Question about mount
rob at landley.net
Wed Mar 29 15:59:44 UTC 2006
On Wednesday 29 March 2006 2:36 am, Alexander Griesser wrote:
> Rob Landley wrote:
> >>>> [lxtc3861 - ~ #] mount /
> >>>> mount: Mounting /dev/hda1 on /system failed: No such device
> >>>> [lxtc3861 - ~ #]
> >>> Out of morbid curiosity, how did you get here without / already
> >>> mounted?
> >> / is mounted, but it is mounted readonly.
> >> In my init-scripts I then try to mount it readwrite.
> > Ok. In theory, that should work. Do you have /proc mounted yet when you
> > try to do this? (That's how I tested. If it doesn't work when /proc
> > isn't mounted, that's a bug I need to fix.)
> It it doesn't work when /proc is _NOT_ mounted, you need to fix it?
> I thought, /proc needs to be mounted to work with /proc/mounts?
It does, but maybe you should be able to remount without /proc? (It'll lose
existing flags, but maybe it should fall through and try to do it with just
the flags you supplied, which should give us a clean failure from the kernel
if it's not mounted. The specific case of "mount -o remount,rw /" in a
recovery situation is what I'm thinking of here.)
> The order of my bootscripts looks like this:
> [lxtc3861 - /etc/rcS.d #] ls -1 | head -3
> [lxtc3861 - /etc/rcS.d #]
> Where S00 does `mount /system`, S01 mounts /proc, /sys, /proc/bus/usb,
> /dev/pts and /var/tmp and S02 does `mount -o remount,rw /dev/root /`.
> So, /proc is definetly mounted before I try to mount /.
> So, BTDT. The result is the same as above with the latest daily
> snapshot. It only works when it gets both parameters.
That's odd. I'm trying to figure out how I can reproduce this...
> > For remount, it's using that as an arbitry identifier name. Notice that
> > you can "remount -o remount,rw harvey-the-wonder-hamster /", and that you
> > can do that with the current mount command, not just mine. :)
> Ah, cool, I didn't know that -> quite funny.
> Sorry again for bringing in some trouble with this thread, as I forgot
> to CC some items immediately to the list :-/
I'm more interested in finding and reproducing the technical issue. Unlikely
to have time to extensively poke at it until this weekend, though...
Never bet against the cheap plastic solution.
More information about the busybox