Re: debian installer on ppc/pegasos
On Wed, Aug 27, 2003 at 11:27:59PM +0200, Geert Stappers wrote:
> On Wed, Aug 27, 2003 at 05:44:45PM +0200, Sven Luther wrote:
> > On Wed, Aug 27, 2003 at 01:41:06AM -0700, Chris Tillman wrote:
> > > On Wed, Aug 27, 2003 at 08:16:07AM +0200, Sven Luther wrote:
> > > > On Wed, Aug 27, 2003 at 12:09:19AM +0200, Petter Reinholdtsen wrote:
> > > > > [Sven Luther]
> > > > > > Then why is it not booting ?
> > > > > >
> > > > > > The kernel boots, loads the cdrom.initrd image, mounts it as root,
> > > > > > mounts the devfs filesystem, and hangs.
> > > > >
> > > > > Perhaps you see the glibc/mklibs/whatever bug making all programs
> > > > > crash? What is the exact messages on the screen before it hangs?
> > > > > Last 4-5 lines would be good. Compare it with #206531.
> > > >
> > > > Ok, here it goes (more or less, typed by hand with the other monitor at
> > > > my side) :
> > > >
> > > > RAMDISK: Compressed image found at block 0
> > > > Freeing initrd memory: 2028k freed
> > > > VFS: Mounted root (ext2 filesystem) readonly.
> > > > Mounted devfs on /dev
> > > > Freeing unused kernel memory: 320k init 40k pmac 8k prep
> > > >
> > > > And a blinking cursor below that, keyboard works, but nothing else
> > > > happens.
>
> It is where the console is used the first time.
> (other message where "kprint", "printk?" ( kernel print))
>
> These symptooms occured to me when I hadn't "virtual console"
> in the kernel. FYI, that was in the 2.0 kernel days.
Apart from enabling devfs and embedding the initrd, this is the same
kernel i normally run (2.4.21 linuxppp_2_4 branch + some pegasos
specific early booting stuff which should not influence anything).
> > > > > > what is supposed to come next ? And how can i check that it
> > > > > > works. Mmm, maybe i will go looking into the /linuxrc, which i think
> > > > > > is where this code is. Is it ok to put some echo or things like that
> > > > > > in there ?
> > > > >
> > > > > Yes, it should be ok to put echos in the linuxrc. I believe it is in
> > > > > rootskel.
> > > >
> > > > Ok, i will try this one.
> > >
> > > A couple of boot arguments to try, may do something, and won't hurt:
> > >
> > > debug verbose
> >
> > Nope, nothing more. I think it is definitively related to the 206531
> > bug.
>
>
> I think it is catch 22 ;-)
>
> The booting computer assumes it can print, but it can't.
>
>
> Check for valid console (/dev/tty ? /dev/console ?)
Nope, no such thing, my /dev is empty in the cdrom.initrd. So, this must
be the problem, i think.
May it be related to not having devfs enabled in the kernel building the
initrd ?
BTW, when i enable devfs, linux doesn't seem to like the /dev/hd* kind
of devices. Is there a way of solving this without using the ugly devfs
harddisk devices ?
Thanks for your attention, i think we are making headway here.
Friendly,
Sven LUther
Reply to: