On Mon, Aug 18, 2003 at 06:19:41AM -0400, Amir Tal wrote:
>
> "installation went fine" means that there were no errors during the
> installation, and no ugly debug or any other output for that matter,
> that may point out that something is wrong. i went just like a normal
> deb package installation, with all post and pre scripts executed like
> they should, and everything indicated that the proccess completed like
> it normaly should..
Since the installation went fine but it still doesn't work, verifying
your work is probably the first action that should be taken. Hopefully
it will be a silly error that I have done more then once, something
along the lines of 1+1==3. In my opinion this would be preferable over
a real problem.
Can you run lilo -v and post the output here?
> > - what is the contents in lilo.conf that "looks ok to you" ?
>
> debian adds a LOT of inline little howto's into all of their conf files,
> and since the resault is a pretty long file, its pointless to send it to
> the list. a based my "assumption" on the fact that i have an entry in
> lilo.conf to the old kernel image as well, and its loading with no
> problem, using the same configuration file.
May I suggest egrep -v '^#|^$' /etc/lilo.conf? I admit that it will
still be about 30 lines or so. Can you post those 30 lines never the
less?
> > - _what_ kind of kernel panic message do you get?
>
> scroll up and read what i wrote. i am being asked to append a correct
> option for "root=" in lilo.conf. current entry points to /dev/hda3,
> which is the correct information for lilo to use (otherwise the older
> image would not load as well..)
>
Can you check what the new kernel reports for its command line?
I don't know if this is doable without a serial console or some such.
Actually, we are talking about a laptop, don't we? In this case, perhaps
you have a null modem cable to attach to another machine? Debugging
kernel problems is way easier with a serial console. However we still
have to verify that a stock Debian kernel is configured with the serial
console option, although I am quite sure that this is the case.
> > - what is the _exact_ message?
>
> ohh, dont make me reboot now, please... :)
> its 1 line, i dont remember the exact syntax, but basicly it goes
> something like "unable to mount vfs filesystem, please append a corrent
> "root=" in lilo.conf .
I believe the stock Debian kernels use initrd. I am not a kernel Guru,
which is why I can not tell how many times VFS gets mentioned in the
boot messages. Please reboot and tell us how successful the initial ram
disk loading was. Do post as many messages as you can, in the exact
order and language. Hopefully someone will be able to see some detail
within those messages. Do note that since one way to solve the problem
is by pointing some extra or missing detail, the exact messages *might*
be important.
--
Shaul Karl, shaulk @ actcom . net . il
=================================================================
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word "unsubscribe" in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]
