Herbert Xu wrote:
> Alexander Supalov wrote:
> >
> > I saw today that Linux kernel 2.0.32 had been released as a Debian
> > package. Is it safe to upgrade the existing Debian.1.3.r4 to this
> > kernel? What about all the libc6 stuff? Should I have it installed or
> > should I better wait until the the next major Debian release arrives?
> > If so, when will this happen?
>
> You should be able to install it onto a bo system.
I had a small trouble when installing, as follows:
I upgraded to a new snapshot of hamm (1997-12-13) and
dselect naturally marked for upgrade everything I had.
I also noticed the new kernel 2.0.32 and installed it.
Now, this is what I believe what happened.
When configuring kernel it asks if I want to make boot disk.
I did want. Then it asks something like "Hmm. You seem to have
new superformat, want to use it?" and I felt I'm taking risks
already and I don't want to answer yes to anything this dubious.
Then it tried to create floppy with old format (and with
non-existing device as well?) but didn't succeed. There was no
obvious way to back up to "Hmmm. ..." or otherwise correct the
situation.
( Later I found a note to the effect that use superformat
since it obsoletes the older one and is better and removes
some /fev/fd* entries. I think that *if* superformat is
safe to use it should be the default. It probably should
be the default anyway, since the alternative does not work.
Why give user possibility to give wrong answer when script
can detect the right one. I also think that configuration
should notice that boot disk was not created and e.g.
not run lilo or at least ask about it.
)
Then configuring kernel asks if I want to use old lilo.config
of let him create a new one. Silly me, I thought that lilo 20
might need something new, let there be new config. A new config
there was, lilo was run, and the system was unbootable. (Well,
maybe I fumbled with it a bit afterwards.)
( Aside from possible lilo version problem, since upgrade from
19 to 20 was unpacked but not configured at this time,
there are some other problems. First, I think that the
default should be to append lines to existing lilo.config
since that presumably worked in the last boot. If lilo 20
is installed then no symlinks should be used or changed.
Maybe that should be the case anyway, since the real
lilo.config may reside in some other place in multi boot
machines? Then running lilo should be optional. If no links
have been broken then not running lilo does not break
anything, only new kernel is not yet used.
The generated lilo.config file was broken, since my
partition, /dev/hda5, was a logical one, and lilo gave
error for that. Is there a way of detecting the partition
type?
)
That was it. Although the consequences were not so nice,
it was mostly my mistake, but I would recommend that you do
not upgrade kernel and lilo without booting in between.
A couple questions about it all:
Is it necessary or even possible to coordinate upgrades
of kernel and bootloader so that their upgrades are not
interleaved?
Is it possible to safely generate bootloader config file
or would it be better to show message
"Will mail more info to you, read it before booting"
and pause to make sure that user reads it. (and mail the info)
t.aa
--
TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to
[EMAIL PROTECTED] .
Trouble? e-mail to [EMAIL PROTECTED] .
