Ok, so here's the story...

I have/had a RH 8.0 box, then tried to update to FC4 from CD's, but
that just kept crashing on the update.  So what I ended up doing is
getting all the deps to run yum and then did a yum update to FC4, that
seemed to work ok. But, before I was going to reboot, I wanted to check
the /boot dir and make sure everything was intact.  Well, I came to
find out that /boot/initrd-2.6.13-1.1526_FC4.img was not present.  So I
ran:

[root@minime boot]# mkinitrd /boot/initrd-2.6.13-1.1526_FC4.img
2.6.13-1.1526_FC4
/sbin/mkinitrd: line 327: 10003 Done                    echo access -w
$t
10004 Segmentation fault      | /sbin/nash --quiet
/sbin/mkinitrd: line 327: 10005 Done                    echo access -w
$t
10006 Segmentation fault      | /sbin/nash --quiet
/sbin/mkinitrd: line 327: 10007 Done                    echo access -w
$t
10008 Segmentation fault      | /sbin/nash --quiet
/sbin/mkinitrd: line 327: 10009 Done                    echo access -w
$t
10010 Segmentation fault      | /sbin/nash --quiet
no temporary directory could be found

With no ideas on what to do, i found there was a mkinitrd update in the
development tree, so i turned on my /etc/yum.repos.d/fedora-devel.repo
for yum. yum grabbed:

Updating:
mkinitrd                i386       5.0.3-1          development
305 k
Updating for dependencies:
glibc                   i686       2.3.90-12        development
4.4 M
glibc-common            i386       2.3.90-12        development
15 M
glibc-devel             i386       2.3.90-12        development
1.9 M
glibc-headers           i386       2.3.90-12        development
579 k

and now i have

[root@minime boot]# uname -a
FATAL: kernel too old
[root@minime boot]# ls
FATAL: kernel too old


on pretty much every command i issue.  Any ideas?????


-Zach

[ Post a follow-up to this message ]

zwilson@gmail.com wrote:

> Ok, so here's the story...
>
> I have/had a RH 8.0 box, then tried to update to FC4 from CD's, but
> that just kept crashing on the update.  So what I ended up doing is
> getting all the deps to run yum and then did a yum update to FC4, that
> seemed to work ok.

Bad idea.....

> on pretty much every command i issue.  Any ideas?????

Boot from the FC4 installation CD into rescue mode, answer a few
questions use chroot and install the kernel from the CD.

Or better yet if you have an active internet connection on the system in
rescue mode;

rpm -ivh
http://download.fedora.redhat.com/p...26_FC4.i686.rpm

The command above is one long line with a space after the '-ivh' portion
(the first line) and before the rest (the second line).


--
Contained within the Microsoft EULA;
This Limited Warranty is void if failure of the Product has resulted
from accident, abuse, misapplication, abnormal use or a virus.

[ Post a follow-up to this message ]

So there is nothing that I can really do from the box in it's current
state?

[ Post a follow-up to this message ]

zwilson@gmail.com wrote:

> So there is nothing that I can really do from the box in it's current
> state?
>

Not really, it's broken.

You might want to use something like the SystemRescueCD and backup
everything that is important.When done start fresh and do a clean
install. If your having problems installing FC4 try do the installation
in text mode. Updating from RHL8 to FC4 is not a good idea, to many
changes in everything starting with the kernel, mkinitrd, initrd and so
on. The 2.4 RHL8 kernel could not deal with all of the changes.

Another thing is the glibc rpm version you installed, if you installed
the wrong one (i686 version instead of the i386 version) then your
system is pretty much hosed.

Also do you realize you were installing developmental RPMS???

For example, from;

http://download.fedora.redhat.com/p...a/RP
MS/
^^^^^^^^^^^^

glibc-2.3.90-12.i386.rpm        12-Sep-2005 12:56       3.6M
glibc-2.3.90-12.i686.rpm        12-Sep-2005 13:42       4.4M
mkinitrd-5.0.3-1.i386.rpm       30-Sep-2005 18:07       305K

And the kernel which was 2.6.13-1.1526_FC5 now as of October 3, 2005;
kernel-2.6.13-1.1592_FC5.i586.rpm

The latest FC4 equivalent packages are;


glibc-2.3.5-10.3.i386.rpm       23-Aug-2005 14:07       4.2M
glibc-2.3.5-10.3.i686.rpm       23-Aug-2005 14:07       5.7M
mkinitrd-4.2.15-1.i386.rpm      20-May-2005 13:35       300K
kernel-2.6.13-1.1526_FC4.i586.rpm       29-Sep-2005 22:48       16M
kernel-2.6.13-1.1526_FC4.i686.rpm       29-Sep-2005 22:49       16M


--
Contained within the Microsoft EULA;
This Limited Warranty is void if failure of the Product has resulted
from accident, abuse, misapplication, abnormal use or a virus.

[ Post a follow-up to this message ]

Yeah, I realize that I was installing the developments packages, that
was how I was trying to fix the mkinitrd crapping out on me. That was
the entire problem, the mkinitrd was seg faulting and I couldn't get a
initrd img created.

Right now I am trying to get the FC4 Rescue Disk to load, but it keeps
crashing on load...so I have a whole nother can of worms to deal with
today.

Thanks for your help.

[ Post a follow-up to this message ]

Yeah, I realize that I was installing the developments packages, that
was how I was trying to fix the mkinitrd crapping out on me. That was
the entire problem, the mkinitrd was seg faulting and I couldn't get a
initrd img created.

Right now I am trying to get the FC4 Rescue Disk to load, but it keeps
crashing on load...so I have a whole nother can of worms to deal with
today.

Thanks for your help.

[ Post a follow-up to this message ]

zwilson@gmail.com wrote:
> Yeah, I realize that I was installing the developments packages, that
> was how I was trying to fix the mkinitrd crapping out on me. That was
> the entire problem, the mkinitrd was seg faulting and I couldn't get a
> initrd img created.
>
> Right now I am trying to get the FC4 Rescue Disk to load, but it keeps
> crashing on load...so I have a whole nother can of worms to deal with
> today.
>
If you cannot load an installation disk on a machine, I would suspect
hardware problems, or BIOS not set to boot from installation disk. Rescue
disk is usually on disk 1 of a distribution.

--
.~.  Jean-David Beyer          Registered Linux User 85642.
/V\  PGP-Key: 9A2FC99A         Registered Machine   241939.
/( )\ Shrewsbury, New Jersey    http://counter.li.org
^^-^^ 11:00:00 up 5 days, 4:02, 4 users, load average: 4.33, 4.23, 4.23

[ Post a follow-up to this message ]

Yeah, I have thought about that, but everything else loads/runs on this
machine (windows 2k, knoppix boot disk, RH 8.0, etc...). So I am
guessing that there is just something in the FC4 rescue/install process
that doesn't mesh with our system.  But I have no idea what that might
be.

[ Post a follow-up to this message ]

zwilson@gmail.com wrote:
> Yeah, I have thought about that, but everything else loads/runs on this
> machine (windows 2k, knoppix boot disk, RH 8.0, etc...). So I am
> guessing that there is just something in the FC4 rescue/install process
> that doesn't mesh with our system.  But I have no idea what that might
> be.
>
Nothing has to mesh with your system. The first disk on a distro can run
even with blank hard drives. Could your rescue disk be faulty?

--
.~.  Jean-David Beyer          Registered Linux User 85642.
/V\  PGP-Key: 9A2FC99A         Registered Machine   241939.
/( )\ Shrewsbury, New Jersey    http://counter.li.org
^^-^^ 11:40:01 up 5 days, 4:42, 4 users, load average: 4.47, 4.26, 4.20

[ Post a follow-up to this message ]

I don't think so, but I suppose it is possible. I burned a 'Rescue
Disk' and Disk 1 of FC4 and both crash immediately following the SCSI
detection.  I even tried to do linux dd and use install my drivers for
my RAID controller (3ware Escalade 7500-4LP), that also created a seg
fault/dump.  This is so bizzare, I have no idea what the problem is.

[ Post a follow-up to this message ]