[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: boot 2.4.x *after* using 2.6.2 ???



* On 2004:02:22:11:26:39-0600 I, Michael D Schleif <mds@helices.org>, scribed:
> OK, I am finally testing v2.6.x kernels on one box.
> 
> Prior to yesterdays 2.6.2 install, I had successfully used the
> following:
> 
>    2.4.23-1-686
>    2.4.24-1-686
> 
> This box uses lilo, and I successfully switched between these kernels at
> boot time.
> 
> Then, I installed this:
> 
>    2.6.2-1-686
> 
> Yes, it booted the first time, and after I got around to loading the
> following, even X and KDE work:
> 
>    mousedev
>    psmouse
> 
> Nevertheless, now I cannot boot any 2.4.x kernels ;<
> 
>    . . .
>    Freeing initrd memory: 5292k freed
>    [ request_module[block-major-8]: Root fs not mounted ]
>    VFS: Cannot open root device "801" or 08:01
>    Please append a correct "root=" boot option
>    Kernel panic: VFS: unable to mount root fs on 08:01
> 
> NOTE: That second line only occurs sometimes, and I cannot find any
> `block' references in /etc/modules.conf -- hence the `[]'.
<snip />

Thank you, David, for pointing me in the right direction.

Yes, lilo was properly configured.

It turns out that something _removed_ the `bootable' flag from /dev/sda1
, and with `ROOT=probe' in /etc/mkinitrd/mkinitrd.conf, the probing
process could not find a boot partition for 2.4.x kernels.

I do not know how 2.6.x kernels find a bootable partition; nor do I know
how the installation process of the 2.6.2 kernel removed that flag.
Please, comment if you feel that this is a bug, and that I ought to
report it.

I do know that setting `ROOT=""' is equivalent to the obsolete
`PROBE=off', which _did_ allow booting the 2.4.x kernels.

However, the correct solution is to use cfdisk and reset the `bootable'
flag on /dev/sda1, and reset `ROOT=probe'.

Now, this box and kernels 2.6.x/2.4.x behave as expected.

Thank you.

-- 
Best Regards,

mds
mds resource
877.596.8237
-
Dare to fix things before they break . . .
-
Our capacity for understanding is inversely proportional to how much
we think we know.  The more I know, the more I know I don't know . . .
--

Attachment: signature.asc
Description: Digital signature


Reply to: