[ale] Problems upgrading to RH 8.0

John Wells jb at sourceillustrated.com
Tue Oct 1 11:58:55 EDT 2002


Alas...didn't have time to fool with it any longer (it was my workstation
at work and I have pressing matters).  Luckily was able to mount via nfs
and back up most of my stuff.

Major pain though.  No indication from the installer that there was
anything wrong whatsoever.  Most likely, win4lin kernel caused this, but
I'm very surprised that the installer didn't recognize something had gone
wrong.

One hint was that the boot disk creation wouldn't work, but I was too
stoked about seeing the new gui and decided to skip it without checking
out error messages.

I'm going to submit a bug report to redhat...


Thanks for your help.

John

John Wells said:
> Chris,
>
> Thanks for the reply.
>
> I did have a few of the older kernels laying around...but the win4lin
> was my current kernel and its source was pointed to by /usr/src/linux.
>
> I was able to boot up using the "linux rescue" option on the cdrom and
> modify grub.conf and add the old kernel to the options, but when booting
> it I get the same GLIBC error message and a few others...then a panic.
>
> If I could chroot using the rescue on the cd, I could install the rpms
> manually.  But, as it is (without chroot) it doesn't recognize any of
> the installed rpms.
>
> Thanks!
>
> John
>
> Chris Ricker said:
>> On Tue, 1 Oct 2002, John Wells wrote:
>>
>>> I upgraded my system to 8.0 this morning and everything seemed to go
>>> smoothly.  However, on reboot, the kernel wasn't found.  Booted up
>>> using "linux rescue" functionality and the kernel was not installed.
>>> I suspect it had a hard time with this because I have a custom
>>> win4lin kernel installed.
>>
>> You should always keep a "stock" kernel installed as well -- it's good
>> for  backup, and it makes things like upgrades work....
>>
>> Did it actually remove your old kernel from the disk, or did it just
>> not put  it in grub.conf / lilo.conf?  If it's still around, add it to
>> the boot  loader manually and boot from it as a first step....
>>
>>> Anyway, I was going to try to use the "linux rescue" shell to either
>>> install the kernel or to back up my critical data, but much of what I
>>> try to do (including the critical "chroot /mnt/sysimage"), gives the
>>> following error:
>>>
>>> -/bin/sh-2.05b# chroot /mnt/sysimage
>>> /bin/sh: /lib/i686/libc.so.6: version 'GLIBC_2.3' not found (required
>>> by /lib/libdl.so.2).
>>>
>>> Am I screwed?  Do these utilities use LD_LIBRARY_PATH to search for
>>> glibc?
>>>  My first instinct was to run strings on the various flavors of
>>> libc.so.6
>>> on the filesystem and try to find the one that supports 2.3 and add
>>> it to my LD_LIBRARY_PATH, but strings gives the above error (lol).
>>
>> It looks your glibc didn't get upgraded either for some reason....
>>
>>> Is there anything I can do to at least transfer my data?
>>
>> Try getting the kernel running first (w/o doing the chroot; use the
>> rescue  utils on the cd itself).
>>
>> Once you do that, boot your system, and pass an option init=/sbin/sash
>> to  the kernel.  That'll boot you into a statically linked shell, so
>> you should  be able to resolve your glibc troubles at that point....
>>
>> later,
>> chris
>
>
>
> ---
> This message has been sent through the ALE general discussion list. See
> http://www.ale.org/mailing-lists.shtml for more info. Problems should be
>  sent to listmaster at ale dot org.




---
This message has been sent through the ALE general discussion list.
See http://www.ale.org/mailing-lists.shtml for more info. Problems should be 
sent to listmaster at ale dot org.






More information about the Ale mailing list