, Mar 24, 2000 at 01:19:59PM +0100, Raphael Hertzog wrote:
> > Le Thu, Mar 23, 2000 at 05:58:45PM -0500, Ben Collins écrivait:
> > > When mkhybrid creates the CD, you pass it the image(s) to use for booting.
> > > One of these is the root.bin from boot-floppies. It does not need to be
> > > recreated for the CD's, just simply:
> > >
> > > gzip -dc root.bin > root.bin.uncompressed
> > > mount -o loop root.bin.uncompressed /mnt
> > > touch /mnt/cdrom_image
> > > umount /mnt
> > > gzip -c9 root.bin.uncompressed > root.bin
> > >
> > > It's quite simple, and then that becomes the root image used for CD
> > > booting.
> >
> > Yes, but mount needs root rights, and I would better not add such a
> > requiry for using debian-cd... is there no other way to get this
> > information ? Is there no way to get the device that has been used
> > to boot ? Or can't we just automatically detect CD roms and look if there's a
> > .disk/info file in it ?
>
> We do already detect cdrom's, but we still ask the user. Problem is you
> can't really know if you've booted off of a CDROM because the root
> filesystem is /dev/ram no matter what, which could be a tftp image,
> floppy, cdrom, zip drive, syslinux, or who knows what else.
So we put a magic file on the CDROM (or whatever media) so
we can tell which device we booted from. Something along
the lines of:
foundit=0
# Mount /proc
mount proc /proc -t proc
# Try to mount up the boot device, and check
# to be sure it is the right one....
for drive in `echo /dev/hd? /dev/sd? /dev/scd?` ; do
if ! mount $drive /mnt -t iso9660 -o ro > /dev/null 2>&1 ; then
continue
elif [ -e /mnt/README.Debian ] ; then
foundit=1
else
umount /mnt
fi
done
if [ foundit = 0 ] ; then
echo "I'm sorry. I could not find the Debian CD-ROM."
while [ 1 ] ; do
sleep 1
done
fi
-Erik
--
Erik B. Andersen Web: http://www.xmission.com/~andersen/
email: andersee@debian.org
--This message was written using 73% post-consumer electrons--
Reply to: