[wplug] kernel not loading after partitioning hard drive

[Vanco, Donald]

> -----Original Message-----
> From: rkalaskar [mailto:rkalaskar at aethon.com] 
> Sent: Wednesday, December 18, 2002 11:34 AM
> To: wplug at wplug.org
> Subject: Re: [wplug] kernel not loading after partitioning hard drive
> 
> 
> Vanco, Donald wrote:
> 
> >>-----Original Message-----
> >>From: Paul Cantalupo [mailto:lupey+ at pitt.edu] 
> >>Sent: Wednesday, December 18, 2002 10:54 AM
> >>To: wplug at wplug.org
> >>Subject: Re: [wplug] kernel not loading after partitioning 
> hard drive
> >>
> >>
> >>"Vanco, Donald" wrote:
> >>
> >>>Did you move the kernels beyond 1024 cylinders when you 
> >>>
> >>repartitioned?  If
> >>
> >>>so - lilo cannot find them anymore...
> >>>
> >>I actually moved the root partition towards the beginning of 
> >>the hard drive
> >>so I don't think this is a problem. In any event, I don't 
> >>think the LILO
> >>version that I am using cares about the 1024 cylinder problem 
> >>as far as I've
> >>been able to understand.  Please let me know if I am wrong in 
> >>assuming this.
> >>
> >>
> >>FYI, my hard drive is approx 40 GB and my root partition 
> >>starts around 14GB.
> >>My root partition *was* started around 20GB before the 
> repartitioning.
> >>
> >	Hmm - what distro?  If Red Hat 7.3 or it's ilk do you 
> know if your
> >partition table / fstab relied on labels as opposed to 
> device calls?  You
> >did copy off fstab and do a print of fdisk -l - right?
> >	The fact that one kernel boots and the other kernel 
> hangs (it's not
> >a LILO thing if you get to "Loading....." - dunno where my 
> head was) tells
> >me that it's likely a partition (or possibly mkinitrd?) thing.
> >
>     I bet this has nothing to do with boot parition 
> size/location. I had 
> the same problem with a  kernel I was compiling. It used to 
> hang after 
>  "Loading ...".  I found the problem to be with the 
> initrd.img.gz.   I 
> used mkinitrd to make the initial ram disk, and the problem 
> disappeared.

	So the question for Paul is - do you have a valid initrd for each
kernel, and does lilo.conf point to it?  If all looks OK - re-run lilo and
you should be golden.

To get an initrd is something like:
mkinitrd -v -f --ifneeded <initrdimage-name>
<kernel-version-to-build-against> 
	-this assumes that /etc/modules.conf has any required modules /
aliases listed in it, otherwise you'll need to specify '--preload <module>'
in the above syntax

(ex: mkinitrd /boot/initrd-2.2.5-15.img 2.2.5-15)  and of course lilo.conf
must reflect this image for this kernel.  the --ifneeded is nice because it
will not build unnecessary initrd's - usually SCSI / exotic storage are the
only reason for initrd's on PeeCees

YMMV
Don