On 25 Jun, Ruud de Bruin shaped the bitstream to say:
> I am trying to upgrade the kernel from RH 5.1 to 2.0.34-1. According the
> information on www.redhat.com it should be easy: download all the RPM's in
> a temp. direcotry an run: rpm -Uvh kern*.rpm.
>
> However, it doesn't work. The system complains about
> /lib/modules/2.0.34/modules.dep does not exist. (I noticed the modules are
> placed in a directory: /lib/modules/2.0.34-1.
> I tried to solve this by putting: ln -s 2.0.34-1 2.0.34, but now I get a
> numer of unresolved symbols after running depmod -a.
>
> How can I solve this (install the latest kernel on RH 5.1)?
>
> Regards, Ruud.
>
> --
> Next time you wave, use all your fingers.
>
>
I sent this response to someone on a nother list with the same problem,
yesterday. I hope it helps... It is exactly how I did it successfully!
It's a Hurricane system...
----
I just did 'em last night. After log in it says the kernel is 2.0.34
but during boot the kernel shows as 2.0.32.
I recompiled the kernel and now it shows as 2.0.34 all the way around.
I've done this twice now (the first time around with 2.0.34 it whacked
my machine and I had to start over with a complete reinstall of 5.0
from the source CD). The trouble had to do with how the modules are
handled. Prior to 34 there was a separate rpm for modules but in 34
the modules are in the kernel rpm itself. Thus when you install the
kernel rpm it goes along fine until it tries to remove the 32 modules
and then it errors out. What I ended up with was a partial install of
an RPM... (Whoa!). So... I recompiled and when it came time for
mkinitrd (I had already installed the new one BTW) I got an error there
because the loop back device module wasn't present (remember the
modules portion of the kernel install bombed!). Here's what I did...
rather than take no for an answer and be stuck with 32 I cp'd the
loop.o device from the old 32 modules into the new 34 modules and re-ran
mkinitrd... Bing-Bam-Boom... it worked and the new init image was right
there in /boot. Edit lilo.conf ro dual boot for recovery, rename the
old linuz, copy the new linuz over, run /sbin/lilo... reboot...
Magic... it worked like a charm.
So... that's my story and I'm stickin' to it!
--
Chuck Mead
[EMAIL PROTECTED]
http://www.moongroup.com/unix/
HOST SYSTEM NOT RESPONDING, PROBABLY DOWN. DO YOU WANT TO WAIT? (Y/N)
This message sent with TkRat v1.1
--
PLEASE read the Red Hat FAQ, Tips, Errata and the MAILING LIST ARCHIVES!
http://www.redhat.com/RedHat-FAQ /RedHat-Errata /RedHat-Tips /mailing-lists
To unsubscribe: mail [EMAIL PROTECTED] with
"unsubscribe" as the Subject.
